California Grasslands: Ecology and Management 9780520933972, 9780520252202

Citation preview

CALI FO R N IA G RAS S LAN D S E C O LO GY AN D MANAG E M E NT

California Grasslands EC OLO GY AN D MANAG E M E NT

Edited by

MARK R. STROMBERG JEFFREY D. CORBIN CARLA M. D’ANTONIO

UNIVERSITY OF CALIFORNIA PRESS Berkeley Los Angeles London

University of California Press, one of the most distinguished university presses in the United States, enriches lives around the world by advancing scholarship in the humanities, social sciences, and natural sciences. Its activities are supported by the UC Press Foundation and by philanthropic contributions from individuals and institutions. For more information, visit www.ucpress.edu.

Library of Congress Cataloging-in-Publication Data

University of California Press Berkeley and Los Angeles, California

California grasslands : ecology and management / Mark R. Stromberg, Jeffrey D. Corbin, Carla D’Antonio. p. cm. ISBN 978-0-520-25220-2 (cloth : alk.paper) 1. Grasslands—California. 2. Grassland ecology—California. 3. Grasslands—California—Management. I. Stromberg, Mark R., 1951– II. Corbin, Jeffery D. III. D’Antonio, Carla, 1956–

University of California Press, Ltd. London, England

QH541.5P7C35 2007 577.409794—dc22

© 2007 by the Regents of the University of California

Manufactured in the United States of America. 10 09 08 07 10 9 8 7 6 5 4 3 2 1

2007012136

The paper used in this publication meets the minimum requirements of ANSI/NISO Z39.48-1992 (R 1997) (Permanence of Paper). Cover illustration: © David J. Gubernick/www.rainbowspirit.com.

C O NTE NTS

CONTRIBUTORS vii PREFACE ix ACKNOWLEDGMENTS xi

1 Introduction 1 JEFFREY D. CORBIN, CARLA M. D’ANTONIO, ANDREW R. DYER, AND MARK R. STROMBERG

Species Composition at the Time of First European Settlement

Ecological Interactions

PAULA M. SCHIFFMAN

11 Genes on the Range: Population Genetics 131

5 Native American Uses and Management of California’s Grasslands 57 M. KAT ANDERSON

KEVIN J. RICE AND ERIN K. ESPELAND

12 Serpentine Grasslands 145 SUSAN P. HARRISON AND JOSHUA H. VIERS

Overview 2 Systematics of California Grasses (Poaceae) 7 PAUL M. PETERSON AND ROBERT J. SORENG

3 Community Classification and Nomenclature 21

6 Ecology of Invasive Non-native Species in California Grassland 67 CARLA M. D’ANTONIO, CAROLYN

JEFFREY D. CORBIN, ANDREW R. DYER,

MALMSTROM, SALLY A. REYNOLDS,

AND ERIC W. SEABLOOM

AND JOHN GERLACH

Resources 7 Water Relations 87

TODD KEELER-WOLF, JULIE M. EVENS,

KIMBERLY J. REEVER MORGHAN,

AYZIK I. SOLOMESHCH, V. L. HOLLAND,

JEFFREY D. CORBIN, AND

AND MICHAEL G. BARBOUR

JOHN GERLACH

History

13 Competitive Interactions 156

8 Mechanisms Determining Patterns of Nutrient Dynamics 94

14 Dynamics of Woody Species in the California Grassland 169 CLAUDIA M. TYLER, DENNIS C. ODION, AND RAGAN M. CALLAWAY

15 Ecology of Native Animals in California Grasslands 180 PAULA M. SCHIFFMAN

16 History and Ecology of Feral Pig Invasions in California Grasslands 191 J. HALL CUSHMAN

VALERIE T. EVINER AND MARY K. FIRESTONE

4 Pleistocene and Pre-European Grassland Ecosystems 37

9 Soil Biology and Carbon Sequestration in Grasslands 107

Late Quaternary Paleoecology of Grasslands and Other Grassy Habitats

LOUISE E. JACKSON, MARTIN

PETER E. WIGAND

STROMBERG, AND KATE M. SCOW

Rancholabrean Mammals of California and Their Relevance for Understanding Modern Plant Ecology

POTTHOFF, KERRI L. STEENWERTH, ANTHONY T. O’GEEN, MARK R.

10 Resource Limitation 119

17 Grazing Ecology of California Grasslands 197 RANDALL D. JACKSON AND JAMES W. BARTOLOME

18 Fire in California Grasslands 207 RICHARD J. REINER

19 Responses to Changing Atmosphere and Climate 218

W. STANLEY HARPOLE, LEAH

JEFFREY S. DUKES AND

GOLDSTEIN, AND REBECCA J. AICHER

M. REBECCA SHAW

STEPHEN W. EDWARDS

V

Policy and Management

22 Exotic Plant Management in California Annual Grasslands 281 JOE M. DITOMASO, STEPHEN F. ENLOE,

20 Grazing Management on California’s Mediterranean Grasslands 233 LYNN HUNTSINGER, JAMES W. BARTOLOME, AND CARLA M. D’ANTONIO

AND MICHAEL J. PITCAIRN

23 Regulatory Protection and Conservation 297 PATRICK A. JANTZ, BERNHARD F. L. PREUSSER, JESSE K. FUJIKAWA, JOSEPH A. KUHN, CHRISTOPHER J.

21 California Grassland Restoration 254 MARK R. STROMBERG, CARLA M. D’ANTONIO, TRUMAN P. YOUNG, JEANNE WIRKA, AND PAUL R. KEPHART

VI

CONTENTS

BERSBACH, JONATHAN L. GELBARD, AND FRANK W. DAVIS

24 Epilogue: Future Directions 319 MARK R. STROMBERG, CARLA M. D’ANTONIO, AND JEFFREY D. CORBIN REFERENCES 323 INDEX 375

C O NTR I B UTO R S

REBECCA J . AICHER University of California Irvine, California

JULIE M . EVENS California Native Plant Society Sacramento, California

M . KAT ANDERSON

VALERIE T. EVINER University of California Davis, California

USDA NRCS, University of California

Davis, California MICHAEL G . BARBOUR

University of California

MARY K . FIRESTONE University of California Berkeley, California

University of California

JESSIE K . FUJIKAWA University of California Santa Barbara, California

Davis, California JAMES W. BARTOLOME

Berkeley, California CHRISTOPHER J . BERSBACH

University of California

Santa Barbara, California RAGAN M . CALLAWAY

University of Montana

Missoula, Montana

JONATHAN L . GELBARD Conservation Value Institute Berkeley, California JOHN GERLACH Environmental Science Associates Sacramento, California

JEFFREY D . CORBIN

Union College Schenectady, New York

LEAH GOLDSTEIN University of California Irvine, California

CARLA M . D ' ANTONIO University of California Santa Barbara, California

J . HALL CUSHMAN

FRANK W. DAVIS University of California Santa Barbara, California

W. STANLEY HARPOLE

JOSEPH M . DITOMASO University of California Davis, California

SUSAN P. HARRISON University of California Davis, California

JEFFREY S . DUKES University of Massachusetts Boston, Massachusetts

V. L . HOLLAND

ANDREW R . DYER University of South Carolina Aiken, California

LYNN HUNTSINGER

STEPHEN W. EDWARDS

RANDALL D . JACKSON University of Wisconsin Madison, Wisconsin

East Bay Regional Parks

Berkeley, California

Sonoma State University Sonoma, California University of California

Irvine, California

California Polytechnic State University San Luis Obispo, California University of California Berkeley, California

STEPHEN F. ENLOE

University of Wyoming Laramie, Wyoming

LOUISE E . JACKSON University of California Davis, California

ERIN K . ESPELAND University of California Davis, California

PATRICK A . JANTZ

University of California Santa Barbara, California

VII

TODD KEELER - WOLF

California Dept. of Fish and Game Sacramento, California

M . REBECCA SHAW The Nature Conservancy San Francisco, California

PAUL R . KEPHART

Rana Creek Habitat Restoration Carmel Valley, California

PAULA M . SCHIFFMAN

JOSEPH A . KUHN

University of California Santa Barbara, California

KATE M . SCOW

CAROLYN MALMSTROM

Michigan State University East Lansing, Michigan

ERIC W. SEABLOOM Oregon State University Corvallis, Oregon

DENNIS C . ODION University of California, Santa Barbara Southern Oregon University, Ashland

AYZIK I . SOLOMESHCH University of California Davis, California

ANTHONY T. O ' GEEN University of California Davis, California

ROBERT J . SORENG

PAUL M . PETERSON

Washington, DC

KERRI L . STEENWERTH University of California Davis, California

MICHAEL J . PITCAIRN California Dept. Food and Agriculture Sacramento, California

MARK R . STROMBERG University of California Berkeley, California

MARTIN POTTHOFF

Smithsonian Institution

University of Kassell

Kassell, Germany

California State University Northridge, California University of California Davis, California

Smithsonian Institution

Washington, DC

CLAUDIA M . TYLER

University of California Santa Barbara, California

BERNHARD F. L . PREUSSER University of California Santa Barbara, California

JOSHUA H . VIERS

KIMBERLY J . REEVER MORGHAN

PETER E . WIGAND University of Nevada, Reno California State University, Bakersfield

University of California

Davis, California

University of California Davis, California

RICHARD J . REINER The Nature Conservancy Chico, California

JEANNE WIRKA

SALLY A . REYNOLDS University of California Berkeley, California

TRUMAN P. YOUNG University of California Davis, California

University of California Davis, California

KEVIN RICE

VIII

CONTRIBUTORS

The Bouverie Preserve Audubon Canyon Ranch

P R E FAC E Grass is the forgiveness of nature, her constant benediction. Forests decay, harvests perish, flowers vanish, but grass is immortal. JOHN JAMES INGALLS

1873

TO

(1833–1900),

1891, “ IN

SENATOR FROM KANSAS FROM

PRAISE OF BLUE GRASS ,” THE KANSAS MAGAZINE ,

Grasses hold a special place in our lives. We rely on them for food, gain visual relaxation from stands of them, enjoy the wildlife that thrives on them, and often plant them just outside the doors to our homes. In California, grasses have defined the modern landscape, creating the “Golden State’s” golden hills, which flush to green for part of each year. California’s famous spring wildflower shows are generally in places referred to as grasslands, and the majestic oaks that have come to symbolize the bucolic rangelands of coast range and Sierra-foothill landscapes are an integral feature of other of our grassland ecosystems. During the late 1990s significant areas of California were set aside as open space, often as mitigation for the explosive population growth of California’s cities or to meet the rising demands for recreational space and watershed protection. Much of the new parklands or protected areas are composed of grassland or savanna habitat. The management of these habitats and rising interest in the diverse elements of grassland ecology throughout the state have helped to inspire a new generation of research scientists and managers. At the same time, the allure of California native grasses and the desire to restore native biodiversity inspired the creation of the California Native Grass Association—recently renamed as the California Native Grassland Association. Some members of this association are involved with the large-scale production of native grass seed for reclamation, erosion control, and ecological restoration, whereas others focus on the prac-

1872

ticalities and challenges of restoration and management. At Stanford University, at the University of California and California State University schools (e.g., Davis, Berkeley, Santa Barbara, Northridge, Sonoma), and at other institutions of higher learning, faculty have established or expanded research programs that address basic and applied elements of California grassland ecology. Many students have taken on the challenges of learning the many species of native and non-native plants in California grasslands and trying to unravel the complex ecological relationships and forces that shape the structure and functioning of these systems. We are pleased to be able to draw on the breadth of this new knowledge by bringing together chapters composed by active research scientists investigating aspects of the ecology, history, and evolution of these ecosystems. It is our intention that this book will provide an overview and detailed synthesis of the extensive research published since the last such summary of grassland research in California (Huenneke and Mooney 1989a). We have organized the book so as to provide a systematic and historical basis for approaching research on California grasslands, followed by in-depth syntheses of recent ecological research and a discussion of management approaches and policy issues. It is our intention that this book be of use to scientists, managers, and planners, who together face the daunting challenge of trying to maintain or guide the restoration, desired composition, structure, and functioning of these ecosystems in a rapidly changing world.

IX

AC K N OWLE D G M E NTS

This book was the culmination of many meetings and discussions between researchers and managers over the past eight years. As a result, many people and organizations provided the intellectual and logistical support that ultimately led to the writing of these chapters. We would like to acknowledge the support of Dan Lufkin and Mike Markkula, who provided financial support for Mark Stromberg early in the development of the book. Discussions with John Menke, Steve Johnson (Packard Foundation and The Nature Conservancy), Val Eviner, Cini Brown, Paul Kephart, Chris Meacham, John Anderson, and others in the late 1990s and early 2000s led to various workshops and meetings that inspired many of the authors of this book. Several organizations supported the discussions and research that ultimately resulted in the book. The Packard Foundation supported a planning effort that outlined the current state of our knowledge of California grasslands and a ten-year research program to address the critical needs for information to manage and conserve our grasslands. Several of the authors and editors were involved in this planning effort, which ultimately helped to inspire this book. The National Center for Ecological Analysis and Synthesis at the University of California (UC) Santa Barbara (UCSB) hosted a working group that also included many of the authors in an effort to compile data on the current distribution and status of native grasslands in California. At UC Berkeley, the California Biodiversity Center and Museum of Vertebrate Zoology provided logistical support as well as critical funding for the authors to meet at the Bodega Marine Lab

to review, discuss, and coordinate the content of the various chapters. The U.S. Department of Agriculture Agricultural Research Service (USDA-ARS), UCSB, and the Schuyler Endowment provided critical financial support to Carla D’Antonio during the long preparation and editing of the book. We acknowledge the help of many individuals who contributed their time to provide reviews of various chapters: David Ackerly, Gene Anderson, Bruce Baldwin, Cini Brown, David Chang, Wayne Chapman, Nona Chiariello, Wendy Chou, Scott Collins, Grey Hayes, Janneke Hille Ris Lambers, David Kang, Peter Kotanen, Bill Lidicker, Mitch McClaren, Guy McPherson, John Menke, Sophie Parker, Daniel Press, Josh Schimel, Ellen Simms, Katherine Suding, Natasha Hausmann, Meredith Thomsen, Stewart Weiss, Erika Zavaleta, two anonymous reviewers, and members of the D’Antonio lab at UCSB. Many of the authors also reviewed chapters other than their own, and we appreciate this extra effort by Jim Bartolome, Hal Cushman, Jeff Dukes, Andy Dyer, Erin Espeland, Val Eviner, Jon Gelbard, Stan Harpole, Kim Reever-Morghan, Rich Reiner, Polly Schiffman, Eric Seabloom, Claudia Tyler, and Kate Skow. Jim Bartolome and Polly Schiffman in particular went beyond the call of duty in their reviewing efforts. During the production of this book, the three editors faced serious medical challenges, job changes, moves across the country, and many delays and challenges. We thank our editor, Chuck Crumly, for sticking with us through these unexpected interruptions, and we thank all of the authors and our various spouses for their perseverance and patience.

XI

ONE

Introduction J E F F R EY D. C O R B I N, CA R LA M. D’ANTO N I O, AN D R EW R. DYE R, AN D MAR K R. ST R O M B E R G

Grassland ecosystems are one of the most recognizable components of California’s wildlands, and one the state’s most important natural resources both from the perspective of biodiversity and economic values. Over 10% of California’s land area is currently grassland, and this habitat type is home to a many of the threatened and endangered animals and plants in the state (Jantz et al., Chapter 23). Yet these ecosystems have been subject to greater manipulation or destruction than any other vegetation type in the state. Ongoing conversion of grasslands to other uses and the incursion of human population centers into formerly intact habitats have placed enormous pressure on these ecosystems at the local and statewide level. In coming decades, further projected population growth, changes in land use, and alteration of climate will bring even greater pressures and make the need to understand the ecological factors that influence the state’s grasslands even more pressing. Partly because of the wide range of uses and challenges that California’s grasslands face, they have been the subject of much research. The last attempt to summarize the state of knowledge of California grasslands came in 1989, with a focus mostly on the annual grassland ecosystem (Huenneke and Mooney 1989b). Since that time, there has been an explosion of research on the ecology, management, and conservation of grasslands of California. Over 1,000 papers were published between 1990 and 2006 that included the terms California, grassland, and ecology in their titles or abstracts. Among the greatest advances in the past 16 years has been a greater understanding of the interactions between annual species and perennial ones, including the native perennial grasses that likely once dominated much of the region. We also know much more about plant-soil relations, the soil microbial community, restoration of native biodiversity, the importance and recruitment of oaks, and the ecophysiology of California grassland plants.

California’s Mediterranean grasslands differ from other important North American grasslands in ways that make their separate consideration worthwhile. Most obviously, the climate experienced by California’s grasslands, with its Mediterranean pattern of cool, wet winters and warm, dry summers (Reever Morghan et al., Chapter 7) contrasts with the Continental climate found in other grasslands. Because of the strong within-season and season-to-season variability in rainfall, the timing and amount of water available to plants may play a more important role in determining the dynamics of species composition, productivity, and nutrient cycling in California than in most other regions of North America. A second difference is that the flora of California grasslands is dominated primarily by species, especially annual grasses and forbs, whose active growth is during the “cool season” in contrast to the warm-season species found in grasslands elsewhere. Furthermore, annual species are more important in California’s Mediterranean grasslands than in most other North American grasslands— indeed, in most grasslands of the world — although the successful invasion of Bromus tectorum (cheatgrass) in Great Basin shrublands is leading to the conversion of millions of hectares of habitat in the Intermountain West into annualdominated grasslands. A final way in which California grasslands differ from most other grasslands is in their domination by non-native species. The conversion from a native- to a largely non-native-dominated flora occurred over the nineteenth and twentieth centuries and is discussed in several chapters herein. In spite of these differences, information learned by studying California grasslands has informed ecological questions that are relevant to ecosystems around the world. Examples include studies of impacts, ecology and control of invasive non-native species (D’Antonio et al., Chapter 6; Stromberg et al., Chapter 21; and DiTomaso et al., Chapter 22), population genetics (Rice and Espeland, Chapter 11), interactions

1

S I D E B A R 1 . 1 H I S TO R I C A L FA C TO R S T H AT I N F L U E N C E I N T E R P R E TAT I O N O F G R A S S L A N D E C O LO G Y BY AN D R EW DYE R

Changes in Climate

As recently as 4,000–6,000 years ago, rainfall amounts in California were consid-

erably greater and supported a more mesic-type flora than the one present today (Wigand et al., Chapter 4; Edwards 1992). As precipitation decreased, the climate shifted to a more xeric, Mediterranean type and altered the temporal distribution of rainfall, thereby creating drier conditions in the summer, shorter growing seasons, and changes in plant demography. The relict populations of Giant Redwood (Sequoia gigantea), Monterey Pine (Pinus radiata), and other species are indicative of a more mesic past, both inland and on the coast.

Loss of Large Herbivores Edwards and colleagues (Wigand et al., Chapter 4; Edwards 1992) used Rancho La Brea tar pit data to conclude that as many as 18–19 large browsing and grazing taxa were present in recent prehistory (
6,000 years before present). The ecological impact of the loss of these herbivores cannot be adequately estimated or fully appreciated. However, despite the evidence of a long history of herbivory on the West Coast, it is very possible that the composition of the California grasslands was not the result of a close association with frequent or heavy grazing (Painter 1995; Hamilton 1997b; Mensing 1998). Edwards’ (1992) data suggested a predominantly browsing herbivore fauna. The ecological significance of a nongrazing evolutionary history lies in the abrupt shift toward grazing management with European settlement. Burcham (1956, 1957) chronicled the use, and likely overuse, of the land and the possible interactions that may have occurred with other factors such as extended drought and the invasion of non-native species at the time of European contact. A likely consequence of the unprecedented severity of grazing that coincided with other biotic and abiotic changes would have been the greatly reduced abundance and distribution of grazing-intolerant species, including some bunchgrasses.

Fire Suppression It is difficult to describe with great accuracy the fire regime in California grasslands prior to European settlement, but it was certainly very different from the regime today (Reiner, Chapter 18). Minnich (1983) documented the tremendous reduction in fire frequency in southern California, as compared to Baja California, due to anthropogenic fire suppression. Arguments in favor of the importance of fire to the health of native bunchgrasses have been made (Menke 1992; Dyer and Rice 1997b; Seabloom et al. 2005), although the net effects at the community level of the reduced fire frequency are poorly understood (but see D. Dyer 2003). Fire was and is a controlling factor in grassland structure and function, capable of influencing standing biomass, nutrient cycling, competitive interactions, and a number of other basic grassland processes (Reiner, Chapter 18).

Influence of Indigenous Cultures The transition from a mesic climate to the one now characterized as Mediterranean was coincident with the development and growth of Native American cultures in California (Anderson, Chapter 5). How rapidly native cultures grew and affected natural disturbance regimes is unknown. However, it is well established that hunting, fire, horticulture, and seasonal movements were part of these societies.

Hydrological Changes

Over the twentieth century, every major river draining the Sierra Nevada moun-

tains, except the Sacramento River, has been dammed for flood control or hydroelectric purposes. The tremendous impact of flood control on the hydrology of the San Joaquin Valley cannot be overstated. Historical records of the eastern Valley floor indicate an unbroken alluvial floodplain covered with oak

riparian habitat stretching from well north of Sacramento to at least Porterville in the south, a distance of some 300 miles (Heady 1988). All water flow is now regulated, riparian areas are fragmented, and, with very few exceptions, all native habitat has been replaced (due largely to agricultural conversion) and few native plant species remain. Furthermore, changes in hydrology following flood control have altered silt deposition, with important ecosystem consequences. Silt deposition renews nutrient availability, restores soil cation exchange capacity, and washes out accumulated salts. Without overland flooding, soils can become impoverished followed by slow degradation of the communities they support. This may have contributed to the long-term reduction in Valley grassland biomass (suggested in Burcham 1957).

Invasion of Non-native Species

California’s habitats, and its grasslands in particular, have been strongly

impacted by the introduction of non-native plant and animal species (D’Antonio et al., Chapter 6). Some 1,050 non-native plant species are listed in the Jepson Manual (Hickman 1993), with the expectation that more will arrive (Rejmánek and Randall 1994). Nearly half of these exotic species are annuals, and their introduction has biased the California flora as a whole towards annual lifeforms (Heady 1988). The “annualization” of the California flora is reflected in California’s grassland communities, many of which have been converted from perennial to annual dominance. The lack of detailed presettlement botanical information greatly compounds the challenge presented by non-native species for interpreting ecological relationships between native species and their environment (Wigand et al., Chapter 4). As a result, an understanding of the processes that have driven changes in community composition and species richness in the state (or variation in impacts of exotic species by region) remains impossible to address.

between plants and soil nutrient dynamics (Eviner and Firestone, Chapter 8; Jackson et al., Chapter 9; and Harpole et al., Chapter 10), and responses to climate change (Dukes and Shaw, Chapter 19). In addition, the uneasy relationship between wildlands and urbanization may be more acute in California grasslands than in any other ecosystem in the world. Huge areas of California grassland and oak savanna are being converted from agricultural lands to low-density suburban and ranchette development or to vineyards. The consequences of these land use changes for open space and biodiversity, as well as the legal and regulatory responses (Jantz et al., Chapter 23), can inform conservationists and policy-makers elsewhere. The broad goal of this book is to present a state-of-the-art synthesis by scientists studying California grasslands on the status of our knowledge of the history, ecology, and management of this important ecosystem. We also hope that it will serve to identify holes in our knowledge and to provide a basis for research for the next generation. The book begins with an overview (Chapters 2 and 3) that presents the

nomenclature, systematics, and classifications of the important plant species and communities. The next three chapters (Chapters 4–6) present the historical context, including the past vegetation composition, the role of native peoples, and the invasion of California grasslands by nonindigenous grasses and forbs with European settlement and into the present. The third section, “Resources” (Chapters 7 – 10), presents a detailed analysis of climatic and soil conditions that play an important role in determining the phenology, distribution, and species composition of grasslands and that are in turn influenced by these same factors. The fourth section, “Ecological Interactions” (Chapters 11–19), discusses population and community ecology, including population genetics, interactions between plant species, herbivory, the role of fire, and global climate change. The final section, “Policy and Management” (Chapters 20 – 23) reviews the management of grazing in public and private lands, the use of restoration science and other integrative management tools to combat exotic species and restore native biodiversity, and the conservation of biodiversity and open space.

INTRODUCTION

3

We have chosen to focus primarily on the grassland ecosystems, which we define as herbaceous-dominated nonagricultural communities, found within the California floristic province. Though important grasslands are found in the Great Basin and Desert provinces and in Sierran montane meadows in California, we do not specifically consider them here. Instead, we focus on the coastal prairie, Coast Range grasslands, and grasslands of the Great Valley region. These grasslands grade into one another and, in some areas, grade into oak savanna and eventually oak woodland habitats. They all share the strong presence of European annual species, a history of use by both native peoples and European settlers, and an evolutionary history that includes isolation from the Great Basin flora by the Sierra Nevada. Grassdominated communities outside the California floristic province are occasionally mentioned, but this volume should not be considered a complete review of those important systems.

4

INTRODUCTION

Any consideration of the ecology of California’s grasslands today must be tempered by recognition of historical factors that influence the patterns on today’s landscapes (e.g., Sidebar 1-1). Our ability to fully understand these grassland ecosystems is, therefore, constrained by events in the past that have altered biotic and abiotic processes in ways that are often difficult to decipher. Like many other ecosystems, contemporary research is taking place in a setting where community composition is still adjusting to potentially profound short- and long-term shifts in biotic and abiotic stresses and ecological relationships. Specific historical factors are dealt with in detail in such chapters as Chapters 4 (Wigand et al.), 6 (D’Antonio et al.), and 18 (Reiner), but the influence of the forces described in these chapters is a backdrop that affects the topics in most of these chapters. We believe that embracing the variability of process, pattern, and history is critical to managing and restoring these structurally simple but ecologically dynamic and complex ecosystems.

OVE RVI EW

TWO

Systematics of California Grasses (Poaceae) PAU L M. P ETE R S O N AN D R O B E RT J. S O R E N G

The grass family (Poaceae or Gramineae) is the fourth largest flowering plant family in the world and contains about 11,000 species in 800 genera worldwide. Twenty-three genera contain 100 or more species or about half of all grass species, and almost half of the 800 genera are monotypic or diatypic, i.e., with only one or two species (Watson and Dallwitz 1992, 1999). Over the last 150 years the grass flora of California has been the subject of considerable attention by botanists. Bolander (1866) prepared the first comprehensive list, recognizing 112 grasses from California, of which 31 were introductions. Thurber (1880) mentions 175 grasses in California, and Beetle (1947) enumerates 400 known species. It is interesting to note that Crampton (1974) recognized 478 grasses in California, and of these, 175 were introduced and 156 were reported as annuals (we report 152 annuals here). We recognize 524 grass species in 144 genera; of these, 233 (44.5%) species in 65 genera are introduced (see Appendix 1), and the remaining 291 (55.5%) species in 79 genera are native. Thirty-seven species are endemic to California. One hundred fifty-two grasses in California are annual; of these 101 are introduced and 51 are native. Obviously the grass flora has been altered by humans, especially over the last 300 years since European settlement. The percentage of introduced grasses is perhaps higher in California than in any other state, simply because there are many different habitats (from 212 feet below sea level in Death Valley to 14,496 feet on top of Mount Whitney) available for colonization of weedy species. In addition, many annual species and genera of Mediterranean origins have found suitable habitats in California (see D’Antonio et al., Chapter 6). To understand the important adaptations within the grasses, a firm grasp of the unique morphological features that define this family is needed. We start this chapter with an introduction to the morphology and ecology of grasses and then discuss the phylogeny (evolutionary

relationships among organisms) of the major tribes of California grasses.

Morphology The most important feature of grasses (Poaceae) is a oneseeded indehiscent fruit (seed coat is fused with the ovary wall), known as a caryopsis or grain (see Figure 2.1; Peterson 2003). The grain endosperm is rich in starch, although it can contain protein and significant quantities of lipids. The embryo is located on the basal portion of the caryopsis and contains high levels of protein, fats, and vitamins. The stems are referred to as culms, and the roots are fibrous and principally adventitious or arising from lower portions of the culms. Silica-bodies are a conspicuous component of the epidermis and are stored in silica short-cells. Many grasses have rhizomes (underground stems) or stolons (horizontal aboveground branches) that allow for vegetative reproduction in perennial grasses. Another important feature of grasses is intercalary meristems; these allow growth well below the apex, typically near the base of the plant. The leaves are parallel-veined and two-ranked with the basal portion forming cylindrical sheaths and the upper portion referred to as a blade. A ligule, located on the upper surface at the junction of the blade and sheath, commonly consists of a flap of tissue or hairs but can be lacking. The primary inflorescence is referred to as a spikelet with one to many two-ranked bracts inserted along the floral axis or rachilla. The lowest two bracts of each spikelet, inserted opposite each other, are called glumes, above which, along the rachilla, are borne pairs of bracts termed florets. Each floret consists of a lemma (lower bract) and palea (upper bract). Within each pair of lemma and palea the highly reduced flowers can be found. Each grass flower usually consists of two or three small scales at the base called lodicules, an ovary with a style and two plumose stigmas, and one to six

7

F I G U R E 2.1. Diagnostic features of a grass (Festuca californica): caryopsis, culm, floret, flower,

and spikelet. Illustrated by Alice R. Tangerini.

(but more commonly three) stamens with basifixed anthers that contain single-pored, wind-dispersed pollen grains. Lodicules function to open the florets during flowering and evidently represent reduced perianth (sepals and petals) segments. Since the morphological features are often cryptic, or occasionally lacking, identification to species is often very difficult and requires a trained specialist.

Ecology Specializations for open habitats and grazing tolerance, highly reduced floral structure, and wind pollination in the grasses have enabled the family to be extremely successful at planetwide radiation and colonization. One notable feature of grasses and other monocots is intercalary meristems that allow individual culms to resprout once they have been removed. Grasses are well adapted to open, marginal, and frequently disturbed habitats and can be found on every continent, including Antarctica. Two major photosynthetic or carbon dioxide (CO2) assimilation pathways can be found

8

OVERVIEW

in the grasses: C3-fixing CO2 by ribulose 1,5-biphosphate (Calvin-Benson cycle, found in all vascular plants), and C4-fixing of CO2, in which the initial product of photosynthesis is not the C3 unit 3-phosphoglycerate but a unit with four C atoms (oxaloacetate). This is produced when CO2 is bound to phosphoenolpyruvate to form four-carbon molecules (oxaloacetate or malate) in the Hatch-Slack cyle. There are corresponding anatomical, physiological, phytogeographical, and ecological differences between these two types. The C3 grasses are well adapted to temperate climates with winter precipitation, whereas C4 grasses are well suited to tropical environments with summer/fall precipitation. The evolution of C4 photosynthesis has allowed grasses to outcompete other plants in warm, tropical and subtropical environments by limiting oxidation (photorespiration) of photosynthetic products (Ehleringer and Monson 1993). All of these features have led to the family’s ability to occupy nearly one-quarter of the earth’s land surface in various climatic environments as the dominant component of grasslands.

bunchgrasses were common only on well-watered floodplains (Wester 1981).

Phylogeny

F I G U R E 2.2. A hypothetical phylogeny of the grass tribes represented

in California based on Soreng et al. 2005. The first numeral indicates the number of genera within a tribe, and the numeral in parentheses () indicates the number of species in California.

A cladogram showing the relationships of the 17 tribes represented in California is given in Figure 2.2. All grasses in the BEP (Bambusoideae, Ehrhartoideae, and Pooideae) clade (all descended from a single common ancestor) and the Californian Danthonioideae are C3, whereas all grasses in the Aristidoideae and Chloridoideae are C4. The Panicoideae have C3, C4, and C3-C4 intermediates, although the majority of the species in California are C4. Historically, the grassland biome has been maintained by a myriad of biotic, climatic, and edaphic effects. First, there usually is a dry season in which grasses and adjacent forest border dry out and become flammable (Axelrod 1985). Repeated fires favor grasses over most tree and shrub species, since they very easily resprout from the base. Second, large herbivorous mammals (e.g., bison and antelope) are instrumental at maintaining and further opening up grassland communities (Axelrod 1985). An often overlooked consequence of grazing animals is their effect on soil compaction, which again favors sod-forming grasses over trees and shrubs. The exact species composition of California’s preagricultural grasslands is not very well documented. Wester (1981) and Holstein (2001) have presented welldocumented accounts based on historical records and current ecological samples of relict vegetation in California. They found these grasslands to be spatially diverse with many different species of the annual or perennial habit. In the San Joaquin Valley, grasslands were apparently dominated by annual species and xerophytic shrubs, and perennial

Despite variation among grass species in inflorescence structure and vegetative morphology, the grass family was probably characterized as a distinct entity in most early cultures. Three hundred years before the Christian era, Theophrastus, a Greek scholar, recognized the grass family and began to teach his students the concepts of plant morphology. The first scientific subdivision of the family was made by Robert Brown (1814), who recognized two different spikelet types between subfamily Panicoideae and Pooideae (Festucoideae). Bentham (1881) recognized 13 tribes grouped in the two major subfamilies. Hitchcock (1935) and Hitchcock and Chase (1951), in their treatments of the grasses of the United States, recognized 14 tribes in these two major subfamilies. The two-subfamily classification was used by most agrostologists for almost 150 years until more modern syntheses were developed. With the infusion of molecular data our present concept and classification of the grasses is changing at a rapid rate. In California we currently recognize eight subfamilies: Bambusoideae, Ehrhartoideae, Pooideae, Arundinoideae, Danthonoideae, Aristidoideae, Chloridoideae, and Panicoideae (GPWG 2001; Soreng et al. 2005), and in these subfamilies we recognize 18 tribes and 44 subtribes (Table 1). A cladogram (see Figure 2.2) of these 18 tribes summarizes the most widely accepted concepts regarding the phylogenetic relationships among the tribes and subfamilies represented in California (GPWG 2001, Soreng and Davis 1998, 2000). The tree is rooted between the PACAD and BEP clades. Three numerically small, tropical subfamilies of grasses, not represented in California, diverge below this root point. In the PACAD clade, a clade containing Panicoideae (Andropogoneae, Thysanolaeneae, and Paniceae) and the Arundinoideae (Arundineae) is sister to a clade containing the Chloridoideae (Cynodonteae, Eragrostideae, and Zoysieae), Aristidoideae (Aristideae), and Danthonioideae (Danthonieae). In the BEP clade the Pooideae (Brachypodieae, Bromeae, Meliceae, Poeae, Stipeae, and Triticeae) is sister to a clade of the Ehrhartoideae (Ehrharteae and Oryzeae) and the Bambusoideae (Bambuseae). The BEP clade corresponds, in part, to the old term “festucoid” grasses used by historical agrostologists. The three most diverse subfamilies in California are the Pooideae with 323 (61.6%) species in 73 genera, the Chloridoideae with 94 (17.9%) species in 30 genera, and the Panicoideae with 80 (15.3%) species in 24 genera.

Panicoideae The Panicoideae are the least diverse of the three major subfamilies represented by the California grasses, and there are no endemic Panicoideae within the state. This paucity of

S Y S T E M AT I C S O F C A L I F O R N I A G R A S S E S

9

TA B L E 2.1 Classification of the Grasses Found in California

Subfamily Bambusoideae Tribe Bambuseae Subtribe Arundinariinae Pseudosasa Subtribe Shibataeniae Phyllostachys

Subfamily Ehrhartoideae (synonym: Oryzoideae)

Subtribe Phalaridinae Anthoxanthum, Phalaris Subtribe Brizinae Briza Subtribe Agrostidinae Agrostis, Ammophila, Bromidium, Calamagrostis, Gastridium, Lachnagrostis, Podagrostis, Polypogon Subtribe Puccinelliinae Puccinellia, Sclerochloa

Tribe Ehrharteae Ehrharta Tribe Oryzeae Subtribe Oryzinae Leersia, Oryza Subtribe Zizaniinae Zizania

Subtribe Poinae Apera, Dissanthelium, Poa Subtribe Alopecurinae Alopecurus, Beckmannia, Phleum Subtribe Holcinae Holcus Subtribe Airinae Aira, Deschampsia, Vahlodea, Ventenata

Subfamily Pooideae Tribe Stipeae Subtribe Stipinae Achnatherum, Hesperostipa, Jarava, Nassella, Piptatherum, Piptochaetium, Ptilagrostis, Stipa Subtribe Ampelodesminae Ampelodesmos Tribe Meliceae Glyceria, Melica, Pleuropogon Tribe Brachypodieae Brachypodium Supertribe Poodae Tribe Poeae Subtribe Torreyochloinae Amphibromus, Torreyochloa Subtribe Aveninae Arrhenatherum, Avena, Cinna, Gaudinia, Graphephorum, Koeleria, Lagurus, Rostraria, Sphenopholis, Trisetum

Subtribe Scribneriinae Scribneria Subtribe Loliinae Festuca, Leucopoa, Lolium, Schedonorus, Vulpia Subtribe Dactylidinae Dactylis, Lamarckia Subtribe Cynosurinae Cynosurus Subtribe Parapholinae Catapodium, Cutandia, Hainardia, Parapholis Supertribe Triticoidae Tribe Bromeae Bromus Tribe Triticeae Subtribe Hordeinae Agropyron, Elymus, Hordeum, Leymus, Pascopyrum, Pseudoroegneria, Secale Subtribe Triticinae Aegilops, Taeniatherum, Thinopyrum, Triticum

TA B L E 2.1 ( C O N T I N U E D ) Classification of the Grasses Found in California

Subfamily Panicoideae Tribe Thysanolaeneae Thysanolaena Tribe Paniceae

Subfamily Danthonioideae Tribe Danthonieae Cortaderia, Danthonia, Karroochloa, Rytidosperma, Schismus, Tribolium

Subtribe Cenchrinae Cenchrus, Pennisetum Subtribe Digitariinae Digitaria Subtribe Melinidinae Eriochloa, Melinis, Urochloa Subtribe Setariinae Setaria, Stenotaphrum Subtribe Panicinae Dichanthelium, Echinochloa, Panicum Subtribe Paspalinae Axonopus, Paspalum Tribe Andropogoneae Imperata, Miscanthus, Saccharum (Erianthus) Subtribe Sorghinae Bothriochloa, Sorghum Subtribe Andropogoninae Andropogon, Schizachyrium Subtribe Anthistiriinae Heteropogon, Hyparrhenia, Themeda Subtribe Tripsacinae Zea

Subfamily Chloridoideae Tribe Cynodonteae Acrachne, Blepharidachne, Dactyloctenium, Leptochloa, Scleropogon, Swallenia, Tridens Subtribe Boutelouinae Bouteloua Subtribe Chloridinae Chloris, Cynodon, Eustachys Subtribe Eleusiniae Eleusine Subtribe Hilariinae Hilaria (Pleuraphis) Subtribe Monanthochloinae Distichlis, Monanthochloe Subtribe Muhlenbergiinae Lycurus, Muhlenbergia, Schedonnardus Subtribe Munroinae Dasyochloa, Erioneuron, Munroa Subtribe Orcuttiinae Neostapfia, Orcuttia, Tuctoria Tribe Eragrostideae Subtribe Cotteinae

Subfamily Arundinoideae Tribe Arundineae Arundo, Phragmites

Enneapogon Subtribe Eragrostidinae Eragrostis Tribe Zoysieae Subtribe Sporobolinae

Subfamily Aristidoideae Tribe Aristideae Aristida

Crypsis, Spartina, Sporobolus Subtribe Zoysiinae Zoysia

NOTE : Based on Soreng et al. 2005. Native taxa are bold, introduced taxa are lightface type. All genera are italicized. A genus is considered native if it includes one or more native species (see Appendix 1 for clarification of native versus introduced species).

species diversity is likely a direct response to the climatic patterns of the past and present, because Panicoideae grasses are best suited to warm and humid environments of tropical and warm temperate zones. The spikelets in this subfamily usually have two glumes and two closely spaced florets; the lower floret is usually sterile, the upper floret without a rachilla extension.

Paniceae The Paniceae in California contain 57 species in 13 genera and are the sister to the Andropogoneae (Figure 2.2). They are characterized by having two-flowered spikelets with membranous glumes, the lower floret staminate or reduced, membranous, and the upper floret perfect and firm. Even though Panicum (10 spp. in California) and Paspalum (4 spp. in California) are large genera in the eastern United States and especially in tropical America, they are very poorly represented in the western United States, where the climate is generally dryer, especially in the warmer months.

Andropogoneae The Andropogoneae are characterized by having fragile racemes of paired spikelets, where there is a sessile and a pedicellate spikelet with differing sexuality. Commonly the pedicellate spikelets are staminate or reduced, and the sessile spikelets are usually perfect or pistillate. Within spikelets, typically the glumes are firm, and the two florets have membranous bracts. The Andropogoneae in California contains 23 species in 11 genera, most of which are uncommon in California grasslands.

Pooideae There are a few morphological synapomorphies (diagnostic characteristics) delineating the Pooideae. In this subfamily trends include parallel-sided subsidiary cells, nonvascularized lodicules with a membranous margin, an epiblast with no scutellar cleft, and the absence of microhairs.

Stipeae In California the earliest diverging lineage in the Pooideae clade (Figure 2.2) is the Stipeae. The Stipeae probably arose in Laurasia (37–24 mybp) since a few fossil reports, e.g., Stipideum and possibly Piptochaetium, are from the Oligocene in North America (Thomasson 1987). Therefore, ancestors of this tribe likely were able to colonize the North American and Eurasian continents before they separated. The Stipeae are characterized as having one-flowered spikelets without rachilla extensions and terete florets that are usually awned near or immediately below the apex and have a well developed, often sharply pointed callus. The Stipeae include three endemics centered in the Sierra Nevada: Achnatherum latiglume from the Transverse Ranges and central and southern Sierra Nevada, A. stillmanii from the northern Sierra Nevada, and Ptilagrostis kingii from

12

OVERVIEW

the central and southern Sierra Nevada highlands. Members of the Stipeae are well adapted to the steppe vegetation in Eurasia and the Americas, where they are often dominant elements. Nassella cernua, N. lepida, and N. pulchra are sometimes dominant in parts of the California grasslands.

Meliceae The next tribe to diverge from the main lineage in the Pooideae clade is Meliceae, also reported as having possible fossils in the Oligocene of North America (Thomasson 1987). The Meliceae have closed sheaths and lemmas that are five- to 13-nerved; short, bushy stigmas; and short, truncate, fleshy lodicules. Melica torreyana is an endemic from the northwestern region, the Sierra Nevada, and central western California. Three other varieties of Melica are also endemic: Melica californica var. nevadensis Boyle from northwestern and central western regions, Sierra Nevada foothills, western Transverse Ranges, and the Tehachapi Mountains; M. geyeri var. aristulata J.T. Howell from the San Francisco Bay Area; and M. stricta var. albicaulis Boyle from the western Transverse Ranges. The endemics Pleuropogon californicus and P. hooverianus occur in marshy areas from northwestern California, Cascade Range foothills, and north and central Sierra Nevada foothills; and from the southern North Coast and northern Central Coast regions, respectively.

Poaeae The largest tribe, Poaeae (184 spp.), includes Poa with 34 species. Poa has diversified throughout temperate, boreal, and arctic regions around the world and occurs on islands of similar habitats in the tropics (Gillespie and Soreng 2005). The Poeae clade, equivalent to supertribe Poodae, is sandwiched between the ancestral Brachypodieae, with only five introduced species, and the sister supertribe Triticoidae, which includes the Bromeae, with 32 species, and the Triticeae, with 45 species. In California, Poa exhibits both high species diversity (34 spp., of these, 28 are native) and a high degree of endemism, with eight species confined to the state (see Appendix 1). Poa is characterized as having rather small, multiflowered spikelets; lemmas that are keeled, unawned, usually five-nerved, commonly with weblike hairs from the dorsal side of the callus, caryopses that are firm with lipid and a short hilum; lodicules that are broadly lanceolate, often with a lateral lobe; leaf sheaths closed above the base between 1/20 the entire length and the top; leaf blades that generally have two rows of bulliform cells (one on either side of the midnerve, these appearing like railroad tracks) and no additional rows of bulliform cells; and blades commonly with naviculate (boatshaped) apices (Soreng 1993; in press a, b). The Poaeae endemics, Poa kecki, P. stebbinsii, and Cinna bolanderi, have originated on “islands” of arctic habitat in the high Sierra Nevada between 1,800 and 4,000 meters. Poa atropurpurea is known only from high-elevation meadows (1,500–2,000 meters) in the Peninsular Ranges and the San Bernardino Mountains. Poa sierrae and P. tenerrima (known only from serpentine outcrops)

occur in the Sierra Nevada canyons and foothills. Poa napensis is known only from mineralized soils around hot springs in the North Coast Ranges, P. kelloggii is known only from the North and Central Coast Redwood forests (not found in Oregon), P. douglasii is known only from the South Coast sand dunes, and P. diabolii is known only from coastal soft scrub over Edna shale in the South Coast. These species of Poa belong to three subgenera and four sections. Poa kelloggii belongs to the earliest-diverging lineage in the genus, Sylvestres, a subgenus that, so far as is known, is endemic and principally confined to rich forests of North America. Poa napensis and P. tenerrima belong to subgenus Stenopoa section Secundae and are closely related to P. secunda, which is perhaps the most common native grass across California, occurring in a wide range of habitats from coast range low-elevation sites to high-elevation Sierra Nevada meadows and Great Basin grasslands. Poa keckii belongs to a complex of short-anthered species of the western cordillera of North America (Beringia), placed in subgenus Stenopoa section Abbreviatae. Poa atropurpurea, P. douglasii, and P. diaboli are members of the diclinous Poa subgenus Poa section Madropoa, which is centered in and mostly endemic to western North America. Several other species of Poa section Madropoa are nearly confined to California; P. piperi, P. pringlei, and P. rhizomata extend into SW Oregon on serpentine, volcanic, and peridotite substrates, respectively. Other endemic species in the Poeae, subtribe Agrostidinae include Agrostis blasdalei, A. hooveri, Calamagrostis ophitidis, and C. foliosa, all from the North and South Coast Ranges, and Dissanthelium californicum (subtribe Poinae), which was previously thought to be extinct but was recently re-collected on Santa Catalina Island. Scribneria bolanderi, the sole species of subtribe Scribneriinae, is endemic to vernal pool habitats in the California Floristic Province, although it also reaches Oregon. The generic relationship between Agrostis and Calamagrostis is somewhat controversial since both are morphologically similar and have one-flowered spikelets. Species of Calamagrostis have rachilla extensions (usually hairy), a callus with hairs, and membranous to chartaceous lemmas, whereas species of Agrostis do not have rachilla extensions, have a callus that is usually glabrous, and have hyaline to membranous lemmas (Peterson and Saarela in press). Current research on Dissanthelium indicates that species in this genus should be subsumed within Poa (Gillespie and Soreng 2005; Refulio Rodriguez, personal communication). Puccinellia howellii (Puccinelliinae), another endemic, is known only from mineral springs in the Yolla Bolly Mountains and the Klamath Range.

Bromeae The Bromeae are characterized as having closed sheaths, lemmas that are bifid or toothed with a subapical awn, hairy apically bilabiate appendages of the ovary, and simple starch grains. In California, Bromus (Bromeae) consists of 32 species; of these, 17 are native. Bromus can be distinguished from other grasses by having connate leaf sheath margins, subapically inserted awns, hairy apical bilabiate appendages of

the ovary, and simple starch grains (Wagnon 1952; Saarela and Peterson in press). This genus is widely distributed in temperate and mountainous regions of the Northern and Southern Hemispheres, and several species are important native forage grasses in California [B. ciliatus, B. richardsonii, B. suksdorfii] (Peterson et al. 2002). Endemics within California include Bromus grandis and B. hallii from the southern Sierra Nevada, Transverse and Peninsular ranges; and B. pseudolaevipes from the San Francisco Bay Area, Outer South Coast Ranges, South Coast, Channel Islands, Western Transverse Ranges, and the Peninsular Ranges (Saarela and Peterson in press). All three of these species were included in Wagnon’s (1952) Pacific Slope Group of Bromus section Bromopsis, where he mentions that B. grandis and B. hallii perhaps share a common origin with B. orcuttianus. Bromus carinatus is a widespread native that occurs in many habitats mostly below 3500 meters. The genus Bromus contains 15 introduced species; many of these are invasive in California grasslands. Bromus diandrus and B. hordeaceus are widespread and dominant or codominant throughout coastal and valley grasslands (see D’Antonio et al., Chapter 6). Bromus tectorum and B. madritensis subsp. rubens (L.) Husn. are more common in the California deserts. These four species of Bromus are listed by the California Invasive Plant Council (Cal-IPC at http://www.cal-ipc.org) as invasive pest plants of concern to wildland habitat.

Triticeae The Triticeae (sister to the Bromeae), or wheat grass tribe, in California includes 45 species in 10 genera. The tribe is characterized by having a true spike inflorescence where all the spikelets are sessile and aligned singly or in groups of two or three along the central rachis; coriaceous glumes and lemmas; ovaries with densely hairy apices; and caryopses with simple starch grains and long hilums. The evolutionary history of this tribe is fairly well known since wheat (Triticum aestivum), barley (Hordeum vulgare), and rye (Secale cereale) are members. The tribe is thought to have originated in Eurasia, possibly during the Miocene, and then radiated to the New World (Blattner 2006). Two species, Elymus californicus and Leymus pacificus, are endemic to coastal California. The former species is known from the North and Central Coast prairies, North Coast Ranges, and San Francisco Bay Area, and the latter is known from the North and Central Coast and the Channel Islands. Leymus condensatus is also a conspicuous associate (culms 1.5 – 3.5 meters tall) of the chaparral and coastal sage scrub in California and Baja California, Mexico, and L. triticoides was perhaps historically dominant on heavier soils in valleys and hillslopes of Central California (Gould and Moran 1981; Holstein 2001).

Chloridoideae The core species in this subfamily share two morphological synapomorphies. All exhibit “Kranz” or C4 leaf anatomy (except Eragrostis walteri Pilg. from South Africa) and most

S Y S T E M AT I C S O F C A L I F O R N I A G R A S S E S

13

have chloridoid bicellular microhairs (broad, short terminal cell the same thickness as the basal cell) present on leaf surfaces (Peterson et al. in press). Other character trends in the chloridoids include a base chromosome number of x
10 (a pleisiomorphy or ancestral characteristic), embryos with nonlinear hilums that are usually punctiform or small with elongate mesocotyl internodes, and two fleshy, vascularized, truncate lodicules (GPWG 2001; Soreng and Davis 1998). However, most of these character trends are seen in sister subfamilies: Aristidoideae, Arundinoideae, Danthonioideae, and Panicoideae. The Eragrostideae is considered the earliest diverging tribal lineage of the chloridoids and is sister to a clade that contains the Zoysieae and the Cynodonteae (Figure 2.2). Character combinations in the Eragrostideae include spikelets with many florets, lemmas with 3 to 13 nerves, and many species adapted to xeric habitats. At this point we have no clear idea as to the relationships among the seven Cynodonteae subtribes (see Table 1). However, we do have good molecular support for maintaining the tribe Cynodonteae and morphological support for all of the seven subtribes (Peterson et al. in press). There are no definitive morphological characters that differentiate the Cynodonteae from the Eragrostideae and/or Zoysieae; the Cynodonteae essentially includes most of the variation present in the entire subfamily. The evolutionary history of the chloridoids as a whole is even more obscure. Thomasson’s et al. (1986) identification of Kranz anatomy in a fossil from a Miocene Ogallala formation in Kansas is the first definitive record. Since more than half of the genera within the Chloridoideae reside in Africa and all larger tribes and subtribes, excluding Muhlenbergiinae, have centers of diversity in Africa, Hartley and Slater (1960) concluded that the subfamily probably originated on the African continent (perhaps during the Oligocene) and spread from that region to other parts of the world.

Cynodonteae The Boutelouinae, Hilariinae, Muhlenbergiinae, and Orcuttinae are clearly North American subtribes, but how their ancestor(s) arrived there is obscure. These subtribes probably ultimately descended from a Laurasian ancestor, given the distribution of Muhlenbergia, i.e., predominantly from the southwestern United States and northern Mexico, and also with six species in China (Peterson 2000; Peterson and OrtízDiaz 1998; Peterson et al. in press; Wu and Peterson 2006). Since the Chloridinae are most species-rich in South America, it seems likely that they originated in that continent and spread northward, although we have no genetic evidence for this. It is very difficult to determine any directional signal from the Monanthochloinae and Munroinae, although these two subtribes are slightly more species-rich in South America, suggesting a southern derivation. The Eragrostideae and Zoysinae are more likely west Gondwanaland groups, although

14

OVERVIEW

the exact timing and routes of migration to both North America and South America are unknown. The Muhlenbergiinae are characterized by having spikelets perfect, staminate, or sterile; occasionally with cleistogenes in the leaf sheaths; inflorescence paniculate of spicate main branches or a single raceme; spikelet-bearing axis disarticulating (falling entire) or persistent; spikelets solitary, rarely paired or in triplets, occasionally secund; glumes awned or unawned; lemmas three-nerved, awned or unawned; and a base chromosome number of x
8–10 (Peterson 2000). The largest genus, Muhlenbergia, has 18 species in California with a single introduced species (Muhlenbergia schreberi). Two species, M. californica and M. jonesii, are endemic to California, the former occurring in the South Coast and Transverse Ranges and the latter known only from the northern Sierra Nevada and the Klamath and High Cascade ranges (Peterson 1993, in press). Seven species (Neostapfia colusana, Orcuttia inaequalis, O. pilosa, O. tenuis, O. viscida, Tuctoria greenei, and T. mucronata) of the eight species in the Orcuttiinae are endemic to California. Unlike most Chloridoideae, which are adapted to summer rains, these annual species are well adapted to winter rains and summer drought. They occur in the western part of the state in vernal pools, an endangered habitat. This subtribe is a unique lineage in the Cynodonteae and is exclusively restricted to the California biome [Orcuttia californica and Tuctoria fragilis (Swallen) J. Reeder extend into Baja California, Mexico].

Introductions There are 43 introduced grass species currently included in the Cal-IPC Invasive Plant Inventory (Cal-IPC), and 28 species in 20 genera do not share any native congeners (species belonging to the same genus). Two of these genera, Brachypodium with five species and Ehrharta with three species, represent introduced tribes, the Brachypodieae and Ehrharteae. Strauss et al. (2006) compared three groups: introduced species that are harmful to California ecosystem, native species, and introduced species that cause relatively little harm to California ecosystems. They demonstrated that the harmful introduced species are more distant phylogenetically from the native species than the benign introduced species are. This is an interesting conclusion, since it implies that Darwin’s naturalization hypothesis and the “the escape from natural enemies” hypothesis are valid; species that are more distantly related to the native community are more likely to become noxious invasive weeds (Strauss et al. 2006). The following grasses are currently on the alert category (species that appear to be expanding their range or species showing signs of being invasive in some areas) published by the Cal-IPC: Brachypodium sylvaticum, Ehrharta longiflora, Spartina alterniflora, S. anglica, S. densiflora, and Stipa capensis. Aegilops triuncialis, Ammophila arenaria, Arundo donax, Ehrharta calycina, and Taeniatherum caput-medusae are reported on the

Cal-IPC list as having a high rating (species that have severe ecological impacts, have moderate to high rates of dispersal, and are widely distributed).

Evolution toward Specialization We can see several overarching patterns in the distribution of native and endemic species. There were repeated specializations to narrowly distributed habitats or restricted edaphic or climate settings: (1) isolated wetlands, including (a) vernal pools, mostly of the Central Valley and adjacent foothills (Orcuttiinae, Scribneria, Phalaris lemmonii, Pleuropogon californicus, Puccinellia simplex [now introduced in Utah]), (b) saline springs (Puccinellia howellii and the rare P. parishii, which is sporadic across the southwestern states) and mineralized soils around springs (Poa napensis), (c) freshwater wetlands and moist mountain meadows (Pleuropogon hooverianus, Poa atropurpurea, P. stebbinsii, Ptilagrostis kingii); (2) sand dunes (Agrostis blasdalii, Calamagrostis bolanderi, Leymus pacificus, Poa douglasii, Swallenia alexandrae); 3) ultramafic substrates (Calamagrostis ophitidis, Poa piperi, P. rhizomata, P. tenerrima) and isolated shales (P. diaboli); (4) alpine and peaks (Alopecurus aequalis, A. geniculatus, Calamagrostis muiriana, Cinna bolanderi, Festuca brachyphylla, Koeleria macrantha, Poa glauca ssp. rupicola (Nash) W.A. Weber, P. keckii, P. pringlei); (5) the California Floristic Province (Achantherum latiglume, A. stillmannii, A. diegoense, A. coronatum); (6) central and south coastal grasslands (Melica imperfecta, Nassella cernua, N. lepida, N. pulchra, Aristida hamulosa, Leymus condensatus, Muhlenbergia microsperma, M. rigens); (7) southern coastal mountains/chaparral and forests (Achnatherum parishii, Elymus stebbinsii, Melica frutescens, M. torreyana, Phalaris californica, P. lemmonii). Many of these species of limited distribution have evolved from more widespread congeners and belong to genera that are species-rich and well established in California.

Appendix 1: A List of the Grass Species Known to Occur in California Intraspecific categories are not included. Bolded names are native, and those marked with an asterisk (*) are endemic. All other species are introduced and naturalized. This list was prepared using the Catalogue of New World Grasses (Soreng et al. 2005), PLANTS (USDA, NRCS 2006), and the Grass Manual on the Web (Barkworth et al. 2006). Also consulted but not completely followed were The Grasses of California (Smith 2006), A Synthesis of the North American Flora (Kartesz and Meacham 2006), and the Jepson Online Interchange for California floristics (JOI 2006). We have not done an extensive evaluation for all possible introductions, since these are continually being added as reports are published. Achnatherum altum (Swallen) Hoge & Barkworth Achnatherum aridum (M.E. Jones) Barkworth

Achnatherum coronatum (Thurb.) Barkworth Achnatherum diegoense (Swallen) Barkworth Achnatherum hymenoides (Roem. & Schult.) Barkworth *Achnatherum latiglume (Swallen) Barkworth Achnatherum lemmonii (Vasey) Barkworth Achnatherum lettermanii (Vasey) Barkworth Achnatherum nelsonii (Scribn.) Barkworth Achnatherum nevadense (B.L. Johnson) Barkworth Achnatherum occidentale (Thurb. ex S. Watson) Barkworth Achnatherum parishii (Vasey) Barkworth Achnatherum pinetorum (M.E. Jones) Barkworth *Achnatherum stillmanii (Bol.) Barkworth Achnatherum thurberianum (Piper) Barkworth Achnatherum webberi (Thurb.) Barkworth Acrachne racemosa (B. Heyne ex Roem. & Schult.) Ohwi Aegilops cylindrica Host Aegilops geniculata Roth Aegilops tauschii Coss. Aegilops triuncialis L. Agropyron cristatum (L.) Gaertn. Agropyron desertorum (Fisch. ex Link) Schult. Agropyron fragile (Roth) P. Candargy *Agrostis blasdalei Hitchc. Agrostis capillaris L. Agrostis densiflora Vasey Agrostis elliottiana Schult. Agrostis exarata Trin. Agrostis gigantea Roth Agrostis hallii Vasey Agrostis hendersonii Hitchc. *Agrostis hooveri Swallen Agrostis idahoensis Nash Agrostis microphylla Steud. Agrostis oregonensis Vasey Agrostis pallens Trin. Agrostis scabra Willd. Agrostis stolonifera L. Agrostis variabilis Rydb. Aira caryophyllea L. Aira elegantissima Schur Aira praecox L. Alopecurus aequalis Sobol. Alopecurus carolinianus Walter Alopecurus geniculatus L. Alopecurus myosuroides Huds. Alopecurus pratensis L. Alopecurus saccatus Vasey Ammophila arenaria (L.) Link Ammophila breviligulata Fernald Ampelodesmos mauritanicus (Poir.) T. Durand & Schinz Amphibromus neesii Steud. Andropogon glomeratus (Walter) Britton, Sterns & Poggenb. Andropogon virginicus L. Anthoxanthum aristatum Boiss.

S Y S T E M AT I C S O F C A L I F O R N I A G R A S S E S

15

Anthoxanthum nitens (Weber) Y. Schouten & Veldkamp Anthoxanthum occidentale (Buckley) Veldkamp Anthoxanthum odoratum L. Apera interrupta (L.) P. Beauv. Apera spica-venti (L.) P. Beauv. Aristida adscensionis L. Aristida californica Thurb. Aristida dichotoma Michx. Aristida divaricata Humb. & Bonpl. ex Willd. Aristida hamulosa Henrard Aristida oligantha Michx. Aristida purpurea Nutt. Aristida schiedeana Trin. & Rupr. Aristida ternipes Cav. Arrhenatherum elatius (L.) P. Beauv. ex J. Presl & C. Presl Arundo donax L. Avena barbata Pott ex Link Avena fatua L. Avena occidentalis Durieu Avena sativa L. Avena sterilis L. Avena strigosa Schreb. Axonopus fissifolius (Raddi) Kuhlm. Beckmannia syzigachne (Steud.) Fernald Blepharidachne kingii (S. Watson) Hack. Bothriochloa barbinodis (Lag.) Herter Bothriochloa ischaemum (L.) Keng Bothriochloa laguroides (DC.) Herter Bouteloua aristidoides (Kunth) Griseb. Bouteloua barbata Lag. Bouteloua curtipendula (Michx.) Torr. Bouteloua eriopoda (Torr.) Torr. Bouteloua gracilis (Kunth) Lag. ex Griffiths Bouteloua trifida Thurb. Brachypodium distachyon (L.) P. Beauv. Brachypodium phoenicoides (L.) P. Beauv. ex Roem. & Schult. Brachypodium pinnatum (L.) P. Beauv. Brachypodium rupestre (Host) Roem. & Schult. Brachypodium sylvaticum (Huds.) P. Beauv. Briza maxima L. Briza media L. Briza minor L. Bromidium tandilense (Kuntze) Rúgolo Bromus alopecuros Poir. Bromus arenarius Labill. Bromus arizonicus (Shear) Stebbins Bromus berteroanus Colla Bromus briziformis Fisch. & C.A. Mey. Bromus carinatus Hook. & Arn. Bromus catharticus Vahl Bromus cebadilla Steud. Bromus ciliatus L. Bromus commutatus Schrad. Bromus diandrus Roth *Bromus grandis (Shear) Hitchc. *Bromus hallii (Hitchc.) Saarela & P.M. Peterson

16

OVERVIEW

Bromus hordeaceus L. Bromus inermis Leyss. Bromus japonicus Thunb. Bromus laevipes Shear Bromus madritensis L. Bromus marginatus Nees ex Steud. Bromus maritimus (Piper) Hitchc. Bromus orcuttianus Vasey Bromus polyanthus Scribn. ex Shear Bromus porteri (J.M. Coult.) Nash *Bromus pseudolaevipes Wagnon Bromus racemosus L. Bromus richardsonii Link Bromus secalinus L. Bromus sitchensis Trin. Bromus sterilis L. Bromus suksdorfii Vasey Bromus tectorum L. Bromus vulgaris (Hook.) Shear Calamagrostis bolanderi Thurb. Calamagrostis breweri Thurb. Calamagrostis canadensis (Michx.) P. Beauv. *Calamagrostis foliosa Kearney Calamagrostis koelerioides Vasey Calamagrostis muiriana B.L. Wilson & Sami Gray Calamagrostis nutkaensis (J. Presl) J. Presl ex Steud. *Calamagrostis ophitidis ( J.T. Howell) Nygren Calamagrostis purpurascens R. Br. Calamagrostis rubescens Buckley Calamagrostis stricta (Timm) Koeler Catapodium rigidum (L.) Dony Cenchrus ciliaris L. Cenchrus echinatus L. Cenchrus incertus M.A. Curtis Cenchrus longispinus (Hack.) Fernald Chloris gayana Kunth Chloris truncata R. Br. Chloris verticillata Nutt. Chloris virgata Sw. *Cinna bolanderi Scribn. Cinna latifolia (Trevir. ex Go|2pp.) Griseb. Cortaderia jubata (Lemoine) Stapf Cortaderia selloana (Schult. & Schult. f.) Asch. & Graebn. Crypsis alopecuroides (Piller & Mitterp.) Schrad. Crypsis schoenoides (L.) Lam. Crypsis vaginiflora (Forssk.) Opiz Cutandia memphitica (Spreng.) K. Richt. Cynodon dactylon (L.) Pers. Cynodon plectostachyus (K. Schum.) Pilg. Cynodon transvaalensis Burtt Davy Cynosurus cristatus L. Cynosurus echinatus L. Dactylis glomerata L. Dactyloctenium aegyptium (L.) Willd. Danthonia californica Bol. Danthonia decumbens (L.) DC.

Danthonia intermedia Vasey Danthonia unispicata (Thurb.) Munro ex Macoun Dasyochloa pulchella (Kunth) Willd. ex Rydb. Deschampsia cespitosa (L.) P. Beauv. Deschampsia danthonioides (Trin.) Munro Deschampsia elongata (Hook.) Munro Dichanthelium acuminatum (Sw.) Gould & C.A. Clark Dichanthelium oligosanthes (Schult.) Gould Digitaria bicornis Digitaria ciliaris (Retz.) Koeler Digitaria eriantha Steud. Digitaria ischaemum (Schreb.) Schreb. ex Muhl. Digitaria sanguinalis (L.) Scop. *Dissanthelium californicum (Nutt.) Benth. Distichlis spicata (L.) Greene Echinochloa colona (L.) Link Echinochloa crus-galli (L.) P. Beauv. Echinochloa crus-pavonis (Kunth) Schult. Echinochloa muricata (P. Beauv.) Fernald Echinochloa oryzoides (Ard.) Fritsch Echinochloa phyllopogon (Stapf) Stapf ex Kossenko Ehrharta calycina Sm. Ehrharta erecta Lam. Ehrharta longiflora Sm. Eleusine indica (L.) Gaertn. Eleusine tristachya (Lam.) Lam. Elymus arizonicus (Scribn. & J.G. Sm.) Gould *Elymus californicus (Bol. ex Thurb.) Gould Elymus canadensis L. Elymus elymoides (Raf.) Swezey Elymus glaucus Buckley Elymus lanceolatus (Scribn. & J.G. Sm.) Gould Elymus multisetus (J.G. Sm.) Burtt Davy Elymus repens (L.) Gould Elymus scribneri (Vasey) M.E. Jones Elymus sierrae Gould *Elymus stebbinsii Gould Elymus trachycaulus (Link) Gould ex Shinners Enneapogon desvauxii P. Beauv. Eragrostis barrelieri Daveau Eragrostis cilianensis (All.) Vignolo ex Janch. Eragrostis curvula (Schrad.) Nees Eragrostis hypnoides (Lam.) Britton, Sterns & Poggenb. Eragrostis lehmanniana Nees Eragrostis lutescens Eragrostis mexicana (Hornem.) Link Eragrostis minor Host Eragrostis pectinacea (Michx.) Nees Eragrostis pilosa (L.) P. Beauv. Eragrostis superba Peyr. Eriochloa acuminata (J. Presl) Kunth Eriochloa aristata Vasey Eriochloa contracta Hitchc. Eriochloa fatmensis (Hochst. & Steud.) Clayton Eriochloa villosa (Thunb.) Kunth Erioneuron pilosum (Buckley) Nash

Eustachys distichophylla (Lag.) Nees Festuca ammobia Pavlick Festuca arvernensis Auquier, Kergue|4len & Markgr.-Dann. Festuca brachyphylla Schult. & Schult. f. Festuca californica Vasey Festuca elmeri Scribn. & Merr. Festuca idahoensis Elmer Festuca minutiflora Rydb. Festuca occidentalis Hook. Festuca roemeri (Pavlick) E.B. Alexeev Festuca rubra L. Festuca saximontana Rydb. Festuca sororia Piper Festuca subulata Trin. Festuca subuliflora Scribn. Festuca trachyphylla (Hack.) Krajina Festuca viridula Vasey Gastridium phleoides (Nees & Meyen) C.E. Hubb. Gaudinia fragilis (L.) P. Beauv. Glyceria borealis (Nash) Batch. Glyceria elata (Nash) M.E. Jones Glyceria fluitans (L.) R. Br. Glyceria grandis S. Watson Glyceria leptostachya Buckley Glyceria occidentalis (Piper) J.C. Nelson Glyceria striata (Lam.) Hitchc. Graphephorum wolfii (Vasey) Vasey ex Coult. Hainardia cylindrica (Willd.) Greuter Hesperostipa comata (Trin. & Rupr.) Barkworth Heteropogon contortus (L.) P. Beauv. ex Roem. & Schult. Hilaria jamesii (Torr.) Benth. Hilaria mutica (Buckley) Benth. Hilaria rigida (Thurb.) Benth. ex Scribn. Holcus lanatus L. Holcus mollis L. Hordeum arizonicum Covas Hordeum brachyantherum Nevski Hordeum bulbosum L. Hordeum depressum (Scribn. & J.G. Sm.) Rydb. Hordeum intercedens Nevski Hordeum jubatum L. Hordeum marinum Huds. Hordeum murinum L. Hordeum pusillum Nutt. Hordeum vulgare L. Hyparrhenia hirta (L.) Stapf Imperata brevifolia Vasey Jarava brachychaeta (Godr.) Peñailillo Jarava ichu Ruiz & Pav. Jarava plumosa (Spreng.) S.W.L. Jacobs & J. Everett Jarava speciosa (Trin. & Rupr.) Peñailillo Karroochloa purpurea (L. f.) Conert & Türpe Koeleria macrantha (Ledeb.) Schult. Lachnagrostis filiformis (G. Forst.) Trin. Lagurus ovatus L. Lamarckia aurea (L.) Moench

S Y S T E M AT I C S O F C A L I F O R N I A G R A S S E S

17

Leersia oryzoides (L.) Sw. Leptochloa dubia (Kunth) Nees Leptochloa fusca (L.) Kunth Leptochloa panicea (Retz.) Ohwi Leptochloa viscida (Scribn.) Beal Leucopoa kingii (S. Watson) W.A. Weber Leymus cinereus (Scribn. & Merr.) Á. Löve Leymus condensatus (J. Presl) Á. Löve Leymus mollis (Trin.) Pilg. * Leymus pacificus (Gould) D.R. Dewey Leymus salinus (M.E. Jones) Á. Löve Leymus triticoides (Buckley) Pilg. Lolium multiflorum Lam. Lolium perenne L. Lolium rigidum Gaudin Lolium temulentum L. Lycurus setosus (Nutt.) C. Reeder Megathyrsus maxima (Jacq.) B.K. Simon & S.W.L. Jacobs Melica aristata Thurb. ex Bol. Melica bulbosa Geyer ex Porter & Coult. Melica californica Scribn. Melica frutescens Scribn. Melica fugax Bol. Melica geyeri Munro Melica harfordii Bol. Melica imperfecta Trin. Melica spectabilis Scribn. Melica stricta Bol. Melica subulata (Griseb.) Scribn. * Melica torreyana Scribn. Melinis repens (Willd.) Zizka Miscanthus sinensis Andersson Monanthochloe littoralis Engelm. Muhlenbergia andina (Nutt.) Hitchc. Muhlenbergia appressa C.O. Goodd. Muhlenbergia arsenei Hitchc. Muhlenbergia asperifolia (Nees & Meyen ex Trin.) Parodi *Muhlenbergia californica Vasey Muhlenbergia filiformis (Thurb. ex S. Watson) Rydb. Muhlenbergia fragilis Swallen *Muhlenbergia jonesii (Vasey) Hitchc. Muhlenbergia mexicana (L.) Trin. Muhlenbergia microsperma (DC.) Kunth Muhlenbergia minutissima (Steud.) Swallen Muhlenbergia montana (Nutt.) Hitchc. Muhlenbergia pauciflora Buckley Muhlenbergia porteri Scribn. ex Beal Muhlenbergia richardsonis (Trin.) Rydb. Muhlenbergia rigens (Benth.) Hitchc. Muhlenbergia schreberi J.F. Gmel. Muhlenbergia utilis (Torr.) Hitchc. Munroa squarrosa (Nutt.) Torr. Nassella cernua (Stebbins & Love) Barkworth Nassella lepida (Hitchc.) Barkworth Nassella manicata (E. Desv.) Barkworth

18

OVERVIEW

Nassella pulchra (Hitchc.) Barkworth Nassella tenuissima (Trin.) Barkworth Nassella viridula (Trin.) Barkworth *Neostapfia colusana (Burtt Davy) Burtt Davy Orcuttia californica Vasey *Orcuttia inaequalis Hoover *Orcuttia pilosa Hoover *Orcuttia tenuis Hitchc. *Orcuttia viscida (Hoover) Reeder Oryza rufipogon Griff. Oryza sativa L. Panicum alatum Zuloaga & Morrone Panicum antidotale Retz. Panicum capillare L. Panicum dichotomiflorum Michx. Panicum hillmanii Chase Panicum hirticaule J. Presl Panicum miliaceum L. Panicum repens L. Panicum rigidulum Bosc ex Nees Panicum urvilleanum Kunth Parapholis incurva (L.) C.E. Hubb. Parapholis strigosa (Dumort.) C.E. Hubb. Pascopyrum smithii (Rydb.) Barkworth & D.R. Dewey Paspalum dilatatum Poir. Paspalum distichum L. Paspalum notatum Flüggé Paspalum urvillei Steud. Pennisetum clandestinum Hochst. ex Chiov. Pennisetum glaucum (L.) R. Br. Pennisetum latifolium Spreng. Pennisetum macrourum Trin. Pennisetum nervosum (Nees) Trin. Pennisetum purpureum Schumach. Pennisetum setaceum (Forssk.) Chiov. Pennisetum villosum R. Br. ex Fresen. Phalaris angusta Nees ex Trin. Phalaris aquatica L. Phalaris arundinacea L. Phalaris brachystachys Link Phalaris californica Hook. & Arn. Phalaris canariensis L. Phalaris caroliniana Walter Phalaris coerulescens Desf. Phalaris lemmonii Vasey Phalaris minor Retz. Phalaris paradoxa L. Phleum alpinum L. Phleum pratense L. Phragmites australis (Cav.) Steud. Phyllostachys bambusoides Siebold & Zucc. Phyllostachys nigra (Lodd. ex Lindl.) Munro Piptatherum exiguum (Thurb.) Dorn Piptatherum micranthum (Trin. & Rupr.) Barkworth Piptatherum miliaceum (L.) Coss. Piptochaetium setosum (Trin.) Arechav.

Piptochaetium stipoides (Trin. & Rupr.) Hack. ex Arechav. *Pleuropogon californicus (Nees) Benth. ex Vasey *Pleuropogon hooverianus (L.D. Benson) J.T. Howell Pleuropogon refractus (A. Gray) Benth. Poa abbreviata R. Br. Poa annua L. *Poa atropurpurea Scribn. Poa bigelovii Vasey & Scribn. Poa bolanderi Vasey Poa bulbosa L. Poa compressa L. Poa confinis Vasey Poa cusickii Vasey * Poa diaboli Soreng & D.J. Keil Poa douglasii Nees Poa fendleriana (Steud.) Vasey Poa glauca Vahl Poa howellii Vasey & Scribn. Poa infirma Kunth *Poa keckii Soreng *Poa kelloggii Vasey Poa leptocoma Trin. Poa lettermanii Vasey Poa macrantha Vasey * Poa napensis Beetle Poa nemoralis L. Poa palustris L. Poa piperi Hitchc. Poa pratensis L. Poa pringlei Scribn. Poa rhizomata Hitchc. Poa secunda J. Presl *Poa sierrae J.T. Howell *Poa stebbinsii Soreng *Poa tenerrima Scribn. Poa trivialis L. Poa unilateralis Scribn. ex Vasey Poa wheeleri Vasey Podagrostis humilis (Vasey) Björkman Podagrostis thurberiana (Hitchc.) Hultén Polypogon australis Brongn. Polypogon elongatus Kunth Polypogon imberbis (Phil.) Johow Polypogon interruptus Kunth Polypogon maritimus Willd. Polypogon monspeliensis (L.) Desf. Polypogon viridis (Gouan) Breistr. Pseudoroegneria spicata (Pursh) Á. Löve Pseudosasa japonica (Siebold & Zucc. ex Steud.) Makino ex Nakai *Ptilagrostis kingii (Bol.) Barkworth Puccinellia distans ( Jacq.) Parl. *Puccinellia howellii J.I. Davis Puccinellia lemmonii (Vasey) Scribn. Puccinellia maritima (Huds.) Parl. Puccinellia nutkaensis (J. Presl) Fernald & Weath.

Puccinellia nuttalliana (Schult.) Hitchc. Puccinellia parishii Hitchc. Puccinellia pumila (Vasey) Hitchc. Puccinellia simplex Scribn. Rostraria cristata (L.) Tzvelev Rytidosperma biannulare (Zotov) Connor & Edgar Rytidosperma caespitosum (Gaudich.) Connor & Edgar Rytidosperma penicillatum (Labill.) Connor & Edgar Rytidosperma racemosum (R. Br.) Connor & Edgar Rytidosperma richardsonii (Cashmore) Connor & Edgar Saccharum ravennae (L.) L. Schedonnardus paniculatus (Nutt.) Branner & Coville Schedonorus arundinaceus (Schreb.) Dumort. Schedonorus pratensis (Huds.) P. Beauv. Schismus arabicus Nees Schismus barbatus (L.) Thell. Schizachyrium cirratum (Hack.) Wooton & Standl. Schizachyrium scoparium (Michx.) Nash Sclerochloa dura (L.) P. Beauv. Scleropogon brevifolius Phil. Scribneria bolanderi (Thurb.) Hack. Secale cereale L. Setaria faberi R.A.W. Herrm. Setaria italica (L.) P. Beauv. Setaria parviflora (Poir.) Kerguélen Setaria pumila (Poir.) Roem. & Schult. Setaria sphacelata Setaria verticillata (L.) P. Beauv. Setaria viridis (L.) P. Beauv. Sorghum bicolor (L.) Moench Sorghum halepense (L.) Pers. Spartina alterniflora Loisel. Spartina anglica C.E. Hubb. Spartina densiflora Brongn. Spartina foliosa Trin. Spartina gracilis Trin. Spartina patens (Aiton) Muhl. Sphenopholis obtusata (Michx.) Scribn. Sporobolus airoides (Torr.) Torr. Sporobolus contractus Hitchc. Sporobolus creber De Nardi Sporobolus cryptandrus (Torr.) A. Gray Sporobolus flexuosus (Thurb. ex Vasey) Rydb. Sporobolus indicus (L.) R. Br. Sporobolus vaginiflorus (Torr. ex A. Gray) Alph. Wood Sporobolus wrightii Munro ex Scribn. Stenotaphrum secundatum (Walter) Kuntze Stipa capensis Thunb. *Swallenia alexandrae (Swallen) Soderstr. & H.F. Decker Taeniatherum caput-medusae (L.) Nevski Themeda quadrivalvis (L.) Kuntze Thinopyrum intermedium (Host) Barkworth & D.R. Dewey Thinopyrum junceum (L.) Á. Löve Thinopyrum ponticum (Podp.) Barkworth & D.R. Dewey Thinopyrum pycnanthum (Godr.) Barkworth Thysanolaena latifolia (Roxb. ex Hornem.) Honda

S Y S T E M AT I C S O F C A L I F O R N I A G R A S S E S

19

Torreyochloa erecta (Hitchc.) G.L. Church Torreyochloa pallida (Torr.) G.L. Church Tribolium obliterum (Hemsl.) Renvoize Tridens flavus (L.) Hitchc. Tridens muticus (Torr.) Nash Trisetum cernuum Trin. Trisetum flavescens (L.) P. Beauv. Trisetum spicatum (L.) K. Richt. Triticum aestivum L. *Tuctoria greenei (Vasey) Reeder *Tuctoria mucronata (Crampton) Reeder

20

OVERVIEW

Urochloa arizonica (Scribn. & Merr.) Morrone & Zuloaga Urochloa texana (Buckley) R.D. Webster Vahlodea atropurpurea (Wahlenb.) Fr. ex Hartm. Ventenata dubia (Leers) Coss. Vulpia bromoides (L.) Gray Vulpia microstachys (Nutt.) Munro Vulpia myuros (L.) C.C. Gmel. Vulpia octoflora (Walter) Rydb. Zea mays L. Zizania palustris L. Zoysia japonica Steud.

THREE

Community Classification and Nomenclature TO D D K E E LE R-WO L F, J U LI E M. EVE N S, AYZ I K I. S O LO M E S H C H, V. L. H O LLAN D, AN D M I C HAE L G. BAR B O U R

Grass-dominated vegetation covers approximately one-fourth of California’s area, and it is well known that virtually all of it has been significantly modified by the invasion of naturalized annual and perennial grasses and forbs. Less commonly understood is the fact that this type conversion resulted in very few extinctions. Bartolome et al. (in press) concluded that—although local extirpation, reduced abundance, and range retraction characterize the status of the once-dominant native taxa—only a few species have retreated to the point of complete extinction. Only four of the 29 taxa presumed to be extinct throughout all of California could have once been components of the valley grassland (CNPS 2001). Native species remain rich in number, even if individually their cover is low. In some areas, their cumulative cover is greater than that of the exotics. This is an exciting time for grassland classification: The number of defined and named community types is rapidly climbing, and the scale at which community types can be recognized is becoming finer and more localized. On the other hand, grassland classification is still unfinished: Defined and mapped types are scattered polygons within a much larger blank matrix that remains undescribed. It is as though we are looking at a jigsaw puzzle of California with at least half the pieces missing. Furthermore, the classification level for existing pieces is not uniform. Some associations have not yet been aggregated into alliances, and some alliances have not yet been subdivided into associations. Following a number of authors, we recognize five major grassland types within the state’s boundaries (Figure 3.1). Their common names are valley/south coastal grassland, north coastal grassland, serpentine grassland, cold desert grassland, and warm-desert grassland. Many synonyms exist for these types. They occupy such large areas, and are so floristically and environmentally different, that each may be the equivalent of classes or orders, in the sense of traditional BraunBlanquet phytosociological classification (see, for example,

Mueller-Dombois and Ellenberg 1974). In addition, each type exists in a precontact and a postcontact condition. Four of them are regional in distribution, but serpentine grassland is azonal and not limited to a single geographic region. Before we tour these major types, we offer several definitions. These definitions are largely our own creation, because the literature is so inconsistent in their treatment.

Vegetation and Community Type Definitions Grassland is vegetation that belongs to the Herbaceous Formation Class, defined by Grossman et al. (1998) as “Herbs (graminoids, forbs, and ferns) dominant (generally forming at least 25% cover; trees, shrubs, and dwarf shrubs generally less than 25% cover).” For any given California grassland community, three to four grass species typically constitute the majority of the aboveground biomass, but forb species richness can be four times that of the grasses and grass-allies (Sims and Risser 2000). More commonly, grassland is described as lacking woody plants. Many grassland taxa extend as an understory beneath foothill savannas and woodlands and beneath riparian woodland. However, these formations differ from grassland by having a distinctive physiognomy, and they will not be discussed in this chapter. Open savannas, however, with ⬍ 10% cover by tree canopies, appear to have no significant effect on grassland communities (Sawyer and Keeler-Wolf 1995; Sawyer et al. 2007), so readers should be aware that information in this chapter also applies to sparsely treed grasslands. Climatically, North American grassland communities are very diverse and have few factors in common (Barbour and Christensen 1993). In California, they occupy habitats where the annual precipitation ranges from a low of 120 mm, in the southwestern part of the San Joaquin Valley, to 2000 mm, in the Coast Ranges of northwestern California (Bartolome et al. in press). The annual precipitation-to-evaporation ratio (P:E) is

21

F I G U R E 3.1. Modern distribution of major grasslands, including areas type-converted from grassland to agricultural and urban uses. Solid lines separate the state into four grassland regions: north coastal (NC), Valley and south coastal (CV), cold desert (CD), and warm desert (WD). The boundaries of north coastal grassland are based on Ford and Hayes (in press); boundaries for the other three major grasslands are based on Kuchler’s (1977) map of potential natural vegetation and on geographic subdivisions recognized by the Jepson Manual (Hickman 1993). The boundaries are not intended to indicate complete dominance by grassland within them, only to indicate the limits of each grassland type. Patches of grassland do occur in other parts of the state, but they are not mapped here.

from 0.7 to 0.3, soils are deep and fine-textured, and the firereturn interval is shorter than a decade (Barbour et al. 1999). Where the P:E ratio ⬎1.0, as in cismontane elevations above 1200 m, herb-dominated vegetation can be present, but only where local hydrology prevents the growth of woody vegetation, such as in meadows. We do not include meadows in this chapter, because they occur in a climate more appropriate for forest vegetation or they occupy seasonally saturated soils (see, for example, Brown 1982). We also omit wetlands: tidal marshes, freshwater marshes, and vernal pools. Prairie is defined in this chapter as a synonym for grassland. The term has been used by some ecologists to denote subhumid grasslands that have a P:E ratio close to 1.0 and are dominated by tall perennial grasses. An example is the tall grass prairie in the central portion of North America, which receives

22

OVERVIEW

500 – 1000 mm precipitation annually (see, for example, Bailey 1995). However, a climatic distinction between prairie and grassland is just as often ignored in the literature, with the term prairie applied almost randomly. Thus, we choose to treat prairie and grassland as complete synonyms, and for that reason the vegetation sometimes called “north coastal prairie” is here referred to as “north coastal grassland.” Steppe is as problematic a term as prairie. In this chapter we define steppe as a complete synonym for grassland, even though others have attempted to define it as different from grassland. Bailey (1995), for example, differentiates steppes from grasslands as being more open, occupying landscapes colder and drier (more temperate and more continental), and dominated by short-stature grasses associated with shrubs and trees as scattered individuals or in clumps. The ratio of

herb-to-woody cover in typical steppe has not been widely defined. West and Young (2000) and Daubenmire (1970, 1975) describe relictual sagebrush steppe in the intermountain region as having 20–80% cover by shrubs and 50–80% cover by grasses and forbs, attaining a combined, total plant cover of 60 –200%. Prominent growth forms in that steppe (in terms of relative floristic richness) are hemicryptophytic herbs and a soil crust composed of mosses, lichens, and algae. The term steppe has been variously applied to bunchgrass vegetation lacking any woody component (for example, to the valley grassland of California), to bunchgrass vegetation with shrubs regularly present (equally often called a “shrubsteppe” or “scrub grassland”), and to bunchgrass vegetation with trees regularly present (also called a “tree-steppe”). The basic classification unit, and often the unit most local in its range, is the association. As initially defined at the International Botanical Congress of 1910, an association (group of stands) is a plant community of definite floristic composition that presents a uniform physiognomy and that grows in uniform habitat conditions (Flahault and Schroeter 1910). Associations are abstractions, and they are represented in reality by concrete examples, called stands. The Ecological Society of America’s Vegetation Classification Panel (Jennings et al. in press) adopted a more American-biased definition of association as being “a vegetation classification unit defined on the basis of a characteristic range of species composition, diagnostic species occurrence, habitat conditions, and physiognomy,” a definition that clearly states that stands within a particular association are expected to be somewhat different from each other. Box 3.1 provides an example of the kind of vegetation analysis that leads to definition of an association. Whereas vegetation types are named after habitats and plant growth forms, associations are named after one or more diagnostic species. Diagnostic species are those taxa most often found in nature as being part of just one particular association. Diagnostic species are not necessarily the dominant species. Those species in the name that share the same stratum are separated by a dash, as for example, the Nassella pulchra–Melica californica association, once common in the Central Valley. Those that occur in different strata are separated by a forward slash, as, for example, the Calamagrostis nutkaensis/Baccharis pilularis association, common along the north coast. The diagnostic species in the tallest stratum is listed first, followed by a slash and then the diagnostic species from a different stratum. As of 2005 (NatureServe 2006), approximately 6,200 associations within the United States had been defined and named. We have yet to describe a large enough number of associations in California to be able to predict, with any rigor, how many associations exist within the state’s boundaries. The usual guesses by vegetation scientists most familiar with the state’s landscape place the number at more than 2,000. Given the fact that grasslands in California occupy such a large portion of the state’s area and that they occur in such a wide range of habitats and floristic regions, we estimate that a statewide inventory would include more than 400 grassland association types.

An alliance is a group of floristically related associations that collectively occupy a larger range than does any single association. Its makeup is broader than any one association (that is, more floristically, structurally, and ecological variable; Jennings et al. in press). Nationally, there are 1,900 alliances, meaning that each alliance contains (on average) about three associations. In California, Sawyer et al. (2007) have so far recognized some 400 alliances, and we predict that more than 50 grassland alliance types will ultimately be recognized when vegetation classification is closer to completion than it is at present.

Valley/South Coastal Grassland Several names have been given to upland herbaceous vegetation in California’s Central Valley: Valley needlegrass grassland (with Nassella pulchra and N. cernua), California prairie, and California annual grassland, among others (Heady et al. 1988 and 1992; Holland 1986; Kuchler 1977; Figure 3.2). “Valley”

A

B F I G U R E 3.2. Nassella pulchra dominating on a serpentine clay loam soil on slopes above the Santa Clara Valley, south of San Jose. (A) Early spring aspect, when the needle grass clumps are easily seen. (B) Late spring aspect, when the characteristic long-awned florets wave in the breeze, and associated species have grown to the height of the bunch grass. Photographs courtesy of Julie Evens.

C O M M U N I T Y C L A S S I F I C AT I O N , N O M E N C L AT U R E

23

BOX 3.1 THE SPOROBOLUS AIROIDES ASSOCIATION

Sporobolus airoides, although restricted in its distribution today, was likely an important part of the valley floor grasslands of the Great Central Valley prior to the onset of agricultural activities. Some of the best relictual stands of S. airoides are in Great Valley Grasslands State Park, a 1144 ha preserve in the northern part of the San Joaquin Valley. One of the authors (Solomeshch) sampled vegetation with 54 plots, each 100 square meters in area. A list of all taxa present and their percent cover was tabulated for each plot. The information was treated to Twinspan analysis (Hill 1979) and then refined by the Braun-Blanquet technique, as described by Mueller-Dombois and Ellenberg (1974). Detrended correspondence analysis (DCA) ordination using Canoco statistical software (Ter Braak and Smilauer 1998) shows the high degree of distinctiveness of each community type (Figure 3.3). Results were used to reconstruct the floristic composition of natural communities before they had been invaded by exotics. Five community types were recognized (Table 3.1), and native perennial grasses and forbs play important roles in four of these types (Solomeshch and Barbour 2006). The Sporobolus airoides association is dominated by tussocks of S. airoides. Associated perennials include Cressa truxillensis, Distichlis spicata, and Frankenia salina and such common and abundant annual exotics as Bromus hordeaceus, B. diandrus, Hordeum marinum ssp. gussoneanum, Vulpia bromoides, and V. myuros. Average plant cover is ⬎95%, and average mature vegetation height is 30–40 centimeters. This is the most floristically rich grassland community in the park, averaging 17 taxa per 100 square meters. Native species include perennial bunchgrass, perennial rhizomatous, and annual grasses (S. airoides, Distichlis spicata, Hordeum depressum, Phalaris lemmonii, Vulpia microstachys), perennial and annual forbs (Achyrachaena mollis, Atriplex fruticulosa, Astragalus tener var. tener, Carex praegracilis, Crassula connata, Dichelostemma capitatum, Hemizonia pungens, Grindelia camporum var. camporum, Epilobium brachycarpum, Lasthenia californica, L. platycarpa, L. glabrata ssp. coulteri, Lotus wrangelianus, Myosurus minimus, Microseris campestris, M. douglasii ssp. douglasii, Triteleia hyacinthine, Trifolium depauperatum var. depauperatum), and shrubs and subshrubs (Frankenia salina, Isocoma acradenia var. bracteosa).

-0.5

2.5

The other four grassland associations in the park are Bromus diandrus, Hordeum murinum ssp. gussoneanum, Distichlis spicata, and Leymus triticoides. The first represents the understory of a Quercus lobata woodland or savanna floodplain, and the fourth the understory of a denser, mixed riparian forest prior to the diversion of flood waters behind dams and artificial levees. The remaining two ( Hordeum and Distichlis) occupy habitats with elevated soil salinity.

-1.0

4.0

FIGURE 3.3. Ordination of grassland plots by Detrended

Correspondence Analysis. The distance between the plots in the diagram reflects the dissimilarity of their floristic composition, measured by their chi-square distance. The scale marks are in multiples of standard deviations. Plots of four grassland community types at Great Valley Grassland State Park in the San Joaquin Valley form tight clusters, reflecting a high floristic similarity among them. The community types are (䉱) Hordeum marinum ssp. gussonianum; (䉫) Sporobolus airoides; (䊏) Leymus triticoides; (䊊) Bromus diandrus. One B. diandrus plot was an outlier because it had experienced an accidental burn 1 year before sampling. From Solomeshch and Barbour (2006).

TA B L E 3.1 Four Grassland Community Types within Great Valley Grassland State Park

Bromus diandrus Grassland (11 Plots)

Sporobolus airoides Grassland (15 Plots)

100 73

100 7

83 8

50 —

8 8

Sporobolus airoides Grassland Sporobolus airoides Bromus madritensis Vulpia myuros Hemizonia pungens Erodium botrys Dichelostemma capitatu

— 46 36 27 27 9

100 87 87 73 73 47

— 8 33 33 8 8

50 — 50 50 — —

— 15 8 — — —

Hordeum marinum gussonianum Grassland Hordeum marinum ssp. gussonianum Vulpia bromoides Frankenia salina Cressa truxillensis

18 9 36 —

100 87 93 73

100 92 100 92

100 — 50 —

23 — 62 —

Distichlis spicata Grassland Distichlis spicata

82

80

100

100

46

Leymus triticoides Grassland Leymus triticoides Grindelia camporum Epilobium brachycarpum Malvella leprosa

27 — 9 —

— 13 13 —

— 8 — —

— — 50 —

100 85 85 77

Other Native Species Amsinckia menziesii Atriplex species Juncus balticus Croton setigerus Helianthus annuus

46 — 18 9 9

7 7 — — —

8 — — — —

— 50 — — —

46 23 23 8 15

Other Exotic Species Bromus hordeaceus Lactuca serriola Senecio vulgaris Rumex crispus Erodium cicutarium Stellaria media Sonchus asper Centaurea solstitialis Carduus pycnocephalus Brassica nigra Silybum marianum

91 73 18 9 36 9 18 — 27 46 18

100 80 53 13 7 13 27 53 27 — —

100 83 17 17 — — — 50 8 — —

100 50 100 50 — 50 50 — — — —

69 62 39 31 54 — 46 54 23 46 15

Bromus diandrus Grassland Bromus diandrus Hordeum murinum ssp. leporinum

Hordeum marinum Grassland (12 Plots)

Distichlis spicata Grassland (2 Plots)

Leymus triticoides Grassland (13 Plots)

NOTE : Each column represents sample plots from the same community type. Rows across show how species were distributed through locations. Numbers are the percent of plots in which the species was found. Boxed numbers highlight species that are associated with a particular community type. Bolded numbers indicate species with ⬎5% cover in more than half the plots containing them. Only the more common species are shown.

is here used in a wider sense, because this grassland extends well upslope to 700⫹ m elevation in foothills and to the coast south of Santa Barbara. Collectively, introduced annual species have attained dominance throughout this type since the early nineteenth century (see D’Antonio et al., Chapter 6), and their permanence was recognized when rangeland ecologist Harold Heady (1988) called them “the new natives.” Barry (1972) estimated that ⬍1% of the precontact Nassella grassland remains. In spite of the dramatic reduction in abundance of native species, we may be able to infer past community composition by examining present-day stands (Box 3.2). Climate is a major factor in determining plant communities within this type. Bromus madritensis ssp. rubens and Erodium cicutarium consistently dominate the semiarid south, whereas Avena barbata, Bromus hordeaceus, and Cynosurus echinatus are prevalent in the north. The most recent invasives are Aegilops triuncialis, Brachypodium distachyon, Phalaris aquatica, and Taeniatherum caput-medusae (Bartolome et al. in press). Another community control is the timing and amount of rainfall in a given year—that is, the annual variability in precipitation (e.g., Heady 1956 and 1958, Pitt and Heady 1978, Hobbs and Mooney 1991, Seabloom et al. 2003a). In years with both fall and winter rains, a plethora of native forbs and some native grasses emerge, including annuals and perennials in the genera Agoseris, Brodiaea, Daucus, Lotus, Navarretia, Trifolium, Triphysaria, Triteleia, and several others. By late spring in such wet years, non-native grasses tend to dominate, especially on fertile, well-drained, deep soils. The most abundant genera include Avena, Bromus, Hordeum, Lolium, and Vulpia. Relatively dry fall or winter periods can give rise to “grass-clover-filaree” years, where Erodium and Trifolium species are especially abundant (see Reever Morghan et al., Chapter 7). Grassland immediately adjacent to wetland vegetation may be enriched in native forbs and grasses, particularly annual species. The shallow edges of vernal pools, for example, feature a matrix of introduced grasses and forbs together with such vernal pool taxa as Castilleja attenuata, Deschampsia danthonioides, Juncus bufonius, Lasthenia californica, Layia fremontii, Triphysaria eriantha, and species of Blennosperma, Limnanthes, Plagiobothrys, Sidalcea, and Trifolium (Barbour et al. 2003). Despite the widespread distribution of vernal pools in all but the wettest and driest parts of Californian grasslands, the shallow (and usually narrow) pool–grassland ecotone is best considered as part of the vernal pool ecosystem, rather than the grassland, and thus will not be discussed in detail in this treatment. Species abundance shifts within any given year as a result of phenological attributes of plants. Winter annuals (Avena, Bromus, Hordeum, and Vulpia) dominate in winter and early spring. In early to late summer, another set of annuals may appear in the mix, including Aristida oligantha, Centaurea solstitialis, Eremocarpus setigerus, Gastridium ventricosum, and aromatic members of the Asteraceae (Calycadenia, Hemizonia, Holocarpha, and Madia species). From our own anecdotal observations, drier years favor native forbs over exotic grasses. This is especially true when rain only comes early, allowing

26

OVERVIEW

non-natives and native alike to germinate, followed by abnormally low precipitation in winter and spring. Many exotics fail to reach maturity under these conditions. Type of disturbance and soil parent material also add to the constraints on community distribution and composition in this grassland habitat (McNaughton 1968; Hobbs and Mooney 1991; Moloney and Levin 1996; Safford and Harrison 2001, 2004; Harrison et al. 2003; Seabloom et al. 2003a; Weiss 1999; Schiffman, Chapter 15; Cushman Chapter 16; Jackson and Bartolome, Chapter 17). For example, gopher disturbance may increase the cover of Bromus hordeaceus in stands on serpentine, and nitrogen deposition from air pollutants may increase the cover of Lolium multiflorum. Serpentine soils and vertisols have higher native species richness and a higher ratio of native to exotic plant cover, compared to the vegetation on other soils, apparently because few exotics can tolerate the stresses associated with such substrata (see Harrison and Viers, Chapter 12). Disturbed sites show lower species richness and a lower ratio of native to exotic cover than relatively undisturbed sites do. Highly disturbed sites near roads are dominated by only one to a few exotic species (Gelbard and Harrison 2003; Keeley 2001). The community types currently defined for valley grassland are typically split into perennial or annual types. Generally, researchers have used 10% relative cover of perennial grasses to be the deciding factor between these two types, perennial communities requiring ⬎10% combined relative cover of all bunchgrasses present. However, we caution that this is by no means a “magic” number, and local site conditions may vary substantially based on cumulative diversity and constancy of native versus non-native species composition across related stands. Nassella pulchra (purple needle grass) has been considered the flagship species of the bunchgrasses (Clements 1934; Kuchler 1964; Heady 1988; Holland 1986), but its past role and extent in the Central Valley and the central and southern coast regions of California are currently being debated (Wester 1981; Brown 1982; Hamilton 1997a; Stromberg et al. 2001; Holstein 2001). Some ecologists have concluded that purple needle grass was not the overall dominant but rather was the most opportunistic, r-selected bunchgrass, which explains why it is the most common and widespread bunchgrass in today’s disturbed grassland. Purple needle grass stands are commonly found where the soils are deep and clay-rich, but they also occur on less fertile serpentine soils (McNaughton 1968; Hamilton 1997a; Gelbard and Harrison 2003; Evens and San 2004). Other bunchgrasses that probably were associated with purple needle grass include Aristida hamulosa, Elymus glaucus, Koeleria cristata, Melica californica, M. imperfecta, Poa secunda, and, in the San Joaquin Valley, Sporobolus airoides (Barry 1972). Modern grasslands in the Valley now include a significant amount of cover from such non-native annual grasses as Avena barbata, A. fatua, Bromus hordeaceus, Hordeum murinum, and Lolium multiflorum. Many associations have been defined within the Nassella pulchra alliance. They range from the north coastal types

BOX 3.2 DETECTION OF LOCAL COMMUNITY TYPES IN THE SACRAMENTO VALLEY

It may be possible to infer the past extent and composition of grassland community types in California by examining stands for the presence of now-muted natives. We propose aboriginal grassland associations by analyzing the patterns of co-occurrence of all taxa in sample plots. Plot data also show that particular assemblages of dominant exotics are associated with particular assemblages of background natives. The native species remaining in the modern valley grassland are numerous and varied enough to distinguish local communities. Grasslands in several locations within a radius of 40 km of Sacramento were sampled in the spring of 2002 (by Solomeshch). A total of 55 plots, each 100 square meters in area, contained a total of 175 taxa of vascular plants, 122 of which (70%) were natives. All plots are dominated by the exotic species Bromus hordeaceus, Erodium botrys, Hypochaeris glabra, Lolium perenne ssp. multiflorum, and Vulpia bromoides, among others. They are also similar in the presence of some native species, such as Juncus bufonius, Lupinus bicolor, Triphysaria eriantha, and Triteleia hyacinthina. But plots differed from each other in the presence of species more restricted to one or two locations and a single habitat type. Each location had a unique assemblage of species capable of being used to define distinct communities. The five communities are briefly described in the following paragraphs. The Hemizonia congesta ssp. luzulifolia association has the highest cover and number of native species. It differs from the other communities in the dominance of Hemizonia and the high diversity and often abundance of Casielleja attenuata, Eryngium aristulatum, Hesperevax caulescens, Lasthenia glabrata ssp. glabrata, Lotus wrangelianus, Pogogyne douglasii, Trifolium willdenovii, and T. bifidum, among other natives. The Lasthenia platycarpha association is characterized by the high constancy of Brodiaea coronaria, Cicendia quadrangularis, Lasthenia californica, Layia chrysanthemoides, and Pogogyne ziziphoroides, among others. Presence of the halophytes Frankenia salina, Lasthenia platycarpha, Lepidium latipes var latipes, and Plantago coronopus indicate saline soil. These first two associations are floristically richer than the other three. On average, there are 24 – 25 native species per plot, two to three times the number of exotics. The association described below is at the other extreme, with only 10 species per plot and with the lowest cover ratio of natives to exotics (0.13). The Trifolium microdon–Nassella pulchra–Distichlis spicata association is the floristically poorest and is further characterized by high abundance of Distichlis spicata, high constancy of N. pulchra, and the presence of Leymus triticoides, Trifolium microdon, and Viola pedunculata. The number and cover of perennial natives are highest in this community, but at the same time, exotics (Bromus diandrus, Taeniatherum caput-medusae, Hordeum murinum ssp. leporinum, and others) have a much higher abundance than in any other community, so the native-to-exotic cover ratio is low. The Triphysaria eriantha– Nassella pulchra– Distichlis spicata association resembles the Lasthenia platycarpha association, except that halophytes are absent. N. pulchra is associated with a different suite of species, including Isoetes orcuttii, Eryngium vaseyi, and Achyrachaena mollis. Finally, the Holocarpha virgata association is characterized by the associates Castilleja attenuata, Centaurium venustum, Clarkia purpurea, Daucus pusillus, Holocarpha virgata, Navarretia intertexta, Triteleia hyacinthina, and Navarretia intertexta, among others. H. virgata is the dominant among the natives (10 – 15% cover). Solomeshch concluded that the native grassland as a vegetation type may be extinct, but its component species largely remain, and their presence can be used to define and distinguish associations. He presumes that native species have not significantly changed their ecological requirements since European-American contact, meaning that if a native taxon was confined to a particular soil series prior to contact, it will still show that pattern of occurrence today. This approach to reconstructing the landscape reveals that modern community diversity is high, even in a relatively small area. It appears that it would be possible to build a classification of California grassland associations based on the presence of native species if Solomeshch’s approach is extended to the rest of California’s grasslands.

N. pulchra/Baccharis piluaris and N. pulchra–Melica californica (NatureServe et al. 2003; Evens et al. 2006) to the serpentine types in the middle Coast Ranges N. pulchra – Astragalus gambelianus–Lepidium nitidum and N. pulcha–Calystgia collina (Evens and San 2004), to N. pulchra – Erodium spp.–Avena barbata in western Riverside County (Klein and Evens 2005). The earliest botanical descriptions of the grassland now occupied by Los Angeles clearly fit the N. pulchra alliance and ecosystem of the Central Valley (Schiffman 2005). Additional communities of perennials are organized around the loosely tufted Elymus glaucus, typically adjacent to oak woodlands on sandy loam soils, especially in the foothills of the Coast Ranges, the strongly tufted meadow barley (Hordeum brachyantherum) near marshes and seeps on heavier soils, and the rhizomatous Leymus triticoides beneath valley oaks on poorly drained floodplain soils and around seeps and drainages (associated with Carex barbarae, C. praegracilis, Juncus spp., and Muhlenbergia rigens; Hamilton 1997a, Shuford and Timossi 1989). The Leymus triticoides alliance is thought to have played a major role in valley grasslands on heavy soils and in ecotones between riparian and marsh habitats (Holstein 2001). Other communities of perennials have Distichlis spicata, Hordeum depressum (a native annual), and Sporobolus airoides on alkaline soils in the San Joaquin Valley, in southern California (Hamilton 1997a), and north to Contra Costa County (Holland 1986; Hamilton 1997a); Aristida purpurea on granitic, sandy soils in southern California (Evens and San 2005); and Elymus multisetus and Melica torreyana on rocky or clayey serpentine soils in the Central Coast Range (Evens and San 2004; Evens et al. 2006). The southernmost portion of the San Joaquin valley, stretching from Kettleman City in the northwest to Bakersfield in the southeast and including the Carrizo Plain, is widely understood to have been covered by a saltbush scrub, with patches of alkali grassland, prior to overgrazing and conversion to irrigated agriculture in the early twentieth century. As mapped and described by Kuchler (1977), it was a semiarid desert scrub dominated by Atriplex polycarpa associated with Ephedra californica, Ericameria linearifolia, Haplopappus racemosus, Isocoma acradenia, Lycium andersonii, and L. cooperi. Where soil salinity is high, A. polycarpa was associated with Allenrolfea occidentalis, Atriplex spinifera, Kochia californica, Salicornia suberminalis, Sarcobatus vermiculatus, and Suaeda moquinii. At higher elevations and on less saline soils, A. polycarpa grew with Artemisia tridentata. Perennial grasses were dominant in small stands with low species richness; they included the rhizomatous Distichlis spicata and the bunchgrass Sporobolus airoides. Today, overgrazing has removed the more palatable shrubs (especially A. polycarpa) and the bunchgrasses; such invasive annual grasses as Bromus madritensis ssp. rubens, Schismus arabicus, and S. barbatus have taken their place. Another arid pocket within the southern San Joaquin Valley is known as the Carrizo Plain, west of Taft and separated from the Central Valley by the Temblor Range. Its climate is arid but otherwise different from that of desert grasslands in northeastern or southeastern California; as a

28

OVERVIEW

result, the vegetation is an arid version of Central Valley grassland, with Lasthenia californica, Pectocarya penicillata, Poa secunda, and Vulpia microstachys particularly abundant, and various rare plants also present (Kimball and Schiffman 2003).

Serpentine Grassland As generalized in the chapter by Harrison and Viers of this volume (Chapter 12), grassland on serpentine outcrops throughout California is similar to Valley grassland. Most serpentine sites support a forbland dominated by natives, with ⬍10% cover by grasses. On grass-dominated sites, the most common native grasses are Elymus elymoides, Melica californica, M. imperfecta, Nassella pulchra, Poa secunda, and Vulpia microstachys (see Figure 3.2). Exotic grasses may constitute ⬎ 50% relative cover. Serpentine forblands most commonly feature native species in the genera Calochortus, Calycadenia, Clarkia, Eschscholzia, Gilia, Hesperolinon, Lasthenia, Layia, Lessingia, Linanthus, Lotus, Madia, Navarretia, Stellaria, Streptanthus, and Trifolium. The first syntaxonomy of serpentine annual grasslands in northern California was proposed by Pilar Rodriguez-Rojo (Rodriguez-Rojo et al. 2001a,b), based on her dissertation research of 36 localities. Three-fourths of those localities were in the North Coast Ranges, the rest in Sierran foothills. She classified the vegetation into two classes, one of which was new (Loto wrangeliani–Vulpietea microstachys). The other was characterized by weedy species from Europe (Stellarietea mediae) and thus had already been recognized in the European literature. The new class was subdivided into one new order, three new alliances, and five new associations. Characteristic native species for the new class are Agoseris heterophylla, Astragalus gambelianus, Calandrinia ciliata, Castilleja densiflora, Centaurium muehlenbergii, C. trichanthum, Cryptantha microstachys, Daucus pusillus, Epilobium minutum, Eriogonum luteolum, Filago californica, Gilia angelensis, G. capitata, Hemizonia fitchii, Layia platyglossa, Lessingia leptoclada, L. nemaclada, Linanthus micranthus, Lotus unifoliatus, L. wrangelianus, Madia exigua, M. subspicata, Micropus californicus, Microseris lindleyi, Minuartia californica, Navarretia intertexta, Plantago erecta, Rigiopappus leptocladus, Trifolium depauperatum, T. microdon, T. microcephalum, and Vulpia microstachys. These species are not, however, endemic to serpentine. The three different alliances corresponded with major habitat differences: open coastal foothills, wooded coastal foothills, and open Sierran foothills. Additional alliances have been proposed for serpentine grasslands in the Central and North Coast Ranges and the Sierra Nevada foothills, including some organized around Calycadenia spp., Lasthenia californica, Plantago erecta, and Vulpia microstachys (Evens and San 2004; Evens et al. 2006). Although serpentine soils are largely seen as refugia for native plants, nearness of roads or urbanized areas can have a significant negative effect on the diversity and abundance of native taxa (Gelbard and Harrison 2003; Safford and Harrison 2001).

F I G U R E 3.4. A stand representative of the Calamagrostis nutkaensis alliance, on ocean-facing bluffs near the Point Reyes lighthouse. Photograph courtesy of Todd Keeler-Wolf.

North Coastal Grassland Along the north coast of California, moderate to high winter rainfall and summer fog produce a maritime variant of Mediterranean-type climate, within which occurs a grassland floristically different from that of the Central Valley. According to a study of coastal grasses by Corbin et al. (2005), 28 – 66% of soil moisture taken up by roots originates as fog drip, softening the stress of summer drought and possibly favoring perennials. Called variously coastal prairie, fescue–oatgrass (Festuca–Danthonia), or coastal terrace prairie (Burcham 1957; Munz 1959; Kuchler 1964; Heady et al. 1988; Holland 1986; Ford and Hayes in press), it has strong affinities to the grasslands of central and eastern Oregon and the Palouse prairie of eastern Washington. Species richness and the amount of cover still provided by natives are higher along the coast than in the Central Valley (Stromberg et al. 2001). Ranchers as early as the 1820s also recognized that forage productivity was higher than in the Central Valley. The northern limit of the north coastal grassland has generally been accepted as the California – Oregon border, although elements of this grassland (e.g., Deschampsia caespitosa alliance) certainly extend much further north. Its southern limit depends upon the investigator. Kuchler (1964), among others, mapped its southern limit as extending only to the San Francisco Bay area, but others contend that it extends thoughout the central coast to San Luis Obispo County (Barry 1972; Elliot and Wehausen 1974; McBride 1974; Stromberg et al. 2001). Some range ecologists claim that fully a quarter of this grassland’s precontact area of 355,600 hectares has been urbanized, making it the largest

urbanized fraction of any major plant community type in the United States. We choose to recognize three alliances of the immediate coast: two alliances of drier and more interior habitats, and a southern alliance from Sonoma County to San Luis Obispo County.

Alliances along the Immediate Coast The Deschampsia cespitosa ssp. holciformis (tufted hairgrass) alliance extends from brackish tidelands to coastal bluffs and terraces. The soils are typically poorly drained, remaining moist long into the growing season. Two associations with this subspecies have been defined at Point Reyes National Seashore (NatureServe et al. 2003) that likely extend beyond the park’s borders: D. cespitosa–Horkelia marinensis on sandy loam and D. cespitosa–Danthonia californica on poorly drained loamy terraces. Farther north, in Humboldt Bay, a third D. cespitosa association near salt marshes includes Distichlis spicata, Potentilla anserina, and the exotics Agrostis stolonifera, Cirsium vulgare, and Holcus lanatus (Pickart 2006). A fourth association, in the Central Valley’s delta area, includes Liliaopsis masonii, Senecio hydrophiloides, Triglochin striatum, and other brackish wetland species. This association often grows on peat at the edges of sloughs and channels adjacent to stands of Schoenoplectus californicus. Stands dominated by Deschampsia cespitosa ssp. cespitosa occur south along the immediate coast to San Luis Obispo County (Stromberg et al. 2001). These appear similar to D. cespitosa var. holciformis stands to the north. Another alliance typical of the immediate coast is that of Calamagrostis nutkaensis (Figure 3.4). Its distribution ranges

C O M M U N I T Y C L A S S I F I C AT I O N , N O M E N C L AT U R E

29

from British Columbia to San Francisco Bay, and its southernmost occurrence is on the foggy upper slopes of San Bruno Mountain (McClintock et al. 1990). C. nutkaensis is a large, tufted grass that has relatively high moisture requirements. Because of its size, it competes well with Baccharis pilularis, can occur within northern coastal scrub vegetation, and is tall enough to create its own fog drip in summer (Corbin et al. 2005). At Point Reyes National Seashore, this alliance is represented by two associations: C. nutkaensis/Baccharis pilularis in upland situations and C. nutkaensis–Carex–Juncus on bottomlands and lower slopes. Festuca rubra is characteristic of a third alliance of the immediate coast. Of all the coastal alliances, it has suffered the greatest degree of historic loss because of agriculture, exotics, and development. The few stands that remain are known from Marin, Humboldt, and Sonoma Counties (NatureServe et al. 2003) occur on somewhat drier and better-drained sites than those with Deschampsia cespitosa or Calamagrostis nutkaensis. Species richness is high, with many forbs and grasses, including Achillea millefolium, Artemisia suksdorfii, Avena barbata, Bromus carinatus, Calamagrostis nutkaensis, Camassia leichtlinii, Danthonia californica, Elymus glaucus, Erigeron glaucus, Eriophyllum lanatum, Festuca roemeri, Gentiana affinis, Koeleria macrantha, Nassella pulchra, Solidago canadensis, and Viola adunca. Stands of Festuca rubra also occur on stabilized sand dunes mixed with coastal dune species such as Ambrosia chamissions and Abronia villosa at sites such as Stone Lagoon, Humboldt County. The seral status of many grasslands along the immediate coast is tied to grazing and fire disturbances and to the history of cultivation (Elliot and Wehausen 1974; McBride 1974; Keeley 2002; NatureServe et al. 2003). Many stands are clearly being colonized by Baccharis pilularis and other coastal scrub elements. The removal of grazing has coincided with a shift toward exotic perennial grasses such as Anthoxanthum odoratum and Holcus lanatus at Sea Ranch in Sonoma County (Foin and Hektner 1986; Peart 1989c). Exotic perennial grasses, including Holcus lanatus, Festuca arundinacea, Phalaris aquatalis, and Dactylis glomerata, are also abundant in other Sonoma County and Marin County grasslands. Stromberg and Griffin (1996) demonstrated that coastal grassland in Monterey County, which had at one time been plowed and cultivated, was rich in exotics and was only (after 60 years) being slowly colonized by natives. The old field community was dominated by Avena fatua, Bromus diandrus, Hypochaeris glabra, Eremocarpus setigerus, and species of Amsinckia, Erodium, and Vulpia. Nassella pulchra and Poa secunda were abundant in relictual sites nearby that had never been cultivated. Other coastal grasslands — and not just those that have been cultivated—have been displaced by what Sawyer and Keeler-Wolf (1995) call “non-native perennial grassland” (see also Jimerson et al. 2000; NatureServe et al. 2003; Pickart 2006; and Ford and Hayes in press). The following examples of non-native grassland associations occur along the moist, immediate coast: Agrostis stolonifera, Aira caryophyllea, and Anthoxanthum odoratum. Common associates, including

30

OVERVIEW

some aggressive species, are Achillea millefolium, Briza maxima, Bromus diandrus, B. hordeaceus, Carduus pycnocephalus, Dactylis glomerata, Danthonia pilosa, Daucus pusillus, Erodium botrys, E. brachycarpa, Eschscholzia californica, Festuca arundinacea, Holcus lanatus, Hypochaeris glabra, Lupinus spp., Phalaris aquatica, Poa pratensis, Silybum marianum, Sisyrinchium bellum, Sonchus oleraceus, Trifolium depauperatum, and T. dubium (Hektner and Foin 1977; Fiedler and Leidy 1987; Holland and Keil 1995; Ford and Hayes in press).

More Interior Grassland Away from the Immediate Coast Hilltop grasslands (balds, glades) support stands in the alliances Festuca californica, F. idahoensis/roemeri, and F. rubra (Evens et al. 2006; Figure 3.5). Some of the best-known stands of F. idahoensis/roemeri occur in the inner North Coast Ranges in the Yolla Bolly–Middle Eel Wilderness Area, where glades 2–20 hectares in size have a rich mixture of native and non-native herbs. The glades are surrounded by Quercus garryana var. garryana woodland with an understory also dominated by F. californica. A Danthonia californica (California oatgrass) alliance occurs on foothill slopes and balds from San Luis Obispo to Humboldt Counties (Stromberg et al. 2001, NatureServe et al. 2003, Hektner et al. 1983; CVIS 2006; Ford and Hayes in press). Particularly large stands in Marin and Sonoma counties also include such bunchgrass species as Elymus glaucus, Melica californica, and Nassella pulchra, and balds in the Santa Cruz Mountains have high cover of the exotic grass Brachypodium distachyon. Ford and Hayes (in press) refer to the oatgrass alliance as “moist native grassland,” and their Twinspan analysis of 450 plots, from Mendocino to San Luis Obispo, shows moderate to high constancy for the following natives: Bromus carinatus, Carex species, Hordeum brachyantherum, Juncus patens, J. phaeocephalus, Oxalis albicans ssp. pilosa, and Ranunculus californicus. Other alliances of balds have been proposed for non-native associations. For example, Arrhenatherum elatius stands that occur from Humboldt County to Lake County and east to Tehama County, with associated species Aira caryophyllea, Dichelostemma capitatum, Lotus micranthus, Rumex acetosella, and Sherardia arvensis (Sugihara and Reed 1987; Sugihara et al. 1987). Jimerson et al. (2000), Hektner et al. (1983), and Evens et al. (2006) have proposed a related Cynosurus echinatus alliance, divided into more than eight associations. Another semiinterior coastal grassland alliance is that of Bromus hordeaceus. Total cover is lowest and forb cover highest for this alliance, among all coastal alliances. It has been split into half a dozen associations by Jimerson et al. (2000), each differentiated by associates that include both exotics (Aira caryophyllea, Erodium species, Taeniatherum caput-medusae, and Bromus tectorum) and natives (Agoseris heterophylla, Cryptantha intermedia, Dichelostemma capitatum, Grindelia camporum, Hemizonia congesta, Ranunculus occidentalis, and

F I G U R E 3.5. Northern coastal grassland (foreground) as a glade or bald, dominated by Festuca roemeri.

Plum Garden glade, Yolla Bolly Mountains. Photograph courtesy of Todd Keeler-Wolf.

species in the genera Clarkia, Linanthus, Lotus, Lupinus, and Trifolium). Driest and warmest coastal sites support the Phalaris aquatica alliance. Associated species with highest constancy and cover include Bromus hordeaceus, Centaurea solstitialis, Koeleria micrantha, and species in the general Clarkia and Galium (Jimerson et al. 2000).

Cold Desert Grassland East of the Sierra-Cascade axis, a continental location produces a cold and dry climate. Elevations are generally ⬎1250 meters, making even summer temperatures lower than those in the warm deserts to the south. The grassland here is ecologically part of the Great Basin or Inter-Mountain region. Although some authors (e.g., Young et al. in press) label them “steppes,” clearly meaning an association of bunchgrasses and shrubs or trees, our experience with this region is that isolated stands of grassland without shrubs or trees also are present. The most abundant and widespread grasses include Pseudoroegneria spicata, Achnatherum thurberi, Elymus elymoides, Festuca idahoensis, Hesperostipa comata, Koeleria macrantha, Leymus cinereus, Pascopyrum smithii, Poa nevadensis, and P. secunda. Many of these species also occur in the Palouse grassland of eastern Oregon and Washington, which shares a similar climate and range of soils and geologic substrates (Young and Clements 1999). The cold desert grassland ecosystem has been significantly modified by the introduction and spread of highly invasive annuals or persistent perennials, in particular Agropyron cristatum, A. intermedia, Bromus tectorum, B. japonica, Elytrigia pontica, Ventenata dubia, and Taeniatherum caput-medusae. Although most cold desert

perennial species can survive infrequent wildfires (Wright and Klemmedson 1965), promiscuous burning and inappropriate grazing throughout the 1800s resulted in a type conversion of Artemisia steppe to annual grassland (Pickford 1932; West and Young 2000, Young et al. in press). The cold desert grassland is the least understood and least inventoried among all other grasslands in California. No detailed classification exists, and only scattered plot data have been collected. Our treatment here is thus provisional and cursory. We focus on four community types: Festuca idahoensis (Idaho fescue), Leymus cinereus (Great Basin wild rye), Poa secunda (desert bluegrass), and Pseudoroegneria spicata (bluebunch wheatgrass). Festuca idahoensis grassland (sensu lato, including several subspecies) occurs in the Modoc Plateau area, the stands often being small and surrounded by Artemisia scrub or Juniperus occidentalis woodland. The most extensive stands occur on north-facing exposures that have had most of their woody cover removed by recent fire. Three examples are (1) Clear Lake Hills in Modoc County, where F. idahoensis and P. secunda are codominants, both with 10–20% cover; (2) the Crater Peak Research Natural Area on slopes of the Klamath Mountains (Keeler-Wolf 1987; Cheng 2004), where the grassland occurs on gabbro-derived soils and is surrounded by a woodland of Cercocarpus ledifolius and Pinus balfouriana; and (3) the northern Warner Mountains, where the grassland is adjacent to subalpine vegetation with Antennaria rosea, Arenaria aculeata, Carex rossii, and Pinus monticola. Leymus cinereus dominates small stands in deep bottomland soil and also on moderately deep slope soils subirrigated with snow melt. The few large stands in California are on the Modoc Plateau, either on recently burned uplands or in semiriparian valley bottoms and toe slopes. The burned slopes

C O M M U N I T Y C L A S S I F I C AT I O N , N O M E N C L AT U R E

31

have resprouts or seedlings of Ericameria parryi, Symphoricarpos species, and Tetradymia canescens. Poa secunda (sensu lato) is widespread and has a broad ecological amplitude. It is part of a taxonomically messy group, variously known as P. canbyi, P. gracillima, P. juncifolia, P. incurva, P. nevadensis, P. sandbergii, P. scabrella, and others. The grassland it forms is common at the edge of wet meadows, with Juncus balticus and Artemisia cana, but it also grows on drier upland slopes where it interfaces with Festuca idahoensis or Pseudoroegneria spicata grasslands. Pseudoroegneria spicata is the one of the most widespread native perennial grasses in northeastern California, but it is seldom a dominant. It occupies the more xerophytic habitats, usually restricted to well-drained, rocky slopes on south aspects. P. spicata usually provides 15–30% cover in grassland patches that occur within a mosaic of such scrub alliances as Artemisia tridentata ssp. vaseyana and A. arbuscula ssp. arbuscula. The largest stands in California are 0.5–2.0 hectares in size in the Skedaddle Mountains and along the western slopes of the Warner Mountains and appear to have a recent fire history (Figure 3.6).

Warm Desert Grasslands These grasslands are distinct from valley grasslands to the west, and their distinctiveness increases to the east and south. Some stands in Antelope Valley of Los Angeles and Kern Counties and in San Felipe Valley of San Diego County are relatively similar to cismontane grassland. Part of the reason for overlapping similarity is that the western portion of the warm desert is still influenced by a Mediterranean-type climate, although annual total precipitation there is lower (⬍250 mm). The warm desert grassland is typically a shrub steppe, but in some areas the native perennial grasses have greater biomass than that of the shrubs (Figure 3.7). We divide warm desert grassland into six types: Pleuraphis rigida (big galleta grass), P. jamesii (galleta grass), Achnatherum speciosum (desert needle grass), Aristida purpurea (three-awn), dune grassland, and saline grassland. Annual forb types are also present and variable, especially in El Niño years. As with the cold desert grassland, invasive species have significantly altered the ecosystem. The most important transformative taxa are Brassica tournefortii, Bromus madritensis ssp. rubens, Salsola tragus, Schismus arabicus, and S. barbatus. The extent and cover of B. madritensis have particularly increased in the past few decades, in part because of its ability to carry fire (Brooks 1999; Brooks and Esque 2000; Brooks and Minnich 2007; Brooks and Pyke 2001). The Pleuraphis rigida alliance is characteristic of grasslands in the Colorado and eastern Mojave deserts, where there is a significant amount of summer precipitation. At the sandy base of bajadas and on dune aprons it may be a codominant with Larrea tridentata, Ambrosia dumosa, and Yucca schidigera; on upper slopes it may be associated with Ephedra nevadensis, Ferocactus cylindrica, and Sphaeralcea

32

OVERVIEW

F I G U R E 3.6. Stand of Pseudoroegneria spicata on convex upper slope

of the Skedaddle Mountains in Lassen County. Photograph courtesy of Todd Keeler-Wolf.

ambigua. Thomas et al. (2004) have sampled, analyzed, and classified several associations within this alliance, the most common being P. rigida/Ericameria cooperi in the Mojave Desert and P. rigida – Acamptopappus sphaerocephalus in the Sonoran Desert. The Pleuraphis jamesii alliance occurs mainly in the northern Mojave Desert at higher, rockier sites than those occupied by the P. rigida alliance. It is unique in being a sod-forming grass but is otherwise similar to its congener in being associated with shrubs. Thomas et al. (2004) have defined several associations: (1) P. jamesii/Ephedra nevadensis, the most widespread association in the eastern Mojave, which includes the shrubs Lycium andersonii, Menodora spinosa, and Castilleja applegatei; (2) P. jamesii–Eriogonum fasciculatum on granite substrates in the eastern Mojave; and (3) the very rare P. jamesii– Muhlenbergia porteri, known only from the Mid Hills and New York Mountains. The Achnatherum speciosum alliance is probably the most widespread of all the warm desert perennial grassland alliances. It ranges all across the Mojave Desert and south to Joshua Tree National Park. It appears to be a successional

F I G U R E 3.7. Pleuraphis rigida grassland east of Superior Lake in the central Mojave Desert. Photograph

courtesy of Todd Keeler-Wolf.

grassland, occupying burned blackbush stands (Coleogyne ramosissima) for several years (Brooks and Matchett 2003). Some stands are ⬎20 ha in extent, as those in Antelope Valley adjacent to Juniperus californica, Eriogonum fasciculatum, and Yucca brevifolia stands. Aristida purpurea grassland is limited to the western margin of the Sonoran Desert, where it is associated with Acacia greggii and Quercus engelmannii. Patches of this grassland occur within a matrix of Bromus madritensis ssp. rubens or Nassella pulchra stands. Sand dunes and sand sheets in the warm deserts of California support several types of open (⬍15% cover) grassland. These include stands dominated by Achnatherum hymenoides (Indian rice grass), Panicum urvilleanum (dune panic grass), and Swallenia alexandrae (Eureka dune grass). Dune panic grass is the most widespread of the three, ranging through both the Mojave and Colorado deserts and present on both isolated dunes and extensive, continuous sand sheets. In contrast, Indian rice grass is restricted to just a few places, such as Eureka Dunes, Owens Lake, and Kelso Dunes, where it forms a very open grassland. Associates include indicators of recently disturbed sites, such as Petalonyx thurberi, Sphaeralcea ambigua, and Salsola tragus. Eureka dune grass is the rarest of the three, restricted to Eureka Valley dunes. Swallenia grassland may be the rarest of all grasslands in California. Its biology and autecology have been described by Pavlik and Barbour (1986a, b). Grasslands of saline desert playas feature Distichlis spicata (salt grass) or Sporobolus airoides (alkali sacaton), the latter previously mentioned in this chapter as a component of grassland in the San Joaquin Valley. On playas, the two grasses are often adjacent to, or mixed with, the shrubs and subshrubs

Allenrolfea occidentalis, Atriplex canescens, A. polycarpa, Isocoma acradenia, Iva acerosa, and Suaeda moquinii. Near springs, the grasses are associated with the perennial graminoids and herbs Schoenoplectus americanus, Juncus cooperi, and Anemopsis californica.

Summary Major grassland types within the state’s boundaries are valley grassland (including coastal southern California grassland), north coastal grassland, serpentine grassland, cold desert grassland, and warm desert grassland. Sometimes these types include shrub or tree associates and could be called shrub-steppe or tree-steppe. In any case, species richness comes primarily from forbs, not from grasses or woody taxa. We chose grassland as the term of choice to apply to lowelevation vegetation dominated by herbs, and reduced the terms prairie and steppe to synonyms. Classification of grassland associations and alliances is rapidly progressing, but it is still far behind the progress that has been made with woody vegetation. We can only guess at the ultimate number of community types that will be defined, named, and recognized in terms of species composition and ecological relationships. We estimate that 400 or more associations eventually will be discernible. Californians today tend to elevate the importance and uniqueness of dramatic landscapes with tall trees that are distant from large cities. As a result, nearby, soft, subtle grassland landscapes were long discounted, disregarded, and degraded. It is only in the past two decades that grassland vegetation has been seen as an important target for conservation and restoration activities, both of which require an understanding

C O M M U N I T Y C L A S S I F I C AT I O N , N O M E N C L AT U R E

33

of the identity, location, and spatial relationships among natural grassland community types. We have tried to show that precontact grassland community types can be reconstructed on the basis of the many native species that remain present, although at low abundance and biomass. Classification of community types based on the cooccurrence of both exotic and native species is also resulting in the recognition of alliances and associations that are important indicators of environmental traits and land use history.

34

OVERVIEW

By examining complete species assemblages (instead of focusing on dominant species), this method of sampling can overcome vegetation change caused by 200 years of overgrazing, the spread of exotics, and the alteration of fire regimes, which have type-converted some grasslands, significantly modified the composition of others, and shifted many from a sustainable, steady-state condition into temporal, seral phases on the way to (for example) coastal scrub, oak woodland, or mixed evergreen forest.

H I STO RY

FOUR

Pleistocene and Pre-European Grassland Ecosystems

Late Quaternary Paleoecology of Grasslands and Other Grassy Habitats Peter E. Wigand

Today as in the past, grasslands, and grassy steppe and chaparral, have been essential and dynamic elements of the western North American ecosystems. Since the appearance of grasses during the Eocene, they have provided both a crucial role in the recycling of nutrients and an important habitat for animal populations that have in many cases coevolved with them. Grasslands and grassy steppes are dynamic ecosystem components that are constantly responding to climate, fire, animals, geomorphic change, and human impact. Grass abundance within vegetation communities, as well as the diversity of grass species, responds to changes in seasonal and annual precipitation and to changes in evaporation rate due to variations in annual or seasonal temperature. This can be seen historically (e.g., the Dust Bowl) but especially prehistorically in the paleobotanical record, where there is abundant evidence that grass abundance, distribution, and diversity have fluctuated significantly during the late Quaternary. At times grasses have been much more, and at other times much less, ample within vegetation communities where they presently occur. In the West, and in California in particular, both pollen and plant macrofossil records provide evidence of the ebb and flow of grasses within late Quaternary vegetation communities. Although there is some phytolith evidence from dinosaur coprolites (Prasad et al. 2005) suggesting the presence of grasses during the upper Cretaceous in central India, the first well-documented appearance of grass pollen (spherical shape and single pore) in the evolutionary record suggests an origin on the Gondwana continent (present-day South America and Africa) shortly before the beginning of the Paleocene (
65 million years ago) (Jacobs et al. 1999). The earliest

unequivocal grass fossils date to the Paleocene-Eocene boundary, about 56 million years ago (Jacobs et al. 1999; Kellogg 2001). After grasslands’ appearance (65 – 50 million years ago), their expansion seems to have been limited until the middle and late Miocene (ca. 20–10 million years ago), when grasslands and grass-rich ecosystems became widely distributed as a result of either lower atmospheric CO2 content, which gave grasses a photosynthetic advantage, or, more likely, climatic changes that created a fire regime suitable for the replacement of woodlands by grasslands (Keeley and Rundel 2005). Keeley and Rundel (2005) suggest that during the middle and late Miocene, climates became more seasonal, resulting in an annual cycle comprising a wet season of high plant production followed by a dry season during which these materials dried. A monsoonal climate coupled with the dry season generated storms with abundant lightning, which ignited fires that cleared the previously dominant forest habitats and paved the way for grassland expansion. During the Miocene epoch, 20 million years ago, grass species developed characteristics that are similar to those of modern grasses, even identifiable to modern genera. In particular, they evolved with herbivore grazing. This is especially true of western North America, where grassy habitats and herbivores characterized the Miocene of much of the region and had coevolved through the Eocene. In addition, grasses seem to have developed the capacity to respond during the same year to dramatic increases in either winter or spring/summer precipitation. The history and environmental relationships of grassy habitats of the West during the late Quaternary is being revealed by three kinds of paleobotanical evidence: pollen, plant macrofossils (primarily seeds), and phytoliths. Each of these provides different yet complementary kinds of evidence regarding the distribution and abundance of grasses, and in

37

some cases of their importance to human and animal populations. Grasses, like all plants that rely upon the vagaries of the wind for pollination, produce relatively large quantities of pollen. These clouds of pollen settle across the landscape and accumulate in lakes, bogs, or other locations favorable to their preservation. Analysis of samples from such places provides a potentially continuous record of the local and regional relative abundance of grasses. Phytoliths (silica concretions that form in the cells of grass plants) provide a record of their actual distribution on the landscape. Because they are deposited in the ground and are buried when the plant dies and decays, phytoliths rarely blow around the landscape. Therefore, they mark the places where grasses actually grew. Grass macrofossils usually occur in contexts where they have been collected either by animals or humans. In most cases, grass macrofossils are in relatively close proximity to the areas where they were collected. This is especially true in the case of small mammals, which have a relatively restricted foraging area. In the case of small mammals, nesting sites are rarely preserved for very long, so they do not provide a long-term record of grasses in a region. However, the indurated nests (middens) of woodrats provide an exception. Urine-encrusted woodrat nests can preserve plant macrofossils, insects, and pollen for tens of thousands of years (Betancourt et al. 1990). The relationship between plant macrofossils collected by ancient peoples and the sources of these materials is a bit more problematic. People can move across great distances in order to collect raw materials and food for their survival. However, in most cases plant materials are used, processed, or stored close to the area of collection (Anderson, Chapter 5). Like both pollen and phytoliths, plant macrofossils must be deposited in places where their preservation is ensured. One exception to this is if the plant materials are charred by fire. In that case they become very resistant to destruction after burial.

vegetation (including grass) have been obtained from a variety of habitats; including coastal estuaries, large and small lakes lying at a wide range of elevations on both sides of the Sierra Nevada/Cascade mountain ranges, bogs and higher mountain meadows, desert springs, caves, ancient woodrat nests (middens), and archaeological sites. In most cases, the archaeological records are either poorly dated or undated and so cannot be used to provide much information regarding the late Quaternary history of grass. However, several of the coastal estuary, mountain lake and meadow, and desert spring records are dated and provide a continuous, and at times detailed, record of grass expansion and contraction in plant communities throughout the West and California in particular. These are supplemented by well-dated and, in a few cases, well-stratified ancient woodrat midden records (Wigand and Rhode 2002). TH E P LE I STO CE N E

Thus far, there are four long pollen sequences recording local and regional vegetation change: •

The three-million-year sequence from Tulelake on the eastern edge of the Modoc Plateau in northeastern California (Adam et al. 1989)

•

The pollen sequence from Owens Lake on the lower portion of the Owens River east of the Sierra Nevada Mountains in southeastern California, comprising an approximately 180,000-year-long section (Woolfenden 1993, 2003), and a lower section extending from the base of this section to over 870,000 years ago (Litwin et al. 1993)

•

•

General Characteristics of the Late Quaternary History of Grasses The paleoecological evidence of episodic increases and declines in grass abundance and changing distributions during the late Quaternary (we will restrict our purview to the last 20,000 to 30,000 radiocarbon years before present [rcyr BP]) in the western Unitied States consists primarily of pollen data. Paleoecological evidence from the late Quaternary of the West suggests that grass dynamics are primarily the result of changes in precipitation, though temperature may at times also play a role. Ideally, for our examination of grassland history, we would examine pollen localities located in the Central Valley in the midst of the grassiest habitats, i.e., the northern portion of the Central Valley. Unfortunately, there are no good palynological (study of pollen) records from such environments. Instead, our pollen records are obtained from environments that, in most cases, do not correspond to grassdominated habitats in the West. Currently, late Quaternary pollen records documenting the dynamics of late Quaternary

38

HISTORY

The
130,000-year-long sequence from Clear Lake in the coast range of northern California (Adam 1981, 1988) The ongoing analysis of a fragmentary million-year long record from the Buena Vista Lake Basin southwest of Bakersfield, California, at the mouth of the Kern River, which will also eventually provide some information on grass history (Wigand 2006, unpublished data)

These records provide an indication of the response of grasses under California’s natural variation in precipitation, temperature, and climate. Variance in the environment ranges from annual (Reever-Morghan et al., Chapter 7) to decadal patterns (e.g., El Niño/La Niña cycles) or to millennial scales (e.g., glacial cycles). Temperature-driven changes based upon orbital-scale climate change underlie many of these precipitation cycles (Ruddiman 2001; Liu and Herbert 2004), but they can also drive grass response by reducing evaporation rates, thereby increasing effective water availability. At scales of tens and hundreds of thousands of years, the earth’s orbital characteristics, including axial precession (precession of the equinoxes), obliquity (or tilt of the earth’s axis), and eccentricity, have resulted in differences in solar insolation affecting global temperature (Milankovitch 1930).

Variations in the amount of heat accumulated in various regions (oceans or land, Northern or Southern Hemisphere) have driven the global climate system (Cronin 1999: 560). Changes in millennial-scale solar insolation directly impact the amount of moisture evaporated from the oceans, the paths of this moisture across continents, its condensation as precipitation, and finally its accumulation as glaciers. Glaciers, once they begin to grow, create their own local climates that further impact the amount and nature of precipitation. Over the span of the late Cenozoic the rise of the Sierra Nevada/Cascade Mountains has further affected long-term grass distribution and abundance through changing the distributions of moisture on opposite sides of their crest. A general rule of thumb is that the Sierra Nevada Mountains have risen roughly 100 meters per 100,000 years since about 3 million years ago. Given orographic forcing of precipitation on the western slopes of these rising mountains, each successive glaciation forced more moisture out of storms west of their crests and increased the rain shadow effect on their lee. The eventual result of this process might suggest the gradual decrease in grass abundance east of the Sierra Nevada crest and its gradual increase at sites on the west slope of the Sierra Nevada Mountains. This has not, however, been observed in the pollen records from Tulelake and Owens Lake. The three-million-year record of grass abundance from Tulelake in northeastern California’s Modoc Plateau clearly reflects the trend to progressively cooler glacial cycles during the last 3 million years (Adam et al. 1989). Currently Tulelake is much smaller than it was during the Pleistocene. In presettlement California, it was surrounded by marsh habitats dominated by sedges and cattails. Now, lowlands beyond the marsh are composed of grassy sagebrush steppe. Whereas intermediate elevations are covered with western juniperdominated woodlands, higher elevations are characterized by pine- and fir-dominated montane forests. Compared with areas further south and east in the Great Basin, the Modoc Plateau region surrounding Tulelake is rich in grasses. During the Quaternary, the pollen record indicates increasing abundance of grass in the Tulelake core. This appears to correspond directly to both increasingly cold and long glacial cycles, as recorded in the
18O of Atlantic Ocean sediment cores (Ruddiman 2001). In the deeper sections of the Tulelake core, greater grass abundance is both low and sporadic in its frequency (Adam et al. 1989). Grass abundance appears to correlate only with the coolest of the glacial cycles prior to 0.6 million years ago. However, at the top of the core it is not only much more abundant, but abundant for much longer spans of time (Adam et al. 1989). The correspondence of increased grass pollen in the Tulelake record with the current timing of late Pleistocene glacial cycles is given additional support by coincident increases in pelagic lake algae during these periods, suggesting a deeper lake (Adam et al. 1989). It appears that the early and later portions of the Holocene (the modern interglacial) have been cool enough to support more relatively abundant grass communities, compared to those found in previous interglacials (Figure 4.1a). At

Tulelake, abundant grass primarily reflects increased effective precipitation due to reduced evaporation rates, although there may have been a slight increase in real precipitation. Especially during the last 600,000 years, grass has been more abundant than in the previous 2.6 million years (Adam et al. 1989). That transition occurred when 100,000-year glacial cycle dominance replaced 41,000-year glacial cycle dominance (Ruddiman 2001). Since 600,000 years ago, glacial cycles have been both more severe and longer in duration, factors that have encouraged the proliferation of grasses in areas where they would normally be rare. At Owens Lake the record of grass is less clear. Counts of grass pollen are very low, and many samples do not even have grass in them. This could indicate poor preservation or simply its rarity. Despite this, there is a general correspondence of grass pollen appearance in both the Tulelake and Owens Lake records. The major difference is that the increase in grasses at Owens Lake between 500,000 and 1,000,000 years ago is much more dramatic than that at Tulelake. The two most abundant periods of grass at Owens Lake during this period are centered at
550,000 and at 750,000–800,000 years ago. Both of these periods correspond to major glacial episodes, as indicated in the
18O record from the eastern Pacific Ocean (Ruddiman 2001: Figure 12.16). The pollen record from pluvial lake Buena Vista, southeast of Bakersfield, contains a discontinuous pollen record spanning the last million years. Although analysis is ongoing, the record indicates several episodes of increased grass pollen during the last 250,000 years, between 370,000 and 380,000 years ago, about 440,000 years ago, 560,000 years ago, and 680,000 years ago. (Wigand 2006, unpublished data). The most dramatic increase occurred at 560,000 years ago, roughly coinciding with the dramatic increase in grass in the Owens Lake record centered at 550,000 years ago. This latter event corresponds with the longest-duration cool, moist episode, as recorded in the Owens Lake pollen record by plants reflecting cooler, moister climate (our diagrams plotted from data presented by Litwin et al. 1993). Therefore, both the Tulelake and Owens Lake pollen records indicate that cooler temperatures and increased effective precipitation, due to reduced evaporation rates during glacial episodes, appear to have been crucial in dramatically increased grass in semiarid woodland and shrub steppe habitats in the currently dry interior regions of the West. Was this also true of the region west of the crest of the Sierra Nevada Mountains? Clear Lake in the coast range of northwestern California is the only published record from western California that extends beyond the latest Pleistocene (Adam 1988: 86). Currently, this lake is surrounded by blue oak (Quercus douglasii)/gray or foothills pine (Pinus sabiniana) forest with small patches of chaparral near its southern shore. Grasses occur primarily as understory in both the forest and chaparral habitats, and as a result are not as abundant as in more open habitats. A 130,000-year-long pollen record from the lake provides a pollen record extending back into the penultimate glaciation (Illinoian) (Adam 1988). Grass pollen

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

39

F I G U R E 4.1a. Grass-to-sagebrush and grass-to-total-terrestrial (grass pollen percentage) pollen ratios for five of the longest pollen records from California (the ratios from the other two long pollen records from California, Clear and Owens Lakes, are not illustrated here). The Exchequer Meadow data are available online from the National Climate Data Center, NOAA, at the North American Pollen Database (NCDC 2007). The Tulare Lake and Playa Vista core 1 data are available online from the Department of Geosciences at the University of Arizona (Davis 2002). The Tulelake data were obtained from Adam and Vagenas (1990: 307). The axis on the right of each plot records the ratio of grass to terrestrial pollen; the axis on the left records the ratio of grass to sagebrush. Courtesy of Dr. Peter E. Wigand.

has usually constituted less than 4% of the terrestrial pollen at the site. During the late Quaternary, there is a clear correspondence between grass pollen abundance and vegetation community structure that can be tied indirectly to climate. Grasses seem to be slightly more abundant during interglacial periods, when oak forests are dominant. This may reflect a slightly more open structure in oak forests, as opposed to the mixed conifer forests that characterized the glacial periods. More light and precipitation was probably available for a grass understory to develop. This is confirmed by greater

40

HISTORY

abundance during interglacial periods of alders, willows, hazel, buckthorn, and composites (Adam 1988). However, increased sagebrush during glacial cycles in the Clear Lake record also suggests either that there were open areas nearby or that the conifer forest may have been more open in some areas. Because values of sagebrush pollen were never greater than 10%, there probably was never a real sagebrush steppe community within the region (Adam 1988). Episodes of greater moisture during glacial periods seem to have resulted in sagebrush decline and grass increase. These

periods may also have been slightly warmer as well, as is suggested by declines of fir during such episodes. In summary, over the span of the Pleistocene, grass abundance became greater in shrubby steppe environments east of the Sierra Nevada Mountain crest during each successive glacial period. This seems to correspond to a shift toward cooler, wetter, and longer-duration glacial cycles. Cooler temperatures during these cycles resulted in greater effective soil moisture. In addition, the wettest portions of these cycles were not during the glacial maxima, but during the slightly warmer onsets and declines (Wigand and Rhode 2002). During glacial maxima, reduced global temperatures caused orographic rainfall to commence and be more intense at lower elevations than is the case today on the western slopes of the Sierra Nevada and Cascade mountains. As a result, much of the moisture that reached the coast of California during the Pleistocene was forced out of Pacific storms before they reached the crest of the mountain ranges. This resulted in dramatically reduced precipitation in the intermountain interior during the glacial maxima, when increased precipitation might be expected. However, reduced precipitation was somewhat balanced by much reduced evaporation rates, due to the lower global temperatures, as well. The ultimate result was a cold, dry continental climate during glacial maxima in the intermountain West. However, just before and after the glacial maxima were periods of warmer temperatures, when the rate and amounts of precipitation wrung out of Pacific storms on the western slopes of the Sierra Nevada and Cascade mountains was reduced. That is, because of warmer temperatures, condensation of moisture occurred at higher elevations and more slowly than during the glacial maxima. More and wetter storms were able to cross the Sierra Nevada/Cascade mountain ranges and bring precipitation to the interior. Although evaporation rates were slightly higher, there was a net gain in effective precipitation. It is during these periods that we see grass expansion in sagebrush steppe and juniper woodlands in the north and in the middle and higher elevation shrub steppe communities in the south (Wigand and Rhode 2002). At the same time, pluvial lakes reached their highest levels (Benson et al. 1990), and glacial advances in the Sierra Nevada reached their greatest late Pleistocene extent (Phillips 1996). The pollen record from Tulare Lake at the southern end of the Central Valley is the longest record currently available from southern California west of the crest of the Sierra Nevada Mountains (Davis 1999). There the pollen record reveals three periods during the last 27,000 years when grasses were relatively more abundant. During the last glacial cycle of the Pleistocene it appears that grass was more abundant between 26,000 and 19,000 rcyr BP than during the glacial maximum 18,000 to 19,000 rcyr BP (Figure 4.1a). This is similar to the record from Clear Lake, where decreased grass abundance corresponded to cooler, drier episodes around the glacial maximum. The transition from the Pleistocene to the Holocene is not well documented. Unfortunately, the longer pollen records

discussed in the foregoing paragraphs have very low-resolution (wide spacing between samples) during the late Pleistocene/ Holocene transition. In some cases, such as Tulare Lake, the Pleistocene/Holocene transition is missing (Figure 4.1a). The Playa Vista 1 pollen record from the Ballona Estuary reveals a late Pleistocene increase in grasses between
12,500 and 11,500 rcyr BP that corresponds to increased pine, and moist-climate shrub pollen values from the same core (Figure 4.1a; Wigand in press, a). We know from pollen and woodrat midden macrofossil records from the northern Mojave Desert that climates were much wetter between 13,000 and 12,000 rcyr BP than they had been during the glacial maximum (Wigand and Rhode 2002). Unfortunately, there is no preserved pollen record between 10,700 and 9,500 rcyr BP from the Playa Vista 1 locality, and only poorly preserved pollen from the Playa Vista 8 locality for this period (Figure 4.1a). Therefore, we do not have a good picture of what may have been happening with grasses during the transition from glacial to postglacial climates in southern California. The record from Exchequer Meadow from the western slope of the Central Sierra Nevada Mountains provides a record of the transition from late glacial alpine/subalpine grassy habitats to Holocene Sierran Montane Forest. Currently, the meadow is dominated by sedges (Scirpus spp.) and rushes (Carex spp.) (Davis and Moratto 1988). During the latest Pleistocene (between
13,000 and 12,000 rcyr BP), Exchequer Meadow was dominated by grasses, sagebrush, and composites. This suggests a climate considerably cooler and drier than that of today. Grasses were more abundant at Exchequer Meadow at that time than at any time later during the Holocene. This is in part due to the fact that slopes surrounding the meadow were still almost entirely dominated by subalpine grasslands. By 10,500 rcyr BP, grass abundance had declined precipitously as sedges and rushes became more abundant in the increasingly wetter meadow. The grasslands surrounding the meadow were slowly invaded by pine and fir as they readvanced into higher elevations of the Sierra Nevada to areas surrounding Exchequer Meadow. These changes signaled both warmer and moister climates, as storms that, during the glacial period, had lost much of their moisture at lower elevations now carried increasing amounts of precipitation to higher elevations. The Tulelake record indicates that during the last glacial cycle grasses were more abundant than during most of the Holocene. In addition, a dramatic increase of grass that occurred at
10,000 years ago at Tulelake may correspond with an episode of effectively globally more moist climate due to very cold temperature conditions that occurred between 11,200 and 10,200 rcyr BP. Allowing for the vagaries of radiocarbon dating and changing deposition rates, this cool episode is recorded in the Tulelake record at 10,000 rcyr BP. This cool episode may also appear in the record from Tulare Lake, where it is recorded as the highest early Holocene grass values centered at 10,000 rcyr BP (Figure 4.1a). At about the same time, the European Younger Dryas event was bringing dramatically lower global temperatures.

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

41

FIGURE 4.1b. Grass-to-sagebrush and grass-to-total-terrestrial (grass pollen percentage) pollen ratios for four pollen records from California that span much of the middle and late Holocene. The San Joaquin Marsh, Shellmaker, and John Wayne Freshwater Marsh records are from coastal southern California and are available from the Department of Geosciences at the University of Arizona (Davis 2002). The Dinkey Meadow record is from the west slope of the central Sierra Nevada Mountains and is available online from the National Climate Data Center, NOAA, at the North American Pollen Data Base (NCDC 2007). The axis on the right of each plot records the ratio of grass to terrestrial pollen; the axis on the left records the ratio of grass to sagebrush. Courtesy of Dr. Peter E. Wigand. TH E HOLO C E N E

The Holocene record of grass in California is documented by substantially more pollen records than during the Pleistocene (Figures 4.1a–d). It is clear that there are both long- (centuryscale) and short-term (multidecade) cycles in grass abundance.

42

HISTORY

FIGURE 4.1c. Grass-to-sagebrush and grass-to-total-terrestrial (grass pollen percentage) pollen ratios for four pollen records from California that span the late Holocene. These pollen records are all from southern California west and northwest of the Los Angeles Basin. The San Nicholas Island data are available from the Department of Geosciences at the University of Arizona (Davis 2002). The data for Zaca Lake, Carpinteria Marsh, and Cleveland Pond are from Mensing (1993). The axis on the right of each plot records the ratio of grass to terrestrial pollen; the axis on the left records the ratio of grass to sagebrush. Courtesy of Dr. Peter E. Wigand.

However, it is only the highest-resolution pollen records that can reveal what seem to be decade-long episodes of dramatic grass increase (only one or two such records currently exist in the West). In addition, the available pollen records show that grass abundance is highly variable. These differences may reflect microclimatic differences as well as the impact of local topography. Finally, it should be noted that the age assignments are not exact. Differences in calculation of the deposition rate of each site due to the position and number of radiocarbon dates

FIGURE 4.1d. Grass-to-sagebrush and grass-to-total-terrestrial (grass pollen

percentage) pollen ratios for three pollen records from California that span the latest Holocene. The Playa Vista and Long Beach Campus pollen records are both from southern California: Playa Vista from the southwestern corner of the Los Angeles Basin, and the Long Beach Campus record from the Long Beach area. The pollen data from these sites are available through the University of Arizona Department of Geosciences (Davis 2002). The Woski Pond record is from the west slope of the central Sierra Nevada Mountains, and the data are available through the North American Pollen Database (NCDC 2007). The axis on the right of each plot records the ratio of grass to terrestrial pollen; the axis on the left records the ratio of grass to sagebrush. Courtesy of Dr. Peter. E. Wigand.

and the errors associated with them can result in slight differences from site to site in the apparent position of specific climatic events. In general, there appear to be several cycles of grass increase during the Holocene that seem to reflect region precipitation patterns, and mirror their change through time. After the conclusion of the early Holocene Younger Dryas event, there was a period of summer-shifted precipitation between about 9,500 and 8,000 rcyr BP, coincident with the post-glacial thermal maximum (Wigand and Rhode 2002). Although this is most strongly manifested in the southern portion of the intermountain West, it is evidenced as far north as the northern Great Basin. A middle Holocene increase in precipitation that occurred between 8,000 and 5,000 rcyr BP was concentrated primarily in southern California during a period of warmer temperatures (Wigand in press, a). A brief cool, winter-wet event centered
5,500 rcyr BP, because of its

brevity, is only seen in higher-resolution pollen records from the West (Wigand and Rhode 2002). The cool, winter-wet Neopluvial period occurred between
4,000 and 2,000 rcyr BP (Wigand 1987; Wigand and Rhode 2002). The evidence for these events is much more pronounced in pollen sites in the northern half of the West than in the southern half. A dramatic increase in precipitation at the same time that much cooler temperatures occurred
2,000 rcyr BP. The impact of this event appears to have been similar throughout the West (Wigand and Rhode 2002; Wigand in press, a). Between
1,600 and 1,000 rcyr BP the West was characterized by another period of wetter climate. However, this period was characterized by warmer temperatures and increased late season (summer) precipitation (Wigand and Rhode 2002). A succeeding period of warm, moist winter climate between
800 and 700 rcyr BP resulted in a brief, though dramatic, increase in biotic productivity in the northern intermountain West and in some areas of the southwestern United States as well. Finally, a period of cool, wet winters during the last phase of the Little Ice Age, between about
350 and 180 rcyr BP, resulted in the most recent surge of grass abundance in the West. Pollen from Tulare Lake in California’s Central Valley provides one of the most sensitive proxy records of Holocene vegetation change for south-central California (Davis 1999). That record is tied to changes in lake level and marsh history of the lake as well (Negrini et al. 2006). At Tulare Lake there was an early and middle Holocene episode of more abundant grass between 10,000 and 5,000 rcyr BP, with a major break about 8,000 rcyr BP (Figure 4.1a). The middle Holocene grass event clearly suggests more significant increases in precipitation than does the early Holocene event. Evidence from the Mojave Desert suggests that the early Holocene event is the result of an increase in summer precipitation (Wigand and Rhode 2002). An increase in grass between 8,000 and 6,000 rcyr BP corresponds to a hypothesized middle Holocene increase in summer precipitation that is revealed in recent reanalyses of two cores from the Ballona Estuary in the southwestern corner of the Los Angeles Basin (Wigand in press, a). The Playa Vista 1 and 8 records from the Ballona Estuary provide some of the most detailed information on both local and regional vegetation change for coastal southern California currently available. Although the timing and magnitudes of individual responses in the two cores are variable, there is clear evidence of a middle Holocene increase in grass around the Ballona Estuary between 8,000 and 6,000 rcyr BP (Figure 4.1a). This corresponds to middle Holocene increases in oak and chaparral species as well (Wigand in press, a). A late Quaternary synoptic climate model reconstruction of southern California climate by Dr. Reid Bryson of the University of Wisconsin suggests that this period was characterized by warmer annual temperature and increased annual precipitation, including increases in both winter and summer precipitation (Wigand in press, a). Farther north, at Exchequer Meadow, there are only slight increases in grass pollen at that time (Figure 4.1a). However, they are insignificant when compared with early Holocene grass abundance at Exchequer Meadow. Yet farther north, at

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

43

Tulelake, this event does not appear in the pollen record (Figure 4.1a). This suggests that this event did not extend much beyond the latitude of central California. Evidence for increased grass in response to a brief, but dramatic cool, winter-wet episode centered around 5,500 rcyr BP is evident only in the higher-resolution records from the Shellmaker and Playa Vista 8 sites from southern California (Figures 4.1a and b). An event at Dinkey Meadow that appears to have occurred at about 4,700 rcyr BP may also be this episode (Figure 4.1b). In this case, however, the episode may simply be shifted as a result of the problems previously mentioned in calculating an accurate chronology for the core. In the Intermountain West the 5,500-rcyr BP event is striking. It can be traced from the eastern Washington Plateau to the northern Mojave Desert. It is manifested in data as diverse as increased vegetation density in eastern Washington, shifts from desert shrub to sagebrush steppe vegetation in southern Oregon, dramatic rises of lake levels in Lake Tahoe, and renewed spring activity in the northern Mojave Desert (Wigand and Rhode 2002). Although several small-scale increases in precipitation have been recorded in the West between 5,500 and 4,000 rcyr BP, the first significant ones occurred as part of what is called the Neopluvial. Three episodes of cool, winter-wet climate centered at
3,700,
2,700, and
2,200 rcyr BP characterize the Neopluvial period (Wigand 1987; Wigand and Rhode 2002). The Tulare Lake, Playa Vista, Shellmaker, and, apparently, Dinkey Meadow sites may all record one or more of the three episodes associated with this period (Figures 4.1a–c). In the interior West these events are much more strongly manifest in the northern half of the region (Wigand and Rhode 2002). Expansions of woodlands and grassy habitats during this period were the greatest since the end of the Pleistocene. These data suggest a cool, winter-wet precipitation pattern that probably originated in the northern Pacific. Although the impact of the dramatic, but brief cold, winterwet episode of climate centered around 2,000 years ago is most dramatic in higher-resolution pollen records of the intermountain West, it also appears as an episode of increased grass in the Tulare Lake, Shellmaker, and John Wayne Freshwater Marsh records (Figures 4.1a and b). The event is also recorded in the Playa Vista 1 and 8 records, but not as an increase in grass. In the Ballona record it is characterized by dramatic regional increases in both pine and sagebrush pollen (Wigand in press. a). Because the increase in these pollen types was so great, they may have masked a similar increase in grass pollen. This episode is recorded in pollen sequences from Diamond Pond in south central Oregon to Lower Pahranagat Lake in southern Nevada (Wigand and Rhode 2002). This event may have been caused by a major volcanic eruption that occurred at that time and was recorded as the highest Holocene sulfur values in ice cores from both Antarctica and Greenland (Wigand in press, a). The dramatic increase of precipitation associated with this event not only affected vegetation distributions but also resulted in dramatic changes in stream flow regimes, desert lake levels, and even shifts of stream channels

44

HISTORY

on the Mojave River (Ely et al. 1993) and the Humboldt River (House et al. 2001) for the next 150 years. In addition, although the increase in grass in southern California does not seem as dramatic in the pollen record as it does in the intermountain West, it must have been significant. Its importance in southern California is hinted at in the charred plant macrofossils recovered from Native American archaeological sites. A dramatic increase in the number of radiocarbon-dated sites around the Los Angeles Basin attest to a sudden increase in native population (Wigand in press, b). Grasses constituted 20 to 25% of the seeds used by these Native Americans at that time (Wigand in press, b). Although some of the grass seeds were of varieties that might be found in sandy, or dune, areas, a significant proportion were of types that are typically associated with the vernal pools found on the Los Angeles Prairie (Wigand in press, b). This suggests that vernal pools may have been less ephemeral at that time and that the grasses around them may have been more abundant. Both factors may have played a role in drawing Native Americans to the Los Angeles Prairie at that time. Perhaps the most interesting event of the last two millennia occurred between
1,600 and
1,000 rcyr BP. During this period, which was contemporaneous with the European Medieval Warm period, warmer climate and decreased winter precipitation characterized the West. However, this period was also characterized by increased late spring through early summer precipitation in the northern half of the region and increased middle to late summer precipitation in the south (Wigand and Rhode 2002). In the northern intermountain West, grass abundance relative to winter-loving plant species such as juniper increased dramatically (Wigand 1987). The effect of this episode on vegetation, animals, and people was nothing short of spectacular. In an area stretching from northern Nevada to eastern Washington, grasses became much more abundant in the sagebrush steppe and semiarid woodlands characteristic of the area at that time (Wigand and Rhode 2002). In response to this, bison expanded into the lush new grazing habitats. Radiocarbon dates from archaeological sites on bison remains records Native American pursuit of these animals into areas where bison had not been since the earliest Holocene (Wigand and Rhode 2002). In the northwestern Great Basin, late spring to early summer–shifted precipitation resulted in the final northward expansion of piñon pine and promoted a major shift from root crops to pine nuts in the native economy. In the eastern Great Basin the shift in seasonal precipitation enabled Fremont horticulturalists to raise maize in areas where previously it could not be grown. Because these people had no major irrigation systems but relied upon flood farming on alluvial fans, they could grow corn only where there was sufficient summer rainfall available. When this period ended,
1,000 to 900 rcyr BP, these peoples disappeared. In southern California the archaeological evidence for the use of grasses from vernal pool areas noted in the previous discussion of the 2,000 rcyr BP cool winter-wet episode continued through this period as well (Wigand in press, b). By
1,000 rcyr BP it appears that Native American populations declined precipitously around the southwestern corner of the

Los Angeles Basin, and evidence for their extensive use of grasses wanes. During the last millennium two wet-climate events are evident in the paleoecological record. A warm, wet climatic event
700 to 800 years ago is most strongly recorded in pollen records currently being analyzed in the northern Great Basin (Sardine Meadow in northeastern California west of Reno, Nevada, and Summit Lake, north central Nevada). This event also appears in the Exchequer and Dinkey Meadow records and in the Tulare Lake record. It is also evident in the results of a recent study of lake levels in the Tulare Lake Basin (Negrini et al. 2006). In northeastern California this period is marked by increased grass abundance and greater spring discharge in the Sardine Meadows south of the Sierra Valley. At Summit Lake there are dramatic increases in aquatic algae productivity, indicating warmer water temperatures, and increases in floating aquatic plant abundance, indicating deeper water. At Grays Lake, Idaho, floating aquatic plants became more abundant relative to littoral plant species, indicating slightly deeper water conditions as well. This event seems to be one that is primarily restricted to a relatively narrow swath across the northern Great Basin stretching from the Sierra Valley in California to Grays Lake. Finally, it is tempting to associate an increase of grass at Tulare Lake during the last 300 years to the cool, moist Little Ice Age (Wigand and Rhode 2002; Figure 4.1a). Tulelake also records the final Little Ice Age cool, moist climate event (Figure 4.1a). However, a record of the Little Ice Age event is more difficult to confirm in the other pollen records available from California. An accurate age assignment to a final episode of grass expansion apparent in the upper sections of many of the cores from southern California, and its correlation to the Little Ice Age event, is almost impossible. However, it is highly probable that many of these grass increases do correspond to the cool, winter-wet Little Ice Age. Further east pollen records from the intermountain West contain abundant evidence of increased grass abundance and expansion of winter moisture-loving plant species during the Little Ice Age (Wigand and Rhode 2002). Although the record of grass abundance from the Holocene appear to be more variable from the West, there is good correspondence to episodes of wetter climate (associated with both warm- and cold-temperature climate) and, in a few cases, to shifts in seasonal distribution of precipitation. This correspondence is evident in late Quaternary grass pollen records from east of the Sierra Nevada Mountains as well. In the more arid regions of the West, plants (and their pollen records) are more sensitive to slight variations in precipitation, so apparent responses can be very dramatic (Figure 4.2). There is also clear evidence that fluctuating precipitation resulting in vegetation change also affected the distribution of the woodrats. The timing, abundance, and spatial distribution of woodrat middens are clear evidence of the impact of climate upon woodrat populations (Wigand and Rhode 2002; Betancourt et al 1990). A comparison of grass pollen from both sediment

cores and ancient woodrat middens reveals some of the same major episodes of grass abundance that were discussed above (Figure 4.2). In addition, the actual periodicity of ancient woodrat midden occurrence reflects the recurrence of climates favorable to woodrats. As previously discussed, the grass events seen in Figure 4.2 were due to a variety of climatic factors. Some of these episodes were related to slightly warmer, wetter climates at the onset and conclusions of glacial maxima (
40,000, 34,000, 21,000, and 10,600 years ago), others were due to increased summer precipitation during episodes of much warmer climate (
9,000 and 1,600 years ago).

Regional Climate Differences From the review of the Holocene portion of this record it is clear that there are regional differences in the response of grass (and other plant species) that suggest regional differences in climate. Differences in precipitation amount, and more importantly in the seasonality of precipitation, are the factors most clearly associated with these changes in vegetation. A comparison of several average monthly precipitation records from weather stations in northern and southern California on both sides of the Sierra Nevada Mountains reveals several clear patterns (Figure 4.3). In northern California January and December tend to be the wettest months. East of the Sierra Nevada Mountains a slightly increased May/June precipitation is clearly evident. This provides an early growing season for grasses. If precipitation increases significantly during this period, as the paleovegetation record clearly indicates that it has, it results in dramatic increases in grass abundance in shrub steppe and semiarid woodland habitats east of the mountains (e.g., between 1,600 and 1,000 rcyr BP). In south-central California maximum monthly average precipitation occurs later in winter than in the north and is centered around February (Figure 4.3). This pattern is similar on both sides of the Sierra Nevada Mountains at lower elevations. Closer to the Sierra Nevada Mountains precipitation is again centered around January, suggesting that the foothills have an earlier onset of heavy winter precipitation than the surrounding lowlands do. East of the Sierra Nevada Mountains at the south end of the Owens Valley, an August/September increase in average monthly precipitation (related to the southwestern monsoon) is evident (Figure 4.3). At the latitude of Blythe, California, in the central Mojave Desert, August and September have the highest monthly average precipitation (Figure 4.3). In southern California west of the Sierra Nevada, winter-dominated precipitation is again the typical pattern (Figure 4.3). This is roughly the current seasonal distribution of precipitation in California. However, as the paleovegetation evidence discussed in the preceding sections suggests, this pattern has shifted significantly in the past. For example, at Lower Pahranagat Lake in the northeastern Mojave between 1,600 and 1,000 rcyr BP, middle to late summer precipitation may have increased significantly (Figure 4). Late spring and early summer rainfall also allowed water levels in marshes to

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

45

F I G U R E 4.2. Late Quaternary grass pollen records from east of the Sierra Nevada Mountains. Pollen records from the more arid side of the Sierra Nevada/Cascade mountain crest are more sensitive to even small variations in precipitation. These records compare pollen from both sediment cores and ancient woodrat middens. Periodicity in the occurrence of ancient woodrat middens has been shown in the past to reflect recurrence of climates favorable to woodrats. The pollen from these records show clear periodicity in grass abundance during the last 50,000 years. Some of these peaks are due to glacial advances or cooler climate episodes (
40,000, 34,000, 21,000, and 10,600 years ago), and others are due to increased summer precipitation (
9,000 and 1,600 years ago). Diagram modified from one in Wigand and Rhode (2002).

remain deeper well into the summer months, promoting a greater abundance of floating aquatic plants (Wigand and Rhode 2002). Episodes of increased grass between 4,000 and 2,000 rcyr BP suggest that the cold, winter-wet climate characteristic of the northwestern Great Basin may have penetrated as far south as the northern Mojave during the Neopluvial (Figure 4.4). Even the warm, wet episode of precipitation between 800 to 700 rcyr BP is evidenced in the Lower Pahranagat Lake record, suggesting that the May/June increase in monthly average precipitation currently seen in the northern Great Basin may also have extended into the northern Mojave during that period.

Conclusions In summary, late Cenozoic grass populations in the West have responded to changes in both long- and short-term variations in climate. On the millennial scale of glacial to interglacial

46

HISTORY

climates these variations have coincided with major shifts, not only in grass abundance and species composition but also in vegetation community composition and structure as a whole. This process was driven by variations in solar insolation (amount of solar radiation reaching the earth), resulting from ongoing changes in the earth’s orbit and tilt as well as variations in solar output (radiation output by the sun), and mountain building. During the Holocene, variations primarily in grass abundance (rather than in species composition) have resulted from changes in precipitation on the scale of decades to centuries caused by variations in solar output. During the last 200 years the history of grass has been complicated by the impact of Euro-American activities and by their animals, primarily cattle and horses. For example, east of the Sierra Nevada/Cascade crest the impact of the horse may have been felt as early as the 1680s, when Spanish horses escaped during the Pueblo uprising. By the 1740s Native American

FIGURE 4.3. Average monthly precipitation records for weather stations near some of the pollen localities discussed in the text. These provide an indication of the differences in the precipitation pattern within California. (Source: Western Regional Climate Center, Desert Research Institute, UCCSN Reno, NV, www.wrcc.dri.edu/climsum.html).

F I G U R E 4.4. Grass-to-sagebrush and grass-to-total-terrestrial (grass pollen percentage) pollen ratios for a pollen record in the northern Mojave Desert northeast of Las Vegas, Nevada. It is the highest resolution pollen record currently published for the western Hemisphere and spans the latest Holocene. Sample spacing is decadal (Wigand, unpublished data).

horse cultures were well established, and grazing impacts upon native grasses were in full swing. These impacts, together with fire, have resulted in significantly changed habitats. Not only have vegetation community compositions changed, but their structure as well. Only in a few isolated places have relatively intact native plant communities with rich grass understories survived the onslaught of cheatgrass and other Eurasian invaders (Figure 4.5). However, with the onset of global warming, even these few communities may be in jeopardy.

Rancholabrean Mammals of California and Their Relevance for Understanding Modern Plant Ecology Stephen W. Edwards

The term Rancholabrean is rightly associated with the spectacular latest Pleistocene fossil assemblage recovered from the Rancho La Brea tar pits in the Los Angeles Basin. But that is only one among many assemblages of late Pleistocene age in California, not to mention a host of localities producing isolated mammal fossils. Savage (1951) defined Rancholabrean as a continent-wide mammalian provincial age, recognized by the appearance in North America, approximately 150,000 years ago, of Bison, an immigrant from Asia (Woodburne and Swisher 1995), and ending with the demise of “charismatic” megafauna such as mammoths and sabrecats at the end of the Pleistocene. By Rancholabrean time the California flora consisted almost entirely of genera (and, at least in woody plants, also of species) that are still extant in the state. There were differences in distribution, and there were combinations of plant taxa that no longer occur together, but the flora and vegetation were nevertheless distinctly Californian and recognizably modern (Edwards 2004). The fossil flora recovered from the Rancho La Brea tar pits, for example (Akersten et al. 1988; Templeton 1964; Warter 1976) is basically of central-southern California aspect. That paleoflora, as well as many others, show that cismontane California was not a Pleistocene arctic

48

HISTORY

waste, but experienced a temperate-maritime climate that served as a refuge for evergreen hardwoods (Johnson 1977). The Rancholabrean fauna, looking at first encounter like something belonging in Africa or Asia, was adapted to a California flora. Mastodons browsed trees and shrubs, horses grazed needlegrass and the other familiar perennial grasses, which must have been more abundant and productive in the more mesic Pleistocene climate than they are in today’s fully Mediterannean regime. Although human hunters probably sealed the fates of many megafaunal species, it is likely that a climatically induced type conversion from lush Pleistocene grasslands to arid Holocene landscapes dominated by native annuals had already diminished megafaunal populations. The relative contributions of climate change and hunting are still debated, but the result is clear. The Rancholabrean fauna, with so many large animals, representing the zenith of the Age of Mammals in North America and what could be considered the true fauna of California, was wiped out forever in the space of about 2,000 years. While the fauna that had coevolved for millions of years with the flora and vegetation of California thus disappeared, the flora and vegetation fared better. Though there are many gaps in the fossil record and in particular herbaceous taxa are very poorly represented, all indications are that late Pleistocene woody species persisted to make up modern associations, and the same is true at least of herbaceous genera. Therefore, when it is evident that a modern native plant species is well adapted to grazing and/or browsing or even to fire, it makes sense to recall the relations that were forged through millions of years of coevolution between late Cenozoic (and especially late Pleistocene) mammals and California native plants. Adaptations affording grazing tolerance among living grassland plants include, among others, widespread occurrences of toxic or unpalatable defense compounds (e.g., Fabaceae, Madiinae, Ranunculaceae, Scrophulariaceae) and basal (or near basal) meristems (probably in all Poaceae). Their presence through numerous genera within families reflects the deep histories of such adaptations. It is unlikely that they

The Larger Rancholabrean Mammals of California E DE NTATA (G ROU N D S LOTH S)

FIGURE 4.5. Grassy, semiarid woodland east of the Sierra Nevada crest.

Today overgrazing and fire have decimated much of the grass cover of the Great Basin at lower elevations. At intermediate and higher elevations, however, precipitation is high enough for native grasses to recover from these impacts, and compete with invading plant species. This photo was taken in the Virginia Mountains just northeast of Reno, Nevada. The grasses are primarily bluebunch wheatgrass (Pseudoroegneria spicata, formerly Agropyron spicatum). Pollen records indicate that prior to the arrival of Euro-Americans most of the vegetation communities in the Great Basin, ranging from the lower sagebrush through the semiarid woodland and upper sagebrush communities, were much richer in grasses. Photograph by P. Wigand.

Megalonyx jeffersoni: Variously reported as black bear to oxsized, and consistently regarded, on the basis of the simplicity of its grinding teeth, as a browser. Nothrotheriops shastensis: The smallest of the Californian ground sloths, grizzly bear–sized at most. Dung deposits in Utah, Nevada, Arizona, and New Mexico have been attributed to this animal. Studying deposits in Arizona, Hansen (1978) identified 72 genera of plants in Nothrotheriops dung. The most abundant taxa were Sphaeralcea ambigua (52%), Ephedra nevadensis (18%), Atriplex spp. (7%), Acacia greggii (6%), Cactaceae (3%), Phragmites communis (5%), and Yucca spp. (2%). A similar study by Thompson et al. (1980) showed a greater percentage of Ephedra (51%), with Rosaceae and Agave the other prominent components. Paramylodon harlani (Harlan’s Ground Sloth): Also known as Glossotherium h., these were the largest edentates in California, ox-sized and weighing up to 3,500 pounds. They were capable of standing on their hind feet and manipulating tall vegetation with massive forelimbs armed with large claws. Their simple, peglike grinding teeth have been interpreted as useful for grazing grass as well as for browsing, but definitive research to elucidate their diet has not been done. U R S I DAE (B EAR S)

suddenly appeared in response to decreased grazing pressures of the Holocene after the megafauna disappeared. But they would have constituted excellent preadaptations for fire, more intense in the arid Holocene, and grazing by hypsodont microtine rodents, rabbits, hares, grazing avifauna, and elk. Much basic research needs to be done on dietary habits of Rancholabrean mammals, but enough data are available (Edwards 1996, 1998) for preliminary indications to be given. This will be done, telegraphically, in the following list of the larger Rancholabrean mammals of California. There is no way accurately to estimate population numbers of extinct mammals, but two lines of evidence suggest that large Rancholabrean herbivores were very abundant and thus must have had dramatic and pervasive impacts on vegetation and flora. First, the array of carnivorous mammals and large scavenging birds (two condors, four vultures, one teratorn, and six eagles at Rancho La Brea) equals or exceeds that of the East African Pleistocene, which exceeded that of the game reserves of East Africa today. There must have been plenty of meat on the hoof to support this diversity. Second, Davis and Moratto (1988) found abundant spores of the dung-consuming fungus Sporormiella in sediments dated to 11,600 rcyr BP at Exchequer Meadow in the Sierra Nevada. They noted: Sporormiella spores are abundant in modern sediments only where introduced grazing animals are plentiful, and they are even more profuse in sediments older than 11,000 yr B.P. in several sites. (Davis and Moratto 1988: 146)

Arctodus simus (short-faced bear): These huge animals, outsizing polar bears and having longer limbs, making them capable of bursts of greater speed, were perhaps the most powerful predators of the Pleistocene world. Convergences in skull structure with felids and cheek teeth less modified for omnivory than those of other bears suggest that the large ungulates of the day may have been prime targets (Kurten 1967; Shaw and Cox 1993). Ursus americanus (black bear): These small bears (by Pleistocene standards) are omnivorous, with emphasis on plants and insects. Ursus arctos (grizzly bear): The diets of these animals are well understood from studies in the northern United States and Canada. They are omnivores, feeding on everything from limpets in the intertidal zone to glacier lily bulbs in the subalpine, with copious supplementation from fish and mammals. In 1862 Brewer (1966) observed the extensive rototilling effects of these animals resulting from digging for bulbs in the south coast ranges. Whether in the Holocene (when it is likely that California’s geophyte flora attained its greatest prominence) or earlier, bulbs must always have been a major food source for grizzlies. CAN I DAE (D O G S)

Canis dirus (dire wolf): This wolf was similar in size to the modern gray or timber wolf but had a larger head, stronger jaws, more massive teeth, and a heavier build, but shorter lower limbs. It was presumably a pack-hunter like its modern relative,

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

49

and thus capable of exciting and stampeding large ungulates, thus increasing the impacts of the latter on the landscape. Canis latrans (coyote) Canis lupus (gray wolf) F E LI DAE (CATS)

Felis concolor (puma): Pumas in California today will consume practically any other mammal, but in terms of kill frequencies they are deer specialists, and that adaptation makes sense for the Pleistocene, given the relatively gracile skeleton (though powerfully muscled) of these extremely nimble cats. Homotherium serum (scimitar cat): These lion-sized cats had smaller sabers but longer limbs than their better-known relative Smilodon and were probably more cursorial. Evidence from Friesenhahn Cave in Texas suggests that young mammoths were a favorite prey (Turner 1997). Lynx rufus (bobcat): Like pumas, bobcats consume a wide range of prey. They are fully capable of killing adult deer and commonly do so. Miracinonyx trumani (American cheetah): Fossils of these gracile cats have been found in Nevada. These are the rarest of fossils in a family that is rare in fossil form to begin with. It is reasonable to suspect that fossils of Miracinonyx will show up in California, especially because prey animals capable of speeds far surpassing any other cats, namely pronghorns, were present in the California Rancholabrean. Panthera leo atrox (American lion): This was the largest felid of the Rancholabrean. Males were about 25% larger than African lions. Grayson (1991) speculates that these were animals of open country, since they are absent from the “forested east,” though the eastern United States was not as forested then as it was historically (Guthrie 1990). Panthera onca ( jaguar): Jaguars originated in Eurasia or North America and later spread south across Panamania to South America. Jaguars persisted in California into historic time, and “roamed the South Coast Ranges between San Francisco and Monterey up to at least 1826” (Jameson and Peeters 1988). In South America their preferred prey include peccaries and tapirs, and since both were represented in Rancholabrean California, one may speculate that jaguars hunted similar species in the north. Smilodon fatalis (sabrecat): This saber-toothed cat was about the size of a female African lion. Stock and Harris (1992) suggest on the basis of the skeleton that these animals were less capable than other big cats of the time in chasing down prey and hence would have depended more upon stalking and ambush. One might speculate, then, that they did not specialize in the fleeter ungulates. TAYAS S U I DAE (P E C CA R I E S)

Platygonus compressus (flat-headed peccary): This animal is a little larger than the extant peccary of the desert Southwest, with longer limbs. Although Grayson (1991) considered this an animal of open habitats, its dentition is low-crowned and adapted for browsing. On analogy with living peccaries,

50

HISTORY

Platygonus may have been an opportunistic feeder, browsing but also rooting for bulbs and taking small animals and carrion. According to Simpson (1980), Platygonus is questionably distinct from Catagonus, the living Chacoan peccary of South America that was only discovered in 1975. Catagonus lives in dry thorn forest and feeds on cacti, bromeliad roots, fruits, and forbs. CAM E LI DAE (CAM E LS)

Camelops hesternus (western camel): This was a large camel, with limbs up to 25% larger than those of the modern dromedary (Webb 1965). Its cheek dentition is higher-crowned than that of modern tule elk, mixed grazer-browsers that consume about 50% grass. Camelops cheek teeth often preserve cementum, which gives extra support for heavy mastication. Wear profiles (mesowear) of cheek teeth are consistent with a mixed-feeding strategy. Dompierre and Churcher (1996) concluded on the basis of comparisons of snout shapes in ungulates that Camelops was a mixed grazer-browser. North African dromedaries are similarly mixed grazer-browsers. Akersten et al. (1988) examined dental boli impacted in molars from Rancho La Brea and found that they contained Bouteloua, Bromus, Festuca, Hilaria, and Sporobolus, these grasses collectively amounting to 10.7% of the identifiable remains. The rest of the identifiable remains were dicotyledonous, but overall, 80% of the material in the boli was unidentifiable as to monocot vs. dicot. Hemiauchenia macrocephala (large-headed llama): A slenderlimbed, long-legged llama with relatively high-crowned cheek teeth, this species has often been interpreted as a swift, opencountry grazer. The native living camelids of the Andes, vicuñas and guanacos, are open-country animals with diets focused on perennial bunchgrasses and forbs. Hemiauchenia is in or close to their ancestry (Webb 1965). Dompierre and Churcher (1996) interpreted Hemiauchenia as a browser based on premaxillary morphology, while MacFadden and Cerling (1996) decided this llama was a grazer, based on carbon isotopes in dental enamel. More recent assessment of carbon isotopes by Feranec (2003) led to the characterization of Hemiauchenia as an intermediate feeder with a preference for browse. CE RVI DAE (DE E R AN D E LK)

Cervus elaphus (elk): Observations of tule elk at Pt. Reyes National Seashore reveal that these large deer are mixed grazer-browsers, consuming about 50% grasses and 50% other, mostly forbs (Gogan and Barrett 1995). Their cheek teeth are less high-crowned than those of cattle, and they lack supporting cementum. Odocoileus hemionus (mule deer): Deer are essentially browsing animals that take very little grass, when the foliage is young and tender. ANTI LO CAP R I DAE (P RONG HOR N S)

Antilocapra americana (pronghorn): Historically, pronghorns are browsers, and this is an important point, because in

popular literature they are regularly pictured as grazers that impact grasses directly and substantially. In fact they prefer shrubs and forbs; like deer, they consume grasses only sparingly and usually when the foliage is young (Yoakum 1980). Capromeryx minor (dwarf pronghorn): This is a diminutive creature, less than two feet (0.6 meter) tall at the shoulder. Anderson (1984: 76) reports it as a grazer, but this is unlikely. High surface-to-volume ratio would probably have necessitated a focus on higher protein values available in browse. If one can take Thompson’s gazelle, an animal of similar size, as an analogue, Capromeryx would have had no more interest in grasses than Antilocapra does.

Bovidae (Cattle, Sheep, and Their Relatives) Bison antiquus (Ice Age bison): These animals were morphologically very similar to living Bison bison and some authorities prefer to classify antiquus as a subspecies. On the basis of plant tissues in dental boli from Rancho La Brea, Akersten et al. (1988) considered this species a browser (only 13.4% monocot). However, the relatively wide muzzle is more characteristic of grazers; mesowear on cheek teeth is consistent with a grazing emphasis; dung attributed to this animal from Cowboy Cave, Utah (Hansen 1980), is dominated by grasses; skeletal evidence suggests bison visitied La Brea only for one month in spring (Jefferson and Goldin 1989) when browse may have been preferred locally; and isotopic analyses reported by Feranec (2004) and Feranec and MacFadden (2000) for Bison in Florida point to a diet ranging from grazing to mixed with an emphasis on grazing. The consensus among those who have studied fossil Bison is that these were predominantly grazing animals. Bison latifrons (giant bison): The largest bison that ever lived, with horn cores up to 7.5 feet (2.3 meters) across, became extinct perhaps as much as 12,000 years before the last antiquus. McDonald (1981) studied the cranial morphology and concluded these were browser-grazers; that is, they preferred browse but did some grazing. In terms of vegetation impact, an analogy can be drawn with moose. McInnes et al. (1992) have shown that exclusion of these large ungulates can lead to decline of herbaceous cover and invasion of clearings by shrubs and trees. Euceratherium collinum (shrub ox): Some investigators have classified these animals, about elk- or cattle-sized, as grazers; but, according to Mead et al. (2003), dung pellets from sandstone rock shelters in the Glen Canyon region of Arizona have been attributed to Euceratherium and contain mostly browse species such as Quercus, Artemisia, and Chrysothamnus. Mesowear on cheek teeth is consistent with a mixed diet emphasizing browse. Oreamnos americanus (mountain goat): Fossils of these animals have been recovered only in the far north, in Lassen and Shasta counties. Ovis canadensis (bighorn sheep): Even in the Pleistocene this species was probably focused in transmontane California.

Symbos cavifrons (woodland musk ox): This species was widespread, from Alaska to Texas, though so far in California it has been found only in Modoc County. Perhaps it is to be expected at higher elevations in the mountains elsewhere. Symbos was bison-sized, but more slender. Its dietary adaptations have not been studied. EQU I DAE (HOR S E S)

Equus conversidens: This species was apparently restricted to the desert counties east of the Transverse Ranges. Equus cf. occidentalis (Western horse): According to Harris and Jefferson (1985) the large sample at Rancho La Brea affords a reconstruction about 4 .5 feet (1.4 meters) tall at the shoulder. The cheek teeth of these animals are as hypsodont as those of any other mammals known. Although isotopic studies suggest some browsing occurred, cheek teeth this hypsodont are intelligible only as adaptations for habitual grazing of grasses. TA P I R I DAE (TAP I R S)

Tapirus (tapir): Jefferson (1989) reports two species of tapir in the Rancholabrean of California. Both occur along the south coast, but T. californicus also has been found at several localities in the central Sierra foothills, while T. merriami has been recovered in Alameda and Contra Costa counties. Living tapirs are wetland/woodland/forest browsers, and the low-crowned dentitions of the fossils resemble those of living species. According to Graham (2003), living tapirs of the New World are very selective browsers that show some preference for colonizing plants that are low in toxic defense compounds. P ROBOSCI DEA (E LE P HANTS AN D MASTOD ONTS)

Mammut americanum (American mastodon): These were medium-sized elephant relatives, 6 to 9 feet (1.8–2.7 meters) high at the shoulder. Their bunodont molars, lacking supporting cementum, were adapted for browsing. Mammuthus columbi (Columbian mammoth): As large as African elephants, these massive animals had high-crowned cheek teeth with closely packed lamellae intermediate in number between those of African and Indian elephants, both of which are browser-grazers. Dung ascribed to mammoths from Bechan Cave in Utah (Haynes 1991) contained over 95% by weight grasses, sedges, and rushes. Dung from Cowboy Cave, also in Utah (Hansen 1980), contained more than 95% grasses, mostly Sporobolus. At Grobot Grotto the dung contained mostly Phragmites. These mammoths have traditionally been interpreted as open-country animals. The immense size and length of tusks on some males, nearly doubling the overall length of the animal, surely would have limited their mobility in a forested environment. Judging from behavior of modern African elephants, California mammoths may have opened up vegetation by trampling and tree felling.

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

51

Effects of Megafauna on California Grasslands Grazing, browsing, and trampling are different but overlapping activities, and the habits of herbivorous mammals sort out on a continuum. Horses, rabbits, and elephants graze, browse, and trample with different foci and intensities. The trampling element is increased as animals are harried by predators, and Rancholabrean California had one of the most awesome arrays of predators known since the days of Tyrannosaurus rex. Grazing, browsing, and trampling have all been demonstrated by contemporary studies to contribute to protection of grassland from encroachment by woody plants. It is likely that such impacts by Rancholabrean megafauna helped to maintain large areas of grassland even during glacial periods when cooler, more mesic conditions favored forests. The origins of California grasslands ultimately cannot be fully understood without considering the fauna that evolved with them. Five million years and more of that history involved a diverse megafauna, a fauna growing larger and more diverse with time. Only 10,000 years involved the anomalously depleted faunal remnants of the Holocene. Study of Pleistocene megafauna is no idle pursuit. Parkman (2002) has made a convincing case that brilliantly polished surfaces high on raised seastacks along the Sonoma coast were mammoth rubbing stations. He has also suggested that some of California’s extant vernal pools may have originated as wallowing basins of Rancholabrean herbivores, and this hypothesis should be investigated. As for the native plants of California’s grasslands, it is likely that they preserve a substantial genetic legacy of their relations with magnificent animals that grazed, browsed, and trampled them not really so long ago.

Species Composition at the Time of First European Settlement Paula M. Schiffman

A very basic question nags at ecologists and habitat managers: What was the species composition of California’s grasslands like at the time of European contact? More specifically, which species were dominant? This question exists because the grasslands were colonized by several invasive plant species soon after European contact (Hendry 1931; Spira and Wagner 1983; Sauer 1988; Blumler 1995; Mensing and Byrne 1998, 1999), and these species rapidly became incorporated into natural landscapes. There are no descriptions of grassland species composition from that early time period, and, amazingly, the invasion went unnoticed. When people finally began to record detailed vegetation accounts in the mid-1800s (e.g., Cronise 1868), invasive plant species were already geographically widespread and ecologically dominant.

Early Records The historical spatial extent of California’s grassland area was enormous (5.29 million hectares; Barbour and Major 1988: 3 – 10). That such a massive invasion could have occurred,

52

HISTORY

without anyone documenting it, is remarkable and perplexing. The Native Americans who lived in this ecosystem for millennia used oral communication to share information. When European diseases and brutality decimated their populations (Preston 2002b), their in-depth knowledge of historical grassland species compositions and community dynamics was largely lost. The first European settlers were not naturalists, and from the very start they tried to dominate, rather than describe, their vast new environment. They simply viewed it as land where opportunities for livestock grazing and cultivation abounded. Moreover, California’s grassland-covered plains and valleys were subtle landscapes that, except in the spring when expanses of colorful wildflowers appeared, were dry, stark places that lacked the visual drama of California’s wave-crashed coastline or imposing mountain ranges. Perhaps the minutiae of grassland species composition seemed trivial when juxtaposed with such huge, wide-open landscapes. Harrison (1982) has noted that as a group, early explorers and settlers were unusually reticent to record their observations and impressions of the North American prairie landscapes that they encountered. He describes this odd phenomenon as “verbal blindness.” Although detailed ecological accounts do not exist, a few early observers did record general descriptions of California grasslands. The writings of Juan Crespí, a Spanish priest who journeyed from Baja California to San Francisco Bay in 1769–1770 and then from San Diego to Monterey in 1770, were full of descriptions of places with “everything very grass-grown” (Crespí 2001: 309). Spanish mission period journals of other early Europeans such as Francisco Garcés, Pedro Fages, Juan Bautista de Anza, Pedro Font, Josef Joaquin Moraga, Francisco Palou, George Vancouver, Georg von Langsdorff, and others also commented on the productive pastoral environments that they encountered (Coues 1900; Priestley 1937; Bolton 1930, 1931, 1966; Paddison 1999). These observers’ accounts of the vegetation were extremely general and it is clear that descriptions such as “good grass,” “much grass,” and “level and grassy” terrain were not used in a strict taxonomic sense. Rather, they were general portrayals of low green vegetation that could be exploited for livestock grazing. In this context, simplifying a diverse assemblage of species — which would have included many graminoids, forbs, geophytes, and even subshrubs — as “grass” made sense. Even if some of the plants were not all grasses, they grew alongside grasses and they were consumed by grazing livestock as well as herds of native elk and pronghorn antelope. References to California’s grass-covered landscapes continued well into the 1800s, as did the botanical imprecision. For example, in an 1847 observation, Edwin Bryant, an American journalist, noted that “[T]he varieties of grass are greater than on the Atlantic side of the continent, and far more nutritious. I have seen seven different kinds of clover, several of them in a dry state, depositing seed upon the ground so abundant as to cover it, which is lapped up by the cattle and horses and other animals” (Bryant 1848: 448). Native clovers (Trifolium spp.) and grasses co-occurred, and

Bryant lumped these two completely different and unrelated taxa together into a single functional group. However, early observers were not so unobservant or naïve to think that grasslands consisted only of grasses. It is evident from their journal accounts that native forbs were abundant in the grassland landscapes through which they passed. For example, on May 7, 1770, when traveling near the Santa Ynez River in what is now Santa Barbara County, Crespí described “a great plenty of white, yellow, red, purple and blue blossoms: a great many yellow violets or gillyflowers such as are planted in gardens, a great deal of larkspur, a great deal of prickly poppy in bloom, a great deal of sage in bloom; but seeing all the different sorts of colors together was what beautified the fields the most” (Crespí 2001: 711). Early descriptions like this one did not include nearly enough information for us to reconstruct the species composition of these landscapes accurately today. Still, it is quite clear that spring-flowering forbs were important, though ephemeral, ecosystem constituents. A little more than a century after Crespí, naturalist John Muir’s writings included reminiscences of great profusions of annual wildflowers in the mid-1800s. He wrote, “The Great Central Plain of California, during the months of March, April, and May, was one smooth continuous bed of honeybloom, so marvelously rich that, in walking from one end of it to the other, a distance of more than 400 miles, your foot would press about a hundred flowers at every step. Mints, Gilias, Nemophilas, Castillejas, and innumerable Compositae were so crowded together that, had ninety-nine percent of them been taken away, the plain would still have seemed to any but Californians extravagantly flowery” (Muir 1894: 339). Although Muir mentioned several annual taxa, his descriptions primarily conveyed a vivid sense of biodiversity rather than an ecologically meaningful accounting of community composition. However, he did remark that “all of the ground was covered, not with grass and green leaves, but with radiant corollas” (Muir 1894: 342).

Clements’ Influence and Recent Interpretations Because of their ephemeral nature, the ecological importance of these annual and perennial forbs was not widely recognized. Frederic E. Clements’ (1934) relict analysis indicated that the perennial bunchgrass, Nassella pulchra (Stipa pulchra and S. setigera; Hamilton 1997a), had been the historical dominant in California’s grasslands. He interpreted the prominence of N. pulchra in some relict grassland fragments as key. Clements’ reputation as a leading twentieth-century ecologist led to the acceptance of his hypothesis among California biologists (e.g., Piemeisel and Lawson 1937; Munz and Keck 1959; Burcham 1961; Heady 1988). However, the relatively mesic and periodically burned fragments that were Clements’ exemplars did not constitute a good representation of the wide range of habitats that supported grassland vegetation in California. In addition, as Hamilton (1997a) convincingly explains, the scientific basis for Clements’ hypothesis was shaky because it relied upon little real data

and several erroneous assumptions. Nevertheless, relatively recent references that discuss California grassland composition and ecology in detail still usually identify N. pulchra as the likely historically dominant species (Heady 1988; Schoenherr 1992; Holland and Keil 1995), and field studies, particularly those focused on conservation and restoration, have continued to give more attention to N. pulchra than to any other native grassland species. However, it has also been suggested that several other perennial grasses (e.g., Poa secunda, Leymus triticoides, Melica spp., Muhlenbergia rigens) were historically more important community constituents in some environments (Keeley 1990; Heady et al. 1992; Holland and Keil 1995; Holstein 2001). But what about the historical importance of forbs? Historical accounts, though limited in ecological detail, did clearly point to an impressive diversity and cover of colorful spring wildflowers. Even Clements recognized perennial forbs as “subdominants” and stated that “even more typical are the great masses of annuals, representing more than 50 genera and several hundred species” (Clements and Shelford 1939: 288). In fact, his description of the springtime vegetation of 1935 bore considerable resemblance to the much earlier descriptions of Crespí and Muir: “the carpet of brilliant blues, oranges, and yellows covered an area approximately 50 miles wide and 100 miles long” (Clements and Shelford 1939: 288). Like other observers, Clements noted the abundance of native annuals and then glossed over their identities as if they were unimportant. Despite his clear acknowledgement of their tremendous percent cover, these plants’ transient nature indicated to him that they had little real ecological value. Clements’ endeavor to draw ecological linkages between California’s grasslands and those of the midwestern United States demanded that he emphasize perennials, especially grasses (Hamilton 1997a), despite the ubiquity of so many annual forbs. The ruderal nature of annual plants (Grime 1979a) was another feature of California’s native forbs that precluded Clements from considering them to be ecologically important. By definition, he viewed climax communities as generally stable associations of species that developed through succession (Hamilton 1997a). So, although vegetation made up of weedy, invasive, non-native annuals including Avena, Bromus, Hordeum, Festuca (Vulpia), and Erodium was considered a “proclimax” community, a stable community dominated by an association of disturbance-adapted native annual plants completely violated his theoretical framework and, therefore, went unrecognized. Today, it is well known that native forbs repeatedly reappear on the same sites for decades, though their covers vary with annual rainfall amounts. In addition, soil disturbances by small burrowing mammals, herbivory, periodic fires, and environmental management by Native people were integral ecosystem processes that had compositional consequences including the promotion of annuals (Blumler 1992; Hobbs and Mooney 1995; Painter 1995; Schiffman 2000; Reichman and Seabloom 2002; Keeley 1990, 2002; Anderson 2005). Surely,

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

53

Monterey Monterey San Luis Obispo Santa Barbara Orange Riverside

ELK HAS CAR SED STA ROS

The grasslands are plotted by code in Figure 4.7.

Alameda San Mateo Madera

LAW JAS JOA

NOTE :

Mendocino Sacramento Solano Marin

HOP COS JEP REY

Hopland Research and Extension Center Cosumnes River Preserve Jepson Prairie Reserve Point Reyes National Seashore Lawrence Livermore National Laboratory Site 300 Jasper Ridge Biological Preserve San Joaquin Experimental Range Elkhorn Slough National Estuarine Research Reserve Hastings Natural History Reservation Carrizo Plain National Monument Sedgwick Reserve Starr Ranch Sanctuary Santa Rosa Plateau Ecological Reserve

County

Code*

Grassland

Mean rainfall (cm/yr) 94.0 44.2 47.5 165.5 36.8 65.2 48.6 55.2 53.0 14.5 38.0 38.1 48.0

Latitude 39o00’ 38o25’ 38o15’ 38o05’ 37o42’ 37o24’ 37o05’ 36o48’ 36o22’ 35o10’ 34o45’ 33o37’ 33o31’

566 932 101,000 2,358 1,616 3,434

2,828 481 1,806

2,165 16,160 634 25,907

Area (ha)

TA B L E 4.1 Characteristics of the 13 Relict Grasslands Included in the Ordination

147 399 346 201 266 395

262 439 248

396 216 228 483

Total native species

9.5 6.0 2.9 6.0 4.5 6.3

4.2 6.4 3.6

8.6 5.6 4.0 9.7

Percent perennial grasses

26.5 57.4 58.6 53.2 41.0 48.4

66.8 44.0 65.7

43.7 50.9 52.2 29.8

Percent annual forbs

the endurance of native annual forbs in California’s grasslands and their apparently adaptive interactions with other organisms and processes reflects their historical ecological significance. In recent years, researchers have used evaluations of historical accounts, floristic surveys, relict analyses, and modern experimental and comparative findings to propose alternatives to Clements’ vision of California’s grassland species composition. Several of these reconstructions have suggested that annual plant species, rather than N. pulchra or other perennial grasses, had been the most ecologically important species in much of southern California and relatively arid inland environments including the Central Valley (Talbot et al. 1939; Twisselmann 1967; Wester 1981; Blumler 1995; Holstein 2000, 2001; Schiffman 2000, 2005). In more mesic areas, annual forbs still constituted a diverse group of plants. Sadly, it is now impossible to truly understand the ecological roles of individual plant species at the time of European contact. Clues to the historical past have been blurred by massive changes caused by the contamination of California’s grasslands by invasive non-native annuals and a wide range of human activities including cultivation, livestock grazing, fire suppression, eradication of the grizzly bear (a keystone species), and habitat fragmentation. So, the degree to which the ecological dynamics in relict grasslands resemble those of historical ecosystems is somewhat unclear. One thing is quite certain, however. These habitats continue to support very large numbers of native species, particularly forbs, just as they did when Europeans first encountered them.

A Relict Analysis Relict grassland floras typically include hundreds of native species in addition to grasses. Therefore, a study of relict floras that focuses on grassland plants of all forms should yield historically meaningful results. For example, this approach can be used to estimate the degree to which the native species compositions of historical grasslands in California resembled each other. Did regional differences in latitude, proximity to the Pacific coast, and rain shadow–producing hills mean that the grasslands of the northern coast or Sacramento Delta bore little resemblance to those of the San Joaquin Valley or southern California? They undoubtedly had some species in common, but how similar were these floras? Were they as monolithic as much of the literature has implied? To address these questions I have compared the native floras of 13 different relict grassland preserves in California (Table 4.1). Comprehensive plant species lists available for each of the preserves were the data sources for the study. The boundaries of these preserves encompass other vegetations in addition to grasslands (e.g., wetlands, chaparral, oak woodlands, riparian forests, and coniferous forests), and they frequently intergrade. Therefore, grasslands typically share some species with adjacent communities, and the communities themselves can be difficult to differentiate. Because there is ambiguity about the definition of “grassland,” my relict

F I G U R E 4.6. Frequency distribution of annual forbs, perennial grasses, and other species in the 13 relict grasslands.

analysis was limited to low-stature native plants that could be considered to be grassland species, at least in a broad sense (grasses, graminoids, annual, biennial, and perennial forbs, geophytes, and subshrubs as indicated by species descriptions in Hickman 1993). Trees and shrubs were excluded from the analysis, as were their parasites and nonwoody plants that, according to Hickman (1993), occur primarily in forests. Multiple taxa differentiated below the species level (subspecies and varieties) were also excluded from the analysis. The analysis encompassed a remarkable number of plant species and indicated that California’s extant grasslands are extremely important reservoirs of biodiversity. A total of 1,348 native grassland species occurred at the 13 sites surveyed. This means that these relict grasslands collectively support about 40% of the state’s total native plant species richness (Hickman 1993). Many of the species in this study occurred at only one or two of the sites, and most of these species were annuals (Figure 4.6). Surprisingly, just 1% of the species were present in all of the study’s grasslands. This small group of ubiquitous species consisted of a perennial herb (Achillea millefolium), 10 annual forbs (Amsinckia menziesii, Calandrinia ciliata, Claytonia perfoliata, Crassula connata, Eschscholzia californica, Lasthenia californica, Lotus wrangelianus, Lupinus bicolor, Mimulus guttatus, and Trifolium willdenovii), an annual graminoid (Juncus bufonius), and just one perennial grass (Nassella pulchra). These findings strongly indicate that, historically, California’s grasslands were habitat for an enormous number of different plant species and that the vast majority of them were not perennial grasses. PC-ORD (MJM Software Design, Gleneden, OR) was used to compute Jaccard distances (Magurran 1988) for the 13 grasslands and to ordinate them in two-dimensional space (Figure 4.7). Separation of the grasslands along the horizontal axis (axis 1) was strongly correlated with percentages of annual forbs and perennial grasses as well as with mean annual precipitation (Table 4.2). Latitude was most highly correlated with the distribution of grasslands along the

PLEISTOCENE AND PRE-EUROPEAN GRASSLANDS

55

TA B L E 4.2 Pearson Correlation Coefficients for the Two-dimensional Ordination of 13 Relict California Grasslands

Correlation coefficients (r) Variable Percent perennial grasses Percent annual forbs Mean annual precipitation Latitude Area Total number grassland species

Axis 1

Axis 2

0.918 0.905 0.670 0.233 0.372 0.138

0.198 0.114 0.218 0.819 0.259 0.245

F I G U R E 4.7. Ordination of 13 relict California grasslands produced

using Jaccard distances. Each grassland site is indicated by a threeletter code (Table 4.1). Variables correlated with most of the separation of grasslands along the two axes are plotted as vectors (percent annual forbs, percent perennial grasses, mean annual precipitation, and latitude).

vertical axis (axis 2). These correlation relationships were plotted as vectors (Figure 4.7). Although Nassella pulchra did occur in all of the grasslands included in this study, this very simple relict analysis of species presence/absence data strongly suggested that, historically, grasslands located in different regions of California had broadly differing species compositions. The ordination showed four geographically distinctive grassland groupings (Figure 4.7). San Joaquin Valley grasslands (represented by Carrizo Plain National Monument, San Joaquin Experimental Range, and Lawrence Livermore National Laboratory Site 300), were characterized by high proportions of annual forbs and relatively few perennial grasses. In contrast, the more mesic coastal prairies at Elkhorn Slough National Estuarine Research Reserve and Point Reyes National Seashore had high percentages of perennial grasses and fewer annual forbs. Latitude is associated with environmental and floristic gradients, and the grasslands of the southern, central, and northern coastal mountains (Starr Ranch Sanctuary, Sedgwick Reserve, Santa Rosa Plateau Ecological Reserve, Hastings Natural History Reservation, Jasper Ridge Biological Preserve, and Hopland Research and Extension Center) were

56

HISTORY

distributed as a generally latitudinal group with moderate proportions of annual forbs and perennial grasses. Finally, a floristically distinctive grassland type occurred in the Sacramento Delta (Cosumnes River Preserve and Jepson Prairie). These northerly grasslands also had moderate levels of annual forbs and perennial grasses. Unfortunately, these relict grasslands now also include many non-native plant species, and they no longer experience the disturbance regimes of the pre-European settlement environment. So it is impossible to estimate the importance (e.g., percent cover) of particular native species at the time of first European settlement. Moreover, historical percents cover of native plants, particularly annuals, would have varied with the year-to-year variation in winter rainfall amounts and other environmental factors. Despite the information limitations caused by such realities, this study’s comparative approach to species presence/absence likely provides an accurate perspective on the historical species compositions of California’s grasslands. If, however, composition is viewed more narrowly (for example, in terms of the presence/ absence of perennial grasses or the presence/absence of invasive non-native species), the relict grassland sites that this study found to be floristically different would seem much more homogeneous. It is clear that by fixating on a few perennial grasses and invasive species, California biologists have been distracted from what was actually an array of compositionally diverse and regionally distinctive historical grasslands.

FIVE

Native American Uses and Management of California’s Grasslands M. KAT AN D E R S O N

A widespread cause of grassland fires was man. With this easy-touse tool, the aborigine was able to create openings in the forest, convert forest to savanna, and change forest and brush to open grasslands. VOGL

Prehistorically, Native peoples around the world spent much of their waking day in grasslands or grassland-woodland ecotones: stalking and driving animals; harvesting edible seeds, bulbs, and greens; setting fires; and domesticating grain plants and ungulates (Vogl 1974). In fact, Homo sapiens’ hominoid antecedents may have evolved out of the grassland savannas of Africa. Human ecologist Paul Shepard says that grasslands are “central to this intricate history of the relationship among seeds, nervous systems, and minds” (Shepard 1996). Most if not all of the great grasslands of the world, from the Serengeti Plains to the prairie bioregion of the contiguous United States, were maintained with fires set by native peoples (Adams and McShane 1996; Wells 1965, 1970). In California it was no different. Fire was the most significant, effective, and widely employed management tool in grasslands. Early descriptions of Indians setting fire to grasslands abound in missionary, explorer, and military diaries. Diaries or early missionary journals offer us some of the richest resources in reconstructing Indian burning in grasslands. One of the best sources is Fray Juan Crespí’s diaries. He traveled with the Portola Expedition, and noted that, on July 21, 1769, near the Santa Margarita River and Ranch House areas, “we came onto very open, rolling knolls and tablelands of sheer soil, everything very overgrown with dry grass, though over most of this march we found it burnt off by the heathens” (Crespí 2001: 287). Upon entering the Salinas Valley on September 28 near what is now Greenfield, his party “followed the same valley and river by a level road, the grass all burned” (Bolton 1927: 199–201, cited in Gordon 1974:29). Proceeding northward from Chualar, Crespí wrote, “The soil is whitish and short of pasture on account of the fires set by the heathen” (Bolton 1927: 199 – 201, cited in Gordon 1974: 29). On October 16, 1769, Crespí’s party, near Corralitos, traveled “very near the beach, and the range of hills which follows, which has good pasture, although it has just been burned by the heathen” (Bolton 1927: 214–216, cited in

1974

Gordon 1974:29). On October 18 they traveled from Santa Cruz, along the coast toward Ano Nuevo: “we descended and ascended four deep watercourses [these would include Wilder, Laguna, and Scott Creeks]. . . . Only in the watercourses are any trees to be seen; elsewhere we saw nothing but grass that was burned” (Bolton 1927: 214–216, cited in Gordon 1974: 30). Crespí made note of many other areas that had burned grass, including what are now La Puente, City of Industry, Saugus, Goleta, Gaviota, La Honda Canyon, Guadalupe Lake, Price Canyon, Pico Creek, Soquel, Scott Creek, Waddell Creek, San Gregorio, San Andreas Lake, Del Monte Lake, and San Jose Creek. Humans migrated into the region that is now called California about twelve to fifteen thousand years ago. At European contact, perhaps 500 tribes, speaking approximately 100 languages, inhabited the state (Figure 5.1). Like indigenous peoples elsewhere, California Indians gathered numerous plants in the grasslands—such as our (modern) state flower, the California poppy, for food and medicine; soaproot for fish poison; and California fescue for thatch for winter houses. Native Californians hunted many kinds of animals in grassy openings, including rabbits for meat and fur blankets, elk for armor and elkhorn spoons, and deer for meat and clothing. Flat or gently undulating grasslands harbored major villages, temporary camps, and work stations. Tribes practiced controlled burning in and around their villages to decrease the risk of fires that might ignite their houses (Pilling 1978). With few trees or brush to obscure views, grasslands provided safety from surprise attacks by hostile tribes or grizzly bears. Granite outcrops on the peripheries of meadows or prairies often could host grinding slicks and bedrock mortars for processing seeds, berries, and other foods. Large grassy clearings, kept open by burning, also formed the most important sites for various California Indian games of physical dexterity, such as archery contests, double ball, shinny, games akin to racquet ball and football, and the hoop

57

F I G U R E 5.1. The territories associated with California Native American language groups.

Names in bold represent language families that comprise two or more languages and multiple dialects. Map drawn by Claudia Graham.

and pole game (Culin 1975). The goal posts or targets were located at the ends of long, grassy fields, as in archery, lacrosse, field hockey, and football today. Grasslands were attractive areas for hosting festivals and ceremonies and as trading grounds for bartering goods. Grasslands also meant plentiful grasshoppers, an important insect food up and down California. The Wintu and other tribes obtained them by burning off large grass patches (Du Bois 1935). All California tribes made fire with a fire-making kit composed of a hand drill or spindle and a hearth plate. Dry grass tinder was often part of the kit used by different tribes

58

HISTORY

(Foster 1944). Most tribes utilized a slow match or torch, which consisted of a tightly packed flammable material that would burn at one end as a constant flame for a considerable period of time. This gave Native Americans the technological capability to burn both small patches and extensive tracts of vegetation in a systematic fashion. Vegetation types that occur as continuous plant cover, such as grasslands, provided a continuous fuelbed. Therefore, human-set fires under the right environmental conditions could conceivably burn uninterrupted for a considerable distance (Anderson 2006). On rare occasions artists drew or painted these Indian-set fires.

Today Native people still visit grasslands to reenact age-old traditions by gathering many kinds of useful plants. The Wukchumni still harvest saltgrass for its delicious salt crystals from the southern San Joaquin Valley, the Karuk still gather wild iris from the Bald Hills for making cordage, and the Mono still dig the rhizomes of sedges found in high mountain meadows to be used as sewing strands for the making of coiled baskets.

The Ethnobotany and Ethnozoology of the Grasslands California’s grasslands were integral to the prosperity of each tribe, perhaps contributing more species to subsistence economies than any other single vegetation type. Native people knew that many kinds of medicinal and food plants required full sunlight. They also understood that many kinds of wildlife need open grasslands for performing mating rituals, facilitating movement, and heightening forage diversity and accessibility. Grasslands supported most types of human activities, providing the open space and plant species diversity needed for hunting, gathering, game playing, trading, dancing, feasting, and other activities and events.

Adornments, Clothing, and Regalia Tongva girls as well as individuals from many other tribes wove wildflowers into their hair, and they wore wreaths and boas of flowers that were gathered from the grasslands. These were worn at special festivals, dances, or other ceremonies (Dakin 1939; Gayton 1976: 83; Powers 1976). Wildflowers included many kinds of Brodiaea, Triteleia, and Dichelostemma, iris (Iris douglasiana), and common monkeyflower (Mimulus guttatus) (Barrett and Gifford 1933; Goodrich et al. 1980). For example, crowns of flowers were worn by young and old of both sexes in the Yokuts culture; armloads of flowers were plucked and danced with to special songs. Anthropologist Anna Gayton (1976) noted of the Yokuts: “Pleasure came from their beauty, their fragrance, and their indication of a plentiful seed harvest to follow.” Dyes for face paints came from the grasslands such as a brown sticky substance on the stems of sleepy catchfly (Silene antirrhina) used to paint Sierra Miwok girls’ faces (Barrett and Gifford 1933). Different tribes wore grasses in their pierced nasal septa and pierced ears. For example, young Sierra Miwok girls and boys wore flowers of non-native quaking grass (Briza minor) in their pierced ears, with the flower head forward, the stem passing through the hole (Barrett and Gifford 1933). Some tribes created women’s skirts of grasses, such as the Chukchansi, who pounded strands of an unidentified long grass called chulochul to make the front side of the skirt (Gayton 1948). Regalia of various tribes were made with grasses and wildflowers. C. Hart Merriam (1955) described the elaborate and beautiful regalia for the Big Head dance of the Wintu: “The head is covered with a grass mat fastened over the head, and in this mat or cap are placed many slender

willow sticks plumed with various colored flowers tipped with white feathers.” The woman’s basketry cap, worn in parts of southern California, was made on a foundation of deergrass (Muhlenbergia rigens) and worn to protect against the chafe of the pack strap (Kroeber 1951).

Animal Products A tremendous diversity of mammals that frequent grasslands provided food including deer, elk, pronghorn antelope, bears, gray squirrels, ground squirrels, and various kinds of rabbits. Deer were the most important mammal for food for many tribes. Grasslands supported large herds of deer and elk, flocks of birds, and, indirectly, mammal predators. John Hazelton wrote to his sister Sarah in 1850 about the diversity of meat for sale in the San Francisco markets: “it is filled with Wild Geese and Ducks of all kinds Deer antilope Elk, in any quantity. Last night I saw four Grisley Bares for sale. This is a grate Countrey. I saw one a few days since hanging up in market weighing 1460 lbs. This is nothing to what is seen some times” (Bloom 1958: 14). A more open country undoubtedly facilitated deer hunting, as numerous early anthropologists and settlers witnessed. The Sierra Miwok would set small fires in the hills around a meadow into which deer went. These men then kept building new fires. As the deer descended to the meadow, they approached the fires from curiosity, and concealed hunters shot them with bows and arrows (Barrett and Gifford 1933). Anthropologist Philip Drucker wrote of the Tolowa in northern California: “Informants maintain that near-by hills were kept clear of brush by annual burning; this also improved the grass, so that deer frequented such clearings and could be shot easily. Late spring, when the old fern was quite dry and the new growth just starting, is said to have been the time for burning off the hillsides to improve the hunting grounds” (Drucker 1937). Many tribes, such as the Ohlone, Tongva, Kitanemuk, Chumash, and Tubatulabal, used fire in conjunction with hunting rabbits (Harrington 1942; Voegelin 1938). Ethnobotanist Jan Timbrook and her colleagues have uncovered references in padres’ journals such as this one: “entered upon some mesas covered with dry grass, in parts burned by the heathen [Indians] for the purpose of hunting hares and rabbits” (Timbrook et al. 1993).

Basketry Basketry’s influence on the California landscape is long, as the craft is probably as old as the first human arrival into the state (Anderson 1993a). Hints of this time depth lie in archeological findings. Basketry fragments have been found in caves in western North America dating to 10,000 years ago (Orr 1956). Basketry was the quintessential craft of native California, being an indispensable item from early childhood to old age. If one were to visit a village in pre-contact California, baskets of many shapes and sizes would be in and around the homes. California Indian basketry reached a

N AT I V E A M E R I C A N U S E S A N D M A N A G E M E N T

59

significance and artistic development unsurpassed anywhere in the world (Kroeber 1925). Today weavers’ hands still sort, debark, split, trim, soak, and dye branches, stems, roots, and rhizomes of various grasses, shrubs, trees, ferns, and sedges to be woven into beautiful baskets. The vast array of plant parts woven into baskets is phenomenal, and each type of habitat produced unique materials. From coastal prairies and the edges of montane meadows came bracken fern (Pteridium aquilinum var. pubescens), prized for its rhizomes (Merrill 1923). Also called “black roots,” the rhizomes were used for decoration by such tribes as the Salinan, Ohlone, Miwok, Western Mono, and Washoe. Bracken fern patches were worked along the borders of mountain meadows, in coastal prairies, or high up in red fir forests. The most important basketry material gathered from grasslands was deergrass (Muhlenbergia rigens). Over half the tribes in the state gathered its flower stalks from valley grasslands, meadows, grassy openings in chaparral, oak savannahs, and open pine forests. It is still widely gathered today. Apparently, deergrass was an important associate in the purple needlegrass (Nassella pulchra) prairie that covered portions of the Central Valley and perhaps the South Coast Ranges in precontact times (Beetle 1947). Deergrass needs partial to full sunlight and is eliminated if too many shrubs or trees invade grasslands and meadows. This native bunchgrass was burned by the Luiseño, Western Mono, Foothill Yokuts, Kumeyaay, and other tribes to keep trees and shrubs from encroaching and to encourage abundant flowering culms, relished for the stuffing of coiled baskets (Shipek 1989; Anderson 1993b, 1996). The line drawn between agriculture and cultivation of “wild” plants is blurred when one examines the spectacular numbers of plant parts needed for basketry: A twelveinch (30 cm) Western Mono gift basket would require over 1,000 deergrass flower stalks; a gambling tray would demand about 3,000 flower stalks; and a cooking basket a quarter more (3,750 flower stalks) (Anderson 1993a) (Figure 5.2). The flower stalks of deergrass were incorporated in many other kinds of California baskets such as bread molds, eating dishes, burial baskets, acorn flour-sifting trays, storage baskets, coiled burden baskets, basket hoppers, and loosely woven bread baskets (Zigmond 1978; Harrington 1942; Bates 1982). This grass swells when immersed in water, helping to make cooking baskets watertight. Some native elders today talk about how deergrass can enhance the flavor of acorn mush being cooked in such a basket. Because of the difficulty in identifying grasses, some anthropologists relied on professional botanists to identify the grasses used in California Indian baskets (Barrows 1967). Interestingly, Sporobolus, a native perennial grass characteristic of the alkaline Central Valley grasslands or desert grasslands, is one of the few grass genera mentioned in the ethnohistoric and ethnographic literature other than Muhlenbergia. Powers (1976), who was visiting with Native Americans on the Tule River in the summer of 1871 or 1872, noted this genus in use. This could have been alkali sacaton

60

HISTORY

F I G U R E 5.2. Aida Icho, Wukchumni, with native walnut dice and a

gambling tray on a foundation of deergrass (Muhlenbergia rigens). About 3,000 flowerstalks were gathered to complete this basket. Courtesy of Yosemite National Park, Frank Latta Collection, Accession YOSE-4937.

(Sporobolus airoides), depending upon the elevation at which it was gathered. It was widespread at one time in alkali flats, especially on the west side of the San Joaquin Valley, and the culms are very similar to those of deergrass (Barry 1972; Stephen Edwards, personal communication, 2000). In describing California Indian baskets Jeanne Carr also mentions the genus Sporobolus (Carr 1892).

Construction Materials Native grasses were utilized by many different tribes to furnish tillers for both construction of dwellings and comforts used inside those dwellings. Uses include bedding, floor coverings, and thatching. Often these grasses were not scientifically identified. A Pomo bed was made by hollowing out the ground to some extent and placing dry grass or tule in the depression. Over this tule, mats, skin blankets, or both were placed (Barrett 1952). The Yana, who inhabited the upper Sacramento River valley and foothills east of the river itself, when hunting some distance from home, built temporary camping-out houses and thatched them with grass (Sapir and Spier 1943). The Wappo, who lived just above what is now Napa and Sonoma in the south to Cloverdale and Middletown in the north, thatched

their homes with grass. The Dry Creek and Cloverdale Pomo thatched winter houses with California fescue (Festuca californica). Giant wild rye (Leymus condensatus) has wide and long leaf blades. These leaves were used for thatching Salinan domed houses on a pole framework (Hester 1978; Harrington 1942). The Owens Valley and Mono Lake Paiute, the Cahuilla, and the Chumash thatched their houses with giant wild rye as well (Hudson n.d.: 20.017: 47; Bean and Saubel 1972; Timbrook 1990). The Modoc and Klamath built sun shelters and covered them with long grasses (Barrett 1910). Michahai and Chukaimina Yokuts snake doctors used cages made of an unindentified twined stiff grass to carry rattlesnakes for a planned rattlesnake ritual (Gayton 1948).

Cordage Cordage can be defined as “the twisting together of separate fiber strands into a single, long twined string or rope” (Mathewson 1985). Making of string or cordage is probably the oldest fiber art in America (Adovasio 1974). The most important cordage fiber plants used by California Indians were dogbane (Apocynum cannabinum) throughout California; the milkweeds (Asclepias fasciularis, A. cordifolia, A. eriocarpa, A. speciosa) used mainly in the central part of the state; iris (Iris douglasiana, I. innominata, I. tenuissima, I. tenax ssp. klamathensis, I. macrosiphon) in northwestern California, and yucca and agave in the southern deserts (Baker 1981; Mathewson 1985). Dogbane grows in grasslands near seeps and springs and seasonal wetlands; the milkweeds grow in grasslands or open woodlands and forests. Dogbane and milkweed have plants with stems that are composed of excellent “bast” fibers. These bast fibers were collected, extracted, and manufactured into many items including nets for fishing, deer and rabbit nets, netting bags, tump lines, slings, flicker feather head bands, hair nets, feather capes, feather skirts, belts, cord belts for women’s aprons, and bowstrings. Areas with cordage plants were periodically burned by the Wukchumni Yokuts, Western Mono, Pomo, and other tribes to decrease accumulated dead material, provide increased access for harvesting, allow greater sunlight to the new growth, and recycle nutrients to the soil. Plants were reputed to grow straighter and taller when burned (Anderson 1993a; Peri et al. 1982). The leaves from at least five kinds of iris were gathered extensively by the Wappo, Karuk, Yurok, Sinkyone, and other tribes in coastal prairies and open woodlands to make cordage by northwestern California tribes in the creation of net headdresses, camping bags, and nets for fish, birds, and small mammals (Baker 1981; Beard 1979; Nomland 1935). Areas where iris grows were burned by the Karuk to produce better growth (Gifford 1939).

Medicines Hundreds of species of plants formed the pharmacopoeia of different tribes, and many of these were gathered from grasslands. The Kumeyaay, for example, collected the leaves of sanicle (Sanicula arguta) on open grassy slopes in what is now Torrey Pines State Park and made a tea for cramps (Shipek

F I G U R E 5.3. Aida Icho, Wukchumni, collecting saltgrass near Guernsey, Kings County. The fingers of her right hand are bandaged in order to prevent the grass from cutting them. 1929. Courtesy of Yosemite National Park, Frank Latta Collection, Accession YOSE-4937.

1991). The Chumash made a tea of the new shoots of Leymus condensatus to treat venereal disease (Timbrook 1990). The roots of Delphinium nudicaule were harvested on the dry, open slopes in foothill woodland and mixed conifer forests by the Pomo, dried and powdered and used to induce sleep (Hudson n.d. 20.202). The Coast Miwok dug the roots of blue-eyed-grass (Sisyrinchium bellum) and made a tea to alleviate a stomachache (Collier and Thalman 1991). Certain plants that grow in grasslands had widespread medicinal use across tribes. One example is yarrow (Achillea millefolium). The Ohlone made a decoction for stomachache and to wash sores. The heated leaves were held over wounds to prevent swelling or held in the mouth for a toothache (Bocek 1984). It was used by the Yokeya Pomo after being charred and sprinkled over burns. An extract of the whole plant was given as a fluid extract for diarrhea (Hudson n.d.: 20.021). The Washoe made an infusion of the leaves and flowers for sore eyes and sick stomach (Hudson n.d.: 20.012). The plant is still used medicinally in northwestern California. The Hupa make a tea of the leaves of yarrow for stomach ailments. The Karuk steep a tea from the dried flowers to cure chills and fevers. The Yurok and Tolowa treat colds and headaches with a tea from the plant (Heffner 1984).

Plant Foods Plants provided 60 to 70 percent of the primary nourishment for most tribes of California, supplemented with meat, fish, and fowl. Condiments were also important. Many tribes gathered the salt crystals from the stems of saltgrass (Distichlis spicata), which grows in grasslands with alkaline soils and in salt marshes (Figure 5.3). Large patches at one time grew in the Central Valley. Saltgrass plants were beaten or burned by the Cahuilla, Tubatulabal, Yokuts, Kawaiisu, Chumash, and other tribes to remove surface incrustations for use as a seasoning or main food (Bean and Saubel 1972; Kroeber 1941; Latta 1977; Timbrook 1990). The Wukchumni Yokuts burned large

N AT I V E A M E R I C A N U S E S A N D M A N A G E M E N T

61

patches of saltgrass in the Central Valley to remove dead material and stimulate new rhizomes and aboveground plants. The diversity of victuals in indigenous diets coming from grasslands was remarkable and included three plant part categories. Grains and seeds provided ample protein. Numerous wildflowers with potato-like underground stems—bulbs, corms, and tubers — provided carbohydrates, dietary fiber, and protein. Leafy greens and stems provided needed vitamins, minerals, and fiber.

Archaeologists, finding these stones deep in excavations, hypothesize that collecting wildflower seeds and grass grains is almost as old as human occupancy of California— over nine thousand years in some areas (Wohlgemuth 2004). Indian women in most parts of California thrust seed beaters over grass or wildflower inflorescences to beat grains and seeds into wide-mouthed baskets. Their contents are then poured into a burden basket. A Belgian goldminer, Jean-Nicolas Perlot, described this phenomenon among the Sierra Miwok:

G R A I N S AN D S E E DS

All of the early high civilizations whose diets are known to us were based on grain-reproducing plants — wheat, maize, or rice (Heizer 1990). California Indians also relied heavily on the grains of over two dozen native grasses such as California brome (Bromus carinatus), Indian ricegrass (Achnatherum hymenoides), and the wild ryes (Elymus glaucus, Leymus cinereus, L. condensatus, L. triticoides) (Hudson 20.017; Shipek 1991; Steward 1933; Zigmond 1981). Early missionary, settler, and anthropology accounts call these grains “wild wheat” or “Indian rice,” alluding to their importance in the Indian diet. The prominence of seeds and grains in the diet is glaringly apparent in diaries, field notes, and other early accounts. Mason (1912: 120) noted of the Salinan along the Central Coast: “Seeds of many varieties were eaten and doubtless formed a considerable item in the Salinan dietary.” Alfred Kroeber recorded in northwestern California: “In summer the Chilula left their permanent homes, near which they fished, and dwelt chiefly on the upper prairelike reaches of the Bald Hills ridge, where seeds as well as bulbs abounded and hunting was convenient” (Kroeber 1925: 138). Robert Heizer wrote of the Patwin and Pomo in coastal California: “These Indians had no real problem gathering food. Natural parklands produced more acorns than could be gathered, there were vast seed-bearing grasslands, rivers and lakes to fish in, deer, elk and waterfowl to hunt” (Heizer 1959: 19). Anna Gayton wrote of the Yokuts: “The sustenance which they [Yokuts] have is of acorns, very savory wild grain, and different kinds of seeds which this country naturally produces with great abundance” (Gayton 1936: 73). Some of the most important wildflowers collected from grasslands and grassy areas in open forests include the achenes of five kinds of mule ears ( Wyethia angustifolia, W. elata, W. helenioides, W. longicaulis, and W. mollis), many different kinds of farewell-to-spring (Clarkia amoena, C. biloba, C. purpurea, C. rhomboidea, C. unguiculata, C. williamsonii), many kinds of tarweed (Hemizonia congesta, H. fitchii, H. paniculata, Madia elegans, M. glomerata, M. gracilis, M. sativa), chia (Salvia columbariae), red maids (Calandrinia ciliata), buttercups (Ranunculus californicus and R. occidentalis), popcornflower (Plagiobothrys nothofulvus), and many others. Archaeological evidence in the form of processing tools such as milling stones (called metates by the Spanish), consisting of a portable flat dish of stone, suggests a very long tradition of wild grain and seed exploitation in the state.

62

HISTORY

During the months of August and September [1851], we often saw Indians coming and going [Perlot spent time between Coultersville and Marble Springs]. It was the time of their harvest; they came to our flats to gather all kinds of seeds, even hayseeds. It is the Indian woman who does this work; she has a big hamper or very open basket, of very fine reeds, and coated with a starch made of powdered seeds and warm water. She holds this hamper with one hand under the grass in seed; then with a sort of fan also made of reed and supplied with a handle, which she holds in the other hand, she pulls the grass over her hamper; the seeds, thanks to the shake given by the fan, are detached and fall; the half-full hamper serves as winnowing-basket to clean them. After which, the Indian woman puts this grain in conical baskets; when she has her load, she puts the whole in her big cornucopia, which serves her as basket, then she returns to the camp to come back again some days later. (Perlot 1985:169)

Seed beating may have enhanced the forb and grass population in a number of ways. As seeds were knocked into another basket, undoubtedly some seeds were scattered around the collection area, perhaps perpetuating the stand. Sam Fox (Eastern Achomawi) told anthropologist Erminie Voegelin in 1936 that when seeds begin to ripen in patches out on the hill, two to three women take sticks and walk “forward and back, forward and back,” flaying plants with sticks; “next year there will be a big crop there because lots of seeds have been scattered” (Voegelin 1942: 176). California provides many case examples of tribes managing grasslands to foster the growth and abundance of herbaceous plants for edible seeds and grains. Anthropologists recorded Indian burning for better wild seed crops among the Achomawi, Atsugewi, Chilula, Chimariko, Coast Yuki, Hupa, Karuk, Kato, Klamath, Maidu, Mattole, Modoc, Mountain Maidu, Nisenan, Nongatl, Shasta, Sinkyone, Wintu, Wiyot, and Yurok (Driver 1939; Silver 1978). Sowing involves broadcasting seeds or grains collected from native plants in an area. Sowing was usually done on recently burned ground. Long before the invention of domesticated agriculture, many tribes understood how to maintain or augment wild plant populations through collecting and sowing seeds and grains of favored species. The Modoc, East Shasta, Achomawi, Northern Maidu and Nisenan tribes of northern California sowed the seeds of different herbaceous plants (Voegelin 1942). The Central Sierra Miwok of the Sierra Nevada cultivated six kinds of seeds with

burning, sowing and harrowing in the Sierra Nevada foothills. Two of these seeds are farewell-to-spring (Clarkia purpurea ssp. viminea) and mule ear (Wyethia helenioides). B U LB S, C O R M S, AN D TU B E R S

Plants with underground swollen stems have great carbohydrate reserves and can withstand summer drought. Many of these underground stems are edible, and Native Americans call them “root crops.” Indian women dug them for foods in great quantity with digging sticks in grass-pine areas, oak savannahs, montane meadows, valley grasslands, coastal prairies, grassy areas along streams, and the first years in the herbaceous flora after a human-set chaparral fire. Some of them are still gathered today. There are dozens of species of herbaceous plants that were dug in these grasslands. For example, the piquant roots of Microseris laciniata found in open oak woodlands were dug by the Pomo and eaten raw with nut bread (Barrett 1952). All kinds of corms in the brodiaea complex in the genera Brodiaea, Dichelostemma, and Triteleia grow in grassy areas, and these were dug and eaten wherever they were found. These corms were sometimes eaten raw, and they had a texture similar to that of a water chestnut, with a slightly nutty flavor. They were also cooked by boiling, steaming, roasting, or baking in an earth oven. At least five kinds of brodiaeas were harvested (Brodiaea coronaria, B. douglasii, B. elegans, B. minor, B. insignis, and B. coronaria) for their edible corms. All of the dichelostemmas were eaten: D. capitatum, D. congestum, D. ida-maia, D. multiflorum, and D. volubile. Five kinds of Triteleia were eaten: T. grandiflora, T. hyacinthina, T. ixioides, T. laxa, and T. peduncularis. At least five species of yampah were harvested for their edible tubers: Perideridia bolanderi, P. gardnerii, P. kelloggii, P. parishii, and P. pringleii. Sanicula tuberosa and soaproot (Chlorogalum pomeridianum) were widely gathered for food (Anderson 2005; Barrett 1952). At least seventeen kinds of mariposa lilies were dug for food by different tribes: Calochortus amabilis, C. bruneaunis, C. catalinae, C. coeruleus, C. concolor, C. flexuosus, C. invenustus, C. kennedyi, C. leichtlinii, C. luteus, C. macrocarpus, C. palmeri, C. pulchellus, C. superbus, C. tolmiei, C. venustus, and C. vestae. These were eaten raw, steamed in an earth oven, or roasted in ashes. Digging was often done judiciously, purposefully leaving tuberous root fragments and bulblets and cormlets behind to regrow. Digging these bulbs and corms may have increased the size of the tract, aerating the soil, lowering weed competition, and preparing the seedbed. GREENS

A great diversity of forbs grow in grasslands, and many of them provided stems and leaves for greens. These included many kinds of clovers (Trifolium spp.), lupines (Lupinus spp.) (boiled in several changes of water to eliminate the alkaloids), and fiddlenecks (Amsinckia spp.). Hancock described havesting and preparation of one of the Amsinckia species

F I G U R E 5.4. Melba Beecher, Mono, harvesting common lomatium (Lomatium urticulatum), a green used for food by various tribes that was gathered in grasslands and oak savannas. Called “wild celery,” the plant is boiled when young, before flowering. Melba says: “You can eat the whole thing. It’s good.” Photograph by Kat Anderson.

to Frank Latta: “The Wobonuch and other tribes on the Kings River ate lots of fiddleneck for greens. They gathered them when they were thick and had lots of leaves and before they sent up a shoot to flower. They put them in a basket of boiling water and kept it boiling with hot rocks until they were cooked. They liked them with ahlet (saltgrass salt)” (Latta 1949). The leaves of woolen breeches (Hydrophyllum capitatum var. alpinum) were gathered in montane meadows by the Mountain Maidu (McMillin 1956). The Ohlone gathered the foliage of sun cups (Camissonia ovata) in grassy areas and ate it raw, boiled, or steamed (Bocek 1984). A plant similar to Scotch lovage — a vegetable cultivated in northern Britain — was eaten in California. It was called wild parsley (Ligusticum grayi), and the leaves were gathered young, soaked in water, and cooked in an earth oven by the Atsugewi. The Maidu also ate the stems and young shoots. Some of the lomatiums were also called “wild parsley” or “wild celery” and formed an important source of spring greens (Figure 5.4). As late as 1973, a Kawaiisu woman gathered a whole sack of the

N AT I V E A M E R I C A N U S E S A N D M A N A G E M E N T

63

common lomatium (Lomatium utriculatum) to be boiled, then fried with grease and salt (Zigmond 1981). The stems of different angelicas (Angelica tomentosa; A. hendersonii) were broken off when fresh and, after the outer rinds had been peeled off, were eaten raw, boiled, or used as a seasoning in soups by the Yana, Pomo, Wintu, and Coast Miwok (Sapir and Spier 1943). Dock (Rumex spp.) greens are high in vitamins A and C, richer in C than citrus juice is, and richer in vitamin A than carrots are. Some of these docks are still gathered by Native Americans today. The young leaves are gathered before flowering, rolled into a ball, sprinkled with salt, and eaten as a snack. The Wappo practiced the burning of the fields and chaparral to keep trails open, stimulate new growth, and encourage the clover, a favorite delicacy (Beard 1979). Other tribes burned the grasslands to enhance clovers and other greens as well (Anderson 2005). Exotic plants were brought to California by Europeans, altering not only the native flora but also the native diet. As wild mustards from Europe and the Mediterranean spread rapidly through California, Native Americans quickly adopted these as peppery greens in the diet; these are the brassicas, related to our domesticated cabbages, kales, and chards. By 1856, wild mustard was even spotted on the Farallone Islands off San Francisco.

Indigenous Influences on Grasslands Most California Indian tribes have been labeled by anthropologists as “hunter-gatherers” because they did not grow domesticated crops. The Mojave, Halchidhoma, Cahuilla, and Quechan in southeastern California were exceptions; they planted corn, beans and squash. Ecologists and resource managers have, by and large, believed that hunter-gatherers, unlike farmers, were environmentally benign, incapable of influencing the productivity, availability, or genetic diversity of natural resources (Lewis and Anderson 2002). In this view, only with development of animal husbandry and agriculture did indigenous peoples emerge from the shadows as a significant ecological force. But given the widespread and long-term management practices of California’s tribes, it makes sense that we place agriculture within a much richer and complex extended history of cultivation. It is highly likely that hunter-gatherers have been a significant ecological factor in grassland ecosystems in California for twelve to fifteen thousand years or more. Their techniques of selective harvesting, tilling, seed beating, weeding, and burning caused ecological effects at different levels of biological organization from the genetic to landscape scales. Thus, domestication of plant species grew out of a set of comprehensive land management systems and complex traditional ecological knowledge that were already in place in many different parts of the world for thousands of years. Human-plant interactions can be appropriately viewed as a continuum from gathering to tending to cultivation to domestication (Figure 5.5).

64

HISTORY

F I G U R E 5.5. Human-Plant interaction continuum. The distinction between “wild” and “domesticated” in plant exploitation is not a sudden or marked change; it is a gradual transition. Native American tribes in California did not domesticate crops, yet they still cultivated wildland environments with an array of management techniques, causing changes at different scales of biological organization. They operated on the middle of a continuum. Adapted from Harris 1989 and Ford 1985. Drawn by Claudia Graham.

Vegetation Changes at the Genetic and Organism Levels Ecologist Harold Heady (1972: 102) argued in the 1970s that on a longer time scale, humans had little to do with the evolution and maintenance of natural grasslands. Human fire setting in grasslands, practiced for millennia, was considered too recent to have been a factor in the evolution of the native grasses (Hatch et al. 1991). Yet it is now recognized that genetic changes in plants can occur in relatively short durations of time through human selection. For example, indigenous burning repeatedly in the same areas may have selected for fire-dependent grass and wildflower species. Fire ecologist E. V. Komarek reminds us that “[t]here is a greater and quicker potential of mutations in a fire environment than in a non-fire one — with all of its evolutionary ramifications” (Komarek 1965: 183). Ecologist Richard Vogl would even go so far as to suggest that grassland fires could be a contributing factor in “creation of the variety of species and ecotypes present in some grassland families” (Vogl 1974: 143). Other indigenous interactions over hundreds of generations also influenced grasslands. Scattering seeds and digging up bulbs and corms and replanting propagules repeatedly in the same areas could have had genetic consequences at

particular gathering sites. Seed beating, burning, and sowing seed would tend to select for specific genotypes that hold up well to and even thrive under human harvesting and management regimes. Archaeologist Bruce Smith suggests that “[t]he open habitats created through burning were probably significant places for plants of economic importance to evolve weedy adaptations” (Smith 1995). One of the oldest forms of tillage in the world is the digging of subterranean organs of wild plants for food and other purposes. Domestication of the potato, as well as yams, arrowroot, taro, and cassava, are at the endpoint of a rich continuum of human interactions with plants with bulbs, corms, and tubers that caused genetic changes. Cases in point are Indian digging of California’s wild lilies and brodiaeas and replanting their propagules. Once dislodged, bulblets and cormlets may take less time to reach flowering size. This type of disturbance may act as a selective force in determining species’ reproductive strategies, affecting the course of plant evolution (Pickett 1976; White 1979). Favored herbaceous plants might be affected at the organism scale after an Indian-set fire: growing taller, more robust, with more leaves, flowers, and seeds, and containing higher nutrient levels. This would affect forage quality for both humans and wildlife.

grassland extent and savanna appearance of the Chumash environment in coastal south-central California (Bicknell et al. 1992; Timbrook et al. 1993). Additionally, repeated burning expanded small patches of grasslands in natural openings of oak woodlands and coastal redwood and mixed conifer forests, creating derived or anthropogenic grasslands that maintained a distinct form and character (Vogl 1974). Under indigenous burning regimes, woodlands and forests often exhibited widely spaced trees, giving better insolation to the forest floor and exhibiting bare mineral soil, heightening seed germination rates of herbaceous plants and ultimately leading to an increase in plant species diversity on an area basis. Studies have shown that disturbance such as fire triggering succession increases diversity. In some of these coastal prairies, rain-fed montane meadows, and woodlands and forests, it is highly likely that Indians burned often; perhaps every one to several years to keep grasslands from converting to woody vegetation (Greenlee and Langenheim 1980, 1990). Native Americans burned grasslands in the spring, summer, fall, and early winter — with summer and fall being most frequently mentioned in the historical literature (Crespí 2001; Timbrook et al. 1993; Anderson 1993a, 2005). Botanist Willis Linn Jepson describes the native influence on creating widely spaced oaks in grasslands:

Vegetation Changes at the Population Level Numbers and densities of favored plant species were encouraged through various kinds of management, including burning, seed beating, sowing, weeding, and tilling. These actions also sometimes expanded the gathering tract for favored species. For example, burning of coastal sage scrub to promote Leymus condensatus was probably important in southern California. Dense populations of plant species at known collection sites would offer a significant reduction in labor and remove uncertainty and time involved in a random or haphazard search for useful plants in the landscape. Indians widened the ecological amplitude of native species in grasslands by introducing them to new areas through sowing seed at traditional collection sites away from villages and adjacent to villages. Thus, the ranges and distributions of favored plant species were expanded through human intervention. Burning grass patches may have changed forage quality, thereby favoring grasshoppers. Frequent fires may have kept seed-eating insect populations low, for example, the insects that feed on the seeds of species of mule ears (Wyethia) and chia (Salvia columbariae).

Vegetation Changes at the Community Level With new paleoecological, anthropological, and historical evidence, it is now clear that Indian-set fires were also an important factor in maintaining and enhancing native grassland distribution, size, and vigor in many parts of California. For example, California Indian tribes expanded the coastal prairies of northern and central California and influenced the

The herbaceous vegetation in aboriginal days grew with the utmost rankness, so rank as to excite the wonderment of the first whites, who repeatedly tell of tying wild oats or grasses over the backs of their riding horses. This dense growth was usually burned each year by the native tribes, making a quick hot fire sufficiently destructive to kill seedlings, although doing little injury to established or even quite young trees. It was, therefore, only in certain years as a result of a combination of local conditions or indirect influences now only to be guessed, such as the character of seasonal rainfall, winds, fire, flood or pestilence, that germinating seedlings were enabled to persist in particular areas. The presence, therefore, of groves of oaks is rather to be explained than their absence. (Jepson 1910)

Leymus condensatus is found on dry slopes and open woodlands below 1500 meters. The grass was useful to many tribes. As mentioned earlier, the long, wide leaves were popular for thatching roofs by tribes in southern California. The Cahuilla and Luiseño also used the main stalks, which were fire-hardened for the main shafts of arrows (Bean and Saubel 1972; Sparkman 1908). The Chumash not only used the stems for arrows, but they also transformed them into tobacco tubes, cigarettes, paintbrush handles, gambling counter sticks, and knives, and the new shoots were made into a tea for venereal disease (Timbrook 1990: 246). The Paiute and Pit River gathered the grains from the inflorescences for food, and it is likely that other tribes ate the grains as well (Steward 1933). Early missionary, explorer, and settler journals frequently describe “a clumped grass that is

N AT I V E A M E R I C A N U S E S A N D M A N A G E M E N T

65

taller than a man on horseback.” This is probably Leymus condensatus. This grass was sometimes burned, probably to promote young, fresh leaves for thatching and abundant grains for food. For example, Crespí noted on August 20, 1769, near Goleta that “We went the whole way over dark friable very level soil, very much clad with very fine grazing and very large clumps of very tall broad grasses, burnt off in some spots and not in others; the unburnt [grasses] were so tall that they topped us on horseback by about a yard” (Crespí 2001: 419). It is conceivable that Indians burned certain areas of coastal sage scrub in southern California to promote L. condensatus, thus, creating a grassland formation in this coastal region that was strictly anthropogenic (Carla D’Antonio, personal communication, 2006).

Vegetation Changes at the Landscape Level Landscapes can be viewed as mosaics of ecosystems, generated by disturbance (Pickett 1976). In addition to natural disturbances, Native people introduced systematic disturbance to maximize plant community diversity. They recognized that certain plant communities harbored a unique array of plant and animal species and that some plant communities, while covering small land surface areas, harbored extremely useful and varied plant life. These vegetation types were maintained in a holding pattern rather than allowed to succeed naturally into a new plant community type. Strategies for maintaining community and habitat integrity included hand-clearing and burning detritus that might alter moisture and soil conditions (which would encourage a new array of plant species to colonize),and handweeding and burning to maintain ecotones around special

66

HISTORY

plant community types such as meadows. Meadows harbor a rich diversity of plant species useful to California Indian tribes. Native Americans prolonged the life of dry meadows through periodic burning (Anderson 2005). Setting fires in the meadows and in the ecotone areas surrounding the meadows decreased shrubs and conifers from encroaching into meadow areas, thus maintaining, and perhaps in some cases enlarging, meadow areas.

Conclusions Grasslands were extremely important to the economies of many Indian tribes throughout California and are still highly significant today. In aboriginal California wildflowers, grasses, sedges, and ferns had dozens of uses. Many of these plants formed part of California’s rich ethnobotany— an overflowing store of biological wealth that was useful to many tribes. These plants succored, fed, sheltered, and clothed Native Americans for thousands of years. Today many kinds of plants coming from the grasslands are still gathered, but some kinds are getting harder and harder to find. Through the introduction of time-tested management techniques, such as burning, tilling, sowing, weeding, and selective harvesting practices of different tribes, California’s grasslands were profoundly shaped by human influence over thousands of years, probably without appreciably diminishing its variety of plant and animal life. This indelible human imprint is reflected in the areal extent, structure, function, and composition of different grasslands throughout California. Thus, our rich grassland heritage is linked with the historical use and stewardship of vegetation by Native Americans. These human influences may be critical to restoration and maintenance of some of these grasslands today.

SIX

Ecology of Invasive Non-native Species in California Grassland CAR LA M. D’ANTO N I O, CA R O LYN MALM ST R O M, SALLY A. R EYN O LD S, AN D J O H N G E R L AC H

Non-native species are widespread and often the dominant plants in California’s grasslands (Biswell 1956, Heady 1977). Many of these species have been abundant since the nineteenth century, whereas others are only now becoming abundant over large areas. The extent to which non-native species dominate grassland sites varies at both local and regional scales, but it is clear that annual grasses of Eurasian origin, in particular, are dominant over enormous portions of the state (Heady 1977), Indeed Heady’s treatment of California grassland in Barbour and Major’s original version of Terrestrial Vegetation of California focuses almost entirely on the “annual type”—a grassland dominated by European annual species. It is impossible to study Californian grasslands without including some non-native species, because they are so widespread and abundant. They are thus an integral part of grassland ecology and management in California and are discussed in most chapters in this volume. Nonetheless, because of the keen interest in harmful non-native species and the clear negative impact of some species on values we gain from grasslands, we will devote this chapter to providing an overview of the nature of these invaders, their distribution within California, factors influencing their abundance, and what impacts have been documented or are postulated to be caused by them. We also review the impacts of plant pathogens in California grassland. While the literature on pathogens is limited, it is an emerging area of interest, and evidence suggests that pathogens may influence and mediate interactions between native and non-native plant species.

Invasion Terminology In the past two decades there has been widespread recognition than many non-native species alter the structure and function of ecological communities in ways that are either economically or ecologically undesirable. These species have been variously called alien, exotic, non-native, nonindigenous,

or invasive. The terms alien, exotic, introduced, non-native, and non-indigenous reflect origin and are used most properly here to indicate species that arrived in California after European contact as a direct or indirect result of human activity. These terms have no implication regarding the impact of any of these species. The term invasive has two current usages. It has been widely used to describe species that negatively influence community or ecosystem properties (see Davis and Thompson 2001)—in other words, species considered to be “weeds.” However, invasive has a more specific demographic meaning, which long preceded its use by “invasion biologists” and which we prefer to retain (reviewed in Rejmanek et al. 2002). An invasive species, in demographic terms, is one that is in the process of increasing in its abundance across the landscape from a point of introduction and has the potential to spread widely. Both “native” and “exotic” species can be invasive. In attempting to standardize invasion biology terminology, Richardson et al. (2000) suggest that the species that managers and “invasion biologists” care most about are those that “transform” aspects of the landscape, and they refer to these as transformer species. They reject the term “invasive” as implying anything about impact and suggest that ecologists focus on “transformer” species. We agree with their terminology and will refer to species that are spreading or have spread since their introduction as invaders or invasive species, irrespective of impact. To refer to impact, we will use the terms harmful, undesirable, or damaging, or we will specifically describe the species’ measured effects without attaching a value-laden word to the described effects. In this chapter we focus primarily on exotic species that are known to negatively influence, or that have been implicated in negatively influencing, community properties or ecosystem processes as defined by the California Invasive Plant Council (Cal-IPC). Some may qualify as transformer species, although in many cases their impacts are not fully understood.

67

California Grassland Subdivisions Although many individual non-native species are widespread across multiple grassland types, there are some strong differences in the abundance and identity of invaders across the major classes of grassland broadly recognized in the state (for community classifications, see Keeler-Wolf et al., Chapter 3). The California grassland had long been divided into two general vegetation types (1) coastal prairie and (2) valley grassland (Heady 1977; Heady et al. 1988). The former occurs in close proximity to the coast and contains a larger proportion of perennial species, while the latter is considered to be largely annual-dominated. Despite the long history of use of these terms, they are a dramatic oversimplification, and many grassland “types” exist even within short distances of each other in the state (Heady et al. 1992, and see Chapter 3). However, for brevity, we will treat grasslands of the California floristic province as falling into three major subdivisions following Jackson and Bartolome (2002): (1) the coastal prairie, (2) coast range grasslands, and (3) valley grassland. The addition of coast range grasslands recognizes the strong mix of native perennial and exotic annual species throughout the Coast Ranges. Although it is not a widely accepted subdivision yet, we find it useful because it distinguishes those strongly mixed perennial
exotic annual grasslands from the more perennial-dominated coastal prairie and the annualdominated valley grasslands. Also, although the general region of Valley grassland includes grass-dominated wetlands such as Sporobolus airoides (alkali sacaton), and Distichilis spicata (saltgrass) grasslands, Leymus triticoides (creeping wildrye) dominated sites and vernal pool grasslands (see Chapter 3), for simplicity we focus our discussion here on upland grassland types. The wetland grasslands are being invaded by a variety of non-native species including Lepidium latifolium (perennial pepper weed), Lolium multiflorum (Italian ryegrass), Crypsis schoenoides (swamp grass) and Glyceria declinata (mannagrass), but, ecologically, they have not been well studied. We also exclude montane meadow grasslands from consideration here.

Non-native Species in the Modern California Grassland Of the approx. 1,100 non-native plant species that have established self-replacing populations in the California flora, at least 300 occur within grassland settings throughout the state. Of these, approximately 66 are listed by the Cal-IPC as of moderate or high concern in terms of being invasive and damaging in grassland habitats (see Cal-IPC 2006 and Table 6.1). The list provided in Table 6.1 is taken from the California Invasive Plant Inventory (Cal-IPC 2006) and reflects both the information available there and our own judgment. For example, we included some moderate- or highseverity Cal-IPC species that Cal-IPC does not list as occurring in “grasslands” but with which one of us has had substantial experience in coastal grassland settings (e.g., Carpobrotus

68

HISTORY

edulis, Table 6.1). We also include some species that are listed as “desert” weeds by Cal-IPC or as having “limited effects” but that, we believe, can have large effects in specific grassland habitats. For example, Erodium cicutarium (red-stemmed filaree) is described by Cal-IPC as “transient” and with “limited effects,” but that is not the case in the San Joaquin Valley and Carrizo Plain grasslands, where this species can have large effects (Kimball and Schiffman 2003; P. Schiffman, personal observations). Other species that we include with an asterisk, such as Bromus tectorum (cheatgrass), are listed by Cal-IPC as occurring in grasslands but are mostly problematic in nongrassland settings. In total more than 90% of the species listed in Table 6.1 are of European or Eurasian origin. They tend to be from genera not represented in the California native grassland flora. For example, a phylogenetic analysis of native versus non-native grasses revealed that “harmful” species (e.g., those in Table 6.1) tend to be less related to the native grassland flora than are nonharmful invaders (Strauss et al. 2006). In California today, much of the coast range and valley grasslands are dominated by European annual grasses in the genera Bromus, Avena, and Hordeum, although other grasses are becoming increasingly common regionally (e.g., Aegilops triuncialis [barb goatgrass] and Taeniatherum caput-medusae [medusahead, formerly T. asperum]). Heady recognized four phases to their arrival with Avena spp., Erodium spp., and Brassica nigra among the very early arrivers (pre-1860s); Bromus mollis (now hordeaceus, soft chess), B. rigidus (now diandrus, ripgut) and Hordeum leporinum (now murinum, foxtail barley) spreading in the 1860s–1870s; and Aira caryophyllea, B. rubens (now madratensis, red brome), and Centaurea melitensis (tocalote) spreading in the late 1800s. He recognizes a fourth phase of recent spread that includes the two grass species A. triuncialis and T. caput-medusae and the forb Centaurea solstitialis (yellow starthistle). Although most of the exotic annual species mentioned in this paragraph occur in all three of the major grassland types that we recognize here, they are most dominant in terms of relative biomass and cover in valley grassland. Indeed, virtually all of the Cal-IPC invaders in valley grassland are annuals (Table 6.1). For a discussion of control methods for some of the problematic valley grassland invaders, see DiTomaso et al. (Chapter 22). Although coastal prairie grasslands also experienced invasion by annual species in the 1800s, today they are increasingly invaded by perennial grasses, also largely of European origin, including Holcus lanatus (velvet grass), Festuca arundinacea (tall fescue), Phalaris aquatica (Harding grass), and Dactylis glomeratus (orchard grass) (see Table 6.1; Hektner and Foin 1977; Foin and Hektner 1986; Peart 1989c). Of the species in Table 6.1, the vast majority of the perennial invaders are problematic in the coastal prairie or coast range grasslands. These include several nitrogen-fixing shrubs such as Cytissus scoparius (Scotch broom), Genista monspessulana (French broom), and Ulex europeus (gorse). Other recent invaders in coastal grasslands include the shrubs Myoporum laetum (myoporum) and Cytissus striatus (striated broom) and the

Myrtaceae Fabaceae Myoporaceae Solanaceae Fabaceae Fabaceae

Eucalyptus globulus Genista monspessulana Myoporum laetum Nicotiana glauca Spartium junceum Ulex europaeus

Chenopodiaceae Asteraceae Aizoaceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Apiaceae

Atriplex semibaccata Carduus nutans

Carpobrotus edulis Centaurea calcitrapa* Centaurea maculosa Centaurea squarrosa Chondrilla juncea

Cirsium arvense Cirsium vulgare^ Conium maculatum^

Asteraceae Asteraceae Liliaceae

Fabaceae

Cytisus striatus

Perennial forbs, subshrubs and grasses Acroptilon repens Arctotheca calendula Asphodelus fistulosus

Fabaceae

Rosaceae Cupressaceae

Woody invaders Cotoneaster spp Cupressus macrocarpa

Cytisus scoparius

Family

Cal-IPC Invaders into CA Grassland

Mod Mod Mod

High Mod High Mod Mod

Mod Mod

Mod Mod Mod

Mod High Mod Mod High High

Mod

High

Mod Mod

Cal-IPC score

Throughout Mostly coastal

Coastal, Islands Throughout North Coastal Throughout Throughout Northern coast Throughout North

Throughout North Coast Coastal

Coastal Coastal Coastal Throughout Coastal Coastal

Coastal, SN foothill Coastal

Coastal Coastal

Distribution

S.E. Europe Europe Europe, Asia, N. Africa

S. Africa S. Europe Eurasia Mediterranean S. Europe

Australia Europe

Central Asia S. Africa S. Europe

Australia Mediterranean New Zealand S. America Mediterranean C. & W. Europe

China Monterey Peninsula S. Europe & N. Africa Portugal

Origin

1800s 1894* 1800s

Early 1900s 1887* 1982* 1950* 1965

1895 1954*

1898 1963 1924*

1853 Mid-1800s 1968* Early 1900s 1848 1894

1960

1850

1850

Date of introduction

Contaminant Accidental Ornamental

Soil stabilization Contaminant Accidental Accidental ? Accidental

Forage Accidental

Contaminant Ornamental Ornamental

Ornamental & erosion control Ornamental Ornamental Ornamental Ornamental Ornamental Ornamental

Ornamental

Ornamental Ornamental

Method of introduction

Cal-IPC Cal-IPC Cal-IPC

(Continued)

USDA Forest Service 2007

Cal-IPC Pitcairn et al. 2002

USGS Cal-IPC Jepson Herbarium & ArizonaSonora Desert Museum n.d. Hilgard (1895) Jepson Herbarium

Cal-IPC Cal-IPC

Cal-IPC Cal-IPC Cal-IPC & Jepson Herbarium

Cal-IPC

Cal-IPC

Cal-IPC Jepson herbarium

Source

TA B L E 6.1 Non-native Species Listed by the California Invasive Plant Council as Being of Moderate or High Concern in California

Euphorbiaceae Poaceae Apiaceae Brassicaceae Poaceae Hypericaceae Hypericaceae Asteraceae Scrophulariaceae Asteraceae Aizoaceae Oxalidaceae Poaceae Poaceae

Euphorbia esula

Festuca arundinacea Foeniculum vulgare Hirschfeldia incana Holcus lanatus Hypericum canariense Hypericum perforatum Hypochaeris radicata Linaria genistifolia Leucanthemum vulgare

Mesembryanthemum crystallinum* Oxalis pes-caprae Pennisetum setaceum Phalaris aquatica

Avena barbata Avena fatua Bellardia trixago

Poaceae Poaceae Scrophulariaceae

Poaceae

Mod Mod Mod

Poaceae Asteraceae

Ehrharta erecta Erechtites spp.*

Annual species Aegilops triuncialis

Mod

Dipsacaceae Poaceae

Dipsacus sp.^ Ehrharta calycina

Mod Mod Limited

High

Mod Mod Mod Mod Mod Mod Mod Mod Mod

High

Mod Mod

Mod High

Mod

Poaceae

Cynodon dactylon

High Mod

Cal-IPC score

Poaceaea Asteraceae

Family

Cortaderia selloana Cynara cardunculus

Cal-IPC Invaders into CA Grassland

Northern throughout Throughout Throughout Coastal

Northern interior Coastal Throughout Throughout Coastal Mostly coastal Interior Coastal Throughout Coastal, some SN C. & S. Coast, Islands Coastal Coastal Mostly coastal

Mostly Coastal Coastal, rarely grassland Coastal Coastal

Coastal Central and S. Coast Throughout

Distribution

TA B L E

Mediterranean & W. Asia S. Europe Europe Mediterranean

S. Africa Africa Mediterranean

S. Africa

Europe Mediterranean Mediterranean Europe Canary Islands Europe Europe Mediterranean Europe

S. Africa Australia, New Zealand Eurasia

Europe S. Africa

Africa

S. America Mediterranean

Origin

6.1 ( C O N T I N U E D )

Cultivated & forage Ornamental Forage & soil stabilization Unknown Unknown

1880*

Contaminant Forage Forage Ornamenta-l?

1914 1888* 1882* 1889*

Ornamental Ornamental Forage

1927* 1906* 1899*

GISI 2005 Cal-IPC

Ornamental Ornamental

Cal-IPC Cal-IPC

Kennedy 1928

USDA pest notes Cal-IPC Cal-IPC

Cal-IPC

CDFA 2007

Accidental

Jepson

Forage Ornamental Cultivated

1500’s

Aiken et al. 2005 Cal-IPC

Cultivated Cultivated

1939* 1886 1911* 1882* 1928* 1940 1881* 1956* early 1900’s

Cal-IPC

Accidental

Cal-IPC Cal-IPC

Cal-IPC

Cal-IPC Cal-IPC & CA Department of Food and Agriculture Jepson Herbarium

Source

1900

1930 1918

1861* 1929

Ornamental Cultivated

Method of introduction

1848 1897

Date of introduction

High

Limited High Mod Mod

Poaceae

Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Poaceae Asteraceae Geraniaceae Geraniaceae Poaceae

Poaceae Poaceae Asteraceae Apiaceae Fabaceae Poaceae

Carduus pycnocephalus Carthamnus lanatus Centaurea melitensis Centaurea solstitialis

Crupina vulgaris Cynosurus echinatus Dittrichia graveolens Erodium cicutarium Geranium spp Hordeum murinum

Lolium multiflorum

Schismus arabicus, S. barbatus Taeniatherum caput-medusae Torilis arvensis Trifolium hirtum

Vulpia myuros

Throughout

Throughout Throughout

Throughout

Throughout, wetter regions Deserts, interior

Northern Throughout Coastal, SN Inland, dry Mostly coastal Throughout

Throughout, but rare in grasslands Mostly coastal Coastal, SN Throughout Throughout

Throughout Throughout

Coastal

C. & S. Europe Eurasia & N. Africa Europe

Mediterranean

Mediterranean

Europe

Mediterranean Mediterranean S. Europe S. Europe & W. Asia S. Europe Europe Mediterranean Mediterranean Europe Europe

Eurasia

Europe Europe

Europe

Accidental ? Accidental ? Forage ? Accidental ?

1936* 1953* 1925*

Mensing and Byrne 1998

Accidental Ornamental? Probably forage or contaminant with barley Forage; also spread by postfire seeding Accidental

Cal-IPC

Cal-IPC

GISI 2005

Roché et al. 2003

Cal-IPC Cal-IPC Cal-IPC Cal-IPC

Cal-IPC

Cal-IPC

Accidental

Accidental Contaminant Contaminant Contaminant

Accidental

Accidental ?

Purposeful? Also spread by postfire seeding

Late 1800’s

1935

1891*

1970s 1912* 1984* 1700s 1862 1904*

1930s 1891 1797 1869

Late 1800s

1848

1861*

NOTE: Only those species with at least some occurrence in coastal, foothill, or valley grasslands are noted. For the full list of Cal-IPC species see www.cal-ipc.org/list_revision/. For distribution, “coastal” means largely coastal counties, SN is Sierra Nevada, “throughout” means occurs in both coastal, interior, and SN sites. The PLANTS database (http://plants.usda.gov/index.html) provided by NRCS was used to classify species into perennial and annual, * refers to biennial species and ^ denotes individual classified as annual-perennial. The Jepson herbarium (*) was used to estimate the date of introduction for species with unrecorded introduction dates on the Cal-IPC website.

Mod

Mod

Limited Mod Mod Limited Mod Mod

Mod Mod Mod High

Mod High

Poaceae Poaceae

Bromus diandrus Bromus madritensis ssp rubens Bromus tectorum

Mod

Poaceae

Brassica nigra

forbs Arctotheca calendula (capeweed). Interestingly, five of the 10 woody invaders are nitrogen fixers, a number disproportionate with the proportion of total woody species that are N fixers in the native flora. That N fixers are disproportionately represented among invasive woody species has been noted at the global scale as well (Daehler 1998). Their restriction to the coastal region, coast ranges and Sierra foothills suggests that they are sensitive to water limitation. Similar to woody invaders, most of the perennial grass invaders are also largely invaders of coastal counties, presumably because of their reliance on greater access to soil moisture than occurs in interior grasslands. Holcus lanatus, for example, is limited in its distribution and invasion success by water availability (Thomsen et al. 2006b). Virtually all of the widespread perennial grass invaders in the coastal grasslands are from Europe. Exceptions are the species in the genus Ehrharta, which are of African origin. The most aggressive of these, E. calycina (veldt grass), is confined to sandy soils in the central coast. The one perennial grass that has become invasive in interior counties is Phalaris aquatica, a species that has been bred for and planted widely because of its broad environmental tolerance (Langer 1990; Bossard et al. 2000).

Presettlement Vegetation It is difficult to determine the precise nature of grassland vegetation in California prior to European settlement, because of the lack of records and the likely variation that existed. For many decades it was widely thought that the preEuropean condition of the California grassland was domination by native perennial bunchgrasses such as Nassella pulchra (purple needlegrass) (Burcham 1957; Heady 1977; Heady et al. 1992). This view apparently came about because of observations by influential ecologists such as Frederic Clements that undisturbed rights-of-way along railroads and roads were typically dominated by N. pulchra (Clements 1934). This view however, was challenged in the late 1990s in two cogent reviews (see Hamilton 1997a; Schiffman 2000; and see Schiffman’s section, Chapter 4). A revised view, gaining acceptance, is that perennial grasses likely dominated the wetter portions of the state’s grasslands such as the coastal prairie, windward portions of the coast range grasslands, and wetter portions of the central valley such as near waterways and marshlands, while annual forbs likely dominated drier valley grassland habitats, including large portions of the Sierra Nevada foothills, interior drier portions of the coastal ranges, and broad terraces around the Central Valley (Schiffman, Chapter 4). Early explorers crossing the Central Valley describe regions that today are dominated by annual grasses as having been a sea of annual forbs (see Sidebar 6.1, Hamilton 1997a), causing some researchers to refer to these early “grasslands” as forblands (Holstein 2001). Other descriptions of the pre-1900 Central Valley region describe tall grasses up to the bellies of the horses (Anderson 2005). It is not clear, however, whether these were introduced, already dominant Avena species or native perennial grasses in wetter

72

HISTORY

valley bottom habitats. Phytolith data do suggest that some valley bottom habitats now dominated by annual grasses were previously dominated by N. pulchra (Bartolome et al. 1986). Other valley bottom habitats may have been dominated by Sporobolus airoides (see Chapter 3). The accounts of European colonists and explorers suggests that the variation driven by topography, climate, and soils that manifests itself as variation in the grassland flora today (see KeelerWolf et al., Chapter 3, and Schiffman, Chapter 4) was similar in the past, although the species composition is likely to have been rather different.

Factors Facilitating Invasion and Conversion The extent of dominance of exotic grassland species in California today testifies to the series of invasions and land use changes that have transformed the region, irrespective of uncertainty about the original composition. A question that has been repeatedly posed is, therefore, What is it about California grasslands that made them so susceptible to invasion? To deduce what factors facilitated invasions and conversion, investigators have had to rely on two imperfect sources of information: (1) historical records and artifacts, which are few in number, and (2) experimental investigations conducted under modern conditions, which likely do not capture historical conditions. Here we summarize what is known about the potential influence of several factors that could have facilitated grassland invasion, but we stress that the answer to this puzzle cannot be definitely known. It is likely that multiple interacting factors were important and that it was the interplay between the conditions on the grasslands and the specific biology of the invaders that led to widespread conversion. Exotic annual domination of the state’s grasslands is presumed to have occurred during the 1860s–1880s (Burcham 1957; Heady 1977). Burcham (1957, 1961) suggests that drought, combined with intensification of crop agriculture and intensive year-round livestock grazing, resulted in a dramatic decline in native perennial grasses over a relatively short period. Diaries of early explorers such as John Muir also suggest that dramatic change occurred relatively rapidly in the mid 1800s (Sidebar 6.1; and see Schiffman, Chapter 4). Native species were presumably replaced with non-native annuals whose seeds had become widespread as a result of transport by livestock, contamination of seed crops, or active planting as forage crops. Livestock grazing was initially implicated as the driving force in conversion of the state’s grasslands to invasive species dominance (Burcham 1957; Heady 1977; Heady et al. 1992). A history of grazing and an overview of impacts are presented in Jackson and Bartolome (Chapter 17) and Huntsinger et al. (Chapter 20). While there is no doubt that grazing intensity was very high in many areas, the tolerance of native perennial grasses or the presumed dominant native annual forbs for intensive grazing has not been well studied. In some valley grassland settings today where native perennial

S I D E B A R 6 . 1 T H E P R E - E U R O P E A N G R A S S L A N D — G R A S S E S O R W I L D F LO W E R S ?

Livestock grazing is frequently blamed for the disappearance of native grasses as dominants in California grassland. Yet the complex topography, climate, and edaphic conditions across the regions of the state supporting grassland today suggest that not all areas considered grasslands today were grass-dominated in the past. In parts of the Central Valley, for example, observations of early explorers and ranchers suggest that annual forbs rather than grasses were the dominant species in the upland grasslands. This sidebar presents examples of the descriptive accounts that support this viewpoint. Because it was the most important entry point into the western side of the San Joaquin Valley, the rangelands of the Rancho San Luis de Gonzaga, which lie at the base of the Pacheco Pass near Gustine, California, were often described in the journals of early visitors. The rancho was established by a Mexican land grant (approximately 180 mile2 or 470 km2) in 1834 to José Estrada (Eigenheer 1976) and was stocked with large numbers of cattle and sheep until the 1860s (Brewer 1966). As recorded by Frank Latta (Mayfield 1993), when Thomas Mayfield and his family first saw the ranch in spring 1850, the display of native wildflowers was very impressive. On their descent from the Coast Range east through the fertile river valley that was later inundated by the San Luis reservoir, they describe a dense cover of threefoot-tall grasses that they called wild oats. As they left the headquarters of the ranch the next day, they crossed the ten-mile-wide plain that separates the foothills from the San Joaquin River basin and described the whole plain as covered in wildflowers rather than grasses: [The plain was] covered with great patches of rose, yellow, scarlet, orange and blue. The colors did not seem to mix to any great extent. Each kind of flower liked a certain kind of soil best and some of the patches of one color were a mile or more across. . . . There were great dens of squirrels. They had thrown the soil up in many places to a height of two feet or more over an area of thirty yards square. Over this area their burrows were thick, and they would stand and bark at us by the hundreds as we approached.

The Mayfields continued on their journey to the Sierra Nevada gold-mining districts, but instead of turning northward and crossing the San Joaquin River via the ferry at the confluence of the San Joaquin and Merced Rivers, the family pushed directly through the San Joaquin River basin near the town of Los Banos. It took them several days to cross the network of sloughs and river channels. When they finally walked out of the river basin onto a “rolling sandy slope country”—probably in what is now central Merced County— they stated that they “saw the same wild flowers we had seen to the west.” The flora and fauna of the Arena Plains are well preserved and documented (Silveira 2000), and its species composition and appearance corroborate Mayfield’s observations. William Brewer described the same area during the drought of the early 1860s as nothing but dust, cracked soil, and dead cattle (Brewer 1966). He also described the behavior of the cattle, which, he said, were prevented from grazing near the San Joaquin River by grizzly bear predation. The native plants had apparently recovered by April of 1868, when John Muir passed through Rancho San Luis de Gonzaga on the way to Yosemite (Muir 1986): Descending the eastern slopes of the coast range through beds of gilias and lupines, and around many a breezy hillock and brush-crowned headland, I at length waded out into the midst of the glorious field of gold. All the ground was covered, not with grass and green leaves, but with radiant corollas, about ankledeep next to the foothills, knee-deep or more five or six miles out. Here were bahia, madia, madaria, burrielia, chrysopsis, corethrogyne, grindelia, etc., growing in close social congregations of various shades of yellow, blending finely with the purples of clarkia, orthocarpus, and oenotheria . . . Hares and spermophiles showd themselves in considerable number, and small bands of antelope were almost constantly

in sight, grazing curiously from some slight elevation, and then bounding swiftly away with unrivaled grace of motion. Yet I could discover no crushed flowers to mark their track, nor, indeed, any destructive action of any wild foot or tooth whatever.

A few days later, in the plains west of the confluence of the San Joaquin and Merced Rivers, Muir conducted the first vegetation survey of the San Joaquin Valley plains (Muir 1974). He recorded the families, number of species, and potential reproductive output of plants in a “one square yard” sample and stated that the numbers of flowers should be doubled to include those that were either in bud or past flowering: Family Gramineae Compositae Leguminosae Umbelliferae Polemoniaceae Scrophulariaceae —? Rubiaceae Geraniaceae

Species 3 2 2 1 2 1 1 1 1

Potential Reproduction 1,000 3,305 2,620 620 401 169 85 40 22

Panicles Heads Flowers Flowers Flowers Flowers Flowers Flowers Flowers

Four years later, in 1872, Muir (Muir 1974) observed the destruction of the native wildflowers as California’s Wheat Bonanza period began (Stoll 1998): The present condition of the Grand Central Garden is very different from that we have sketched . . . fortuneseekers . . . began experiments in a kind of restless, wild-cat agriculture. A load of lumber would be hauled to some spot on the free wilderness where water could easily be found, and a rude box-cabin built. Then a gang plow was procured, and a dozen mustang ponies . . . and with these hundreds of acres were stirred as easily as if the land had been under cultivation for years [note: plowing was done during the early part of the wet season after the soil had been softened by the fall rains and the seed of the native plants had germinated], tough perennial roots being almost wholly absent. Thus a ranch was established, and from these bare wooden huts, as centers of destruction, the wild flora vanished in ever-widening circles.

grasses are absent and may never have been dominant, continuous grazing by livestock today appears to promote native forb dominance (Marty 2005). In other settings, grazing promotes exotic forbs and grasses while also promoting some native perennial grasses (Hayes and Holl 2003a). Modern studies of grazing unfortunately provide limited insight into its role 150– 200 years ago. Current distributions of native perennial grasses suggest that a history of crop agriculture is the best predictor for the occurrence or lack thereof of native perennial grasses in places where they likely could have been dominant based on soils and climate. For example, Stromberg and Griffin (1996) found that coast range grasslands sites that had been tilled for agriculture had virtually no cover of native grasses, no matter how long they had been left fallow. Likewise, Tyler et al. (in preparation) found that a history of deep tillage was the single best predictor for lack of native perennial grasses in some coast range grasslands in Santa Barbara County. Holstein (2001) and detailed historical analyses of individual sites such as in Sidebar 6.2, also suggests the important role of agriculture in altering central California grasslands. Tilled sites are today almost entirely dominated by exotic annual grasses and forbs. Unfortunately, no region wide records of land converted to agriculture are available, so reconstruction of the role of tillage must be done on a site-by-site basis as in Sidebar 6.2. It is known, though, that grain fields in the Central Valley and elsewhere were plowed after fall rains had softened the soil to the appropriate depth and the seeds of the native species had germinated (Stoll 1998). This would eliminate this flush of native seedlings. Today this same technique is used in cropping systems to eliminate the seed bank of weeds in a process known as “flushing the seedbank” (Grundy and Bond 1998). In addition, crop rotations were not used historically; instead, the same fields would be sown annually to grain for up to two decades in a process known as “land killing” (Eigenheer 1976). When crop yields declined to levels where they were no longer economically viable, former grain lands were converted into rangelands. The proportion of annual grasslands today that went through this process is unknown but could be substantial in many counties. Schiffman (2000) (and Chapter 15) argues that the abundance of burrowing animals in valley grasslands was so high that soil disturbance likely had selected for an annual lifestyle both prior to and after European colonization of California. Using observations from early explorers as far back as the late 1700s, she builds a case that burrowing animals were likely abundant throughout the state for centuries. Their disturbance activities set the scene for widespread invasion by European annual species because most of these species are highly disturbance-responsive and have high fecundity. They are thought to have evolved with a 40,000-year history of association with human disturbance in Eurasia and thus were preadapted to take advantage of a highly disturbed environment. Their success today could be explained by their high seed production, faster germination, and more plastic growth, compared to native annual species, and their ability

to set seed even in dry years (Heady 1977; Ewing and Menke 1983; Reynolds et al. 2001; and others). Recent experimental research has shown that soil disturbing animals can promote non-native species in California grasslands (see Schiffman, Chapter 15, and Cushman, Chapter 16). Bartolome and Gemmill (1981) and others (Murphy and Ehrlich 1989) have argued that the success of non-native species in the California grassland could simply be due to their faster early season growth than that of native species, their tolerance of drought, and their ability to produce seed under a wide range of conditions including severe drought (Ewing and Menke 1983) rather than to disruption of ecological resistance (sensu Elton 1958) via livestock grazing or agriculture. However, several recent studies in coastal prairie or coast range grassland have shown that exotic annual grass species are depressed by well-established stands of native perennial grasses and invade poorly into native stands unless disturbance disrupts the native perennial species (Seabloom et al. 2003b; Corbin and D’Antonio 2004b). Likewise, the non-native forb Centaurea solstitialis (yellow star thistle) does poorly in stands of mature native perennial grasses but readily invades annual grassland (Roché et al. 1994; Reever Morghan and Rice 2005). These experimental studies suggest that some sort of stressor would have been necessary to cause the widespread decline of native perennial species and allow exotic annual species to invade in the more mesic sites where perennials likely dominated. Recently, it has been suggested that the Native peoples of California burned the pre-European grasslands regularly or created grasslands by regular burning of sites that would otherwise become dominated by woody species (Anderson 2005). In some sites burning may have occurred every 2–3 years (see Reiner, Chapter 18). While some native bunchgrasses can tolerate occasional burning (Chapter 18), it is not clear that they all could tolerate a high frequency of fire. In general, fire is considered to be a disturbance that creates windows of opportunity for species invasion (D’Antonio et al. 1999). The high frequency of burning of grassland sites may thus have made them susceptible to invasion by non-native species, many of which tolerate fire (Chapter 18). A recent series of papers by Malmstrom et al. (2005a, b, 2006) suggest another intriguing hypothesis as to what might have contributed to the decline of native perennial grasses where they were once dominant in California. This recent research, which is discussed in more detail later in this chapter, indicates that non-native grasses, such as Avena fatua L. (wild oats), can act as disease facilitators. The presence of Avena substantially increases the incidence of barley and cereal yellow dwarf virus infection in neighboring native bunchgrasses (Malmstrom et al. 2005b). The origin of these viruses is not known, but if they are abundant, non-native grasses can increase disease prevalence in the native grasses. The combination of non-native grasses and viruses could then have added to the stressors already affecting native grasses as livestock grazing, drought, and crop agriculture began to impact the landscape.

E C O L O G Y O F I N VA S I V E N O N - N A T I V E S P E C I E S

75

S I D E B A R 6 . 2 L I V E S TO C K O R A G R I C U LT U R E ? A C A S E S T U DY O F L A N D U S E AT A S I N G L E S I T E

As the quote from John Muir in Sidebar 6.1 implied, conversion of grassland to agriculture likely contributed to the disappearance of hectares of native grassland in the Great Valley. Nonetheless, controversy remains as to the role of livestock compared to crop agriculture in contributing to the conversion of California grassland to exotic annual domination. Statewide data on the use of land for grazing versus crop agriculture during the eighteenth and nineteenth centuries are lacking. It often is possible, however, to reconstruct land use history at individual sites and infer from these historical records what factors contributed to the current condition of the land. Following is an example of one such historical reconstruction from an old ranch that today supports some exceptional grassland patches as well as patches of exotic-dominated grassland. Peñasquitos Canyon in central western San Diego County was established as a ranch in 1769 and grazed by either sheep or cattle from 1769 until 1966 (Christenson 2004). The ranch was grazed by sheep belonging to the Mission from establishment until 1823, when the area was granted to Francisco Maria Ruiz. The vegetation of the original ranch was largely coastal sage scrub and chaparral. In 1833 Ruiz requested and received additional lands to the west of the ranch house that included grasslands on the coastal terrace and broader valley bottoms nearer to the coast because the bulk of his initial grant could not support his sheep and cattle. After Ruiz, the ranch passed through a series of Mexican and later American owners until 1921, when it became an addition to the Sawday-Sexton Ranch, which was among the largest cattle operations in the Southwest. After this it was grazed solely by cattle until 1966, when ranching in Los Peñasquitos Canyon ceased because the land was sold to a local developer. This ended a two-hundred-year history of sheep and cattle ranching. In addition to the history of livestock grazing, portions of the canyon experienced some history of crop agriculture and other disturbances, such as roads, that could have contributed to exotic species invasions. Plowing for bean farming and later alfalfa occurred in portions of the valley bottoms and mesa top and was well developed by the 1920s. This continued in some areas into the 1950s and 1960s. In addition to the disturbance of tillage, the ranch included a main transportation/road corridor beginning in the early 1800s. At this time the main cart road from San Diego to the east passed through the canyon, and later the road became the main stage and mail route to the east. It is well known that roads act as corridors for weed invasion, in California grasslands as elsewhere (e.g., Harrison et al. 2002; Gelbard and Harrison 2003). Today this area is part of the Los Peñasquitos Canyon Preserve, which extends westward from the City of Poway to the Los Peñasquitos Lagoon, located on the coast immediately south of the City of Del Mar. The Canyon is steeply walled and relatively narrow as it runs through a broad coastal terrace near the ranch house, but it widens substantially as it nears the coast. The vegetation along the streambed near the ranch house is sycamore woodland, while the vegetation on the canyon walls and coastal terrace is a mix of sage scrub and chaparral types interspersed with vernal pools (Stephenson 2002; City of San Diego 2003). A band of grassland runs along both sides of stream between the sycamore woodland and the upper edge of the toe of the canyon wall. These grassland bands grow wider as the canyon opens up to the coastal terrace, and it is portions of these that were cropped. An aerial photograph from 1928, was taken during the Sawday-Sexton Ranch period, shows dense networks of cattle trails in the canyon and extensive areas of bare ground along the toe of the canyon walls plus areas of bean field. In contrast, aerial photographs taken in 1975 and 1999 after ranching operations ceased show little bare ground (AMEC Earth & Environmental Inc. 2005).

76

HISTORY

Seventeen of the thirty-eight species of grasses collected in the canyon are natives and most are native perennial grasses (Friends of Los Peñasquitos Canyon Preserve 1997). Stands of native grasses are dense in some areas today. Remarkably, despite a two-hundred year history of intense grazing, agriculture, a cart and stage road that functioned as an invasion corridor, numerous extreme droughts (Burcham 1981), and a near-desert precipitation regime, Nassella pulchra stands occupy the grassland bands just downstream of the ranch house that were never plowed. In contrast, exotic annual grasses dominate the wider sections of the canyon nearer the coast, where crop agriculture was more intense (J. Gerlach, personal observation). Barry noted what may be a similar phenomenon in the La Jolla Valley near Oxnard (Barry 1972:82). The persistence of Nassella pulchra and other native perennial grasses in Los Peñasquitos Canyon in the face of two hundred years of intensive grazing and drought and its loss from the wider and more intensively farmed sections of the canyon suggest that the conversion of California’s grasslands is a more complex process than overgrazing during drought years (Burcham 1957, 1981) and later seed limitation (Seabloom et al. 2003b). The dominance of exotic annual grasses only on the land disturbed for farming suggests the importance of tillage or deep soil disturbance in promoting persistent stands of exotic annual grasses.

As a final note about why invasion and conversion occurred, we point out that native annual species in California grassland tend to be forbs (e.g., Keeley et al. 2003), while the most abundant non-native annual species tend to be grasses. Assuming that winter/spring annual grasses are generally better competitors for resources than winter/spring forbs, this difference in life form could have contributed to the rise to dominance of non-native annual species. The relative rarity of annual grasses in the California flora could help to explain the susceptibility of our grasslands to this type of invader, but why annual grasses did not diversify in California is an evolutionary question beyond the scope of this book.

Impacts In this section we provide examples of the different sorts of ecological impacts that plant invaders have had in California grasslands and how they may interact with plant pathogens to influence grassland composition. A brief overview of some invader impacts is also provided in DiTomaso et al. (Chapter 22) along with a discussion of specific control tactics. We focus on those non-native invaders that are viewed as “damaging” although in most cases it is impossible to place dollar values on the measured ecological impacts. The impacts of feral pigs are discussed in Chapter 16. Although introduced species in California grassland are known to incur costs to both private and public land

owners, there have been no systematic attempts to summarize these costs. In a general sense costs include lost revenue from having to alter livestock grazing practices, direct costs of control efforts, and costs incurred by having to repair damage done by these species. The single invader for which market costs are perhaps best documented in California is Centaurea solstitialis. For a discussion of these costs see Chapter 22, Jetter et al. 2003, and Jetter 2005). For a discussion of costs in Idaho, where this species is also a problem, see Hartmans et al. (1997). In addition to market costs, non-native species can incur ecological costs, including those associated with lost “ecosystem services.” These are difficult to translate directly into dollar values, but because of their importance to the ecology of these systems, it is these nonmarket impacts that are summarized in the remainder of this section.

Effects of Non-native Plants C OM P ETITIVE E F F E CTS I N F LU E NCI NG S P EC I E S C OM P OS ITION

The importance of competition as a structuring force in California grassland is reviewed in detail by Corbin et al. (Chapter 13). Here we will highlight the importance of competition for resources between native grassland species and introduced annual and perennial grasses.

E C O L O G Y O F I N VA S I V E N O N - N A T I V E S P E C I E S

77

In California grasslands today, native species almost always co-occur within a matrix of exotic species. If native species exist as well-established perennials in dense stands, they may be less affected by the generally dense exotic annual seedlings than native plants growing alone (e.g., Corbin and D’Antonio 2004b). In general, exotic annual seedlings vastly outnumber native seedlings, so the environment for native seedling establishment is highly competitive (Biswell 1956; Heady 1956; MacDonald et al. 1988; Heady et al. 1992). One native seedling may have several hundred to several thousand neighboring individuals within a 10-centimeter radius (Major and Pyott 1966; Young and Evans 1989). The perceived competitive environment created by European species has spurred much investigation of interactions among native perennial grasses and European annual grasses (Nelson and Allen 1993; Dyer et al. 1996, 2000; Dyer and Rice 1997b, 1999; Eliason and Allen 1997; Hamilton et al. 1999). While interspecific interactions are likely important for native forbs as well, there have been fewer studies on them (Cook 1965; Carlsen et al. 2000; Kimball and Schiffman 2003; Gillespie and Allen 2004). Competition from exotic annual grasses has been demonstrated to be important during all stages of the life cycle of the native perennial grass Nassella pulchra (Hamilton et al. 1999). However, most evidence from California grassland studies points to the seedling stage as the period of highest interference from exotic annuals (Bartolome and Gemmill 1981; Jackson and Roy 1986; Dyer et al. 1996; Dyer and Rice 1997b, 1999), a result supported by grassland studies elsewhere (Grubb 1977; Weiner and Thomas 1986; Foster 1999). Once established, native perennial grasses appear to be successful competitors and can survive in a diverse grassland community for many years despite the presence of exotic annual plants (White 1967; Jackson and Roy 1986; Dyer and Rice 1997b; Corbin and D’Antonio 2004b). Though failure to establish as seedlings in the presence of competing annuals appears to represent a major limitation for populations of native perennial grass species, recent investigators have argued that seed limitation, rather than competition, is the most important factor limiting native grass establishment (Seabloom et al. 2003b). Physiological differences in germination, growth rate, nutrient and water uptake, and reproductive allocation between exotic annual and native perennial grasses all likely contribute to the challenges that native perennial grass seedlings face among competing exotic annuals. In a field experiment conducted in Coast Range grassland, germination of native perennial species was delayed by 2 – 5 days compared to annual species (Jackson and Roy 1986). Likewise, in a greenhouse study, Reynolds et al. (2001) found that seeds of exotic annual grasses germinated much more rapidly than those of several species of native perennial grasses collected from a coastal prairie site. This slight delay could lead to stronger competitive suppression by annuals on native perennial seedlings (Abraham et al., unpublished). Another greenhouse study also found that N. pulchra seeds

78

HISTORY

germinate more slowly and attain a lower density when sown with exotic annuals than when sown alone (Bartolome and Gemmill 1981). In contrast, Robinson (1971) found field germination of N. pulchra seeds to be unaffected by the presence of non-native annual grasses. Once seeds have germinated, native grass seedlings generally have a slower winter growth rate and greater belowground biomass allocation relative to exotic annuals ( Jackson and Roy 1986; Holmes and Rice 1996). At the end of one growing season, Holmes and Rice (1996) found Bromus diandrus (ripgut brome), an exotic annual, to have twice the aboveground dry weight of N. pulchra. With their earlier development, exotic annuals may effectively deplete soil moisture before seedlings of native species have a chance to do so, or they may reduce light and nutrients for native seedlings (Ross and Harper 1972; Bartolome and Gemmill 1981; Hamrick and Lee 1987; Fossum 1990; Dyer and Rice 1999). Dyer and Rice (1997b, 1999) found that seedling survival, growth, and culm production in N. pulchra individuals were negatively affected by a range of exotic annual species densities and that the latter significantly decreased both available light and soil moisture (down to 30 cm). Other studies have shown that the presence of exotic annual grasses negatively influences inflorescence number and seed output of California native perennial grasses (Gordon and Rice 1992; Hamilton et al. 1999; Carlsen et al. 2000).

E F F ECTS OF WO ODY I NVADE R S ON EC OSYSTE M STR UCTU R E

Both native and non-native shrubs and trees can invade California grassland habitat (see Table 6.1 and Tyler et al., Chapter 14). If they are not controlled by burning or cutting, they can alter the distribution of biomass, thereby affecting fuel structure, carbon storage, habitat for animals, and microhabitat for further plant growth. Nitrogen-fixing woody invaders also alter soil N pools (see below). Perhaps the most common shrub to invade grasslands statewide is the native invader, Baccharis pilularis (coyote brush). It has been observed to invade California grasslands very rapidly and can attain a closed canopy rapidly (e.g., Hobbs and Mooney 1986, Williams et al. 1987). It responds favorably to potential global changes such as increases in springtime precipitation (Williams and Hobbs 1989, Zavaleta 2006). Once established it alters understory composition (Hobbs and Mooney 1986), ecosystem carbon storage (Erika Zavaleta, unpublished), and standing biomass/fuel distribution. As with woody plant invasion into grasslands elsewhere, we anticipate that conversion will affect C and N cycling and storage, hydrological properties of the site, productivity, and microclimate (e.g., Breshears et al. 1998; Schlesinger and Pilmanis 1998; Huenneke et al. 2002). Some woody invaders also can act as nurse plants for slower-growing woody species such as oaks. For example, both Salvia mellifera and Artemisia californica, two coastal sage scrub species, facilitated establishment of Quercus douglassii

(blue oak) in savanna settings in Santa Barbara County (Callaway 1992).

A LTE R E D C OM M U N IT Y R E S P ON S E S TO P E RTU R BATION S

Non-native grasses and forbs are prolific seed producers (Heady 1977; Heady et al. 1992; Bartolome 1976; others), and many have a persistent seed bank in contrast to native species (Rice 1989b). As a result, the presence of these species has altered community responses to disturbance. For example, disturbances by native rodents are a common feature of California grassland (Hobbs and Mooney 1985, 1991; Schiffman 1994, 1997, and Chapter 15). Many authors have found that non-native species rapidly colonize these disturbances (Hobbs and Mooney 1991; D’Antonio 1993; Schiffman 1994) even when native species are present in the surrounding community. Colonization can be either from the seed bank or from seed rain. A recent study of colonization of artificial gopher mounds in coastal prairie grasslands with mixed native and non-native species found that the success of native species on mounds is a negative function of the seed rain of non-native species: native species recruit only below a threshold density of exotic seeds (DiVittorio et al. 2007). Another way in which the presence of particular nonnative species in California grasslands has altered community responses to disturbance is livestock grazing. In addition to the direct effects of grazing discussed in Chapters 17 and 20, indirect effects may also occur. For example, Callaway et al. (2001, 2003) found that the spiny invader Centaurea melitensis (tocalote, or malta starthistle), like its relative C. solstitialis, is suppressed by both native perennial grasses and exotic annual species. It is also negatively impacted by simulated grazing (Gerlach and Rice 2004). However, with simulated grazing (e.g., clipping), competitive relationships between C. melitensis and the native perennial grass species shifted: Centurea’s growth overcompensated for herbivory at the expense of the native grasses. Because the response relied on the presence of mycorrhizal fungi, they hypothesize that C. melitensis is somehow able to parasitize the mycorrhizal network in N. pulchra roots. Regardless of the mechanism, such research suggests an indirect mechanism though which grazing and the presence of a particular non-native invader can damage native plant species and favor the spread or increase of an undesirable non-native species. Fire was likely a common disturbance in California grassland during the pre-European period (see Reiner, Chapter 18; Anderson 2005), although it is not known specifically which species were selected for or against by indigenous burning. The widespread occurrence of non-native species has altered the response of grasslands to burning. Although recurrent fire can be used to control some exotic annual grasses, it also appears to promote some exotic forbs (Chapter 18). Abundant exotic annual grasses can also increase the fuel loading around the base of native perennial grasses, potentially resulting in increased fire frequencies and reduced survival of native bunchgrasses (Reiner, Chapter 18).

Atmospheric nitrogen deposition has increased in California as a result of human activities (Fenn et al. 2003a). This ecosystem-scale perturbation has been shown to favor non-native grasses at the expense of native species in several California ecosystems (Hobbs et. al 1988; Huenneke et al. 1990; Weiss 1999; Jefferies and Maron 1997; Yoshida and Allen 2001, 2004; Wood et al. 2006). Hence we can expect non-native species to increase in grassland and shrubland sites affected by the addition of nitrogen. Elevated soil nitrogen can also be the result of enhanced biological N fixation due to native or non-native shrub invasions into grasslands. Such N additions can lead to dominance of sites by non-native annual grasses (Maron and Connors 1996; Maron and Jefferies 1999; Weiss 1999; Alexander and D’Antonio 2003; Wood et al. 2006).

E F F ECTS ON SOI L MOI STU R E

Water plays an overarching role as a critical limiting resource in California environments (see Chapters 7 and 10). In sites that were likely dominated by annual forbs prior to European colonization, little is known about the effects of the replacement of forbs by annual grasses on soil water potential, infiltration, or deep water storage. Eremocarpus setigera (doveweed), a late-season native annual, reduced soil moisture reserves to a greater extent than invasive C. solstitialis, also a late season annual (Gerlach 2004). Unpublished field data with another late-season annual, Hemizonia increscens, shows similar impacts as E. setigera. The main difference between these native late season annuals and the exotic one is that because of its much faster growth rate, C. solstitialis uses its water earlier in the summer. Where perennial grasses have been replaced by exotic annual species, the phenology of ecosystem production has been dramatically altered. Annual grasses germinate with fall rains and grow rapidly during the late winter and early spring, when soil moisture is generally abundant. They flower almost synchronously and die by May. Most of their root mass is in the top 30 centimeters of the soil profile (Holmes and Rice 1996). By contrast, native perennial grasses can extend their growth and phenology into the summer months; they typically have much deeper roots and transpire soil water throughout the summer (e.g., Jackson and Roy 1986; Borman et al. 1992; others). Holmes and Rice (1996) and Gerlach (2004) found that exotic annual grasses cannot take up all of the water passing through the rooting zone during winter rains. Further, their lack of transpiration during the summer has left a reservoir of deep soil water in sites without perennial species. In southern Oregon grasslands, Borman et al. (1992) found that perennial grasses drew down soil moisture 50% more than exotic annual grasses. The remaining deep soil moisture in annual-dominated sites, both there and in much of California, is now being accessed by C. solstitialis (Gerlach 2004). This invasion is particularly noteworthy because C. solstitialis causes a neurological disease in horses and is unpalatable to livestock. Enloe et al. (2004) found that C. solstitialis used more deep soil water

E C O L O G Y O F I N VA S I V E N O N - N A T I V E S P E C I E S

79

than either annual grass alone or perennial grasses alone. Gerlach (2004) argues that this deep-rooted summer annual invader will cost millions of dollars in lost water in California watersheds because it interferes with groundwater recharge by depleting deep soil moisture.

E F F E CTS ON SOI L N UTR I E NTS

Water and nutrient availability can be strongly linked. Nonnative species in California grassland have been found to affect many aspects of the soil environment. Approaches to studying their impacts include both observational studies involving an invasion chronosequence, and experimental studies using planted plots or pots. The plot studies have largely involved planting plots of known composition — either natives alone, exotics alone, or mixed— onto a uniform background. Evaluations of invasive species’ effects on N cycling from throughout the globe have demonstrated that those species with the largest effects are nitrogen fixers invading sites entering sites where N fixers were previously rare (D’Antonio and Corbin 2003; D’Antonio and Hobbie 2005). In coastal prairie, invading nitrogen-fixing shrubs have been found, using an invasion sequence approach, to affect nitrogen pools and N mineralization (e.g., Maron and Jeffries 1999; Haubensak et al. 2004; Haubensak and Parker 2004). Maron and Connors (1996) and Maron and Jeffries (1999) studied a native invading N fixer, Lupinus arboreus, that has recently expanded into coastal prairie patches where it had not previously occurred. It elevates soil N, and when it dies, this high nitrogen promotes fast-growing exotic annual grasses to the exclusion of native species. This species was introduced to northern California beyond its natural range limit and has been found to alter vegetation composition in invaded sites in a similar manner to that shown by the Maron and Connors study (Andrea Pickart, Lanphere Christensen Dunes, personal communication). The rapidly invading European shrub Genista monspessulana (French broom) also appears to elevate soil nitrogen and N mineralization (Haubensak 2001; Haubensak et al. 2004), which could promote fast-growing non-native grasses after its removal (Haubensak and D’Antonio 2006). Its close relative, Cytissus scoparius (Scotch broom), has also been demonstrated to enhance rates of N mineralization in Washington coastal prairie (Haubensak and Parker 2004). Such effects presumably also occur in California, where this species is becoming widespread. While these effects of N-fixing shrubs may be profound, they are limited to coastal grasslands or perhaps Sierra foothill environments, where N-fixing invaders are becoming more common. The effect of exotic annual grasses on N cycling has been more difficult to determine, perhaps because effects are more subtle. On the basis of differences in plant growth and nitrogen allocation between exotic annual grasses and native perennial grasses, one would expect to find some biologically significant differences (see Chapter 8). For example, sites dominated by annual species have extremely high seedling

80

HISTORY

densities after fall rains (Heady 1977; Major and Pyott 1966; Bartolome 1979). These grasses experience high rates of selfthinning and turnover (Savelle 1977; Bartolome 1979), potentially resulting in a large flux of nitrogen through decomposing material of these young seedlings (Eviner and Firestone, Chapter 8). This should not occur in native perennial grasslands, where seedling densities after fall rains are generally low (Savelle 1977). The importance of this mechanism for nutrient turnover during the growing season should vary with both the absolute and the relative abundance of introduced annuals. Eviner and Firestone (Chapter 8) review mechanisms, such as differences in litter quality between introduced annual and native perennial grasses, that should lead to differences in N cycling. However, reported results have been subtle and contradictory. Corbin and D’Antonio (in prep.) found slight differences in measured of N cycling rates between planted plots of exotic annual, exotic perennial, and native perennial grass species, with native perennial plots showing slightly higher rates of N cycling overall compared to annuals. One clear result, however, was that plots with only exotic annual grasses had the highest amounts of nitrate leaching losses compared to native perennial plots. Eviner and Hawkes (in prep.) had similar results in planted grassland mesocosms. By contrast, Parker et al. (in prep.) found opposite results in a more southerly California grassland: N cycling rates were higher in exotic annual grassland than in native perennial plots, but N leaching losses were greater in perennial than annual plots. Eviner and Firestone suggest that the differences between sites may be the result of hydrology. An unusual effect of an invader on soil pH and cations was measured in a coastal grassland site, where the invading South African perennial subshrub Carpobrotus edulis (highway iceplant) was found to decrease soil pH from an average of 5.6 to 4.1 in two grassland sites in Santa Barbara County (D’Antonio 1990). Calcium was also depressed in the acidified soils. In coastal grassland in southern California, the annual iceplant Mesembryanthemum crystallinum was also found to affect soil cations. Vivrette and Muller (1977) found that this plant concentrated salts from throughout the soil profile into its leaves and then deposited those salts on the soil surface during annual senescence, thereby inhibiting germination of potentially co-occurring grassland species.

E F F ECTS ON SOI L M ICROB E S

With the increased development of tools to evaluate soil microbial community composition, there has been a widespread increase in interest in how invading species alter the soil community (see review by Ehrenfeld and Scott 2001; Ehrenfeld 2003). Those few studies that have been done in California grasslands suggest a persistent effect of land-use history and an effect of non-native plants on microbial communities as detected by phospholipid fatty acid (PLFA)

analysis (Steenwerth et al. 2003; Batten et al. 2006a; and see Chapter 9). This technique evaluates community composition by assessing variation in membrane chemistry that accompanies microbial taxon differences. Although PLFA profiles can change as a result of physiological changes in microorganisms rather than true compositional changes (Trish Holden, UCSB personal communication), the changes observed during the temporal sequences of invasion examined by Batten et al. (2006a) generate fairly discrete assemblage clusters, suggesting that some invaders can alter microbial composition in a persistent way. The Steenwerth et al. (2003) study demonstrates a strong persistent effect of soil tillage on grassland microbial assemblages, but it also demonstrates differences in microbial composition between exotic annual pastures and native perennial grass pastures that had both experienced a history of past tillage. The native perennial grass “old fields” that they studied had been seeded to these species, while the exotic annual grass pastures had been left fallow without seeding. The divergence in composition between them suggests that plant species can alter microbial communities. Using planted assemblages of native and non-native species, Hawkes et al. (2006) demonstrated that exotic annual grasses reduced the diversity of mycorrhizal fungi associated with native perennial grass roots. Whether this directly and negatively affects their growth remains to be tested. In the same tubs, Hawkes et al. (2005) demonstrated that exotic annual grasses altered the composition of ammonium-oxidizing microbes, thereby reducing rates of N mineralization. This could in turn affect plant growth. While these studies of microbial compositional in California grassland suggest that plants can cause changes in microbial community structure, we do not yet know what these changes mean in terms of restoration or succession on these sites. Hawkes et al. (2005) suggest that total annual N availability budgets may be altered by microbial compositional change and thus provide a positive feedback to fast-growing, N-demanding invaders, but they do not have evidence for this. Batten et al. (2006b) found no differences in soil aggregate structure or other soil characteristics in serpentine soils where microbial composition was being altered by invasion of C. solstitialis or A. triuncialis, suggesting that larger-scale ecosystem function changes have not yet occurred in response to microbial change.

Interactions with Plant Pathogens Microscale agents that influence grasslands include not only soil microbes but also a broad range of microorganisms such as viruses, rusts, smuts, and bacteria that can have pathogenic influences on plant hosts. Although these microorganisms are integral parts of grassland ecosystems, their presence is easily overlooked, and as a result our understanding of plant-pathogen interactions in grasslands is even less welldeveloped than that of plant–soil microbe dynamics. Recent work in the California grasslands indicates that two different

pathogen groups may play substantial roles in furthering the establishment of invasive grasses, but little is yet known about the ecological significance of most other grassland pathogens, including most pathogens of animals. Here we outline some types of interactions that might generally be expected to occur between invaders and pathogens and then discuss in more detail the two California grassland pathogen systems in which pathogens appear to facilitate invader success, either directly (seed head fungi and A. triuncialis (goatgrass); Eviner and Chapin 2003c) or indirectly (barley and cereal yellow dwarf viruses and Avena fatua; Malmstrom 1998; Malmstrom et al. 2005a, b, 2006). We use the term pathogen to indicate a microbe that has the capacity to reduce host growth, fecundity, or survivorship. Disease represents the manifestation of these effects in a host. We use virulence to indicate the capability of a pathogen to overcome host resistance and to infect it. Aggressiveness describes the extent of a pathogen’s negative influence on a host, once infected, and tolerance indicates the host’s reciprocal capacity to function normally despite infection. It is important to note that not all microbes are pathogens and that a pathogen, which by definition possesses the capacity to harm hosts, may exert this influence only under certain conditions. Newly introduced or emerging pathogens have the capacity to strongly influence California grasslands, although few cases have been documented. The best contemporary example may be that of Phytophthora ramorum, a newly emergent pathogenic oomycete responsible for sudden oak death throughout California (Rizzo et al. 2005). It is a virulent, aggressive pathogen with a broad host range, causing mortality in more than 40 plant genera including Quercus (oaks) and Lithocarpus (tan oaks). If death of these canopy species results in gaps in which grassland species can establish themselves, then P. ramorum—mediated mortality may facilitate grassland expansion in some areas. Where Q. agrifolia is a part of grassland savanna environments, the loss of this canopy tree due to P. ramorum may facilitate invasion by undesirable species that appear to thrive in the generally richer soils under savanna oaks (Callaway et al. 1991; Rice and Nagy 2000). The potential influence of invading pathogens on populations of grassland mammals and birds merits attention as well. For example, the influence of West Nile virus and other introduced mosquito-borne arboviruses on the survivorship of grassland bird populations is unknown and warrants investigation.

P OTE NTIAL I NVADE R S LI M ITE D BY PATHO G E N S

The ecology literature suggests that pathogens play a critical role in controlling potential invaders (Keane and Crawley 2002; Mitchell and Power 2003; Colautti et al. 2004). For example, the enemy release hypothesis predicts that a species is more likely to become invasive when it escapes control by pathogens and other natural enemies present in its original range. The biotic resistance hypothesis expresses the reciprocal idea: that

E C O L O G Y O F I N VA S I V E N O N - N A T I V E S P E C I E S

81

attack by native pathogens is likely to reduce the chance that an exotic species becomes invasive in a new community. Although evidence for these hypotheses is mixed (Colautti et al. 2004), they form the rationale for biological control programs. For example, the California Department of Food and Agriculture has imported Puccinia jaceae (Mediterranean rust) and several Mediterranean insects to California in an attempt to control C. solstitialis (see DiTomaso et al., Chapter 22). However, the extent to which these organisms will control this species or influence other grassland species is unknown. I NVAS IVE P LANTS B E N E F ITI NG DI R E CTLY F ROM PATHO G E N S

Pathogenic organisms may directly facilitate the establishment of invasive species, as interactions between barbed A. triuncialis and a specialist seed head fungus (Ulocladium atrum) in the California grasslands demonstrate (Eviner and Chapin 2003c). U. atrum frequently infects A. triuncialis seed heads and exerts pathogenic effects—by reducing seed count, individual seed mass, and the proportion of germinable seeds. However, by breaking down A. triuncialis’s woody seed head, U. atrum also enhances A. triuncialis seed germination rates. As a result, despite pathogenic effects, U. atrum infection results in an increase in end-of-season aboveground biomass of A. triuncialis because of higher plant densities. U. atrum infection thus substantially facilitates local patterns of A. triuncialis establishment. This example demonstrates not only that pathogenic organisms can directly other nonnative species, but also that the distinction between pathogen and mutualist can be blurred. NON-NATIVE I NVADE R S ALTE R I NG PATHO G E N DYNAM ICS AN D B E N E F ITI NG I N DI R E CTLY

Pathogens may also indirectly facilitate the establishment of non-native species. For example, invading species that introduce new pathogens into a community or alter the dynamics of existing pathogens can act as disease facilitators and benefit from pathogen-mediated apparent competition. In the California grasslands, evidence indicates that invasive exotic grasses, such as A. fatua, can act as disease facilitators: The presence of these exotics substantially increases the incidence of barley and cereal yellow dwarf virus infection in neighboring native bunchgrasses (Malmstrom et al. 2005b). Barley and cereal yellow dwarf viruses (Luteoviridae: BYDVs and CYDVs, hereafter B/CYDVs) are a group of six ssRNA viruses persistently and obligately transmitted by aphids, such as Rhopalosiphum padi L., the birdcherry-oat aphid (Irwin and Thresh 1990; Miller and Rasochova 1997). The viruses are generalist pathogens, causing systemic infection in more than 150 species of Poaceae, including most commercial cereals and many wild grasses (D’Arcy 1995). B/CYDVs were first identified in California in 1951 (Oswald and Houston 1951, 1953), but their origin is unknown. Currently, B/CYDVs are widespread in California (Griesbach et al. 1989, 1990a, b) and can infect most native and exotic grass species (Malmstrom 1998).

82

HISTORY

Avena fatua, as well as other exotic annual grasses such as Bromus hordeaceus L. (soft chess), attract cereal aphids such as R. padi and stimulate their fecundity (Malmstrom et al. 2005b). As a result, springtime aphid populations in stands containing these annuals can be orders of magnitude greater than in stands in which the annuals are absent. In parallel, B/CYDV incidence in native perennial grasses that exist with exotic annuals can more than double (Malmstrom et al. 2005b) compared to the incidence in monotypic stands. Because B/CYDV infection can substantially stunt young bunchgrasses and reduce their fecundity and survivorship (Malmstrom et al. 2005a), these findings suggest that virusmediated apparent competition has the capacity to influence interactions among exotic and native grasses. Because many bunchgrass populations are seed- and recruitment-limited (Dyer and Rice 1997b; Hamilton et al. 1999; Seabloom et al. 2003b), virus-mediated mortality of young plants could contribute to population-level declines of native grasses. Although exotic hosts are also strongly affected by infection, their life histories may provide greater population-level buffering against B/CYDV losses (Malmstrom et al. 2005a). Exotic annual grass stands shed infection each year because B/CYDVs are not seed-transmitted (Irwin and Thresh 1990), whereas infection may persist for longer periods in native perennial grass populations. Furthermore, because annual grasses tend to occur in very high densities (Heady 1958), neighboring uninfected individuals compensate for reduced fecundity of those infected. In addition, the well-developed seed banks of many exotic species (Rice 1989b) are likely to permit stand recovery even if substantial losses in seed production occur in a given year. Such buffering is less likely to occur in native populations, which tend to experience more interspecific than intraspecific competition (Dyer and Rice 1997b) and have poorly developed seed banks (Major and Pyott 1966). The evidence for virus-mediated apparent competition in modern settings and the probable differences in populationlevel consequences of infection for native and exotic grasses suggest that B/CYDVs have the capacity to facilitate the establishment of exotic grasses in California and their continued dominance (Malmstrom 1998; Malmstrom et al. 2005a, b). Critical questions that remain to be explored include the long-term history of B/CYDVs in California; the spatial patterning of virus pressure and virus-host interactions; the nature of interactions between vectors and vector predators; the influence of virus pressure on interactions between different native grass species; and potential positive consequences of infection for natives.

Summary Grasslands in California have undergone a radical transformation in their composition over the past two centuries. While coastal prairie, coast range, and wetter portions of the central valley were likely dominated by native perennial grasses, drier interior grasslands were probably dominated by native annual forbs. A combination of drought, the expansion of crop agriculture with eventual abandonment of

croplands, and year-round livestock grazing all likely contributed to the decline in native grassland species. Introduced annuals, primarily winter-growing grasses and forbs, became dominant during the 1800s, when rates of land use change were very high. It is also possible that introduced pathogens that reduce vigor and seed set of native grasses contributed to their replacement in some areas. Today grasslands throughout the state are heavily invaded by non-native species, primarily of European or Eurasian origin. Annual grass species that are common throughout the state and appear to have become widespread prior to the last century include Avena species, Bromus diandrus and B. hordeaceus, and Hordeum leporinum. Recently spreading grass invaders include Aegilops triuncialis and Taniatherum caput-medusae. Recently spreading forbs include those in the genus Centaurea. Coastal prairie grasslands appear to be becoming increasingly invaded by non-native perennial grasses and, in some areas, by invasive nitrogen fixing shrubs. Non-native plant species are today an integral part of grassland structure, but their impacts are not always desirable. Those species listed in Table 6.1 are considered by the California Invasive Plant Council to inhibit growth of native species, reduce livestock or wildlife forage value, or profoundly

alter ecosystem structure and function in California grassland settings. The documented ecological impacts of nonnative species include competitive suppression of native grass seedlings and adults, replacement of native species following disturbances, alteration of soil water and nutrient resources, and alterations in soil microbial processes. The reversibility of many of these impacts or their importance in restoration needs further study. Plant pathogens in California grasslands have also been little studied and have the potential for a variety of both positive and negative impacts. There is increasing evidence that, like soil microbes, pathogenic microorganisms may play influential roles in shaping grassland structure and composition today. In some cases viruses may decrease the survivorship of native grass seedlings, and recruitment of native seedlings in exotic-dominated communities may be further reduced by the capacity of exotics to increase virus prevalence. In other cases, pathogenic fungi may directly facilitate the establishment of exotic grasses. It is possible that some pathogenic microorganisms may exert positive influence on native populations, but such cases have not yet been documented. Ongoing research on microorganism dynamics in Californian grasslands will bring further insights to this area.

E C O L O G Y O F I N VA S I V E N O N - N A T I V E S P E C I E S

83

RESOURCES

SEVEN

Water Relations K I M B E R LY J. R E EVE R M O R G HAN, J E F F R EY D. C O R B I N, AN D J O H N G E R LAC H

California’s Mediterranean-type climate has very discrete wet and dry seasons that control the availability and use of soil moisture by plants. For the graminoid species that dominate the state’s grassland ecosystems, as well as most other constituents of the community, the strong seasonality of precipitation and water availability is the dominant force influencing their phenological patterns and annual productivity (Chiariello 1989). In this system, the timing of rainfall and of temperatures favorable for growth for vegetation are out of phase. As Major (1988) succinctly described: [The state’s climate] combines the very worst features of arid and humid climates. It is extremely hot and arid in summer, and extremely cool and humid in winter. The supply of water and the need for it are exactly out of phase. The productivity of natural, zonal vegetation of course reflects these climatic disabilities. The growing season is limited by the cool temperatures of winter as well as the summer drought.

temporally, spatially, and as the species traits of the vegetation community change. We then discuss the influence of climatic conditions, including precipitation, on the productivity and species composition of grasslands. Finally, we discuss how the shift in community composition due to invasion of non-native species, including Centaurea solstitialis (yellow starthistle) into inland grasslands (D’Antonio et al., Chapter 6) has influenced soil moisture dynamics in California.

Patterns of Water Availability in California’s Mediterranean Climate Water availability in California grasslands varies as a result of temporal and spatial differences in moisture inputs and evapotranspiration, soil physical factors, and characteristics of plant traits that influence water uptake.

Temporal Variation In this way, California’s grasslands are distinct from grasslands in North American and other temperate regions. Temporal and spatial variation in water availability play a dominant role in the species composition, productivity, and competitive interactions of California’s grasslands. For example, the strong seasonality of rainfall — abundant during the winter but scarce during the summer — likely explains the strong bias in the grassland flora toward annual species (Seabloom et al. 2006). Additionally, differences in water availability between coastal and inland grasslands likely contribute to the relative importance of perennial species in coastal habitats (Elliott and Wehausen 1974; Corbin et al., Chapter 13). In this chapter, we discuss the ecological factors that influence the availability of water to California grassland plants, including abiotic and biotic influences. Specifically, we describe the extent to which soil water availability varies

The most obvious climatic variable influencing the patterns of moisture availability in California grasslands is the strong seasonality in the period of rainfall. The existence of a subtropical zone of high pressure off Oregon in the Pacific Ocean in the summer prevents outbreaks of cold marine air from polar regions from reaching the state, resulting in a summer drought. The zone of high pressure moves further south during the winter, permitting storms to reach California and ending the drought. The net result is a strongly Mediterraneantype climate in most of the state, characterized by cool, wet winters and hot, dry summers. The period of moisture input is mostly concentrated between October and April, when, in most years, over 95% of precipitation occurs. Even within the winter rainy season, extended periods without precipitation are regular features of California’s climate. Null (2006) documented that every year between 1950 and 2006 experienced a period of at least eight days

87

F I G U R E 7. 1 . Seasonal (July-June) precipitation in San Francisco and Los Angeles from 1914

through 2005. Lines indicate trend lines for each location.

without rain during the otherwise wet months of December and January. The mean drought duration was 19 days; the longest was 42 days. There are indications that these “midwinter droughts” may influence community composition in grasslands by favoring perennial grasses or forb species at the expense of annual grasses (Pitt and Heady 1978; Hamilton et al. 1999; Corbin et al., Chapter 13). For example, Hamilton et al. (1999) found that perennial grasses such as Nassella pulchra are less susceptible to simulated midwinter drought (35-day duration) than annual grasses. The more developed root systems of perennial grasses and forbs may make them better able to tolerate such periods of low soil water availability. In addition to temporal variation in moisture inputs, soil moisture levels are also influenced by climatic conditions and biotic activity. The balance between water inputs and losses via evapotranspiration determine the availability of water to vegetation over the course of a growing season (Box 7.1). Here evapotranspiration is the sum of evaporation from surfaces, such as soil or plants, and transpiration loss through plant stomata. Potential evapotranspiration in California grasslands is highest in the warmer summer period and lowest during the winter (Major 1988). During the winter and early spring portions of the wet season there is little plant canopy area in grasslands, evaporative energy is low, and precipitation is abundant. As a result, soil moisture levels are highest. Because precipitation is generally more than adequate for plant growth during the winter, there is a soil water surplus (Major 1988). In the mid-late spring, as both temperature and day length increase, a dramatic increase in plant growth results in higher amounts of evapotranspiration and creates conditions of soil water deficit—that is, potential water loss via evapotranspiration is greater than actual water loss (Chiariello 1989). As water becomes more limited in the late spring, plant growth and evapotranspiration start to decrease; annual plants set seed and die, while perennials that cannot access deep soil water

88

RESOURCES

stores or other water sources such as fog go dormant (Chiariello 1989). As the dry season progresses, plants deplete soil water stores. Summer and fall are marked by high loss of moisture via evaporative demand and no precipitation. Surface soils are typically very dry, though moisture content may be maintained at deeper soil horizons, depending on the summer activity of vegetation. Where deep-rooted vegetation can tap into deeper soil water stores, water pools over 1 meter below the soil surface may be depleted during the summer (Holmes and Rice 1996; Dyer and Rice 1999). The soil moisture that is lost over the summer via evaporation and evapotranspiration is typically recharged by the fall and winter rains. In years when wet season precipitation is low, however, there may not be full recharge of soil moisture stores. In these dry years, the surface soils may be recharged to maximum moisture content, but the deep soil layers may never reach field capacity (Box 7.1). A water balance diagram for the growing season developed by Major (1988) suggests that California grasslands experience an annual water deficit. This water deficit may be as low as 106 millimeters in wetter areas on the northwest coast and as high as 1,089 mm in the driest sites in the California deserts (Major 1988). However, water availability to plants is limited during the hottest, driest part of the year, so total water loss through evapotranspiration is lower than it would be if California grassland soils were saturated all year. Against the backdrop of the strong seasonality of precipitation in the state, the region also experiences significant year-to-year climatic variation (Figure 7.1). For example, although the mean seasonal precipitation in San Francisco and Los Angeles is 533 and 378 mm per year, respectively, these means do not show the tremendous range in precipitation totals. Seasonal precipitation in San Francisco has ranged over sixfold over a 22 year period, from 182 mm in 1976–1977, to 1,199 mm in 1997–1998, respectively. In Los Angeles, the range is eightfold, from 112 mm in 2001 – 2002

B OX 7.1 CALCULATING SEASONAL SOIL MOISTURE AVAILABILITY

In years where winter rains completely recharge soil water, we can estimate the date at which soil moisture stores begin decreasing. We can roughly determine when the net water loss begins by plotting cumulative rainfall less cumulative potential evapotranspiration for the period from February 1 through June 1. Figure 7.2 shows this type of plot for Davis in the Sacramento Valley of central California from 1995 through 1998. It illustrates the dramatic differences between years in the amount of available soil moisture and when this soil moisture becomes available. The date where the plot crosses the zero line indicates the point in time when plant transpiration begins depleting stored soil moisture, and this date is also highly variable between years. Potential evapotranspiration values, which represent the maximum soil depletion rate, can be obtained in electronic form from many weather stations that are connected to the California Irrigation Management System (Snyder and Pruitt 1992). These calculations can be further refined by determining the potential water storage in soil. Tables in the back of USGS Soil Surveys provide estimates of the storage capacity of each soil type in a region. Storage capacity is based on soil texture and porosity. If we assume that all of the precipitation would infiltrate if given a sufficient volume of soil for its storage, the volume of the soil storage reservoir, expressed as depth in millimeters, is simply the amount of precipitation less the soil’s available water storage capacity. A rough approximation of the earliest date of lethal dehydration can then be calculated by dividing the volume of the soil storage reservoir by cumulative potential evapotranspiration rate and adding the resulting number of days to the date at which evapotranspiration began to outstrip cumulative inputs (zero line in Figure 7.2).

400

200

0

-200

-400

-600

-800

1995 1996 1997 1998

February

June

FIGURE 7.2. Cumulative rainfall minus cumulative

potential evapotranspiration from February 1 to June 30 at Davis, California.

to 946 mm in 2004–2005. The years of greatest rainfall coincide with strong episodes of El Niño/Southern Oscillation (ENSO). In such years, higher than usual sea surface temperatures in the tropical waters of the central and eastern Pacific Ocean disrupt the position of the jet stream and alter worldwide precipitation patterns. The eight wettest years in San Francisco, and eight of the ten wettest years in Los Angeles, were strong or moderate ENSO seasons, as categorized by the Western Regional Climate Center (WRCC 2004). These above-average rainfall years can have a great influence on community composition and productivity in California grasslands. Global climate change could potentially increase

the frequency of ENSO events (McCarthy et al. 2001; Timmerman et al. 1999) and create dramatic shifts in composition of California’s grasslands. Years during which precipitation is well below normal, particularly when they occur in succession, can also have significant influences on the productivity and community composition of grasslands. At least eight multiyear droughts have occurred in California since 1900. Droughts that exceed three years are uncommon, though recent occurrences include 1929–1934, 1947–1950, and, most recently, 1987 – 1992 (Figure 7.1). Severe droughts in 1850–1851 and in 1862–1864 have been implicated in the shifts in species composition in

WA T E R R E L A T I O N S

89

many grasslands from domination by native perennial bunchgrasses to domination by introduced annual grasses and forbs (Heady 1988; Major 1988; D’Antonio et al., Chapter 6). Such droughts will undoubtedly continue to influence grassland structure and function in the future, including productivity, species composition, and the range and survival of nonnative introduced species.

Spatial Variation Although the seasonal pattern of precipitation is consistent throughout much of the state, particularly within the range of grasslands, there are strong geographic differences in precipitation amounts each season. Precipitation inputs are consistently lower in southern parts of the state than in central and northern California (Figure 7.1). Mean annual precipitation also decreases as distance from the ocean increases, though not as dramatically as the north-south gradient (Major 1988). Distance from the ocean can also influence water availability, although differences in summertime conditions along a coast-inland gradient may be even more significant than differences in winter precipitation. Evaporative loss of water during the summer from interior grasslands is substantially greater than from the maritime coastal habitats, where summertime coastal fog reduces summer temperatures and evaporative energy (Major 1988). There are also differences in summertime moisture inputs between coastal and inland habitats. Coastal habitats are regularly bathed in fog during the summer (Azevedo and Morgan 1974; Ingraham and Matthews 1995; Dawson 1998), providing regular, if small, precipitation inputs that are not available to habitats outside the zone of coastal influence. This moisture input provides a substantial proportion of the water found in the tissues of perennial species in California coastal grasslands and forests during the summer drought (Ingraham and Matthews 1995; Dawson et al. 2002; Corbin et al. 2005). Finally, topography can influence precipitation; for example, the low precipitation amounts for the Carrizo Plain grasslands occur because it is located in the rain shadow of the central Coastal and Transverse Ranges.

Soil Physical Factors: Texture and Soil Depth Soil physical factors also influence how much water a plant has access to and can use. Soil water availability to vegetation is highest when the soil is deep, plant roots have access to a large volume of soil, and soil water is held loosely, allowing free movement of moisture to plant roots (Singer and Munns 1999). In sandy soils, water is held loosely; as long as water does not drain too rapidly from these soils, it is available for easy uptake by plants (Barbour et al. 1999). Because of this loose hold on water, sandy soils tend to have lower water storage capacity than finer-textured soils. Soils with high clay content, high humus content, or both are able to store larger amounts of water, but this water is more tightly held than in sandy soils and may not be as readily taken up by plants

90

RESOURCES

(Barbour et al. 1999). Research from other systems suggests that productivity in dry climates is higher in coarser-textured soils because these soils experience lower evaporative loss, while productivity in wet climates is higher in finer soils because these soils lose less water to leaching (Noy-Meir 1973). The amount of annual precipitation separating a “dry” climate from a “wet” one has been proposed as 37 cm (Sala et al. 1988) or 80 cm (Epstein et al. 1997) for the grasslands of the Central United States. Because California has a very different rainfall pattern than the Central United States does, defining it as wet or dry presents its own challenges. Soil survey data from Glenn and Tehama counties in Northern California suggest that soils that have the highest percent clay content and, thus, the highest water-holding capacity, are the most productive, but this may be a simplistic assumption (Gowans 1967; Begg 1968). More research is needed before we understand the full extent of influence that soil texture has on soil water availability in California grassland soils. Soil depth also influences plant water uptake. Shallower soils mean less total volume of stored soil water and, therefore, less plant-accessible water. Belcher et al. (1995) observed a linear relationship between soil depth and plant aboveground biomass, reflecting the decrease in soil resources, including moisture, as soil depth decreases. Dyer and Rice (1999) found that restricted soil depth resulted in lower total soil water availability and greater water stress of Nassella pulchra; grasses in deeper soil produced more culms and produced those culms earlier, while grasses in shallower soil needed an extra growing season before they produced their first culms.

Plant Traits: Rooting Patterns and Phenology Constituents of grassland communities vary in their rooting depth and architecture, so the relative dominance of different species can influence patterns of moisture availability. Rooting depth determines how much of the soil water profile a plant can access and deplete. Rooting depth, in turn, is closely related to the phenology of the various species. Species such as cool-season annual grasses, which do not survive the summer drought, do not develop roots that access deeper soil water stores. By contrast, species such as warm-season annuals and perennial species, which must maintain tissues through the summer drought, typically develop deeper root systems that can access water in deeper soil layers. Non-native cool-season annual grasses typically concentrate their roots in the top 30 centimeters of the soil (Hull and Muller 1977; Holmes and Rice 1996). Most of these species finish their growth cycle at the end of the rainy season. This dormancy at the start of the dry season is built into annual grass physiology, and experimentally watering annual grass plots does little to delay their timing of seed formation and senescence (Jackson and Roy 1986). As a result, under annual grass – dominated sites, soil moisture availability in deeper (⬎60 cm) depths is relatively high (e.g., Borman et al. 1992,

Holmes and Rice 1996; Dyer and Rice 1999; Enloe et al. 2004; Gerlach 2004). For example, water relations studies on natural areas in the San Dimas Experimental Forest found that the soil moisture under an exotic spring-flowering annual grass (Lolium multiflorum) remained above the permanent wilting point at depths from 60 cm to 2.1 meters and was at field capacity at depths from 2.1 m to 3.6 m (Rowe and Reimann 1961; Hill and Rice 1963). However, not all cool-season annual species share the same rooting patterns as the annual grasses. For example, Erodium botrys, an introduced annual forb, allocated 30 – 35% of its root biomass to soil depths below 50 cm (Gordon and Rice 1992), and therefore could, presumably, deplete deeper soil moisture reservoirs than annual grasses could. Warm-season forbs and grasses go through much of their growth cycle during the driest part of the year, so they require deeper rooting systems to tap into deep soil water stores. A few studies have looked at native summer-flowering annual species such as Stephanomeria virgata (Rowe and Reimann 1961), Holocarpha virgata (Green and Graham 1957), Hemizonia congesta (Chiariello 1989; Huenneke et al. 1990), and Eremocarpus setigerus (Gerlach 2004) and found that summer-flowering annuals use much more soil moisture during the summer than spring-flowering annuals. The warm-season native grasses Aristida oligantha and A. ternipes var. hamulosa produce root systems with both deep and shallow roots, with the deeper taproot of A. oligantha extending down 122 cm or deeper into the soil profile (Laude and Meldeen 1958). These plants grow throughout the summer, flower in the fall, and use deep soil moisture stores. Cutting A. oligantha’s taproot at 60 cm below the soil surface in late July stops access of the plant to deep soil moisture stores and results in wilting of the plant (Laude and Meldeen 1958). Rowe and Reimann (1961) reported that a native summer-flowering herbaceous dicot (Stephanomeria virgata) depleted soil moisture levels below 60 cm to a greater extent than the cool-season annual grass Lolium multiflorum. Cool-season perennial grasses typically have a longer growing season than cool-season annual grasses, but generally do not stay active as far into the dry season as the warmseason plants do. In order to allow access to deep soil water in the first part of the dry season, they need to produce a root system that extends deeper into the soil profile than that of cool-season annual grasses. Thus, the root systems of perennial grasses are distributed throughout at least the top 60 cm of the soil profile. Hull and Muller (1977) compared the non-native annual Avena fatua to the native perennial Nassella pulchra and found that A. fatua roots are primarily located in the upper 30 cm of the soil profile while N. pulchra roots extended down to 1 m in depth. Holmes and Rice (1996) found that perennial bunchgrasses (N. pulchra and Elymus glaucus) depleted soil moisture during the summer to a greater extent than cool-season annual grasses did. The ability of perennial grasses to access such deeper stores of water during the summer, however, may be influenced by competitive interactions during the winter (Corbin et al.,

Chapter 13). Dyer and Rice (1999) found that plots containing perennial grasses and plots containing a mix of perennial and annual grasses fully depleted the soil water in shallow soil (⬍65 cm deep soil plots at Jepson Prairie (Solano County)). In the same study, when perennial and annual grasses were grown together in deep soil, they depleted shallow soil water stores but left water available below 60 cm. When annual-grass competition was removed, however, the perennial grasses used water much deeper in the soil profile (60 – 120 cm) (Dyer and Rice 1999). Thus, they concluded that competition from annual grasses can impair the ability of perennial grasses to develop deep root systems and access deep soil water stores. Cool-season perennial grasses in the dry inland regions of California generally go dormant during the summer, with little or no green aboveground tissue and no growth. However, some of these perennial grass species have been observed to break dormancy in the fall even before the fall rains occur (Laude 1953). This means that their deep root systems remain active and able to access soil water for use in breaking dormancy and greening up. Breaking dormancy before the rainy season could allow these grasses to take advantage of the nutrient pulses released by the first fall rains. Perennial grasses in coastal sites, may have different patterns of root distribution and activity. Corbin et al. (2005) found that coastal perennials are able to use water from summer fog as a water source and concentrate water uptake in the top 10 cm of the soil profile. Even though the soil was wetter deeper into the soil profile, the perennial grasses in their study areas maintained high shallow root biomass and active uptake near the soil surface to take advantage of regular fog deposition. Thus, these coastal grasses are less likely to deplete deep water sources as long as shallow soil water is available.

Effects of Water Availability on Vegetation Composition and Productivity Because of the economic importance of grassland ecosystems to rangeland activities, researchers have tried to develop methods to predict productivity of grasslands from climatic conditions (e.g., Duncan and Woodmansee 1975; Pitt and Heady 1978; George et al. 1989). Pitt and Heady (1978) used stepwise multiple regression to relate weather conditions (including monthly temperature and rainfall values) and community composition in annual grasslands at Hopland Field Station, and they found broad support for the use of climate as a predictor for grassland productivity (NPP, or standing crop in annual grasslands) and community composition. Interestingly, total seasonal rainfall was not a good predictor of annual NPP. Instead, the timing of the rainfall was more important than the total amount; grassland standing crop was positively correlated with rainfall levels in the mid-winter months of December, January, and February but negatively correlated with rainfall levels in March and April.

WA T E R R E L A T I O N S

91

TA B L E 7.1 Variation in Average Annual Precipitation and Spring Peak Standing Crop for California Grasslands along a North-South Gradient

Location Northern sites Central sites Southern sites NOTE:

Annual precipitation (cm) 150–160 65–100 16–20

Peak standing crop (kg/ha) 30–35 20–45 5–15

Data summarized from Bartolome et al. (1980).

(Abundant precipitation in the spring is associated with cool weather in the Hopland region, thereby decreasing productivity.) Climatic conditions at the time of germination also have strong influences on NPP of annual ecosystems. Pitt and Heady (1978) found that NPP was positively correlated with temperatures in October and November. Because temperatures during the December – February period are cool enough that growth rates slow considerably (Pitt and Heady 1978; Gulmon 1979), late rains or dry autumns may influence the period during which productivity approaches its highest levels. Temperatures at Hopland influenced NPP, as well (Pitt and Heady 1978). Standing crop in March and June was positively correlated with warm mean temperatures from November through February. Standing crop in June was also negatively correlated with dry conditions in the months October, November, March, and April. Finally, productivity is influenced by latitude. The variation in rainfall amount from north to south creates a gradient in maximum productivity, with Northern California sites tending towards higher peak standing crop than Southern California sites (Table 7.1). The quantity of rainfall and temperatures at the time of germination affect both seedling numbers and species composition in annual grasslands (Murphy 1970; Pitt and Heady 1978). Climatic conditions early in the season have been shown to influence seedling densities, as well as the relative abundance of seedlings within the annual community. The existence of significant variation in early growing season conditions, and the responses of standing crop and species’ relative abundances to these climatic conditions, have led to the widespread acceptance that the relative cover of different components can be predicted from the conditions at germination (e.g., Talbot et al. 1939; Pitt and Heady 1978). According to this “grass, clover, filaree year” framework, annual grasses are favored in years when germinating rains begin early (while temperatures are still warm) and moisture input through the autumn and winter is regular. Other community components are favored in years where germinating rains are delayed into November or December (when temperatures are colder) or when fall rainfall is sparse.

92

RESOURCES

The absence of droughts during the growing season, particularly during the fall and spring, encouraged dominance by grasses; the exact importance of particular grass species varied according to other climatic conditions such as freezing temperatures during the winter. Warm and dry conditions after germination (October and November) and drought in the spring were associated with Erodium (filaree). Dry autumns with rainfall in the late fall or early winter encouraged clovers, though this effect was, most likely, due to decreased production of annual grasses rather than due to a positive effect on clovers. The effects of climate change on grassland community composition and productivity are the subject of increasing attention (Dukes and Shaw, Chapter 19). A three-year study at Jasper Ridge (Santa Clara County) measured grassland responses to conditions simulating climate change, including increases in precipitation, and saw strong interactions among global change components (Shaw et al. 2002). Increases in net primary productivity due to increased precipitation as a single factor were counteracted by increases in CO2 in multifactor manipulations (Shaw et al. 2002). Increased precipitation also changed functional group abundance (Zavaleta et al. 2003b). Changes in air CO2 concentration, temperature, and precipitation resulted in a 50% increase in the abundance of forbs, which would greatly change the structure and function of California’s grasslands. (For more detail on the interaction of increased precipitation with N deposition and climate warming as they might affect grassland soil water potential and productivity, see Dukes and Shaw, Chapter 19). In addition to the Jasper Ridge climate change study, Thomsen et al. (2006b) found that increasing the length of the growing season by artificial spring rainfall events increased grassland susceptibility to invasion by a coldseason European perennial grass. Suttle and Thomsen (Suttle et al. 2007) did not find strong effects of elevated winter rainfall on grassland productivity. Further research will help us develop a deeper understanding to predict how new changes, and their subsequent effect on water relations, will change the California grassland communities.

The Legacy of Community Conversion on Soil Moisture Patterns California’s Central Valley grasslands currently consist of a mix of annual and perennial grasses and forbs, with nonnative annual grasses dominating in many areas. However, research suggests that perennial bunchgrasses were once a prominent component of many of these grasslands. It is important to note, however, that recent research suggests that not every site currently dominated by non-native annual grasses was once a perennial grassland (Hamilton 1997a; Holstein 2001). In the sites that were once perennial grasslands, the introduction of non-native annual grasses, in conjunction with drought, overgrazing, and conversion of grasslands to agriculture, resulted in loss of most stands of native perennial grasses (Burcham 1957; Dasmann 1973;

Heady 1988; Menke 1989). Because of the difference in rooting pattern and soil water access between perennial and annual grasses, the conversion of California grasslands from perennial to annual systems may have resulted in hydrological changes to grassland soils and, most importantly, an increase in deep soil moisture. The increase in available soil moisture following grassland conversion in sites that previously were dominated by perennial grasses may have promoted the establishment and spread of new invaders (i.e., “Fluctuating Resource Theory of invasion”; Davis et al. 2000). One invader that may have taken advantage of increased soil moisture is the deep-rooted invasive forb Centaurea solstitialis (yellow starthistle). C. solstitialis is able to use deep stores of soil water, and may deplete even deeper stores of soil moisture than those used by perennial grasses. For example, Enloe et al. (2004) created communities dominated by C. solstitialis, annual grasses, or perennial grasses and measured soil water content at depths from 30 to 150 cm over three growing seasons. They found that soils under C. solstitialis contained a water content of 18.35 ⫾ 0.24%, while under annual grasses the soil water content was 22.76 ⫾ 0.24%, and under perennial grasses the soil water content was 19.74 ⫾ 0.28%. Thus, soils under C. solstitialis were significantly drier than soils under annual grasses, with soil moisture levels under perennial grasses intermediate between the two (Enloe et al. 2004). In a study in southwest Oregon, Borman et al. (1992) compared the depletion of soil moisture by annual grasses, perennial grasses, and C. solstitialis. They found that plots dominated by annual grasses had around 50% higher soil moisture than plots dominated by perennial grasses or C. solstitialis. Similarly, Gerlach (2004) found that perennial grasses extract more deep soil water than annual grasses do and, in some studies, have been shown to extract as much deep soil water as C. solstitialis. Thus, the invasion of California’s Central Valley perennial grassland sites by annual grasses may have left deep soil moisture resources untouched and promoted the invasion of these sites by C. solstitialis. Simberloff and Von Holle (1999) call the process whereby early invaders facilitate invasion by later species, eventually resulting in the total loss of native species, “invasion meltdown.” Because research has shown a positive correlation between C. solstitialis invasion success and soil water availability during late spring and summer (Dukes 2001a), it appears that the conversion of California grasslands to native annual grass dominance and subsequent widespread starthistle codominance is a case of invasional meltdown.

We may be able to reverse the decrease the availability of deep soil moisture and reduce C. solstitialis invasion by restoring grasslands with native perennial grasses. Both C. solstitialis and Nassella pulchra allocate early resources to root growth (Thomsen et al. 1989; Roché et al. 1994; Holmes and Rice 1996; Gerlach et al. 1998; Dyer and Rice 1999). Studies have found decreased cover and higher water stress in C. solstitialis growing in shallow or dry soils (Roché et al. 1994; Sheley and Larson 1995). Thus, if restoration of grasslands with perennial grasses can deplete deep soil water stores, we may be able to reduce the invasion success of invaders such as C. solstitialis (Borman et al. 1992; Roché et al. 1994). Care must be taken, however, to ensure that restoration with perennial grasses is not undertaken as a blanket approach to restoring all C. solstitialis sites but, instead, is used when research suggests that these perennial grasses were once an important component of the vegetation in that site; some sites may require different restoration approaches to manage C. solstitialis and return the site to something more similar to its original plant community.

Conclusion Soil water availability in California grasslands varies both temporally and spatially, with vegetation characteristics such as rooting depth also having an effect on soil water access and depletion. Thus, changes in rainfall amount and timing, amount of transpiration loss through vegetation, or shifts in community composition may all influence soil moisture availability and, in turn, influence the productivity and species composition of grasslands. Because the effects of different plant species can change water availability, shifts in community composition due to invasion of non-native species have the potential to change soil moisture dynamics. We suggest that replacement of perennial grasses with more shallowly-rooted annual grasses in inland areas may have created an increase in deep soil moisture stores. This then resulted in a soil that was particularly at risk from invasion by deep-rooted Centaurea solstitialis (yellow starthistle), resulting in additional changes to soil moisture dynamics.

Acknowledgments We would like to thank Meredith Thomsen, Andy Dyer, and Carla D’Antonio for their helpful comments and advice on this chapter.

WA T E R R E L A T I O N S

93

EIGHT

Mechanisms Determining Patterns of Nutrient Dynamics VALE R I E T. EVI N E R AN D MARY K. F I R E STO N E

Nutrient availability is a critical controller of plant species composition and productivity and is influenced by the interaction of a number of biotic and abiotic factors (Chapin et al. 2002). In this chapter, we explore the controls over nutrient pools and fluxes across different California grassland types. We will begin with an in-depth discussion of the exoticdominated annual grassland type, where most biogeochemical studies have occurred. Patterns in these annual grasslands will then be compared to those in native grasslands and oakgrassland matrices. Finally, we will review the roles of organisms other than plants in influencing nutrient dynamics.

General Patterns of Nitrogen Pools and Fluxes in California Annual Grasslands Most nutrient studies in California grasslands have focused on nitrogen (N), probably because it is the most common limiting nutrient to plant productivity in these grasslands (Hoglund et al. 1952; Woolfolk and Duncan 1962; Jones 1963; Harpole et al., Chapter 10). The relatively few comprehensive nitrogen budgets in California grasslands (Jones and Woodmansee 1979; Jackson et al. 1988; Center et al. 1989) report a wide range of N pools (Table 8.1) and fluxes (Table 8.2) within and across studies, but they also highlight some strong consistent trends (Table 8.1). Most N is stored as soil organic N (concentrated in the top 4 cm of the soil profile), with less than 10% of ecosystem N in plant and microbial biomass at peak standing plant biomass in May (Jones and Woodmansee 1979; Jackson et al. 1988; Herman et al. 2003). The top 10 cm of soil contain 85% of live root biomass N and 75% of microbial biomass N (Jackson et al. 1988; Jackson et al. 1989), with most of that within the top 1–5 cm (Evans et al. 1975; Woodmansee and Duncan 1980). The distribution of plant N in belowground vs. aboveground tissues has been estimated to range between 0.9:1 and 1.33:1 (Jackson et al. 1988; Center et al. 1989) (Table 8.1).

94

Nitrogen inputs and outputs are highly variable across years and sites (Table 8.2). Inputs through biological N fixation largely depend on the prevalence of legumes, although inputs of 0.18–12 kg N/ha/yr can be seen even in grass-dominated sites (Table 8.2). There is a lack of comprehensive data on total atmospheric N deposition because most measurements of N deposition do not include all sources of wet, dry, and fog deposition (Fenn et al. 2003b). Measured inputs of atmospheric deposition into most grassland sites are less than 2 kg N/ha/year (Jones and Woodmansee 1979; Center et al. 1989; Takemoto et al. 1995; Air Resources Board 1995), but modeling studies estimate that total deposition is more likely to range from 4–7 kg N/ha/year in most grassland sites, with higher levels (9–12 kg N/ha/year) in some parts of Southern California (Fenn et al. 2003b). Nitrogen losses can occur through leaching, gaseous loss, and soil erosion. Estimates of N losses through leaching range from 0.18–63 kg N/ha/year, with high year to year variability. Typical leaching rates of grass-dominated sites are at the far low end of this range; the high estimates come from systems dominated by legumes and from studies using tank lysimeters (Jones and Woodmansee 1979; Davidson et al. 1990; Maron and Jeffries 2001). Most studies from intact grasslands measure N leaching in the range of 1 – 4 kg N/ha/year. The most comprehensive study on N leaching in California grasslands demonstrated that, over 20 years, mean NH4
leaching was less than 0.1 kg N/ha/yr, while NO3 leaching averaged 1.59 kg N/ha/yr, and ranged from 0.18 to 3.6 kg N/ha/yr (Lewis et al. 2006). Few studies have closely followed leaching of dissolved organic N (DON), but at some time points, up to 87% of leached N can be in the form of DON (S. Parker, personal communication), suggesting that N leaching may be greatly underestimated in these grasslands if only NO3 or NH4
are measured. Annual estimates of N gas loss are, to our knowledge, unknown, but peak seasonal instantaneous rates range from less than 1 to 13 ng/cm2/hr

TA B L E 8.1 Range of Pools of N (kg N/ha) at Peak Standing Biomass in the Spring in Grass-Dominated Sites

Pool Aboveground live plant Aboveground litter Live roots Dead roots Soil microbial biomass Available N Soil organic N

TA B L E 8.2 Range of Fluxes of N (kg N/ha/yr)

Flux

Flux (kg N/ha/yr)

N fixation, grass-dominated N fixation, legume-dominated Atmospheric deposition

0.5–12 50–200 1–17.8 (most estimates range from 2.0 to 7)

Pool size (kg N/ha) 33–80 11–28 20–80 17–29 10–133 1–44 2,000–6,000

NOTE:

Values based on Jones and Woodmansee 1979, Woodmansee and Duncan 1980, Jackson et al. 1988, Center et al. 1989, Dahlgren et al. 1997.

(0.0001–0.0013 kg N/ha/hour) (Hungate et al. 1997c; Herman et al. 2003), and likely do not significantly contribute to annual N loss (Herman et al. 2003). Nitrogen inputs and outputs are usually small compared to internal N cycling rates (Table 8.2). Nitrogen turnover is rapid in annual-dominated systems, with particularly large seasonal changes in plant and litter N pools (Jackson et al. 1988; Jackson et al. 1989; Davidson et al. 1990). At senescence, 63 – 77% of aboveground plant nitrogen is translocated to seeds, and only 23 – 37% remains in the aboveground litter, while N does not appear to be retranslocated from senescing roots (Woodmansee and Duncan 1980; V. T. Eviner and C. E. Vaughn personal communication). Root litter tends to fully decompose within one year (Savelle 1977; V. T. Eviner, personal observation), while it takes roughly 2.5 years for aboveground litter to decompose completely (Heady et al. 1991), with loss of 59 – 79% of aboveground litter mass within the first year (Savelle 1977; Jones and Woodmansee 1979). Litter N loss does not necessarily follow the pattern of litter mass loss (Dukes and Field 2000) because decomposing litter frequently accumulates N (Center et al. 1989; Hart et al. 1993). Some studies have shown nearly complete turnover of litter N within a season (Jackson et al. 1988), while others show virtually no litter N loss despite large litter C and mass losses (Henry et al. 2005). Nitrogen budgets of California annual grasslands have indicated that decomposition and N fixation often cannot meet plant demands for N, suggesting that N is made available from seedling thinning throughout the growing season (Woodmansee and Duncan 1980; Pendelton et al. 1983; Vaughn et al. 1986; Center et al. 1989; Heady et al. 1991). As mentioned above, about 70% of aboveground N is in seeds at the time of plant senescence. On average, there are 60,000 germinable seeds per m2 at the start of the growing season (Bartolome 1979), with the number reaching as high as 300,000 (Young et al. 1981). Over 90% of these seeds germinate shortly after the first significant rainfall, but within the first seven weeks of the growing season, 50–75% of these seedlings die (Bartolome 1979, Young et al. 1981). These are

N release through aboveground litter decomposition N release through root litter decomposition N release through seeding thinning Net N mineralization Total plant N uptake N leaching N sediment lossb N gas loss

47–71 32–65 40–104 57–125 68–119 0.18–13 (most sites 1 – 3)a 0.044–0.91 (20 year average of 0.38) No annual loss numbers available, but rates at fall wet-up range from 1–13 ng/cm2/hr

NOTE: Values based on Shock et al. 1984, Vaughn et al. 1986, Davidson et al. 1990, Hart et al. 1993, Takemoto et al. 1995, Air Resources Board 1995, Fenn et al. 2003, and unpublished data aTank lysimeter studies report as high as 63, but likely overestimate. bN sediment loss estimate based on 20 year sediment export data (Lewis et al. 2006) at the Sierra Foothills Station (Yuba County), along with measurements of soil %N from this site (Davidson et al. 1990).

very young seedlings containing almost no structural compounds and thus are a substantial source of labile carbon and nutrients, leading to an early season N pulse that exceeds the total N content of aboveground litter (V. T. Eviner and C. E. Vaughn, personal communication). Despite the death of half of the seedlings within the first two months of the growing season, seedling density is high at the onset of winter, at 20,000 – 40,000 individuals per m2. Thinning continues through the growing season, resulting in 8,000–20,000 individuals per m2 by the end of the growing season in late spring (Heady 1958). Again, these dying seedlings provide a relatively labile source of N, such that over the growing season, N inputs from seedling thinning are 66–170% of N inputs from decomposition of litter senesced at the end of the growing season (V. T. Eviner and C. E. Vaughn, unpublished data). The large amount of N released from senesced litter and seedling thinning is retained through the high uptake rates of plants and microbes. While NO3 uptake by biota is considerable (Jackson et al. 1989; Schimel et al. 1989; Davidson et al. 1990), both the plant and soil microbial communities take up substantially more NH4
than NO3 (microbes take up 4–5 times more NH4
than NO3, plants 30–250% more). The soil microbial community can take up N at a much

MECHANISMS DETERMINING NUTRIENT DYNAMICS

95

higher rate than plants (up to 5 – 11 times more 15NH4
and 2–3 times more 15NO3 over 8–24 hours) (Jackson et al. 1989; Davidson et al. 1990). It is important to recognize that these conclusions are drawn from short-term competition experiments (24 hours or less), and that inorganic and microbial N pools can rapidly turn over in these soils. The mean residence time (MRT) in bulk soil has been calculated as 1.6 days for NH4
, 8.4 days for NO3 (Herman et al. 2003), and 17 days for bacterial biomass N (Herman et al. 2006). Each time the microbial N turns over, a fraction of it is taken up by the plants, so that over time, an increasing proportion of a given N pulse will be seen in plant biomass (Jackson et al. 1989). Total pools of N in plant versus microbial biomass can vary substantially. Different studies have shown that N pools are higher in the microbial biomass than in plant biomass (Jackson et al. 1988), higher in the plant pool (Jones and Woodmansee 1979), or that the relative size of these two pools can change seasonally within a given site (Jackson et al. 1989).

Seasonal Dynamics in California Annual Grasslands While a static budget can provide a quick glance at the important nitrogen pools and fluxes in California grasslands (Tables 8.1, 8.2), it can be a misleading representation of this system. N pools and fluxes vary strongly across seasons, years and sites. While many factors are responsible for high variability in N dynamics, the strong seasonality in these patterns is one of the most pronounced features of California grasslands, and on a gross scale, these seasonal patterns are similar across sites and years (Vaughn et al. 1986; Center et al. 1989). These grasslands are strongly influenced by a Mediterranean climate, which is marked by precipitation in the fall through the spring, with temperatures being low in the winter and high in the summer (Figure 8.1a). Seasonal patterns of ecosystem processes are strongly controlled by the relationship between temperature and moisture. Because the timing of ideal temperature and moisture conditions can vary greatly across sites and years, we will refer to seasons as defined by the moisture and temperature conditions shown in Figure 8.1a, rather than assigning them to months of the year (but see Box 8.1 for an example of the timing of these environmental conditions at one site).

Climate and Plant Growth The onset of precipitation marks the beginning of the growing season by stimulating plant germination, and temperature during the rainy season controls plant growth rates (George et al. 1988). Because ideal temperature and moisture conditions are out of phase with one another (Evans et al. 1975; Major 1988; Reever Morghan et al., Chapter 7) (Figure 8.1a), there are only brief periods in the fall and spring when temperature, light, and moisture are favorable to plant growth (Figure 8.1b) (Evans and Young 1989; Chiariello 1989). After a brief peak of rapid plant growth in

96

RESOURCES

the fall, temperatures decrease and aboveground growth decreases and stays relatively low through the winter, but belowground growth continues (Savelle 1977; Heady et al. 1991). Both aboveground and belowground plant growth rates increase again as temperatures rise in the early spring, and within a few weeks of this temperature increase, most annual root growth has occurred (Jackson et al. 1988). Aboveground growth rates rapidly increase through spring until the system begins to dry out, and the plants begin to senesce (Figure 8.1b). At some sites with a prevalence of late season annuals (e.g., on serpentine soils), a modest amount of plant growth can continue through most of the summer (Chiariello 1989).

Seasonality of N Pools and Fluxes The patterns discussed in this section, and illustrated in Figure 8.1, are best guesses at general seasonal N dynamics in California grasslands. Given the high variability in nutrient dynamics, there have been relatively few studies across sites and years (Heady 1991) that can be used to generalize patterns of nutrient dynamics. In addition, while some parts of the growing season (e.g., fall wet up and peak biomass) have been intensively studied, data is relatively sparse at other points in the growing season, so these relationships should be viewed as hypotheses to be tested with further studies. SUMMER

Because most N cycling within the soil derives from litter and newly dead seedlings, we will start at the end of the season, when litter and seeds are produced. As the hot, dry summers begin (Figure 8.1a), most of the dominant annual plants have senesced, and in these senesced plants, 63–77% of aboveground N is in seeds, while the rest is in litter (V. T. Eviner and C. E. Vaughn in preparation). There is no evidence that plants retranslocate N from their roots during senescence (Nambiar 1987; Gordon and Jackson 2000), so root N contents likely stay stable during seed and litter production. Plant nutrient uptake in most annual-dominated interior grasslands is likely minimal (Figure 8.1c) due to the senescence of a large proportion of the vegetation. However, in grasslands with high proportions of summer annuals which germinate in the spring and grow after others set seed (Bartolome 1989), plant N uptake during the summer can be up to 10 kg N/ha, approximately 8% of plant uptake between October and June (Chiariello 1989). It is often assumed that microbial activity is minimal during these dry months, but there is ample evidence that integrated over the whole summer, microbial pools and fluxes can play important roles in the annual cycling of N (Figure 8.1c). When Hart et al. (1993) added 15N to soils in early summer, 9–15% of added 15N was taken up by the microbial biomass by the end of summer, and soil microbial N (as determined by a chloroform fumigation method) has been shown to reach its annual peak by the end of the summer (Jackson et al. 1988; S. Parker, personal communication).

F I G U R E 8.1. Seasonal variations in climate, plant and microbial activity, and N dynamics. Other than the temperature and precipitation data

(1a), all graphs are approximations of general seasonal patterns based on available studies, many of which do not provide frequent enough sampling dates to determine the exact shape of the graphs. In some cases, studies show different patterns, and these graphs are based on the most prevalent patterns across studies. Graphs are based on (Biswell 1956; Savelle 1977; Jones et al. 1977; Jones and Woodmansee 1979; Woodmansee and Duncan 1980; George et al 1988; Jackson et al. 1988, Jackson et al. 1989; Schimel et al. 1989; Center et al. 1989; Davidson et al. 1990; Heady et al. 1991; Hungate et al 1997; Maron and Jeffries 2001; Cheng and Bledsoe 2002; Herman et al 2003; Corbin and D’Antonio 2004a; Lewis et al. 2006; Eviner and Vaughn unpublished data; J. Corbin and C. D’Antonio, personal communication). (a) Temperature and precipitation (adapted from Evans et al. 1975, and updated with data across California grasslands from the database provided by California Irrigation Management Information System). (b) Plant growth. (c) Plant N uptake and microbial N uptake. (d) N release from litter and seedling thinning. (e) Gross rates of N mineralization and nitrification. (f) N leaching and gas loss.

While litter mass loss is relatively slow over the summer (Savelle 1977), 35% of root litter N and 20% of aboveground litter N is released during the summer (Jackson et al. 1988) (Figure 8.1d). Some litter decomposition may be mediated by photodegradation, which is responsible for litter mass loss in other semiarid to arid regions (Moorhead and Callaghan 1994; Austin and Vivanco 2006). Seed N may be lost or recycled through the system through high granivory rates during the summer affecting 1–75% of seeds (reviewed in Heady et al. 1991).

While net and gross rates of mineralization and nitrification are low throughout the summer (Figure 8.1e), integrated over a whole summer, they can lead to significant accumulations of soil inorganic N, particularly NO3
(Jones and Woodmansee 1979; Jackson et al. 1988; Maron and Jeffries 2001; Corbin and D’Antonio 2004a; S. Parker, personal communication). Together, these data indicate that microbes are active at some times or at some locations in the soil profile during the summer, despite low soil water potentials. The frequent presence of morning dew at many grassland sites,

MECHANISMS DETERMINING NUTRIENT DYNAMICS

97

BOX 8.1 SEASONALITY IN CALIFORNIA GRASSLANDS

Because the timing of temperature and moisture conditions varies greatly across sites and across years, it is difficult to generalize nutrient cycling patterns by calendar month. Generalizations based on the environmental conditions at the time of measurement yield a better understanding of N cycling dynamics. However, we recognize that it would be helpful to assign these patterns to a certain month. We use an annual-dominated grassland in the northern California inner coastal range (Mendocino County) as an example of how months of the year typically line up with the seasonal dynamics depicted in Figure 8.1 (although it is arguable whether a “typical” temporal pattern exists). Germinating rains (wet-up) usually begin any time between late September and mid-November. The fall activity peak (Figure 8.1) extends until temperatures dip down, usually in mid- to late-December. Winter is defined as the period when temperatures remain low, and it usually extends until mid to late February, when temperatures slowly increase, increasing plant and microbial activity (“early spring”). By mid to late March, most belowground growth is completed, and temperatures increase more rapidly, leading to rapid aboveground plant growth. Precipitation typically decreases and ceases in April. Plant growth extends until the soil moisture begins to be depleted (typically mid- April), and plants begin to set seed. In a typical rainfall year, mid-season plants begin to senesce by mid- April to early May (late spring), and summer begins once most plants have senesced (not including summer annuals) (typically mid-May to early June).

noted in personal observations and personal communications, may episodically stimulate microbial activity at the litter and soil surface during some summer months. Summertime microbial decomposition and N cycling activity are likely to be even higher at sites with substantial fog inputs, which can support late-season plant growth in coastal grasslands (Corbin et al. 2005). Both dew and fog can result in short-term wet-dry cycles, which have been shown to increase microbial activity (reviewed in Fierer et al. 2003b). Because both gaseous and leaching losses of N are highly dependent on water availability, N loss is likely to be minimal during the summer months (Figure 8.1f).

96 hours, coinciding with germination of many plant seeds (Hungate et al. 1997c). At this time, fungal biomass is 3 – 4 times higher than bacterial biomass (Hungate et al. 1997c). The rains induce a flush of litter decomposition (Figure 8.1d), with litter leaching accounting for 5% of litter mass loss (Savelle 1977). Rapid germination and death of 50 – 75% of germinated seedlings (Bartolome 1979; Young et al. 1981) also provide a flush of labile N and C into the system within the first 2 – 7 weeks of the growing season (Figure 8.1d). These inputs of labile substrates from litter and dying seedlings seedlings may partially account for high rates of N mineralization and nitrification (Figure 8.1e) (Davidson et al. 1990; Maron and Jeffries 2001; Herman et al. 2003). N cycling rates can also be stimulated by wet-up induced lysis of microbial biomass, resulting in substantial short-term (1 – 4 days) increases in C and N mineralization (reviewed in Fierer et al. 2003b). Increases in soil water potential also stimulate nitrification rates (Stark and Firestone 1995). Despite high initial concentrations and production of inorganic N, N consumption is greater than production (Herman et al. 2003) because of high microbial immobilization. This high microbial immobilization is not enough to prevent N loss (Figure 8.1f), but N leaching in this system is generally low, even in response to the addition of 100 kg N/ha (Jones et al. 1977). At wet-up, and for a few weeks following wet-up, NO3 leaching is at its seasonal peak (Jones et al. 1977; Vaughn et al. 1986; Jackson et al. 1988; Davidson et al. 1990; Maron and Jeffries 2001). At some sites, N leaching is seen every year (e.g., Hopland Experimental Station, Mendocino County), while other sites (e.g., San Joaquin Experimental Range, Madera County) have no measurable N leaching in some years (Jones and Woodmansee 1979; Davidson et al. 1990; Maron and Jeffries 2001). It has been assumed that leaching rates peak shortly after wet-up, then rapidly decrease, becoming negligible within 1–2 months (Jones et al. 1977; Jackson et al. 1988; Davidson et al. 1990), but a 20 year study in the Sierra foothills (Yuba County) consistently detected a second leaching peak in mid to late winter (Lewis et al. 2006). Nitrogen gas losses are often negligible (Jackson et al. 1988), even in response to the application of 100 kg N/ha (Jones et al. 1977). Instantaneous measures of N gas losses are usually low (1–2 ng/cm2/hr) and even when they occasionally reach high levels (13 ng N/cm2/hr), these fluxes are shortlived and have negligible impact on ecosystem N budgets (Hungate et al. 1997c; Rudaz et al. 1991; Herman et al. 2003). NO fluxes increase within 30 minutes after wet up, but decrease after 48 hours, while N2O fluxes increase for 192 hours after wet up (Hungate et al. 1997c).

FA LL W ET-U P

The first significant fall rains (15 mm) (Figure 8.1a) mark the beginning of the plant growing season (Figure 8.1b) (Heady 1977). Microbial N immobilization is at its peak at wet-up (Figure 8.1c) (Jones and Woodmansee 1979; Schimel et al. 1989; Joffre 1990), with large increases in bacterial and fungal biomass within 24 hours after wet-up, and peaking at

98

RESOURCES

FA LL/ EAR LY W I NTE R

Due to ideal temperature and moisture conditions (Figure 8.1a), early fall is a period of rapid root and shoot growth (Figure 8.1b) and high N uptake by plants (Jones and Woodmansee 1979; Jackson et al. 1988; Center et al. 1989) and microbes (Hart et al. 1993) (Figure 8.1c). Early fall is

also marked by high loss rates of litter mass and litter N (Figure 8.1d) (Savelle 1977; Jackson et al. 1988), particularly from roots (Heady et al. 1991). Steady seedling thinning also provides a continuous source of labile N (Figure 8.1d). Inorganic N levels can be low (Jackson et al. 1988) to moderate (Vaughn et al. 1986) because rates of microbial NH4
consumption are greater or equal to production rates (Herman et al 2003). However, there is a shift from net N immobilization to net mineralization from early fall through winter (Jones and Woodmansee 1979; Schimel et al. 1989; Davidson et al. 1990; Maron and Jeffries 2001; Herman et al. 2003). These changes in microbial activity, along with lower plant N uptake as temperatures decrease, account for rising concentrations of soil inorganic N levels as winter begins (Vaughn et al. 1986; Jackson et al. 1988). Despite the fact that inorganic N seems to become more available as fall progresses, N leaching losses are low throughout late fall/early winter (Figure 8.1f) (Jones and Woodmansee 1979; Davidson et al. 1991; Maron and Jeffries 2001), and N gas losses are low throughout the remainder of the growing season (Figure 8.1f) (Herman et al. 2003).

W I NTE R

While aboveground plant growth is generally minimal over the winter, root growth rates vary widely (Figure 8.1b). Root growth may be depressed in the colder months (Center et al. 1989), may continue at a steady slow rate (Savelle 1977; Center et al 1989), or may remain high throughout the winter (Heady et al. 1991). These variations in root growth patterns are most likely influenced by variations in winter temperatures across sites and years but may also be attributed to different patterns of root growth across species and/or difficulties in accurately measuring root growth. Microbial N immobilization (Hart et al. 1993) can also remain high. Winter months are a time of high rates of seedling thinning (Figure 8.1d) (Woodmansee and Duncan 1980), potentially facilitated by frost-related dieback. Aboveground litter mass loss continues through the winter, or even peaks at this time (Center et al. 1989). Total aboveground litter N release is highly variable in the winter. During some years, litter nutrients accumulate due to high rates of N and P immobilization, and in other years, total litter N and P pools decrease, despite increasing concentrations of litter N and P (Woodmansee and Duncan 1980; Hart et al. 1993). The fall through early winter is the period of peak root litter mass loss, with rates greatly decreasing through the winter (Heady et al. 1991). Interestingly, the soil depths at which peak soil decomposition activity occur shift through the growing season. In early winter, decomposition is greatest at a depth of 20 to 30 cm because of higher soil temperatures at this depth. Through the winter, peak decomposition activity shifts from the deeper soil layers toward the soil surface, peaking in the top 10 cm by spring (Heady et al. 1991). Like microbial N immobilization, microbial activity can be high during this season (Jones and Woodmansee 1979), and although little data is available on N

production rates, microbial NH4
production has been reported to roughly equal consumption (Herman et al. 2003). Nitrogen gas loss is low during the winter (Figure 8.1f) (Herman et al. 2003), even in response to the addition of 100 kg N/ha (Jones et al. 1977). These same fertilizer inputs also resulted in negligible N leaching during the winter (Jones et al. 1977), and whereas most studies have suggested low N leaching during the winter (Jones and Woodmansee 1979; Davidson et al. 1990; Maron and Jeffries 2001), the only comprehensive N leaching study to date (Lewis et al. 2006) has shown that a second peak of leaching occurs late in the winter (mid to late January) (Figure 8.1f) in the Sierra foothills (Yuba County). EAR LY S P R I NG

Early spring is marked by continued precipitation and a modest increase in temperatures (Figure 8.1a). As temperatures increase, so do rates of plant N accumulation (Jackson et al. 1988) (Figure 8.1c). By the time temperatures begin to increase at a faster rate—approximately 2–4 weeks after temperatures begin to increase (Figure 8.1a)—most root growth (Figure 8.1b) and plant N uptake (Figure 8.1c) have occurred, with 82% of seasonal N uptake completed by this time, even though only 45% of the annual biomass accumulation has occurred (Jackson et al. 1988). This increase in temperature also stimulates a second seasonal peak in microbial activity (Figure 8.1c), which tends to be higher than the first peak in the fall. Increasing temperatures also enhance the loss of litter mass and N (Figure 8.1d) (Savelle 1977; Jackson et al. 1988), although net accumulation of litter N has also been seen at this time (Hart et al. 1993). Net N mineralization also increases in the spring (Maron and Jeffries 2001; Corbin and D’Antonio 2004a; Cushman et al. 2004) (Figure 8.1e). Low NO3- availability is evident from the early spring through the end of the growing season (Vaughn et al. 1986; Jackson et al. 1988), suggesting that N consumption rates increase as steeply as N production rates. High microbial N uptake during this period can limit N availability to plants (Schimel et al. 1989; Jackson et al. 1989). N losses are minimal at this time of year (Figure 8.1f), likely due to high biotic N uptake. S P R I NG / P EAK P HYS IOLO GY

As temperatures rise (Figure 8.1a), there is rapid accumulation of aboveground plant biomass (Figure 8.1b) (Jackson et al. 1988), so that roots, which made up 60% of plant biomass toward the end of the winter, only constitute approximately 14% of total plant biomass by peak plant biomass (Jackson et al. 1989). There is also a slight increase in microbial biomass N at this time (Jackson et al. 1988), but the plant N pool is greater than the microbial N pool for the only time during the growing season (Figure 8.1c) (Jackson et al. 1989). Root litter mass loss is minimal through late spring (Heady et al. 1991), and aboveground litter mass loss begins to trail off as well (Figure 8.1d) (Center et al. 1989; Eviner 2001). For the first time in the growing season, an accumulation of newly dead seedlings is evident.

MECHANISMS DETERMINING NUTRIENT DYNAMICS

99

LATE S P R I NG / P EAK B I OMAS S

Late spring is marked by peak plant biomass, an exhaustion of soil moisture (Reever Morghan et al., Chapter 7), and plant senescence. The duration of plant and microbial activity into late spring largely depends on the availability of soil moisture and therefore on both climatic conditions and soil waterholding capacity. However, regardless of late season precipitation, most annual plants will senesce in the early summer months (Jackson and Roy 1986). At this time of year in the Sierra Foothills Research Station (Yuba County), Jackson and colleagues (1988) found 4.9% of ecosystem N in the microbes, 3.4% in live plant biomass, and 1.3% in dead plant material. As the system dries out, microbial CO2 respiration decreases because of moisture limitation (Fierer et al. 2003a; Eviner 2004). Similarly, rates of mineralization (Herman et al 2003; Eviner et al. 2006) and nitrification decrease (Eviner et al. 2006), with nitrification being particularly sensitive to decreases in soil moisture (Davidson et al. 1990; Eviner et al. 2006).

VA R I ATION S I N N DYNAM ICS I N CA LI FOR N IA G R A S S L A N DS ACROS S S I T E S AN D Y E AR S

Thus far, we have presented the general seasonal patterns of N dynamics in California annual grasslands, but there is substantial variation in the timing and magnitude of these patterns from year to year and site to site (Woodmansee and Duncan 1980; Center et al 1989; Evans and Young 1989). Seasonal patterns of gross mineralization can be better predicted by soil moisture than temperature (Herman et al. 2003), and these mineralization rates largely determine seasonal dynamics of nitrification and N gas fluxes (Herman et al. 2003). However, temperature plays a critical role during the wet season, as wetter, milder winters will enhance not only the magnitude of plant growth, litter decomposition and N turnover, but also their variability (Woodmansee and Duncan 1980; Center et al. 1989). Fluctuations in temperature and moisture can also alter the rates of biogeochemical cycles. When numerous dryingrewetting cycles occur in the fall, C mineralization will significantly decrease and can remain depressed for up to six weeks (Fierer and Schimel 2002). Changes in weather patterns, particularly in the fall, can also affect plant species composition (Talbot et al. 1939; Pitt and Heady 1978; Reever Morghan et al., Chapter 7), which can impact N cycling (see subsequent discussion). Precipitation patterns are also likely to greatly influence N leaching patterns. Site differences are also a key factor in determining nutrient dynamics, as extremely different grasslands types are associated with changes in geology and climate across California’s landscape. Similar to variations across years within a given site, climate patterns can be directly responsible for changes in nutrient dynamics across sites. Climate differences across sites can also influence the importance of nutrient limitation to plant productivity, as nutrient limitation becomes a stronger determinant of plant production with increased

100

RESOURCES

water availability (reviewed in Harpole et al., Chapter 10). Site differences in soil type and land use history are also strong drivers of biogeochemical patterns (reviewed in Jackson et al., Chapter 9). Soil texture and organic matter content can greatly impact storage, cycling, and movement of water and nutrients and can also alter human land use patterns and community composition of plants, microbes, and soil fauna — all of which contribute to marked site-to-site differences in the patterns and controls over nutrient dynamics (reviewed in Jackson et al., Chapter 9). For example, grasslands on serpentine soils can be markedly different from those on the more prevalent sandstone-derived soils. These serpentine soils have lower nutrient and water availability and can have toxic concentrations of heavy metals, leading to large changes in the biomass and composition of the flora and fauna and to differences in how belowground biota and processes respond to global changes (reviewed in Harrison and Viers, Chapter 12). General seasonal controls over plant and soil dynamics are roughly similar between sandstone and serpentine communities, but serpentine communities often contain later-phenology plant species, which extend the growing season and typical time of nutrient uptake (Chiariello 1989). Not only is nutrient availability lower in serpentine soils, but lower rates of gross N mineralization and higher microbial N immobilization in serpentine versus sandstone soils further reinforce low plant N availability (Hungate et al. 1997). Serpentine soils have also been found to have lower active and total fungal biomass than sandstone soils, possibly due to increased fungivory by nematodes (Hungate et al. 2000c).

Pools and Fluxes of Other Nutrients There have been far fewer studies investigating pools and fluxes of other nutrients, but we will briefly highlight how some of these are similar to, or different from N dynamics. As with plant available N, plant demand for phosphorus (P), sulfur (S), potassium (K), and calcium (Ca) must partially derive from seedling thinning rather than simply from litter decomposition, suggesting repeated cycling of plant uptake and mineralization throughout the growing season for all of these nutrients (Center et al. 1989). Litter release of P and S are similar to N release (Center et al 1989), and as with N, soil availability of both P and S peaks early in the growing season, then decreases with increased plant uptake (Vaughn et al. 1986). Sulfur shows similar temporal leaching patterns to N (Vaughn et al. 1986), while P has a relatively closed cycle (Woodmansee and Duncan 1980). While plant growth in most temperate ecosystems is limited by N availability, in California grasslands N, P, and S can all be limiting to plant growth, depending on soils and plant species composition (Martin 1958; Jones and Martin 1964; Jones et al. 1970; Menke 1989). Other nutrients show seasonal patterns that are very different from those of N, P, and S. Potassium release from litter is mainly controlled by early season leaching rather than microbial mineralization, and soil K concentrations show

inconsistent seasonal patterns from year to year (Vaughn et al. 1986; Center et al. 1989). Soil magnesium (Mg) availability also varies with time, without consistent seasonal patterns across years (Vaughn et al. 1986). In contrast to all of the other nutrients discussed, both litter and soil calcium concentrations do not demonstrate large seasonal fluctuations (Vaughn et al. 1986; Center et al. 1989).

Effects of Vegetation Composition on Nutrient Dynamics It has been well documented that plant species can alter almost every aspect of ecosystems from soil physical structure to hydrology to the activity and distribution of other organisms. Plant-induced changes to any of these ecosystem characteristics can have large effects on nutrient dynamics (reviewed in Eviner and Chapin 2003b). In this section we investigate how vegetation composition influences nutrient dynamics. S H I F TS I N P LANT S P ECI E S C OM P OS ITION I N CALI FOR N IA AN N UAL G RAS S LAN DS

Plant species composition in California annual grasslands can vary greatly across sites (Harrison and Viers, Chapter 12), years (Talbot 1939; Heady 1958; Pitt and Heady 1978), management practices (Jackson and Bartolome, Chapter 17; Reiner, Chapter 18; Huntsinger et al., Chapter 20; DiTomaso et al., Chapter 22), and due to invasions of exotic species (D’Antonio et al., Chapter 6). Such shifts in plant species composition can affect the timing and magnitude of most N pools and fluxes (Jones 1963; Jones et al. 1977; Hungate et al. 1996; Canals et al 2005; Eviner et al. 2006). These differences across species can be so marked that they completely mask the generalized seasonal trends in N dynamics discussed in previous paragraphs (Eviner et al. 2006). Species effects on N cycling can often be predicted based on a suite of traits, including litter chemistry, exudation rates, and species effects on microclimate (Eviner and Chapin 2003b; Eviner et al. 2006). Litter N content is often presumed to be the main determinant of a plant’s effect on N cycling, and thus, shifts in the relative dominance of legumes vs. grasses can have marked impacts on N cycling (Center et al. 1989; Eviner et al. 2006). Changes in N pools and fluxes also occur as a result of shifts in dominant grass species due to their differences in labile C inputs and microclimate effects (Eviner et al. 2006). Although litter chemistry, microclimate effects, and labile C can be used to predict the ecosystem consequences of shifts in vegetation composition, these traits often fail to predict the impacts of forbs on N dynamics because many forbs contain high levels of secondary chemicals, which are low or absent in other grassland species. For example, plant species effects on potential nitrification in the spring can be predicted well by litter C:N ratios for most species (Eviner et al. 2006). However, nitrification rates associated with Erodium botrys (filaree) are substantially lower than would be predicted by

litter C:N ratios because of Erodium’s high concentration of phenolics, which inhibits nitrifiers (Eviner and Chapin 2005). Similarly, litter decomposition of Amsinckia douglasiana (fiddleneck) is very different from what would be predicted based on traits that predict decomposition of other California grassland dominants. Amsinckia is a known hepatatoxin of cattle because of its high concentrations of pyrrolizidine alkaloids. These compounds, which inhibit oxidative enzymes in the livers of cattle, similarly inhibit the microbial enzymes that break down litter, thus greatly decreasing decomposition rates of Amsinckia litter (Eviner 2001). In California grasslands, the effects of plant species on N cycling change seasonally because the relative importance of plant traits that control N cycling changes seasonally. For example, species effects on net N mineralization are determined by soil temperature and moisture in the late fall, labile C inputs and live biomass in the winter, and litter C:N ratios and species effects on soil moisture in the spring (Eviner et al. 2006). Many of the relatively new invaders in California grasslands (e.g., Centaurea solstitialis [yellow starthistle], Taeniatherum caput-medusae [medusahead], Aegilops triuncialis [goatgrass]) are active later in the season than the previously dominant exotics, such as Avena sp. and Bromus sp., and may be responsible for shifts in the season and soil depth at which peak N cycling occurs. Similarly, native summer annuals can greatly extend the phenology of plant N uptake (Chiariello 1989) and likely soil N cycling as well. The relative effects of plant species on N dynamics change not only seasonally but also with time elapsed since plant establishment. Labile C inputs and species effects on microclimate can alter N dynamics within the first year of a species being at a site, but the impacts of plants change substantially over the years as their “short-term” effects interact with their litter chemistry and effects on soil chemistry and structure. When shifts in vegetation composition occur, the impact of the previous plant community on litter and soil chemistry can persist and influence N dynamics for years after the species have been replaced by other plants (V. T. Eviner, unpublished data). Thus, to understand the effects of vegetation change on nutrient dynamics, it is critical to consider both the former and the current plant community composition. Shifts in plant species composition not only influence N pools and fluxes but can also greatly affect the importance of N in controlling plant productivity. For example, shifts between grass-dominated and legume-dominated grasslands determine whether N or another nutrient (e.g., P, S) limits plant growth (Jones and Martin 1964; Jones et al. 1970). C OM PAR I SON OF E XOTIC AN D NATIVE-D OM I NATE D G RAS S LAN DS

Over the past two centuries, California grasslands have experienced widespread replacement of native species with exotic annual species (Crampton 1974; Bartolome et al. 1986; Schiffman, Chapter 4; D’Antonio et al., Chapter 6). Most comparative studies have focused on differences between

MECHANISMS DETERMINING NUTRIENT DYNAMICS

101

exotic annual grasses and native perennial bunchgrasses, particularly Nassella pulchra (purple needlegrass). In general, perennial bunchgrasses allocate more energy to roots than annual grasses (50% in perennials vs. 39% in annuals), and while most root growth occurs in annual grasses by early spring, root growth is more evenly distributed over the growing season in perennials (Heady et al. 1991). While 92 – 94% of annual roots are within the top 30 cm of soil (50% within the top 5 cm), only 70% of perennial roots are within the top 30 cm, most of these at the 8 – 30 cm depth (Savelle 1977). The deeper root distribution of perennials increases their access to water late in the growing season, allowing for longer retention of green tissues, longer growing seasons, and greater surface soil water potentials (Brown 1998,; Reever Morghan et al., Chapter 7; Corbin et al., Chapter 13). These differences in root distribution can potentially alter N cycling rates, as well as the depth distribution of N pools and fluxes. Nitrogen cycling rates may also differ between annual and perennial grasses as a result of differences in the amount and quality of nutrient and carbon inputs. Perennial grasslands lack the high N input through seedling thinning that is common in annual grasslands; while 45% of shoot production turns over during the growing season in annual grasses, only 5% turns over in established perennial systems (Heady et al. 1991). An exception may be in the establishment of perennial stands, when nutrients from thinned seedlings may be an important resource for establishing perennial individuals (V. T. Eviner and C. E. Vaughn, unpublished data). Perennial grass litter decomposes more slowly than that of annuals (Savelle 1977; Heady et al. 1991), and the quantity of litter N input by perennials is also lower because more N is recycled within the perennial plant, rather than entering the decomposition pathway (Woodmansee and Duncan 1980). Based on all of these factors, we would expect higher rates of N cycling in annual than perennial grasslands. It has also been hypothesized that perennial grasslands should have lower rates of N leaching early in the growing season because of lower N cycling rates and rapid initiation of plant growth and N uptake in response to fall rains (Bartolome 1989). Comparisons between plots planted to annual vs. perennial grasses in California yield contrasting results and, even within individual studies, do not consistently support the hypothesis that annual systems have higher rates of both N cycling and N leaching (Table 8.3). As expected based on first principles, some studies have seen lower N leaching in perennial vs. annual grass stands (Maron and Jeffries 2001). Corbin and D’Antonio (personal communication), studying experimental plots in a Marin County coastal prairie, found lower N leaching in perennial stands despite equal or higher rates of net N cycling, lower 15N immobilization, and lower plant N uptake by the perennials. Other studies confirm that 15N immobilization is extremely low in stands of perennial grasses (Hooper and Vitousek 1998), and that net N cycling rates are higher in

102

RESOURCES

perennial than annual-dominated plots. In contrast, S. Parker (personal communication), studying experimental plots at Sedgwick Reserve (Santa Barbara County), found that leaching was higher in perennial than annual grass stands, despite the fact that N cycling rates, soil inorganic N concentrations, and N gas losses were lower in perennial systems. Other studies confirm low inorganic N levels associated with perennial plots (Welker et al. 1991; Seabloom et al 2003b), and lower gross rates of nitrification in perennial stands (Hawkes et al. 2005). The one consistent result across these studies is that the effects of plant species on N leaching are opposite of their effects on N cycling and inorganic N pools (Table 8.3). There are a number of potential explanations for these patterns: •

Annual and perennial stands may have peak plant and microbial activity at different soil depths, so that sampling N dynamics in the upper 10 cm of soil cannot account for leaching patterns.

•

Hydrology may account for inconsistencies in N cycling vs. leaching rates. In well-drained soils, available nutrients may be expected to quickly appear in leachate. However, poorly drained soils can allow for a significant lag between soil N production, uptake, and leaching, decoupling these processes over time (S. Parker, personal communication). Annuals and perennials may also greatly differ in their effects on hydrology.

•

While N leaching inversely correlates with microbial biomass N (Table 8.3), microbial N immobilization, on its own, cannot account for the disjunct between patterns of N cycling and N leaching, since a higher microbial biomass does not seem to decrease soil inorganic N concentrations in the summer and fall (Table 8.3), the time at which N produced would be vulnerable to leaching during the early season rains.

Differences in patterns across studies (Table 8.3) may be due to site-related factors, such as soil nutrient capital, soil texture, hydrology, climate, or the species identity of the annual and perennial plants. Different annual grass species can differ in their effects on N dynamics (Eviner et al. 2006), so there is no reason to expect that all annual or all perennial grasses should have similar effects on ecosystem dynamics. Clearly, in order to resolve these conflicting results within and across studies (Table 8.3), we need to rethink our assumptions about N dynamics in these systems and carefully design our experiments to encompass multiple soil depths and hydrology. In addition, we may gain important insights into controls over N dynamics by measuring both net ( production  consumption) and gross ( production) rates of N cycling. In the same set of plots in the same season (although during different years), gross rates of nitrification and microbial biomass N were higher in annual than in perennial plots (Hawkes et al. 2005), while net rates were lower in annual

Location of experiment (county)

Native perennial grasses studied

Fall leaching Microbial biomass N Fall plant N uptake Exotic annual grasses studied

Inorganic N

Nitrification

N mineralization

Summer Fall Winter Spring Summer Fall Winter Spring Summer Fall Winter Spring

Marin

PA PA PA PA PA PA PA PA PA PA PA PA PA PA PA Avena barbata, Bromus diandrus, Vulpia myuros Agrostis oregonensis, Festuca rubra, Nassella pulchra

Corbin and D’Antonio pers. comm. (net rates)

Santa Barbara

Bromus hordeaceus, Bromus madritensis, Hordeum murinum Bromus caritinatus, Elymus glaucus, Nassella pulchra

PA PA PA PA PA PA PA PA PA PA PA PA PA P2.706)[k(Zk)2 ⫺ 2.706]

(Eq. 10.2)

where k is the total number of reviewed studies, and Zk is the mean standard normal deviate of the k studies, which can be

conservatively estimated by taking Z ⫽ 1.645 for significant studies (e.g., authors report effects of treatments as p ⬍ 0.05) and Z ⫽ 0 for nonsignificant studies when exact p values are not given (Rosenthal 1979). For these analyses we excluded four studies that did not report statistical significance of their results; conservatively including these studies as nonsignificant decreased our estimates of X by less than 5%.

R E S O U R C E L I M I TA T I O N

123

0.8 0.6

2 1

Effect Size 0.4

3

1

6 4 1

6

21

0.2

Many of the studies we found had experimental design issues that do not allow for full interpretation of nutrient limitation in grassland communities (for resource addition experiment design see Eviner et al. 2000). Resources were often confounded because they were not added independently. For example, N and S were added together but not separately in some studies. Also, often other limiting factors were not controlled (e.g., water), and may have influenced results, and finally, effects on community composition were often not reported, making it difficult to quantitatively compare how different nutrients affected composition.

1

5

Productivity Responses

2

12 1

124

RESOURCES

1

1

Treatment

NPS

NPKCa

NPK

NPO

PS

NKS

PO

NP

NS

NH

NO

KP

KS

P

1

S

N

H

O

6

1

Ca

0

Despite the shortcomings mentioned previously, the 20 resource addition studies (Table 10.1) we examined show that resource limitation is a general phenomenon in California grasslands. Ten treatments (resource addition combinations) had necessary replication to test for significance of effect sizes (Figure 10.1). Of those, only addition of O (“other,” typically micronutrients) and NS had effect sizes that did not differ significantly from zero (Figure 10.1). Several studies found greater than 100% increase in biomass with resource addition. Fourteen studies reported statistical significance of their results; all of these found a significant response to one or more added resources. In terms of a conservative estimate of a fail-safe number (X), it would require 182 null response studies to make the overall effect of resource addition insignificant. It is highly unlikely that this many unreported null results exist. Examination of the results of our meta-analysis suggests that limitation by both N and P is common in California grasslands. Six experiments added N and P factorially; 3 found significant effects of N addition alone, 3 of P alone, and 4 of N and P together (significant N⫻P interaction). P addition showed the largest effect size of any singleresource addition (Figure 10.1). Fifteen of sixteen experiments reporting significance found a significant effect of adding P alone or a significant P⫻N or P⫻K interaction (failsafe number, X ⫽ 209). Eight of nine experiments found a significant effect of N addition alone or a significant N⫻P, N⫻K, or N⫻S interaction (X ⫽ 55). Although N and S showed similar effect sizes to each other (Figure 10.1), evidence for S limitation is less strong: three of four experiments reporting significance found a significant effect of adding S alone (X ⫽ 5), and because these manipulated only sulfur, comparisons within sites of different nutrients cannot be made. Limitation by nutrients such as K may also be important (Harpole and Tilman 2007), but few studies manipulated K alone, so this could not be evaluated here. Our examination of the average vegetation response to resource addition on serpentine versus nonserpentine soils showed that the average effect was smaller on serpentine soils (0.12 vs. 0.28), although not strongly so (p ⫽ 0.076). The low sample size for serpentine studies (4) makes these results tentative, but they may be consistent with greater

FIGURE 10.1. Mean effect sizes for additions of different resource

combinations across all experiments. H⫽water, O⫽other (Mg, micronutrients). Effect sizes greater than zero correspond to increased productivity with resource addition. Error bars represent ±SE of mean experiment effect size; treatments without error bars were represented by only one experiment; numbers indicate sample size. For treatments with sample size ⬎ 1, only O and NS did not differ significantly from zero.

importance of limitation by other factors (e.g., metal toxicity) on serpentine soils. For example, low Ca:Mg ratios are typical of serpentine soil and have been suggested to play a role in the relative lack of exotic grasses in serpentine areas (see Harrison and Viers, Chapter 12). We found no significant latitudinal trend in baseline productivity as measured by the mean yield of the control plots of each experiment (p ⫽ 0.23), counter to what would be expected if precipitation were the main factor limiting productivity along this gradient (Figure 10.2a). However, because studies were performed in different years (1948–2000) and with different methods, high variability among studies may be due to high interannual variation in climate as well as variation in harvesting methods or timing. Effect size of resource addition did significantly increase with latitude (Figure 10.2b), suggesting that soil mineral resource limitation increases as water limitation decreases. We found no significant effect of distance from the coast on effect size variation. To assess the generality of limitation by multiple resources, we additionally categorized each treatment in terms of the number of resources added. For example, one added resource might be N or P addition alone, while two added resources might be N and P or N and K added together, and three added resources might be NPK. In some cases, several micronutrients were added together as a mixture designated “other.” In these cases, we categorized this as a “single resource” if no one was also applied separately. California grasslands appear to be generally limited by multiple resources: on average, the response to two or more

0.35

0

0

11

18

0.15

Effect Size 0.2 0.6

R2 =0.14, P=0.027

Effect Size 0.25 0.30

80 40

B

1

0.20

120

Precipitation (cm)

300 100

Yield (g/m2)

0.40

A

46

– 0.2

1

34

35

36 37 38 Latitude (deg)

39

40

2 3 Number of Added Resources

4

F I G U R E 10.3. Linear increase in effect size with increased numbers of simultaneously added resources. Effect size is the log ratio of treatment/control biomass. Error bars are ±SE. Only one experiment added four resources separately, consequently no error bars shown.

F I G U R E 10.2. (a) No significant relationship between baseline

productivity (left axis) and latitude in spite of precipitation gradient (dashed line, right axis). Points are mean productivity ⫾SE of control plots of all experiments at a given latitude. Dashed line is 40-year average annual precipitation. (b) Effect size (log ratio of treatment/control biomass) increases with latitude. Points are mean effect size for each treatment from each experiment at a given latitude. Dashed line at zero effect size corresponds to no effect of nutrient addition; positive values correspond to increased yield with resource addition. Points are jittered for clarity.

resources was greater than the response to addition of a single resource (Figure 10.1), and of the 11 factorial experiments reporting significance tests, 7 had significant interactions (4 appeared to be single-resource limited); the fail-safe number, X, was 38. Strict co-limitation — where a productivity response occurs only with additions of two or more resources, but not with any single resource—has also been shown but was not common in the data set (Harpole, unpublished data). Effect size increased linearly with the number of added resources (Figure 10.3). This relationship remained significant even when we excluded the single study in which four resources were added together. On average, each added resource resulted in an increase in the effect size, suggesting that limitation by multiple resources may be more common than generally assumed. Conversely, it also suggests that studies that add fewer resources are more likely to observe smaller or no effect of resource addition and will be therefore less likely to assess resource limitation adequately.

Functional Group Responses Nine resource addition field studies recorded functional group responses for various resource addition treatments (Table 10.1). For each study we calculated the relative abundance of grass, forb, and legume functional groupings for each treatment based on biomass or % cover depending on what was reported. Because relative cover and relative biomass are strongly correlated, use of mixed response units in this case

is less likely to violate assumptions of the meta-analysis. The grass category was typically dominated by exotic annual grasses (most commonly Bromus hordeaceus). In rangeland studies, the most common forbs were Erodium spp., and clovers (Trifolium spp.) were the most common legumes. As our response metric, we calculated the change in relative abundance of each functional group to resource addition: log ratio of relative abundance with treatment to relative abundance of control (Eq. 10.1). Resource addition can alter species and functional group composition (see Dukes and Shaw, Chapter 19). In the studies surveyed (Table 10.1), nonlegume forbs decreased in relative abundance in response to additions of N, P, S, and “other” (Mg ⫹ micronutrients) (Figure 10.4), P⬍0.01. Grasses increased in relative abundance in response to N additions (P⬍0.01), while legumes showed a positive response to additions containing P and S (P⬍0.01) and a negative response to N (P⬍0.01). These results are consistent with an N-P resource competition tradeoff between grasses and legumes. Grasses should be more N-limited than legumes, which can fix their own N but as a consequence have higher P demands (Tilman 1988). Changes in relative abundance reflect both changes in resource limitation and changes in competitive relationships after resource addition. Forbs may be better competitors for N than exotic annual grasses but poorer competitors for light, which becomes more limiting as productivity increases with added resources. Limitation of forbs, grasses, and legumes by different resources may contribute to their coexistence in grassland communities (Tilman 1988). In those studies that extended over multiple years, the initial fertilization response differed from the long-term response. The initial positive response by legumes to S and P addition was followed by a positive response by grasses the following year (Bentley and Green 1954, Bentley 1958, Jones et al. 1990): increased N fixation in the legumes may subsequently increase N availability to the grasses.

R E S O U R C E L I M I TA T I O N

125

F I G U R E 10.5. Effect size of livestock weight gain in response to nutrient addition treatments. Error bars are ±SE.

Figure 10.4. Change in the relative abundances of forb, grass, and legume functional groups to nutrient addition treatments. “O” ⫽ Other (Mg and micronutrients).

Aggressive competition for limiting resources is often cited as an important mechanism for explaining why some nonnative species have become dominant in certain habitats, yet the resources for which species compete are rarely identified (Parker and Reichard 1998; Shea and Chesson 2002). Although native and exotic species may differ in their responses to changes in resource availability, particularly where they belong to different life-history groups, too few studies presented species-level responses to resource addition to allow a quantitative comparison of native versus nonnative species responses. One study (Haubensak 2001) found that exotic annual grasses were released from N limitation with N addition at one site where native perennial grasses showed no response to N addition. At Haubensak’s second site, exotic annual grasses and native perennial grasses were both co-limited by N and P and neither showed a response to either nutrient alone. Since most studies of N addition in California grasslands show increases in annual grasses compared to forbs, it is likely that life form is more important than origin in determining which becomes dominant. Overall we were not able to test the generality of how native versus exotic species within life form groups, or across all groups combined, responded to resource addition in the surveyed studies. A recent greenhouse pot study demonstrated stronger responses to nutrient additions in one non-native perennial grass, Holcus lanatus, compared to other exotic and three native perennial grass species (Thomsen et al. 2006a). This study concluded that responses were species-specific and not clearly associated with “origin” within this life form group (perennial grasses). Because fertilization can promote invasion and dominance of exotic annual grasses (Huenneke et al. 1990), researchers have experimented with “reverse fertilization” to re-establish native species. This typically involves the addition of labile carbon as a means to decrease N availability to plants. The results have been mixed. For example, Alpert and Maron (2000) reported lower exotic grass biomass following

126

RESOURCES

sawdust addition, but Corbin and D’Antonio (2004a) saw no decrease in exotic grass biomass or increase in native biomass in a nearby coastal prairie experiment (Corbin et al., Chapter 21).

Trophic Responses Because resource supply can control trophic interactions, we additionally examined those studies that reported the effects of resource addition on other trophic groups beyond plants. This included only four studies of livestock (herbivores) response to resource addition (Table 10.1). We calculated effect sizes for livestock weight gain for each resource addition treatment Eq. 10.1). One objective of resource additions to rangelands has been to increase production of N-fixing legumes (which can be P-limited), thereby increasing the protein (N) content of cattle forage with the goal of increasing livestock weight gain (Jones 1974). Consistent with a stoichiometric switching of limiting resources with trophic level, in one study (Jones et al. 1990) livestock weight gain relative to control treatments increased with P and P⫹S addition (Figure 10.5). Responses to S (5 experiments) and N⫹S (2 experiments) were mixed and, on average, not significant (Figure 10.5). Interactions with other trophic groups can affect the response of plants to resource addition. For example, Hobbs (1988) found increased gopher activity in fertilized plots, leading to disturbances that altered species and functional group composition in following years. Mycorrhizal infection increased plant biomass response to NH4⫹ addition but either decreased or did not affect biomass response to NO3 (Yoshida and Allen 2001). Resource addition is also expected to change the nature of plant-mutualist interactions by changing the cost ratio of the elements being exchanged (Sterner and Elser 2002). For example, when soil nutrients are scarce, it may be “cheaper” for a plant to provide C that it acquires by photosynthesis to mycorrhizae in exchange for nutrients than it is to invest in producing additional roots. Fertilization, by increasing the supply of nutrients, might reduce the cost of acquiring nutrients via roots relative to trading with mutualists. Rillig et al. (1998) found support for

this hypothesis, mycorrhizal colonization decreased with N and/or P addition, and increased with higher CO2.

Summary Even though vegetation in California’s Mediterranean climate experiences a strong drought during the summer, evidence for the importance of water as a limiting resource is mixed. Experimental manipulations of precipitation amounts have resulted in either increased productivity (Hamilton et al. 1999; Seabloom et al. 2003b) or no change (Dukes et al. 2005). However, comparison of annual grassland productivity and precipitation across many years provides stronger evidence for water limitation of grassland productivity (Murphy 1970; Pitt and Heady 1978). Therefore, water can be realistically assumed to be a primary limiting resource in California grasslands. The degree to which it is limiting will vary greatly with timing, intensity, and frequency of rainfall events, which in turn will be influenced by latitude, elevation, topography, and distance from the coast. Nutrients appear to limit production in almost all studies where they have been manipulated. Nitrogen, phosphorus, and sulfur all appear to be limiting nutrients in several settings. The largest effects of nutrient addition across studies appear to arise from the addition of phosphorus. Nonetheless, N limitation was found often, as was S limitation, and addition of these resources individually produced similar biomass responses. Their relative importance is dependent on site as well as, possibly, on the degree of water limitation. There were not enough studies of individual added nutrients or the appropriate meta-data (soil characterizations, slope, aspect, land use history) to rigorously evaluate how P or N or S limitation will vary across sites. Future studies might focus on how soil age, weathering, and site features affect the degree and type of limiting nutrients. Overall, treatments that added multiple resources found stronger, typically nonadditive, responses than treatments that only added one resource. Based on our results, California grasslands appear to be a classic example of multiple resource limitation as has been found in other grasslands (Brenchley and Warington 1958). In spite of the potential importance of

resource limitation in structuring plant communities, there are few studies that have examined limitation by multiple resources and their interactions in natural systems; studies are often restricted to a single resource, or to N versus P additions, which often leads to an underestimation of the possibility of limitation by other resources (Downing et al. 1999). In general, future experiments that explore limitation by water and its interactions with other resources would be valuable. Also, given the apparent ubiquity of multiple resource limitation in these grasslands, we need better information on the generality of limitation by resources other than N, P, and S, and also on possible co-limitation. These are all issues that are relevant to resource addition experiments in general, not just to the California studies we examined. Other approaches to studying resource limitation in California that have promise include the use of tissue nutrient ratios to assess N and P limitation (Koerselman and Meuleman 1996; Gusewell et al. 2003). Multiple resource limitation also highlights the need for more competition studies that utilize resource competition theory (Tilman 1980, 1982; Grover 1997; Chase and Leibold 2003). Resource ratio theory is particularly applicable to predicting the responses of species to changes in resource supply (Tilman 1982; Miller et al. 2005; Harpole 2006). The seemingly widespread phenomenon of limitation by multiple resources, combined with the functional and trophic group responses to N versus P or S, suggests that California grassland ecosystems might be particularly well suited to the application of stoichiometry as a theoretical framework for understanding the relationships between species and limiting resources (Sterner and Elser 2002). Understanding the factors that limit growth is essential to a mechanistic understanding of species interactions, species diversity and abundance patterns, and ecosystem processes and is critical for predicting the responses of California grasslands to projected global change (Fenn et al. 2003a; Hayhoe et al. 2004).

Acknowledgments We thank Katharine Suding, Valerie Eviner, Wendy Chou, Carla D’Antonio, and Jeffrey Corbin for comments on earlier drafts.

R E S O U R C E L I M I TA T I O N

127

E C O LO G I CAL I NTE RACTI O N S

ELEVEN

Genes on the Range Population Genetics K EVI N J. R I C E AN D E R I N K. E S P E LAN D

The California grassland represents one of the most thoroughly invaded ecosystems on the planet (Mack 1989), and so, perhaps, it is not surprising that much of the early work on the population genetics of component species focused on invasive species—in particular, annual grasses. Much of the early work by Subodh Jain and Robert Allard at the University of California at Davis represented direct offshoots of their agronomic work on long-term evolutionary dynamics of barley mixtures. In an effort to understand how selection may alter population genetic structure within inbreeding species (Jain and Allard 1960; Allard and Jain 1962), they expanded their research programs on the evolution and genetics of selfing species to include other species of Mediterranean annual grasses such as wild oats (Avena spp.) and annual fescues (Vulpia spp.). The grassland species they focused on were annuals that were easy to germinate and grow, and they demonstrated interesting and suggestive patterns of genetic variation on both a regional and local level. At the University of California at Berkeley, Herbert Baker was quick to recognize that many of the invasive species that were the scorn of most California botanists actually represented fascinating model systems for understanding the dynamics of colonizing species and the speed at which evolutionary change may occur. In fact, the symposium volume The Genetics of Colonizing Species, co-edited by Baker (Baker and Stebbins 1965), still remains an insightful synthesis of population ecology and population genetics; many of the areas in plant population biology that represent current “hot topics” can trace their roots to this truly seminal volume. Interest in the population biology and ecological genetics of native species within California grasslands has grown rapidly in recent years. In particular, highly variable success in the attempts to restore native components of California grassland ecosystems (Stromberg et al., Chapter 21) has generated a real need to understand patterns of local adaptation. The restoration practitioner’s perennial question of “How

local is local?” is a difficult one to answer and represents a situation in which the information needed for practical application and policy currently exceeds our basic scientific knowledge. A good answer to this question requires better knowledge in a number of areas such as historical and current variation in gene flow, scales of adaptation, spatial and temporal variation in the form and intensity of selection, and the relative role of biotic and abiotic selective agents (Sidebar 11.1). Unfortunately, relatively few studies of native species combine modern molecular techniques to estimate gene flow with field-based approaches, such as common gardens and reciprocal transplant experiments, to estimate local adaptation. This combined approach is necessary to understand the relative strengths of gene flow and selection in the formation of adapted ecotypes. Although much remains to be done to understand fundamental processes in the population genetics of California grassland species, the work reviewed in this chapter has provided many important insights into processes of microevolutionary change in plant populations. Basic population genetics concepts, ranging from co-adapted gene complexes to evolution of phenotypic plasticity, were initially proposed and developed by researchers working in the California grasslands. Ironically, the potential for restoring native species within the system may depend on these insights gained by studying tractable “model” invasive members of California’s grasslands.

Genetics of Colonization The Origins of Invasive Species and the Role of Genetic Bottlenecks It is almost axiomatic that the genetic diversity of an entire species is not present within the populations invading a new range (see Cox 2004 for examples). Typically, the initial

131

S I D E B A R 1 1 . 1 A D A P TAT I O N A N D R E S TO R AT I O N

The following recommendations for taking population genetics into account during restoration of native biodiversity were adapted from McKay et al. 2005. 1. Collect locally if possible. a. Seed collected from sites dissimilar (either in known or unknown ways) from the restoration site may be maladapted to the restoration site. b. Populations close to the site will be less detrimentally affected by gene flow from the restoration site than if non-local seed is used. c. Increasing the possibility of gene flow between the restoration site and local native populations will reduce the possibility of genetic drift and increase the ability of the restored population to respond to selection. Populations often vary in their ability to interbreed, and populations further away from one another are more likely to be limited in their ability to exchange genes. 2. Look for seed collection sites that match the restoration site in climatic and environmental conditions. a. Combine elevation and composite climate indices to establish environmental similarity. b. Information on climatic zones can be found in the Western Garden Book (Brenzel 2001), and a large “seed-zone” literature exists for tree species. 3. Determine the breeding system(s) of the species under restoration. a. Highly selfing species are less likely to create problems of unwanted gene flow to neighboring populations, are less likely to suffer from inbreeding depression, but are also more likely to be locally differentiated. b. Highly outcrossing species are less likely to be locally differentiated, but are more likely to suffer from inbreeding depression and Allee effects. If seed is poorly selected, there may be problems with gene flow to neighboring native populations. c. Simple bagging experiments can show whether a species is self-compatible, while pollen-ovule ratios (Cruden 1977) can also give rough estimates of outcrossing rates. 4. Determine the ploidy level of the species. a. Differences among populations for ploidy levels can occur, and can often prevent interbreeding. These differences can, in effect, turn a single restoration population into many small inbreeding populations. b. Chromosome counts for species can be found in the Jepson manual (Hickman 1993), or a search of the cytogenetics literature may indicate whether this is a cause for concern in the particular species under restoration. 5. Minimize “unconscious” selection when supplementing with commercial seed. a. Select the commercial grower using the same criteria as one would use for other seed collection sites (see 2). b. Harvest from the entire commercial population and over multiple years. 6. Make an effort to increase collective knowledge about the effects of population genetics on restoration success. a. Track the performance (growth, survival, reproduction) of individuals of known seed sources within the restoration project. b. Make this information available to other researchers and practitioners.

introduction event creates a significant genetic bottleneck; genetic variation of the introduced populations is but a small subsample of the genetic variation in the species’ home range. The amount of genetic variation within an introduced population may determine its capacity to adapt and spread within its new range, so the study of genetic bottlenecks is of fundamental interest and has implications for management strategies. The breeding system of the species is often a prime determinant of the severity of this bottleneck in the initial colonizers. Outcrossers are expected to exhibit less-pronounced bottlenecks, because outcrossing species typically exhibit higher levels of individual heterozygosity as well as greater levels of within-population genetic variation than inbreeders do. The severity of the bottleneck can also be greatly reduced if multiple introductions occur during the initial establishment phase of the invader. In fact, depending on their origins, multiple introductions can radically transform the genetic architecture of the species in its new range. For example, if individuals from genetically distinct subpopulations in the home range are combined in the invading population, genetic variation within the colonizing population may actually be greater than the genetic variation found within any of the source populations in the home range (Cox 2004). Identifying the origins of invaders is also important for deducing transport vectors and assessing the potential for preadaptation in the introduced populations. Colonizers from climatic zones or habitat types similar to the site of introduction are expected to be more successful than genotypes that have not evolved under similar conditions (preadaptation). Identification of invader origins may also be of great practical importance for developing biocontrol programs for the invasive species (DeBach and Rosen 1991). The development of allozyme markers in the late 1960s (Lewontin and Hubby 1966) facilitated efforts of California researchers to trace the origins and colonization dynamics of introduced species in our grasslands. In particular, patterns of genetic variation in Avena barbata (slender wild oats) were examined in both California and Mediterranean populations. Allozyme markers and inherited morphological traits were used to understand the relative importance of chance colonization, genetic drift, and selection in changing genetic structure between home range and new range populations of A. barbata. The earliest study by Clegg and Allard (1972) focused on differences in the regional patterns of genetic variation between California and Mediterranean populations. They sought to test whether differences in genetic structure were random in nature (i.e., caused by chance colonization and genetic drift) or nonrandom and correlated with environmental parameters (i.e., the result of selection). Using rather sparse sampling from a very large putative home range, the authors concluded that California populations probably represented a random sample of the genetic variation from the entire home range. Clegg and Allard (1972) also concluded that selection along climatic gradients, not random drift, created the regional pattern in California. Patterns of regional differentiation in California between drier/warmer

locations (the “xeric genotype”) and wetter/cooler locations (the “mesic genotype”) formed nonrandom subsets of the original genetic sample from the home range. They further presented the more general (and more controversial) conclusion that allozyme variation in this system was not neutral to selection. In a later extension of this idea, de la Vega (1996) argued for the non-neutrality of allozymes (as well as other molecular markers) and used examples from the A. barbata system as evidence that molecular polymorphisms can be involved in processes of adaptation. A more thorough genetic survey of both California and Mediterranean A. barbata populations was conducted by Garcia et al. (1989) to further circumscribe the source of California populations. In contrast to Clegg and Allard (1972), they found that California A. barbata did not comprise a random sample of its home range in the general Mediterranean region. Rather it appeared that California A. barbata populations originated from southwestern Spain (a conclusion supported by Perez et al. 1991); southwestern Spain was a region of intense trade activity with the New World around the time the species was introduced to California. Garcia et al. (1989) argued that the climate of southwestern Spain is similar to that of California, so introduced genotypes may have been preadapted to the general environmental conditions present in their new range. They then compared genetic structure in some detail between populations in Spain and California by examining variation in alleles as well as single-locus and multilocus genotypes. Their examination of allelic variation indicated that a significant bottleneck occurred during the introduction of A. barbata, because California populations contain only a small fraction of the infrequent or rare alleles present in Spain. This study, and subsequent work by de la Vega et al. (1991), emphasized that although Spanish and California gene pools are very similar in allelic frequencies, there are pronounced differences in the way the alleles are associated with each other in forming multilocus genotypes. In particular, the “mesic” and “xeric” multilocus genotypes that dominate California populations are not present in Spain. Both Garcia et al. (1989) and de la Vega et al. (1991) considered this reorganization of alleles to be strong evidence for the role of selection in creating regional and local genetic structure in A. barbata as it colonized and spread throughout California. Bottlenecks may be especially severe in inbreeding species introduced more recently than long-term residents such as A. barbata. Multiple introductions, which might reduce bottleneck effects, are less likely to have occurred in a very recently introduced species. This expectation appears to be supported by recent research on the genetic diversity and origins of Aegilops triuncialis (barbed goatgrass). Introduced to California in the early 1900s from a home range spanning the entire Mediterranean region and extending into the Middle East, A. triuncialis has recently begun to spread rapidly throughout northern California grasslands. Using a series of chloroplast and nuclear molecular markers, Meimberg et al. (2006) examined genetic diversity in several California populations relative to home range populations. Because

G E N E S O N T H E R A N G E : P O P U L AT I O N G E N E T I C S

133

A. triuncialis is used in wheat breeding, an exceptionally extensive collection of germ plasm from the home range was possible. From this survey it appears that A. triuncialis underwent a major bottleneck during its colonization of California rangelands. Meimberg et al. (in press) could identify only two distinct genotypes in California, which correspond roughly to Coast Range and Sierra Foothill colonization events. The authors speculated that the capacity of A. triuncialis to spread may rely on a combination of phenotypic plasticity and mutation in genes underlying quantitative traits. As previously noted, theory predicts that genetic bottlenecks should be less pronounced in outcrossed species (Conner and Hartl 2004). Although empirical data on invasive outcrossers is relatively scarce, the studies that are available generally support the predictions. A study by Wu and Jain (1980) of coastal California populations of the outcrossing perennial grass Anthoxanthum odoratum (vernal grass) examined variation in three allozyme loci. They found significant polymorphism for all three enzyme systems, and average heterozygosity for the three loci was relatively high (0.24–0.43). In studies on Trifolium hirtum (rose clover), Molina-Freaner and Jain (1992) reported that there was little indication of a reduction in genetic variation of California populations relative to home range populations in Turkey. In fact, they found that relative genetic variation within California populations was higher than in Turkish populations and that the number of multilocus genotypes was also higher in California. Because this species was purposefully introduced as a forage species for range improvement (Love and Sumner 1952), it is likely that there was a concerted effort to incorporate high genetic diversity in the initial plantings. In addition to a high genetic diversity of the initial introductions, Molina-Freaner and Jain (1992) also suggested that increased outcrossing rates, which foster gene flow among different subpopulations, may reduce genetic bottlenecks during future colonization events. In a rare test of the importance of genetic variation in colonization success, Martins and Jain (1979) monitored the demography of experimentally established roadside colonies of T. hirtum over a seven-year period. Although genetic polymorphism per se was not a significant determinant of colony persistence, they found that certain rose clover genotypes were more likely to germinate and successfully reproduce in some years than others. This significant year-by-genotype interaction in fitness suggested that different genotypes are better adapted to certain types of years. The retention of an array of “year specialist” genotypes in the seed bank may facilitate colony survival and spread and may thus represent a selective response to a growth environment that is highly variable from one year to the next.

Genetic Variation in Peripheral Populations: Genetic Drift and Founder Effects The inherently patchy distribution of many plant populations within California grasslands has led many researchers to speculate on how metapopulation processes of colonization

134

ECOLOGICAL INTERACTIONS

and extinction might increase the likelihood of founder effects and genetic drift. In annual species, where the dynamics of colonization and extinction are especially pronounced, determining the relative importance of selection versus random genetic drift has proven difficult. One strategy has been to examine populations that, because of their relative isolation or peripheral location, would be more likely affected by founder effects and thus express differentiation resulting from random genetic drift. In an early study of populations of Avena fatua (wild oats) within prune orchards, Jain and Rai (1974) capitalized on the agricultural practices within the orchards to examine the effects of population subdivision, extinction, and recolonization on genetic polymorphism in both morphological traits and isozyme markers. Although not conducted within California grasslands, the study is worth noting because of the relevance of its results to a more general understanding of how metapopulation dynamics can influence patterns of genetic variation in grassland annuals. Cultural practices within the orchards reduced A. fatua populations to small colonies clustered around the base of each tree that were isolated from other such colonies in the orchard. Two orchards were studied that, because of differences in tillage operations, were distinct in terms of relative colony size; the average census sizes of A. fatua populations in one of the orchards were about twice that of populations in the other orchard. Citing data on restricted seed dispersal, inbreeding, and low seedling survival between colonies, the authors argued that this system closely approximated Wright’s model of a large metapopulation subdivided into a series of isolated small populations. Overall, the authors found that there was a large degree of genetic differentiation among colonies with many colonies monomorphic for genetic markers. As might be expected for a series of small isolated populations, this mosaic of genetic differentiation was random in nature and apparently dominated by drift effects. The authors found no significant variation among genotypes along clinal gradients within the orchard, suggesting that selection was not important and that drift effects predominated in structuring genetic spatial patterns. In addition, measurements of fecundity among individuals within the colonies indicated that high reproductive variance within the colonies significantly reduced effective population size, another factor that would increase the likelihood of drift effects. Although these studies were conducted in prune orchards, it is not difficult to imagine that drift could play a similar dominant role in more natural grassland populations that are patchily distributed, isolated by low gene flow, and undergo localized extinction and colonization. Further examination of the potential importance of drift effects during founder events was undertaken by Jain and Rai (1980) in their comparison of genetic variation in morphological traits and allozyme markers between central and isolated roadside populations of A. barbata. In general, their results supported the importance of drift effects in creating spatial differentiation between large central populations and smaller, peripheral roadside colonies of A. barbata. Roadside

populations were more likely to be monomorphic for genetic markers and exhibited much stronger differentiation among populations. Both of these results suggest the importance of founder effects. The authors, citing evidence presented by Endler (1977), further noted that genetic effects of initial founding events can persist for many generations even in the face of substantial gene flow. The persistent influence of founding events on genetic variation was further demonstrated in a later study by Rai and Jain (1982) on the effects of gene flow on spatial genetic variation in A. barbata. In this study, natural populations of A. barbata at four different grassland sites did not suggest any bottlenecks in population size and allelic frequencies remained stable over a three-year period. Overall, these studies on the importance of random processes such as genetic drift and founder effects caution against the assumption that selection is the predominant force in forming the genetic architecture of many grassland species. Particularly for inbreeding annuals undergoing repeated extinction and colonization, the combination of restricted gene flow and small effective population size would suggest that genetic drift is an important “player” in the evolution of grassland metapopulations.

Breeding Systems Inbreeding Effects Inbreeding depression can be a cryptic barrier to successful colonization. In particular, colonizing plants that are outcrossers can suffer from inbreeding depression because of lack of mating partners. Measuring inbreeding depression includes not only an examination of seed set within inbred plants but also following the progeny of these plants through their life cycle and comparing them to outcrossed lines. In a study comparing seed performance of selfed and outcrossed Trifolium hirtum (rose clover; Molina-Freaner and Jain 1993), germination between the two inbred and outcrossed plants was the same, but survival was significantly less for seed produced from the selfed plants. Similar results were found for Delphinium hesperium (western, or coast, larkspur), D. recurvatum (valley, or recurved, larkspur) and D. gypsophilum (gypsum-loving larkspur); selfed plants had seed set as high as outcrossed plants, but seeds produced from selfing had lower germination rates (Lewis and Epling 1959). In studies within the Limnanthes (meadowfoam) genus, it appears that the species L. floccosa (woolly meadowfoam) evolved an inbreeding mating system in addition to considerable phenotypic plasticity in order to be a more successful colonizer than its congeners (Jain 1981). While it might be expected that successful colonizers are inbreeders, evolutionary trajectories do not always operate as expected. Lupinus succulentus (arroyo lupine ) is an outcrossing species with heterozygote advantage, as opposed to its noncolonizing relative Lupinus nanus which has similar outcrossing rates and similar amounts of genetic variability (Harding and Barnes 1977). Outcrossing species are subject to

genetic drift; several colonial populations of L. succulentus were found to be fixed for some genetic markers (Harding and Barnes 1977). When the colonization environment is variable, genetically diverse colonizers are more successful, as variable environments facilitate the maintenance of heterozygote advantage and an outcrossing breeding system (Lewontin 1965).

Gene Flow and Breeding Systems Gene flow, or the migration of genotypes among populations, is the primary microevolutionary force of sufficient strength to counter the effects of natural selection. The balance between gene flow and selection is important to understand because the likelihood of local adaptation occurring depends on the relative effects of selection and gene flow (Kirkpatrick and Barton 1997). Surveys of population structure (e.g., McKay and Latta 2002; Kittelson and Maron 2001) have indicated that even under significant rates of gene flow, selection is often strong enough to maintain population divergence and adaptation. Although topography and other environmental factors create dispersal barriers to gene flow patterns in plants, an important determinant of gene flow is the plant’s breeding system. Using many examples from California grasslands, Jain (1976) summarized some of the potential effects of plant breeding system on population structure. In general he found that reduced gene flow in inbreeding species results in greater among-population genetic divergence and lower levels of genetic variation within populations (e.g., fewer polymorphic loci and reduced heterozygosity). He also suggested that reduced gene flow in selfing species facilitates local adaptation because selection does not have to be as strong to form ecotypes if gene flow is low. The breeding system of a population or species can have manifold effects on local neighborhood structure and the likelihood that a mosaic of subpopulations, or demes, will develop (Levin and Kerster 1971). Knapp and Rice (1996) found that regional and local allozyme variation in the less-selfing native perennial grass Elymus glaucus (blue wildrye) conformed to expectations for a species with a more mixed mating system. Although E. glaucus is more self-pollinating than outcrossing (Cronquist et al. 1977), Knapp and Rice (1996) still found substantial variation within populations, with an average of 31.4% of loci polymorphic and an average of 1.4 alleles per locus. However, other aspects of population structure reflected the reduced gene flow associated with inbreeding populations. For example, gene flow was relatively low (Nm
0.205), and most of the genetic variation was distributed among populations (54.9%). In contrast, a survey by Hamrick and Godt (1989) found that genetic variation in outcrossing plant species occurred primarily within populations and averaged around 85.2%. Dyer and Rice (1997a) studied the potential importance of differential dispersal between pollen and seeds in creating genetic structure in the native bunchgrass Nassella pulchra (purple needlegrass). Using phenological data from a common garden and molecular markers (i.e., randomly

G E N E S O N T H E R A N G E : P O P U L AT I O N G E N E T I C S

135

amplified polymorphic DNA, RAPDs) in field populations, they examined how genetic differences scaled along highly localized gradients in soil depth up to the landscape level. They argued that local selection among soil microenvironments is facilitated by highly restricted gene flow resulting from limited seed dispersal distances. In contrast, weak genetic structure at the landscape level reflected the much greater dispersal distance of pollen; they argue that the detection of genetic structure is highly dependent on the spatial scale of analysis. Both E. glaucus and N. pulchra are important species for restoration in California grasslands; these two studies suggest that patterns of gene flow in mixed mating systems complicate the determination of scales of adaptation. Genetic variability within mixed mating systems does not appear to be related to gene flow. In studies of Lupinus (lupine), Harding et al. (1974) found that genetic variability within populations did not determine outcrossing rates, although outcrossing rates were correlated with the degree of self-compatibility within each population. In studies of genetic variation in A. barbata, Rai and Jain (1982) predicted degrees of patchiness in known genotypes (demonstrated by variation in phenotype) based on measured seed flow and outcrossing rates among populations. The variety they found among the populations was significantly more than indicated by the predictions based on measured gene flow. The ability of populations to accept pollen moved from other populations varies among many California native forb species, such as Holocarpha macradenia (Santa Cruz tarplant) (Palmer 1982), Eschscholzia californica (California poppy) (Cook 1961), and Nemophila menziesii (baby blue-eyes) (Barr 2004). In experiments performed by these researchers, plants from some populations of these species exhibited potential outbreeding depression with reduced or no seed set when fertilized with pollen from other populations. Geologic processes may have influenced gene flow, enabling plants in regions of large land mass movement, such as the southern North Coast Ranges, to retain their ability to exchange genes across populations (Barr 2004). Nagy’s (1997) field experiment tested whether plants receiving foreign pollen (coastal  inland, inland  coastal) were able to produce seed. In this case, only the coastal subspecies Gilia capitata ssp. chamissonis (dune gilia) was able to produce seed from ovules fertilized with foreign pollen, whereas inland populations of Gilia capitata ssp. capitata were not able to hybridize when the other subspecies was the pollen donor. These results were repeated in a greenhouse study. In experiments in which native pollen was diluted by foreign pollen, dilution resulted in a smaller seed set for both subspecies. Simple geographic distance is an obvious barrier to gene flow among populations. In a Washington state study of molecular variation in the California vernal pool native Navarretia leucocephala, Boose et al. (2005) found that populations generally differed more strongly from one another the farther apart they were. Pools a few hundred meters apart were not differentiated from one another for the RAPD molecular markers used, and estimates of number of migrants

136

ECOLOGICAL INTERACTIONS

among populations were surprisingly high (14 migrants per year), especially given the island-like nature of the habitats in which these plants are found. Molecular differentiation was apparent when distances between pool populations became greater than about 1500 meters. Interfertility also varied by geographic distance in Streptanthus glandulosus (jewelflower, Kruckeberg 1957), resulting in a revision of the genus (Kruckeberg 1958). Mayer et al. (1994) found high degrees of population differentiation using enzyme analysis, even within the most widespread species of Streptanthus. Estimates of the number of migrants between populations led to the conclusion that most of these serpentine endemics were vulnerable to genetic drift because of lack of gene flow. These researchers found a classic “isolation by distance” pattern within the Streptanthus species tested in these experiments. Morphology and interfertility patterns closely tracked the differentiation found in the enzyme analysis. In studies of Limnanthes (meadowfoam) species, the allozyme-measured genetic distance between populations closely tracked their geographic distance (Jain 1981). Genetic diversity within a population can interact with inbreeding depression to make a population more or less likely to interbreed with other populations. In a common garden experiment in Riverside, populations of different numbers of individuals (2, 5, 10, and 20) of Raphanus sativus (wild radish) in three levels of relatedness (full-sib, half-sib, and unrelated) were established (Goodell et al. 1997). A paternity exclusion analysis using allozymes determined the degree of outcrossing that occurred among the constructed populations. Populations greater in size than two were indistinguishable from one another in the amount of out-sired seed they produced. The degree to which relatedness affected the proportion of out-sired seed differed among the three experimental trials and differed among the population sizes, but in population size 2, half-sib populations were always in the group that had the greatest proportion of out-sired seed. Gene flow events are easier to detect in small populations, and also easier to detect in genetically uniform populations, because cryptic gene flow is less likely to occur (Devlin and Ellstrand 1990). Because R. sativus has a self-incompatibility system with at least 32 alleles, genetically uniform populations were expected to have the highest out-siring rates. In addition, distance between populations was small (100 m), and therefore, the inability of small populations to effectively attract pollinators was probably a muted effect. Nonetheless, this experiment showed that demography and relatedness can affect gene flow among populations and this effect varies among years.

The Genetics of Inbreeding In trying to explain the diversity of breeding systems in plants, there has been much speculation as to the impact of breeding system on the genetic structure of plant populations. In particular there has long been an interest in how breeding system may influence the capacity of plant populations to

respond to natural selection. Plant evolutionary biologists have argued that inbreeding, by reducing both gene flow among populations and genetic variation within populations, should facilitate close adaptation to current selective regimes (Stebbins 1957). It was reasoned that lower gene flow among selfing populations would be less likely to swamp adaptive differentiation driven by selection. However, it was also assumed that the concomitant reduction in within-population variation would reduce the evolutionary potential of a population to respond to new selective challenges and might result in populations, or even species, becoming evolutionary “dead ends.” Research on selfing annuals in the California grassland was pivotal in testing theoretical predictions that inbreeding should increase among-population variation while reducing within-population genetic diversity (Allard et al. 1968). Early work by Knowles (1943) on populations of Bromus hordeaceus (soft chess) across a climatic gradient from coastal to interior sites in California supported some of the predictions but countered others. Knowles found clinal variation in quantitative traits such as flowering time, plant height, and tillering capacity as well as sharp among-population differences along the cline. These distinct differences among populations were in accord with theoretical predictions, but populations of B. hordeaceus also exhibited substantial within-population variation. Strong differences among families, indicative of within-population genetic variation, were found for all quantitative traits measured. In stark contrast to the theoretical predictions of genetic uniformity within populations of inbreeders, Knowles concluded from his study that most, if not all, B. hordeaceus individuals within each population were genetically distinct. Subsequent work on B. hordeaceus by Jain et al. (1970) also found substantial variation among and within populations for quantitative traits although levels of within-population variation differed sharply among populations. Interestingly, Jain et al. (1970) did not detect a clinal gradient in variation and they speculated that the day length regime of their common garden may have masked phenotypic expression of clinal variation. The unexpectedly high rates of within– population variation found in both of these studies on B. hordeaceus was corroborated by work in A. fatua in which similar differences among families were detected for morphological, phenological and fitness traits (Imam and Allard 1965). In A. fatua, statistical tests of family differences indicated that when families were compared for multiple traits, essentially every family differed from every other family, indicative of substantial genetic variation within populations. In addition to these differences among families, Imam and Allard (1965) also detected significant variation among individuals within full sib families, which suggested that there may be heterozygosity within the parent individuals. To test for this possibility, they established high- and low-selection regimes on flowering time, height, and growth habit (e.g., prostrate vs. erect). After only a single generation of selection, they found

significant differences between low- and high-selection lines for all characters. They interpreted this rapid response to selection as strong evidence that individuals within inbreeding populations may still be heterozygous for quantitative traits. They concluded that phenotypic differences among full sibs within families of selfing species are not only caused by environmental or developmental factors but also by segregation of distinct genotypes. A series of studies by Allard and his co-workers on the highly selfing native annual grass Vulpia microstachys (small fescue) further challenged the assumption that highly inbred species should exhibit little genetic variation within populations (Kannenberg and Allard 1967; Allard and Kannenberg 1968; Adams and Allard 1982). Focusing on V. microstachys as a model system for studying the effects of extreme inbreeding on genetic structure, Kannenberg and Allard (1967) initiated study of this native annual grass by collecting from populations located in three climatic regions in California: Inner Coast Range, eastern edge of the Central Valley, and the Sierra Foothills. Using extensive progeny testing from these field collections, they examined mating system variation among populations as well as genetic structure. In general, V. microstachys populations were found to exhibit extremely low rates of outcrossing (usually much less than 1%) with no detectable hybrids in the examination of over 20,000 individuals. Despite this extremely high level of selfing, a hierarchical analysis of phenotypic variation indicated that the greatest variation in traits such as flowering time, height, and tillering rate was at the among-family level, a result suggesting significant genetic variation within populations despite almost complete inbreeding. The authors also noted that expression of potential outcrossing (i.e., frequency of plants with extruded anthers) varied significantly among populations and may be related to site quality such that outcrossing rates increase in favorable sites. Significant variation among individuals was also detected in some families, but it was not possible to determine whether this variation reflected genetic segregation rather than variation in developmental homeostasis among homozygous families. The strong differences among families within a population occupying a spatially heterogeneous environment prompted Kannenberg and Allard (1967) to argue that this type of localized genetic structure may allow V. microstachys to adapt to selection varying over very small spatial scales. Using electrophoretic markers, Adams and Allard (1982) confirmed the highly selfing nature of V. microstachys. However, they also detected what they called “bursts of outcrossing” in one of the populations that increased the frequency of heterozygotes from less than 1% to more than 9% in a single generation. Kannenberg and Allard (1967) had previously noted that because most individuals are homozygous for different alleles at many of their loci, any outcrossing event tends to produce a unique multiple heterozygote. Withinpopulation diversity of homozygotes, when coupled with these sporadic episodes of outcrossing, could generate a significant amount of genetic diversity that could help to

G E N E S O N T H E R A N G E : P O P U L AT I O N G E N E T I C S

137

explain the high amounts of variation seen in this highly selfing annual. Repeated assertions that fine-scale spatial genetic variation in V. microstachys represents the action of highly localized selection was not tested until a reciprocal transplant study at the University of California’s McLaughlin Reserve in the Inner Coast Ranges of northern California (Jurjavcic et al. 2002). These authors were interested in understanding how gradients in stress (i.e., a serpentine soil gradient) and competition interact to determine V. microstachys distribution and abundance. They manipulated competition in three habitats (rocky serpentine, serpentine meadows, and nonserpentine grasslands) and also reciprocally transplanted seed sources among all three habitats. If V. microstachys distribution is dependent on fine-tuned genetic differentiation, then each habitat seed source should perform best in its “home site.” Contrary to the expectations of previous researchers, local adaptation was not strongly expressed in the study. Local adaptation was suggested by a higher emergence and survival of rocky serpentine genotypes in their home sites, but growth and survival did not show any seed source effects. Instead, variation in these fitness parameters across habitats was the result of phenotypic plasticity and not genetic differentiation. It thus appears that although V. microstachys may represent a model system for studying inbreeding, it is not an exemplar for demonstrating how restricted gene flow in selfers may facilitate highly localized adaptation.

Variable Outcrossing: Causes and Consequences Many outcrossing species are capable of self-fertilization, and the degree to which a species exhibits outcrossing may be variable. The ratio of outcrossing to inbreeding within a population informs both the genetic variability of the population and its ability to exchange genes with neighboring populations. Using two populations located in the Sierran foothills and one population near Cache Creek Canyon (Yolo County), Weil and Allard (1964) examined two phenotypic characters under single gene control in Collinsia heterophylla (Chinese houses). By choosing two sites within each population to conduct their experiments, they were able to examine the degree of variability in outcrossing rates both within and between populations. While other (multigenic) phenotypic traits showed large differences in genetic variability within populations and between sites, the differences were not attributable to variable outcrossing rates among populations and sites. In a common garden experiment performed on seed collected from a variety of locations near Lake Berryessa, Horovitz and Harding (1972) studied the degree of outcrossing indirectly in Lupinus nanus ssp. nanus, recording floral characteristics that would make each plant more or less able to attract flying insects that would serve as cross-pollination vectors. As measured in an ex situ experiment, outcrossing rates varied considerably among individual plants (from 0–0.844), but less so on a population level (0.467–0.828).

138

ECOLOGICAL INTERACTIONS

Outcrossing rates also varied over the season within phenotypes, generally increasing over the 60 day experimental period. While only maternal outcrossing rates were observed in this study, male outcrossing rates would also certainly be affected in either parallel or opposing manners. In natural populations of L. nanus ssp. nanus as well as the subspecies menkerae, apricus, latifolium, and vallicola, outcrossing rates were still variable in three populations of ssp. nanus, ranging from 0.5  0.2 to 0.83  0.05 (Harding et al. 1974). All subspecies exhibitied variation among populations for outcrossing rates; some by a factor of two and others by a factor of ten. A survey of 29 ruderal populations of Lupinus succulentus (arroyo lupine) in the Coast Ranges from Northern to Southern California, the Transverse ranges, and inland to the Sacramento Valley (Harding and Barnes 1977) showed heterozygosity rates ranging from 0.009–0.548 and outcrossing frequencies ranging from 0.138–0.971. The study also found year effects in outcrossing rates. There was nonetheless a strong correlation between outcrossing rates and genetic diversity within populations. In contrast, Harding et al. (1974) found no correlation between outcrossing rates and genetic diversity within populations of Lupinus. Differences in ploidy level within and among populations of the same species may also influence outcrossing rates. In studies of Delphinium gypsophilum (gypsum-loving larkspur), entire populations were found fixed for one ploidy or another, but some populations contained mixtures of both ploidy levels found within the species, which would significantly affect outbreeding rates between and within populations (Koontz and Soltis 2001).

Population Differentiation Local Molecular Variation Gene flow among populations is often measured by examining molecular markers; however, it has been difficult to demonstrate a correlation between adaptation and molecular genetic markers in California grassland species. It is important to note that molecular markers portray a long history of gene flow, and a recent interruption of historical gene flow may not be detected by an analysis of current molecular variation within populations. The most rigorous method of measuring current gene flow is by conducting paternal exclusion analysis on molecular markers within populations. However this technique has its flaws because it underestimates outsiring events (Goodell et al. 1997; Devlin and Ellstrand 1990), often fails to incorporate year-to-year variation in gene flow among plants, and is labor-intensive. While molecular differences among populations are often used to inform decisions on the conservation value of these populations, it is important to remember that molecular markers do not always diverge in parallel with phenotypic or adaptive characteristics, which are ultimately the basis for sustaininable plant populations. For example, Linhart (1988) found adaptive differences

among vernal pool populations of Veronica peregrina (purslane speedwell), this differentiation was not reflected in electrophoretic assays of enzyme systems (Keeler 1978). Instead, differences between center and edge habitats in terms of absolute polymorphism was apparent, with edge individuals exhibiting more electrophoretic variation than center individuals. This paralleled the morphological tests of Linhart (1988), who found more variation among individuals inhabiting edge habitats. Knapp and Rice (1998) found significant differentiation among populations of N. pulchra (purple needlegrass) in both morphological and isozyme traits when plants were grown in a common garden. However, in cluster analysis, populations grouped differently depending on which measure was used, indicating a lack of parallel between phenotypic traits and molecular markers. There are some examples in which variation in genetic markers is correlated with adaptive variation. In a reciprocal transplant experiment of known genotypes of A. barbata, Jain and Rai (1980) found a small but significant advantage of home genotypes vs. nonlocal genotypes. This example of genotypic variation paralleling local adaptation is also suggested in the myriad work of Allard (see the discussion of coadapted gene complexes). One example is an examination of 14-locus genotypes (Allard 1996) found in A. barbata across two very different regions of Spain and across California. The presence of reorganization of common alleles in Spain into new genotypes in California may indicate a link between these complex genotypes and adaptation to growing conditions in California. Molecular markers can be used to detect differences in heterozygosity among populations. In the case of Trifolium amoenum (showy Indian clover), an inland population (Occidental, Sonoma County) known to have been founded by a single individual was compared to a coastal population (Dillon Beach, Marin County) of this rare plant, and was also compared to eight populations of two congeneric species (Knapp and Connors 1999). As expected, the inland population showed reduced heterozygosity compared to the coastal population, while the within-population heterozygosity of the coastal population compared favorably to that of one of the comparison congeneric species. The heterozygosity of the other comparison species was so low as to have a significantly different mating system; probably primarily inbreeding, in contrast to the other two outcrossing species.

Local Adaptation Differences among populations are somewhat simple to test for: Collect seeds from several populations, grow them in a common garden, and observe whether the resulting plants differ from one another on a population level. Without growing the populations together at a single site, it is impossible to tell whether morphological differences are due to a plastic response to the environment, to maternal

environmental effects, or to genetic differences. Demonstrating that genetic differences among populations are adaptive is more arduous in that genotypes must do better in their home environment than genotypes from other environments. Many studies have shown population differentiation without testing the adaptive nature of the differentiated traits. While differences among populations in seed dormancy are common, examples including Eschscholzia californica (California poppy; Montalvo et al. 2002), Nemophila menziesii (baby blue-eyes; Cruden 1974), and Blepharizonia plumosa (big tarweed; Gregory et al. 2001), the genetic link of dormancy cues to environmental differences between the populations have not been shown. Population differences in seed dormancy and leaf pigmentation in Eschscholzia californica (Espeland and Myatt 2001), and allozyme heterozygosity in Trifolium amoenum (showy Indian clover; Knapp and Connnors 1999), and isozyme diversity within species of the Streptanthus (jewelweed) complex (Mayer et al. 1994) might be adaptive but also may be a result of a genetic drift. Furthermore, small, isolated populations are less likely to have genetic and thus phenotypic diversity and are more likely to be differentiated from other populations in a nonadaptive manner.

Coadapted Gene Complexes and Local Adaptation An important question in population genetics revolves around the relative importance of selection at the single-gene level vs. its potential effect on groups of genes or even whole blocks of chromosomes. Franklin and Lewontin (1970) argued that reduced rates of recombination can cause correlations among allelic states such that selection acts on correlated or interacting sets of loci (i.e., epistatic selection). Theoretical studies on inbreeding (Jain and Allard 1966) indicated that close inbreeding strongly reduced recombination rates among all loci (even unlinked loci) and resulted in dynamics similar to those of very tight linkage among genes. An early experimental examination of the effects of inbreeding on the evolution of allelic correlations was conducted by Clegg et al. (1972) on two populations of Hordeum vulgare (wild barley). They found that significant departures from random association among loci increased over generations in both populations and that a similar multilocus gametic type was in excess in both populations by the twenty-sixth generation. The authors argued that these two results indicated a strong role for selection in the development of highly interactive, coadapted gene complexes; a concept first introduced and developed by Dobzhansky (1970). In an extension of the work by Clegg and Allard (1972) on adaptation of multilocus genotypes of A. barbata to regional climate, Hamrick and Allard (1972) reported on the microgeographic variation of multilocus genotypes across a local moisture gradient at a site near Calistoga, Napa County, California. They reported that the gradient appeared to be a mosaic of microsites differing in moisture availability, and they argued that specific sets of alleles were correlated with

G E N E S O N T H E R A N G E : P O P U L AT I O N G E N E T I C S

139

these drier and wetter microsites (the “xeric” and “mesic” genotypes, respectively). The fact that alternate sets of alleles were associated with each genotype suggested the possibility that selection for coadapted gene complexes may be occurring at a very local scale in this microhabitat mosaic. Citing the apparent association of multilocus “xeric” and “mesic” genotypes with localized variation in soil moisture, Allard et al. (1972) proposed that these genotypes represent the product of selection for co-adapted gene complexes. They argued that a gametic-phase disequilibrium exists in these populations and that the coadapted gene complexes of the “mesic” and “xeric” genotypes are maintained by both selection and reduced recombination caused by inbreeding and/or linkage. From an ecological perspective, all of these studies were problematic. The determination of mesic and xeric microhabitats was not supported by independent and quantitative assessments of variation in soil moisture, while mesic and xeric microsites were determined by casual observations and phenotypic variation in plant vigor and appearance. In addition, the use of observations of phenotypic variation in A. barbata growth and vigor in the field to confirm mesic and xeric genotypes confounded genetic and environmental determinants of phenotypic expression. For example, field observations indicated that putative mesic genotypes flowered earlier, were taller, produced more tillers, and had higher seed production than xeric genotypes, but these differences could also be caused by environmental variation in site conditions. In an effort to demonstrate that this phenotypic variation in the field had some genetic basis, Hamrick and Allard (1972) conducted a common garden experiment. Using two populations of each genotype, they found that mesic and xeric genotypes grown in a common garden were significantly different in flowering time, height, and tiller number but not in seed production. Assuming that maternal environmental effects were not significant, this study confirms that, except for seed production, the suite of phenotypic characters used to characterize mesic and xeric genotypes did have a genetic basis. Unfortunately, however, the approach used by Hamrick and Allard (1974) to select mesic and xeric genotypes from the field was again based on phenotypic variation in A. barbata. To relate the genetic variation found in the common garden to ecotypic adaptation to soil moisture, the authors should have first independently determined variation in soil moisture (e.g., gravimetric water content or soil moisture potential) and then collected and categorized plant material as to xeric or mesic on the basis of the soil moisture data. In an expansion of the original study, Hamrick and Holden (1979) examined a much larger number of sites and attempted to characterize in some detail the mosaic of putative mesic and xeric microsites. Although the authors suggested that some soil data were taken, the quantitative classification of xeric and mesic microsites was again solely determined by phenotypic variation in A. barbata height and flowering panicle density. The use of phenotypic expression in the

140

ECOLOGICAL INTERACTIONS

field to designate what is a xeric or mesic microsite again assumes that genotypes are closely tracking soil moisture variation; independent assessment of soil moisture variation is necessary to test this assumption. Although the authors argued that selection has shaped genetic structure within the soil moisture mosaic, they diverged from the conclusions of Allard et al. (1972) and admitted that there was insufficient evidence to confirm that epistatic selection was occurring and maintaining coadapted gene complexes in the xeric and mesic genotypes. Clearly, a good test of the reality of fine-scaled adaptation to soil moisture variation would be provided by a field experiment in which mesic and xeric genotypes are reciprocally transplanted between putative mesic and xeric microsites. Although this conclusive type of experiment has never been done on the microsite level in A. barbata, Jain and Rai (1980) used a reciprocal transplant approach to test local adaptation to regional climate. In a study that was rather remarkable in its scope and duration, they found some evidence for local adaptation at a regional level, but the results indicated weak selection. In a long-term experiment, many of the nonlocal colonies that they established persisted quite well throughout the course of the study (5 – 8 years depending on site), and the authors noted that there was often very little shift in genotypic frequency over time. Also, a shorter-term study of two common gardens established in a coastal and interior valley site did not detect strong regional adaptation. However, there was an indication of weak selection for the mesic genotype in the cooler coastal site and selection for the xeric genotype in the warmer interior site. Overall, the study indicated some regional adaptation, but selection was not as strong as expected. The results raise the question of whether selection is strong enough to create fine-tuned adaptation to the soil moisture mosaic at the Calistoga site in spite of gene flow among the microsites. As previously noted, reciprocal transplant experiments at the small spatial scale of the Calistoga site have not been done, but Gardner and Latta (2006) recently compared the relative fitness of mesic and xeric genotypes in a reciprocal transplant experiment across a regional rainfall gradient in northern California. Using the University of California Hopland Research and Extension Center (Mendocino County) as a “mesic” planting site and the University of California Sierra Field and Extension Center (Yuba County) as a “xeric” planting site, they found that the mesic genotype was more fit in both mesic and xeric sites. Although these relative fitness results are not strong evidence for local adaptation, previous work by Latta et al. (2004) has identified a suite of characters in the mesic and xeric genotypes that would suggest adaptation in A. barbata to variation in microsite productivity and soil moisture. They found that mesic genotypes were more competitive than xeric genotypes and that, primarily because of greater seed size, seedling size and adult fecundity were also greater in the mesic genotype. On the other hand, the xeric genotype exhibited greater adaptation to soil moisture stress by

expressing a greater root mass ratio and allocating significantly more root biomass deeper in the soil.

Phenotypic Plasticity and Adaptation to Heterogeneous Environments A hallmark of the California grassland ecosystem is its wide variation in weather patterns from one year to the next and pronounced spatial heterogeneity at multiple scales. Two alternative strategies that may allow plants to adapt to such rampant environmental heterogeneity are selective genetic differentiation (i.e., “classic” local adaptation) or an environmentally induced phenotypic response (i.e., adaptive phenotypic plasticity). In systems with fine-grained heterogeneity, where temporal variation is rapid relative to an individual’s generation time, theory predicts that adaptive plasticity should be the primary adaptive response to a variable selective regime (Levins 1968). The potential importance of phenotypic plasticity in plants as an alternative strategy to genetic differentiation in response to selection was first summarized by Bradshaw (1965). Subsequent reviews by Schlichting (1986) and Sultan (1987) have further emphasized the importance of phenotypic plasticity as a major force in plant evolution by highlighting additional concepts such as correlated trait matrices and the capacity of selective buffering by plasticity to maintain genetic variation. Researchers were quick to appreciate the potential importance of phenotypic plasticity as an adaptive strategy in California grasslands, especially in introduced species that expressed low amounts of heritable variation because of genetic bottlenecks and inbreeding. Marshall and Jain (1968) compared phenotypic plasticity between Avena fatua and A. barbata in two greenhouse studies that examined phenotypic responses of both grasses to variation in soil characteristics and levels of intraspecific and interspecific competition. Because A. fatua had been shown to be more genetically variable than A. barbata, they hypothesized that the genetically less variable A. barbata should exhibit greater plasticity. This prediction of an apparent trade-off between genetic differentiation and plasticity was supported by results indicating that the genetically depauperate A. barbata was more plastic across environments for essentially all traits measured. Compared to the typical agricultural habitat of A. fatua, Marshall and Jain (1968) further suggested that the grassland environment of A. barbata may be more unpredictable and thus selects for greater plasticity. The importance of plasticity in generating regional patterns of phenotypic variation was studied in the highly selfing annual grass Bromus rubens (red brome) by comparing variation in field populations with progeny raised in a controlled environment (Wu and Jain 1978). They found that only a small amount of the regional phenotypic variation in traits ranging from flowering date to seed weight could be attributed to genetic variation (12 – 27%), while the bulk of variation (73 – 88%) represented plastic responses. As with

A. fatua and A. barbata, a comparison of B. rubens and B. hordeaceus suggested that plasticity is greater in the more genetically depauperate B. rubens while regional adaptation in the more genetically variable B. hordeaceus results from local genetic differentiation (Jain 1979). These congeneric comparisons support the concept of alternate adaptive strategies and suggest a cost to plasticity (i.e., “jack of all trades, master of none”). However, although a cost to plasticity is often assumed, it is rarely demonstrated, and only recently have approaches been developed that may be able to get at this very difficult question (De Witt et al. 1998). In a series of field experiments, Platenkamp (1990, 1991) examined the importance of phenotypic plasticity in the demography of an introduced perennial grass species Anthoxanthum odoratum in two sites that differed in soil moisture availability. Using reciprocal transplants of both seeds and clonal replicates, he examined the relative importance of genetically based local adaptation and adaptive plasticity in structuring mesic and xeric populations in a coastal prairie. In an examination of seed and seedling traits (Platenkamp 1991), variation in germination seemed to exhibit some genetic basis because seeds from the mesic site exhibited higher dormancy regardless of planting site. However, there was no indication of genetically based local adaptation. Overall, phenotypic plasticity was the primary determinant of differential seedling survival and eventual fecundity within mesic and xeric sites. In another experiment, Platenkamp (1990) reciprocally transplanted clonal replicates of A. odoratum between the mesic and xeric sites to more precisely examine the relative importance of phenotypic plasticity and genotypic differentiation in local adaptation. Similar to results from the study on seed and seedling characteristics, there was no indication of local adaptation (i.e., no site-byclone origin interactions). Although the relative importance of genetic differences and plasticity were somewhat trait dependent, fitness variation, measured by mortality and cumulative fecundity over a 3-year period, was almost entirely caused by phenotypic plasticity. Platenkamp (1990) suggests that the failure to demonstrate a clear pattern of local adaptation may reflect the recent introduction of this species to the site as well as year to year fluctuations in the soil moisture gradient. The importance of phenotypic plasticity in this species was further demonstrated by a study of the effect of neighbors on A. odoratum fitness (Platenkamp and Foin 1990). Although the density and biomass of competitive neighborhoods strongly influenced growth and cumulative reproductive output of clonal replicates, there was no difference among clones in response to neighbors. This lack of genetic variation in response to neighborhood composition indicates that response to competition in this species is primarily plastic. Germination timing and seed dormancy are life history traits that can have far reaching effects on both year-to-year variation in community composition of California grasslands (Heady 1958; Bartolome 1979; Corbin et al., Chapter 13) and population dynamics (Rice 1989b). A study by Jain (1982) on

G E N E S O N T H E R A N G E : P O P U L AT I O N G E N E T I C S

141

regional variation in seed dormancy in several annual species suggested that for the grasses tested, there was a genetic adaptive response to the probability of significant summer rainfall. However, Jain (1982) also found that heritability for dormancy was low for some species and that phenotypic plasticity in germination was also important. A study by Rice (1985) on germination cueing in Erodium brachycarpum and E. botrys demonstrated that entirely plastic responses in seed germination had significant fitness consequences in terms of both survival and reproduction. He found that Erodium (storksbill or filaree) seeds germinated more readily under summer soil temperature regimes characterized by large diurnal fluctuations in temperature. This type of temperature environment is usually associated with reduced litter cover and grass density, which, in turn, represents a favorable site for Erodium survival, growth, and reproduction. This type of germination cueing represents an adaptively plastic response in germination behavior that has been documented in other habitats characterized by spatial or temporal variability in “windows of opportunity” for seedling recruitment (Venable and Brown 1988). In addition to within-generation plastic responses in germination, there can also be “trans-generational” plasticity, in which the environment of the maternal plant can influence the germination behavior of progeny. Although the potential adaptive nature of these maternal environmental effects is just beginning to be explored, it is clear that they can often exert significant effects on progeny fitness, especially at early stages of development (Mousseau and Fox 1998). An example from the California grassland is provided by a study on how maternal effects in Nemophila menziesii (baby blue-eyes) can influence seed characters and, in turn, the fitness of the progeny (Platenkamp and Shaw 1993). They found that the maternal competitive regime had a strong influence on all the seed characteristics that they studied. Seed weight was reduced when the maternal plant was subjected to strong competition. In addition, seed dormancy and time to germination were increased in progeny from mothers that experienced competition. Delayed emergence time has been shown repeatedly to result in a competitive disadvantage to the seedlings that emerge later (Miller 1987; Rice and Dyer 2001; Ross and Harper 1972). Additionally, increases in seed dormancy can effectively lengthen the generation time of a population, which, in turn, can reduce fitness by decreasing rates of population increase (Lewontin 1965; Rice 1989b). The authors noted that the influence of the maternal competitive environment on all these seed traits represented a crossgeneration extension of the negative effects of competition. Finally, recent interest in the evolution of plasticity (Via et al. 1995) has highlighted the fact that phenotypic plasticity, like any trait, can evolve if there is sufficient genetic variation for plasticity within the population. One of the earliest studies on genetic variation in plasticity in plants was conducted by Jain (1978), examining inheritance of phenotypic plasticity in Bromus hordeaceus. Using a parent-progeny regression

142

ECOLOGICAL INTERACTIONS

approach, he found significant genetic control of plasticity for two populations that differed in genetic polymorphism (as determined by allozyme markers). Surprisingly rare, these types of studies that demonstrate the heritability of plasticity are critical for arguing the evolution of alternative strategies of adaptation to environmental heterogeneity.

Adaptation to Abiotic Gradients California, with its great diversity in climate and soil types, provides a mosaic of conditions under which plants must thrive. As reviewed in previous paragraphs, some species use adaptive phenotypic plasticity to perform well in a variety of environments, whereas others depend on extremely localized adaptation to microenvironmental variability. Broader abiotic gradients have also been the subject of some research on ecotypic variation in California plants.

Serpentine Soil Gradients Plants that grow on serpentine soil, in particular, have attracted much attention because of the harsh soil conditions (low in calcium and high in toxic metals) and high biodiversity on these soils. Serpentine soils can occur as islands of varying sizes within a matrix of more hospitable soils (see Harrison and Viers, Chapter 12). Early work by Kruckeberg (1951) demonstrated serpentine tolerance within races of Streptanthus (jewelweed, subspecies glandulosus secundus and glandulosus typicus) in container experiments conducted outdoors on populations of seeds collected from serpentine and nonserpentine habitats. Soil type did not affect seed germination but did affect growth in these species. In the same experiments, Gilia capitata segregated into serpentine-tolerant and -intolerant races even at the germination stage. In these experiments, Gilia capitata and Achillea borealis (yarrow, or, milfoil) were grown on both serpentine and nonserpentine substrates (the Streptanthus was grown only on serpentine). Serpentine-tolerant races of G. capitata had better germination and growth in nonserpentine soil than on their home soil. Similar results were achieved for A. borealis, and some nonserpentine strains showed partial tolerance to serpentine soils. These experiments tested plant-soil relationships to germination and growth and did not test adaptation in situ or in competitive environments. In a study comparing sandstone and serpentine ecotypes of Bromus hordeaceus, Frietas and Mooney (1996) found that serpentine ecotypes performed better under water stress, and both ecotypes performed poorly on sandstone soils when stressed for water. As mentioned previously, work by Jurjavcic et al. (2002) found serpentine adaptation at the seed germination stage on rocky substrates in Vulpia microstachys. Gradients can also occur within the serpentine soil type: Lasthenia (goldfields, an outcrossing genus) has evolved in a dichotomous manner (to mesic, ionically challenging soils, or to xeric, benign soils) on serpentine soils multiple times

(Rajakaruna et al. 2003a). Within each subspecies tested (L. californica ssp. californica and L. californica ssp. gracilis), races with differing flavonoid compounds correlated with each soil moisture type were found. The flavonoid compounds did not correspond with phylogenies created with nuclear ribosomal ITS regions (Rajakaruna and Bohm 1999; Rajakaruna et al. 2003c), indicating that the flavonoid compounds corresponding to soil type have arisen multiple times during adaptation to soil type. These two races (mesic and xeric) are found in close proximity at Jasper Ridge, and thus are an example of highly localized adaptation in an outcrossing species where low rates of selfing reduce the chance for the formation of co-adapted gene complexes. It is interesting to note that although flavonoid compound signatures were specific to soil type, and each drought-adapted race did have higher fecundity in its own soil type than the mesic soil type (Rajakaruna et al. 2003a), no direct, mechanistic explanation has been provided for a relationship between flavonoid compound signature and adaptation to either drought stress or soil fertility.

Coastal vs. Inland Climatic Regimes The coast, with its moderate temperatures and summer moisture in the form of fog, is very different from the inland areas of California with hot, dry summers and colder winters, and this abiotic gradient has also been the subject of the study of ecotypic variation in California grassland species. Coastal and inland ecotypes have been found in Gilia capitata, Bromus hordeaceus, and Eschscholzia californica (California poppy) by performing reciprocal transplant experiments (Nagy and Rice 1997; Knowles 1943; Leger 2004). Coastal forms of E. californica tend to be prostrate and perennial with yellow flowers, whereas inland forms can be a mix of perennial and annual plants with flower color ranging from yellow to orange. These differences between forms persist when they are grown in common garden environments (Leger 2004). It is likely that many factors contribute to the evolution of coastal and inland ecotypes of a species, because many abiotic factors differ between the coastal and inland environments in California. For example, water may be more available in the summer on the coast via fog compared to inland (Corbin et al. 2005). In addition, the coast is windier than inland, encouraging more prostrate growth forms. Biotic factors, such as pollinator availability, that are influenced by abiotic factors are also likely to come into play when assessing the adaptive nature of the coastal vs. inland ecotype.

Adaptation to Biotic Selection Evolution of Competitive Interactions As has been true for several other research areas in population genetics, the first studies of the evolution of competition were conducted on agronomic relatives of California

grassland species. Allard and Adams (1969) used a composite cross of four barley varieties initiated by Suneson (1956) to examine changes in the frequency of each variety in response to intergenotypic competition. They found that after 18 generations of mass propagation, selection appeared to preserve genotypes that interacted in a synergistic fashion. They argued that this type of synergistic behavior is not predicted from genotype performance in pure stands and may be an important factor in maintaining genetic diversity within populations that contain mixtures of genotypes. The role of genetic polymorphism in affecting the outcome of competition as well as the coevolutionary history of competitors was examined for the introduced oat species Avena fatua and A. barbata (Yazdi-Samadi and Jain 1978). Collecting material from both mixed and pure stands of each species, they set up competition trials using populations with high and low levels of polymorphism. In general, higher levels of polymorphism favored either species when in competition with a monomorphic population of the other species; competition between monomorphic populations was not as severe. The authors noted that the experiments indicated a higher probability of coexistence between monomorphic populations and that this was consistent with field observations, because most mixed stands of these species are composed of populations with low genetic variability. They also noted that there was low stability and more unpredictable competitive outcomes in mixtures of populations collected from pure stands in the field. They suggested that less competitive determinism in allopatric populations may reflect a lack of coadaptation and coevolutionary history between competitors. Further evidence for the potential importance of coevolution of competitors was provided by Martin and Harding (1981) in a study on competition between allopatric and sympatric populations of Erodium cicutarium and E. brachycarpum (storksbill or filaree). In addition to competitive mixtures containing allopatric and sympatric populations of each species, they also examined mixtures of transposed populations. Transposed mixtures were created by mixing populations from sympatric sites. For example, in their collection of Erodium from various field sites, Martin and Harding (1981) collected material from two sympatric sites: Tehachapi and Turtle Rock. To create transposed mixtures, they planted E. cicutarium from Tehachapi with E. brachycarpum from Turtle Rock and vice versa. Results strongly supported the importance of coevolutionary history in predicting outcomes of interspecific competition in Erodium: Reproductive rates of sympatric mixtures were consistently higher than those of mixtures of allopatric populations and suggested evolution of resource partitioning. Further, reproductive output from transposed mixtures indicated that two of the three tested populations of E. brachycarpum responded specifically to the gene pool of its sympatric interspecific competitor.

G E N E S O N T H E R A N G E : P O P U L AT I O N G E N E T I C S

143

As a counterpoint to these studies indicating substantial selective response to competition, recall that, as noted earlier, Platenkamp and Foin (1990) found no selective response in Anthoxanthum odoratum to variation in its competitive environment. In a series of neighborhood manipulations, they examined response of A. odoratum genotypes to intraspecific competition (allopatric and sympatric populations) as well as interspecific competition from Holcus lanatus. Reproductive output of A. odoratum decreased with competition, but this reduction in reproduction did not differ among plants from different competitive neighborhoods; the increase in fitness appeared to be entirely caused by phenotypic plasticity.

Other Biotic Interactions as Agents of Selection Direct interactions between plants certainly drive evolution, but indirect competitive interactions such as differential response to disease (Lawrence 1945) and to local pollinator activity (Horovitz and Harding 1972) may also serve to differentiate populations. Tests on Deschampsia caespitosa (Lawrence 1945) showed that resistance to rust varied by the seed source population and also by the location in which the plants were grown. Plants collected from Sierran meadows died from rust when grown at Stanford University, but all California collections were rust-resistant when grown at Mather field and at a high-elevation location. In the case of Lupinus (lupine; Horovitz and Harding 1972), differences among populations were found in blue reflectance, honey guide patterns, flower length, and number of pollen grains per flower. Differences in outcrossing rates among these populations were positively correlated with low pollen production, but flower size and honey guides were unrelated to the degree of outcrossing found within populations. Soil microbial communities are another biotic factor that may exert selective pressure on populations. Although this biotic factor on tree population dynamics is beginning to receive some scientific attention, the research investigating soil microbial community as a selective influence on California’s grassland plants has yet to be conducted (but see Batten 2004).

144

ECOLOGICAL INTERACTIONS

Conclusions Many of the important contributions to basic theory of plant population genetics were developed in California using species collected within the state. Important concepts, such as co-adapted gene complexes and speed and spatial scale of adaptation, were examined. Given the agronomic background of the scientists studying these concepts, this seminal work was primarily performed using convenient exotic species (primarily annuals). These exotic species were mostly invasive, and the research has contributed much to global scientific knowledge about the basic population genetics of plant invasions. Most population genetic studies in California grasslands have not focused on native species. Thus, genetic aspects of persistence in a fragmenting habitat and species cohesion within a dynamic landscape have not been examined. Studies that have observed adaptation have generally focused on abiotic selective factors, and there is a paucity of work on how ecology (particularly interspecific interactions) may drive evolution in California grasslands. Nevertheless, of all the North American systems under study, California’s grasslands have probably had the most (but still sparse) attention paid to their competitive interactions and how they may drive evolution. There is still little understanding of important genetic processes for native species used for restoration. It is clear that populations within a species can be genetically different from one another in ways that might be important, and research has yet to determine how genetic differentiation affects the short- and long term success of restored populations. Restoration projects successful over the long term are few and typically use only a handful of individual species. It has yet to be determined how population genetics affects the persistence of native populations within the California grasslands, as many of these populations are prone to decline through land conversion and further habitat fragmentation.

Acknowledgments We would like to thank David Ackerly, Jeff Corbin, Andy Dyer, and Ellen Simms for helpful comments on the manuscript.

T W E LV E

Serpentine Grasslands S U SAN P. HAR R I S O N AN D J O S H UA H. VI E R S

California’s serpentine grasslands are well known for their brilliant spring displays of wildflowers including Lasthenia californica (goldfields), Layia platyglossa (tidy-tips), Gilia tricolor (birdseye gilia), Eschscholzia californica (California poppies), and other native annual forbs. Their harsh, nutrient-poor soils have long been viewed as refuges for native grassland species that have largely disappeared elsewhere under the onslaught of exotic annual grasses such as Avena, Bromus, and Lolium spp. and land use change (Heady 1958, 1988; Kruckeberg 1984). Because of their relatively high complement of native species, it has been suggested that serpentine grasslands are our best model system for understanding native Californian grasslands in general (Murphy and Ehrlich 1989). Their paradoxical native-dominated status, however, raises the question, “If they are so different from other grasslands now, then how similar were they in the first place?” Like so many aspects of native Californian grassland ecology, the question is an exercise in historical guesswork. Regardless of how representative they are of other grasslands, serpentine grasslands are a source of great fascination to native plant enthusiasts, entomologists, ecologists, and other scientists and naturalists. They are the subject of intensive conservation and restoration efforts, especially in the San Francisco Bay area. This chapter provides an overview of the state’s serpentine grasslands, including their ecological peculiarities as well as the features that they may once have shared with other, now vanished, grassland ecosystems. This chapter draws heavily upon ecological studies that have been conducted in the serpentine grasslands at university field stations: the University of California’s McLaughlin Reserve (Napa, Lake, and Yolo Counties) and Sedgwick Reserve (Santa Barbara County), Stanford University’s Jasper Ridge Biological Preserve (San Mateo County), and the nearby Coyote Ridge study site (Santa Clara County).

What Is Serpentine? Geology and Soils The mantle of the earth is made up of rock that is richer in magnesium and iron, but poorer in silica, calcium, potassium, and phosphorus, than the rocks that make up the bulk of the earth’s continental crust. This dense material, known as peridotite, is seldom exposed on the terrestrial surface. However, tectonic plate convergence and subduction, or the collision and disappearance of one plate beneath another, typically lead to the formation of mountains in which rocks of the oceanic crust and upper mantle are uplifted and exposed. In California there are discontinuous bands of mantle-derived rock in the Klamath Mountains, the north and south Coastal Ranges, and the Sierra Nevada foothills. The occurrence of these rocks marks the zones of the ancient (Cretaceous and Jurassic) collisions between the western edge of North America and a succession of smaller tectonic plates that assembled California (McPhee 1994). The term serpentine technically refers to a mineral produced by the metamorphosis of peridotite under high pressure and the presence of water, conditions that occur during plate convergence and subduction. The mineral was given its name because of the slippery, green, snakeskin-like appearance of serpentinite, rock composed of serpentine. Ecologists use the term serpentine more broadly to include peridotite, serpentinite, and related ultramafic (high in magnesium and iron) rocks, plus the soils formed from these rocks. Technically, socalled serpentine exposures in California often include large amounts of peridotite and partly serpentinized peridotite as well as serpentinite. Throughout the world, serpentine rocks and soils are conspicuous for the stunted, sparse, distinctive vegetation they support. The ultimate explanation for the infertility of serpentine is that terrestrial life evolved on the earth’s continental crust and thus is adapted to higher levels of Ca, N, P, and K and lower levels of Mg than are found in the soils

145

formed from the mantle. Many or most plant species in any given region (serpentine avoiders) cannot grow on serpentine, and those that grow both on and off of it (serpentine tolerators) are often found to consist of differentiated races, with the serpentine races showing specialized adaptations related to ion uptake and sequestration. In some parts of the world there are numerous serpentine endemics, or species restricted to these soils. Cuba, New Caledonia, the Mediterranean region, and California are among the world’s hotspots of serpentine endemism, with several hundred species unique to serpentine in each of these locations (Kruckeberg 1984; Brooks 1987; Baker et al. 1992; Roberts and Proctor 1992; Safford et al. 2005).

Reasons for Exclusion The relative amount of calcium and magnesium is widely regarded as the most defining property of serpentine soils, with Ca:Mg ratios typically being less than 1:1 in contrast to the ratios much greater than 2:1 found on more typical soils. Serpentine soils are consistently Mg-rich, and high Mg levels reduce the availability of Ca to plants. Furthermore, the effects of excess Mg on plants are most severe when Ca levels are low. Nutrient amendment studies have demonstrated strong positive effects of Ca addition on the growth of both adapted and nonadapted plant species on serpentine (Brooks 1987). Differential performance under variable Ca and Mg treatments often distinguishes serpentine-tolerant from -intolerant species or ecotypes (Walker 1954; Walker et al. 1955 Madhok and Walker 1969; MacNair and Gardner 1998). Natural variation in Ca:Mg is often a strong predictor of compositional changes within serpentine vegetation, such as the prevalence of exotic versus native species in Californian serpentine grasslands (McCarten 1992; Armstrong and Huenneke 1992; Harrison 1999b). Nutrient (nitrogen, potassium, and phosphorus; NPK) deficiencies, metal (especially Ni) toxicity, and low water availability have also been found to play important roles in excluding plants from serpentine soils. Interestingly, although few studies have found that simple NPK addition can reverse the infertility of serpentine, all of the exceptions come from serpentine grasslands in the San Francisco Bay area. Turitzin (1982) found that N and P, but not Ca, limited the growth of the exotic annual grass Bromus hordeaceus and the native annual grass Vulpia microstachys in serpentine grassland soils at Jasper Ridge. Huenneke et al (1990) found that addition of N and P at Coyote Ridge caused the exotic grasses Bromus hordeaceus and Lolium multiflorum to invade, total biomass to increase, and diversity to decrease. Recent evidence suggests that the increasing invasion of serpentine grasslands in the Bay area by annual grasses, especially Lolium, is caused by atmospheric N deposition from automobile emissions (Weiss 1999, 2003).

Adaptation to Serpentine The precise mechanisms that have been found to allow plants to tolerate the unusual chemistry of serpentine soils differ from species to species (see Brady et al. 2005 for a recent

146

ECOLOGICAL INTERACTIONS

review). Many species are divided into races, or ecotypes, that vary in their degree of serpentine tolerance (Kruckeberg 1951, 1967). Classic work by Walker (1954) (Walker et al. 1955) showed that selective Ca uptake enabled serpentineadapted sunflowers to grow in extremely Ca-poor solutions. Tolerance may also relate to the ability to mobilize nutrients in a Ca-poor or metal-rich environment; the activity of root phosphatase, which functions to make P available to the plant, was highest at low Ca concentrations in serpentinetolerant grasses, whereas in nontolerant species this enzyme was inhibited by low Ca (Willett and Batey 1977). In Festuca rubra, root phosphatase activity was differentially sensitive to Mg and Ni concentrations in serpentine-adapted and nonadapted populations (Johnston and Proctor 1981, 1984). In other ecotypes or species, tolerance of serpentine has been attributed to the ability to exclude Mg and Ni or to accumulate them in a nontoxic form. For example, serpentine populations of Festuca rubra maintained lower internal Ni concentrations than did their nonadapted conspecifics when grown in solutions with high levels of Ni (Nagy and Proctor 1997a, b). Hyperaccumulation of Ni has been found in many tropical serpentine species, but only in Thlaspi montanum and Streptanthus polygaloides in North America (Brooks 1987). Adaptation to water stress is another facet of serpentine tolerance. Neighboring serpentine and sandstone populations of the exotic annual grass Bromus hordeaceus showed ecotypic differences in responses to water stress (Freitas and Mooney 1996). The native annual grass Vulpia microstachys showed ecotypic differentiation between populations on rocky serpentine outcrops and those in grasslands on either serpentine or nonserpentine soil, suggesting that the serpentine ecotype is actually adapted to rocky conditions (e.g., low soil water capacity) rather than to a low Ca:Mg ratio or other characteristics specific to serpentine (Jurjavcic et al. 2002). An even more complex case of ongoing evolution in response to serpentine soil exposure is the common native annual forb Lasthenia californica (goldfields). This species is divided into ecotypes one of which occurs on wet clay-rich soils (“Race A”) and the other on shallow rocky soils (“Race C”) within central Californian serpentine habitats (Rajakaruna and Bohm 1999; Rajakaruna et al. 2003). Race C flowers 7–10 days earlier than Race A. The two races have distinct biochemical and morphological characters and are not completely interfertile. Race A is also found on nonserpentine clay soils with high Mg and Na content, ranging from central California to northwestern Mexico, while race C is found on soils with higher Ca and K and ranges from southern Oregon to central California. Both ecotypes appear to selectively exclude excess Ni, but Race A tolerates high foliar Mg and Na, while Race C maintains constant foliar levels of these elements. The sulfated flavonoids that distinguish Race A may play some role in ion uptake and sequestration (Rajakaruna and Bohm 1999; Rajakaruna et al. 2003). The existence of cryptic species (L. californica and L. gracilis), each of which may have similar A and C races, complicates the story still further (Rajakaruna et al. 2003).

Adaptation to serpentine may be rapid, as shown by studies of Bromus hordeaceus (Freitas and Mooney 1996) and Avena fatua (Harrison et al. 2001). Both of these Mediterranean annual grasses were introduced to California within the past several centuries (D’Antonio et al., Chapter 6). Serpentine populations of both species performed better on serpentine than did nonserpentine populations. Grown in a common, nonserpentine soil, Avena from serpentine populations produced more but smaller seeds and had higher root-to-shoot ratios than did their nonserpentine conspecifics (Harrison et al. 2001). Molecular studies may soon be able to identify particular genes responsible for serpentine adaptation (Bradshaw 2005), which in turn could yield exciting new insights about the classic question of how serpentine endemic species originate (Stebbins 1942; Kruckeberg 1954, 1957; MacNair and Gardner 1998; Brady et al. 2005).

Endemism and Rarity California’s serpentine endemic plants have been enumerated by Kruckeberg (1984) and Safford et al. (2005), both of whom arrive at figures of over 200 species or subspecies that are strongly restricted and another 200 – 400 that are weakly restricted to the substrate. All of these taxa are endemic to California as well as to serpentine. Most of them are found in the Klamath Mountains and northern Coast Ranges, and are found in woodlands, chaparral, or barrens communities rather than in grasslands. Like Californian plant diversity in general, endemic diversity is positively associated with the state’s strong south-to-north gradient of increasing rainfall (Harrison et al. 2001, 2006a– c). The majority of Californian serpentine endemics are herbaceous dicots (e.g., Clarkia, Cordylanthus, Lessingia, Linanthus, Mimulus, Navarretia, Streptanthus spp.), geophytes (e.g., Allium, Calochortus spp.), or sedges (Carex spp.), although there are some serpentine endemic shrubs and trees (e.g., Arctostaphylos, Ceanothus, Quercus, Cupressus spp.), a few grasses (e.g., Calamagrostis ophiditis), and several ferns. Many or most of the herbaceous endemics belong to certain genera that have their centers of diversity in the state and are considered classic examples of neoendemism — that is, groups that diversified rapidly in California since the present Mediterranean climate began to develop in the middle Pliocene (Raven and Axelrod 1978). Serpentine endemics typically have small geographic ranges and can be considered naturally rare (Harrison and Inouye 2002). Of the 1742 designated rare, uncommon, or endangered plants in California, 612 are associated with special substrates: 282 with serpentine, 103 with granite, 89 with carbonate rocks (limestone and dolomite), 77 with volcanics, and 61 with alkaline soils (Skinner and Pavlik 1994). Of the 246 serpentine-endemic taxa tabulated by Safford et al. (2005), fully 194 are rated by the California Native Plant Society’s rare plant inventory (Skinner and Pavlik 1994) as “rare or uncommon,” and 111 of these are state and/or federally listed species.

Rare and endangered species in Californian serpentine grasslands are almost exclusively confined to the San Francisco Bay area (USFWS 1998). This is because other parts of the state either have no serpentine (Central Valley, deserts), or little serpentine grassland (Klamath Range), or the grasslands have very few serpentine endemics (Sierra Nevada foothills, southern Coast Range); or there are serpentine grasslands that contain some endemics, but these habitats are not under intensive pressure from urbanization (northern Coast Range, e.g., Lake and Napa Counties). The rare and endangered species of serpentine grasslands are discussed in a later section.

Distribution and Composition of Serpentine Grassland Serpentine Vegetation in California Grasslands comprise only a small fraction of the vegetation on serpentine in California. Conifer woodland interspersed with open rocky barrens dominates the endemic-rich serpentine and peridotite exposures of the Klamath Mountains, as well as the northern Sierra Nevada. Chaparral, often a rich mixture of many codominant shrub species, is the most prevalent vegetation type on serpentine in much of the Coast Ranges and the Sierra Nevada foothills. Moving from north to south, grasslands appear on serpentine in the northern Coast Ranges, such as Mendocino, Napa, and Lake Counties, usually on small islands of alluvial or colluvial serpentine soil within more widespread serpentine chaparral or woodland. Grasslands abruptly become prevalent on serpentine in the North Bay, such as in Marin, southern Sonoma, and southern Napa Counties. In the San Francisco Bay area as a whole, over 60% of the vegetation on serpentine is grassland, consisting of a mixture of native perennial bunchgrasses, native forbs (mostly annuals), and exotic species (mostly annual grasses) (USFWS 1998). Rather than being confined to shallow slopes and alluvial valley bottoms, as is the case farther north, serpentine grasslands from the Bay area southward are usually found on steep slopes or ridges with abundant protruding rocks. Grasslands are also prevalent on serpentine soils in the dry interior parts of the southern Coast Ranges, although the more coastal and higher-elevation serpentines of the south coast usually support chaparral or woodland. Some of the southernmost serpentine grasslands occur at Vandenberg Air Force Base and the University of California’s Sedgwick Reserve in Santa Barbara County, where the geographic distribution of serpentine in California reaches its southern limit. Southern Californian serpentine grasslands resemble those of the San Francisco Bay area in being found on steep slopes with many protruding rocks. They are dominated by exotic annual grasses and contain few rare or endemic species. Moving southward in the Sierra Nevada, serpentine vegetation goes from conifer woodland to shrubland and then,

SERPENTINE GRASSLANDS

147

F I G U R E 12.1. Serpentine geology and serpentine grasslands in California. Courtesy of

J. H. Viers.

from about Mariposa County south, consists largely of grassland at low elevations. The southern Sierran serpentine grasslands are even more exotic-dominated and endemic-poor than those of the southern Coast Ranges. However, there are occasional stunning displays of native wildflowers (e.g., Clarkia, Calochortus) on the foothill serpentines, even as far south as Fresno County (Latimer 1984). Intersecting the state geologic map (Jennings 1977) with the state vegetation map (California Land Cover Mapping and Monitoring Program 2003) it is possible to obtain an approximate map of serpentine grasslands in California (Figure 12.1). Even though this map represents serpentine grasslands as points, it necessarily exaggerates their true

148

ECOLOGICAL INTERACTIONS

area in order to make them visible. Also, some inaccuracies are inevitable given the coarse resolution of both input maps. In particular, many areas of sparsely vegetated serpentine “barrens,” such as the huge New Idria barrens of San Benito County, are classed as serpentine grasslands on this map. Nonetheless, the map gives a rough depiction of the geographic distribution both of serpentine and of grasslands within serpentine. This map analysis indicated that of the 5752 square km of serpentine in the state, 7% (or 417 square km) is annual grassland, and another 4% is blue oak (or blue oak and gray pine) woodland, which usually has a substantial grassland understory. Coniferous forests comprise 54% of serpentine vegetation, while chaparral covers 22%. The remainder is

urban (1%), agricultural (1%), or a variety of minor vegetation types (e.g., montane hardwood, 7%). BOX 12.1 WHERE TO SEE SERPENTINE GRASSLANDS

Serpentine Grassland Composition Grasslands on serpentine are best known not for their grasses, nor even for their serpentine endemics, but for their dense displays of native (nonendemic) annual forbs; grassland may thus be something of a misnomer, and some authors prefer meadow or prairie (see Keeler-Wolf et al., Chapter 3; Schiffman, Chapter 4). Typical serpentine grasslands in the Coast Ranges feature dense early-spring displays of Lasthenia californica mixed with dozens of other native annual forbs such as Layia, Gilia, Linanthus, Microseris, and Eschscholzia. Native bunchgrasses, especially Nassella pulchra and Poa secunda, are present but seldom abundant. Later in the season, exotic annual grasses (e.g., Avena barbata, Bromus hordeaceus, Taeniatherum caput-medusae) become dominant, but late-season native forbs such as Calochortus, Clarkia, and Navarretia spp. are interspersed, as is the native annual grass Vulpia microstachys. Bunchgrasses such as Nassella pulchra may appear in almost pure stands on serpentine ridges in the coastal zone (Fiedler and Leidy 1987; Heady 1988). Stands dominated by Nassella lepida are found on serpentine in the foothills of the Sierra Nevada and interior Coast Ranges (Sawyer and KeelerWolf 1995). Festuca californica may be abundant in shaded understory situations, and Elymus elymoides is common at the edges of chaparral. The exotic grass Bromus hordeaceus (soft chess) is found in virtually all serpentine grasslands and is often among the most abundant species, especially in southern California. Stands of exotic Lolium multiflorum (Italian ryegrass) are prevalent on deep serpentine-derived clay soils and in coastal environments. Dense carpets of exotic Aegilops triuncialis (barbed goatgrass) appear to be spreading rapidly in the central and southern North Coast Range serpentine grasslands. In a quasi-random sampling of 109 serpentine localities around the state, Harrison et al. (2006c; and unpublished data) found that almost none of the 25 sites that were grassland had more than 10% cover by native grasses. The only native grasses that ever comprised a substantial fraction of total cover were Nassella pulchra, Elymus multisetus or elymoides, and Vulpia microstachys. In addition, Poa secunda, Melica californica, and Melica imperfecta were widespread but never very abundant. Exotic annual grasses were found in all grassland sites and often reached more than 50% of total cover toward the end of spring. Bromus hordeaceus was by far the most common exotic grass, while other widespread species included Avena barbata, Lolium multiflorum, Bromus madritensis, Bromus diandrus, and Vulpia myuros. In even the most detailed vegetation maps at the regional or county scale, “serpentine grassland” is typically treated as a single unit, and its distribution is obtained by overlaying soil maps with coarse vegetation maps as demonstrated in this chapter. Several efforts are now underway to create more

Some sites where interested grassland ecologists and members of the public can visit excellent examples of serpentine grasslands include Edgewood Natural Preserve, San Mateo County; Inspiration Point, San Francisco Presidio National Park; Serpentine Prairie, Redwood County Park, Alameda County; Ring Mountain Preserve, Marin County Open Space District; Missimer Snell Valley Preserve, Land Trust of Napa County; and the Bear Valley Conservation Area, Colusa County. In addition, the Sedgwick, McLaughlin, and Jasper Ridge university reserves all have docent-led hikes and other public programs. For more information, see http://nrs.ucop.edu/reserves/sedgwick/moreinfo.html (Sedgwick), http://nrs.ucdavis.edu/mclaughlin.html (McLaughlin), and http://jasper1.stanford.edu/home/ (Jasper Ridge).

detailed classifications of serpentine grasslands in California based on either “indicator” species (Rodriguez-Rojo et al. 2001b) or dominant species (Sawyer and Keeler-Wolf 2007).

Ecological Patterns and Processes Slopes, Soil Depth, and Species Composition One of the first ecologists to become fascinated by the distinctive features of serpentine grassland was McNaughton (1968), who established that this community was both more diverse and less productive than the adjacent nonserpentine grassland at Stanford University’s Jasper Ridge. Along a topographic gradient from cool, moist northern to warm, dry southern exposures, he found that dominance shifted from Nassella pulchra to Bromus hordeaceus in the serpentine grasslands and from Avena fatua to Bromus diandrus and B. hordeaceus in the nonserpentine grasslands. Overall, the serpentine grasslands showed both much less dominance by exotics as a group and much less dominance by any single species. McNaughton (1968) speculated that lower productivity and dominance, together with fine-scale heterogeneity in soil depth and microtopography, led to the higher diversity of serpentine grassland. Within San Francisco Bay area serpentine grasslands as a whole, McCarten (1992) similarly found that slope, aspect, and soil depth caused considerable small-scale variation in species composition. Deep soils with high moisture content sometimes support stands of such bunchgrasses as Calamagrostis ophiditis, Elymus glaucus, Festuca idahoensis, Elymus

SERPENTINE GRASSLANDS

149

elymoides, Koeleria macrantha, Poa secunda, and Nassella pulchra. However, deeper soils are also more invaded by exotic annual grasses. Shallower, rockier soils support forb-dominated “wildflower fields” with Plantago erecta, Lasthenia californica, Minuartia californica, Microseris douglasii, Layia platyglossa, and Hemizonia congesta. Outcrops of exposed rock, embedded within deeper soils, support many of the characteristic endemic and rare species of Bay area serpentine grasslands (USFWS 1998; also see following paragraphs). In the serpentine grasslands of the McLaughlin Reserve in northern California, Harrison (1999b) found that the Ca:Mg ratio was the best correlate of the percentage of native species, with greater native dominance at low Ca:Mg values, although the level of P was a better predictor of total biomass. Slope was not important, in contrast to neighboring nonserpentine grasslands, where the percentage of native species was higher on cool than warm slopes. Very similar patterns were found in later work in the same region (Safford and Harrison 2001; Gelbard and Harrison 2003). Although these studies did not measure soil depth directly, they found that the percentage of native species in serpentine grasslands was considerably higher on hillsides, where soils are shallow and rocky, than on flatter valley-bottom sites, where soils are deeper and more clay-rich (Gelbard and Harrison 2003). In southern California’s Sedgwick Reserve, Gram et al. (2004) showed there were strong compositional gradients associated with hummocks of shallow rocky serpentine within a matrix of deeper grassland soils, even though the Ca:Mg ratios of hummocks and matrix did not differ. Of the 27 most common plant species sampled along hummock-to-matrix transects, eight were hummock specialists and eight were matrix specialists, as well as seven edge specialists and four generalists. The hummock specialists included Chaenactis glabriuscula, Eriogonum fasciculatum, Gilia achilleifolia, Lotus strigosus, Melica imperfecta, Minuartia douglasi, and Poa secunda. Matrix specialists included Nassella pulchra, Hordeum murinum, and Bromus hordeaceus. Edge specialists included both Lasthenia californica and Plantago erecta, illustrating that some characteristic serpentine grassland species can tolerate neither the harshness of rocky serpentine outcrops nor the tall stature of grasslands on deep soils, but require intermediate conditions.

Resource Partitioning and Competition Serpentine grassland species vary in their phenology, use of shallow and deep soil resources, and allocation to root versus shoot biomass; the latest-flowering annuals are the longest lived, attain the highest biomass, but have the lowest proportional reproductive effort, and tap into the deepest and most persistent water (Gulmon et al. 1983; Mooney et al. 1986; Gulmon 1992). Complementarity in resource use among functional groups in serpentine grasslands has been shown in diversity manipulation experiments (Hooper and Vitousek 1998). In their natural setting, serpentine grassland species segregate spatially along gradients of soil depth, water

150

ECOLOGICAL INTERACTIONS

availability, and nitrogen (Reynolds et al. 1997). Nonetheless, competition both within and between functional groups has been shown through competitor removal, cross-seeding, and pot or microcosm experiments (e.g., Moloney and Levin 1996; Reynolds et al 1997; Dukes 2002b). Plant communities on low-fertility substrates such as serpentine are sometimes thought to be less influenced by interspecific competition than those on more fertile soils, and several recent studies from the grasslands at the McLaughlin Reserve are consistent with this. Removal of competitors did not enhance the emergence, survival, growth, or reproduction of an experimentially planted native annual grass (Vulpia microstachys) in sparse grasslands on rocky serpentine outcrops, although it did so in more productive grasslands on deeper, presumably more fertile, serpentine and nonserpentine soils (Jurjavcic et al. 2002). These data support the hypothesis that at least some serpentine substrates are refuges from competition. In a second study at McLaughlin, the exotic bunchgrass Dactylis glomeratus was experimentally planted into sets of serpentine and nonserpentine plots that varied naturally in species richness. A competitor removal treatment was included across the natural richness gradient. Invasion success on non-serpentine soil was higher in plots with naturally higher species richness, but only if the plots were uncleared, suggesting that competition with the existing community increased with richness and reduced the success of this invader. By contrast, on serpentine soil, success of the same grass was poorer and was unaffected by richness unless the plots were cleared, in which case the invader performed better in plots that had had higher initial richness. This suggests that within serpentine, high species richness is an indicator of soils that are both more fertile and potentially more invasible if competition is suppressed (Williamson and Harrison 2002).

Annual Rainfall Species composition and abundances in the forb community of serpentine grasslands shows striking annual variation, and Gulmon (1992) found that this is due in part to germination patterns; nine annual forbs and one perennial grass responded highly variably in their germination to seasonal rainfall, although they responded little to soil or litter cover. Yearly rainfall patterns also affect forbs indirectly by affecting the degree of dominance by exotic annual grasses. In the serpentine grasslands at Jasper Ridge, several years of aboveaverage rainfall led to an increase in Bromus hordeaceus and shifted the dominance among native annuals from Plantago erecta to Lasthenia californica (Hobbs and Mooney 1991). However, several years of drought reduced the abundance of Bromus hordeaceus even more on serpentine soil than on nonserpentine soil, probably because on serpentine soils it could not grow deep enough roots to survive drought (Armstrong and Huenneke 1992). Other exotics tolerated the drought well on both soils (e.g., Brachypodium distachyon) or were

equally negatively affected on both soils (e.g., Lolium multiflorum). These observations suggested, at least weakly, that the annual grass species best adapted to serpentine (e.g., Vulpia microstachys) are also best adapted to drought. Native bunchgrasses were little affected by the drought, and native geophytes (e.g., Dichelostemma capitata) responded with a one-year time lag.

Herbivory Few studies have addressed the role of natural herbivory in serpentine plant communities, perhaps in part because casual observation is not promising. Similar to plant communities on other unproductive soils, plants on serpentine are characterized by having small, tough leaves with resins, spines, and/or hairs. While these traits may confer resistance to herbivory, they are likely adaptations to nutrient and water stress that incidentally act to discourage herbivory. In one of the few existing studies, Hobbs and Mooney (1991) found that excluding aboveground herbivores had no effect on the plant species composition of the Jasper Ridge serpentine grassland. Some studies indicate that grazing by livestock (Jackson and Bartolome, Chapter 17; Huntsinger et al., Chapter 20) may have a beneficial role in the management of serpentine grasslands. Huenneke et al. (1990) noted a strong tendency of native annual forbs to disappear from ungrazed experimental plots, which were overtaken by both exotic annual and native perennial grasses, while annual forbs continued to be abundant in cattle-grazed areas outside these plots. Similarly, three studies at different sites in the McLaughlin Reserve region reached the conclusion that cattle grazing tended to increase the diversity of native annual forbs in grasslands on serpentine soils, while it increased the diversity of exotic annual forbs in the already exotic-dominated grasslands on adjacent non-serpentine soils (Safford and Harrison 2001; Gelbard and Harrison 2003; Harrison et al. 2003). The simplest explanation for these soil-specific effects of grazing is that low-statured native annual forbs, the most likely group of plants to benefit from moderate grazing, are abundant on serpentine and scarce on nonserpentine soils. Grazing benefits short-statured native forbs by reducing the dead material (thatch) produced by exotic annual grasses (Heady 1958, 1988). Positive grazing effects have also been seen in a study of native-rich coastal prairies (Hayes and Holl 2003a), while grazing has little effect on the composition of more typical, exotic-dominated Californian grasslands (Jackson and Bartolome 2002).

Fire In eastern North America, periodic fire appears to be necessary to maintain open serpentine “barrens” (grasslands), and fire suppression leads to the conversion of these barrens to pine forest (Tyndall and Hull 1999). No such tendency for fire to expand the occurrence of serpentine grasslands, nor for natural succession to convert serpentine grasslands or barrens

to woody vegetation, has been recorded in California. It remains possible that historic burning or grazing caused some permanent conversion of serpentine chaparral to grasslands, but this is completely unknown at present. Fire is an important tool for restoration and management of Californian serpentine grasslands threatened by exotic annual grasses (DiTomaso 2000; Seabloom et al. 2003a; Weiss 2003), as discussed subsequently. Unfortunately, prescribed burns (Reiner, Chapter 18) aimed at reducing the abundance of these grasses must take place in May or June, when the grasses have produced but not yet dropped their seeds, but the peak season for wildfires is in late summer and early fall. Prescribed fires in spring may be detrimental to certain lateseason annuals in serpentine grasslands, such as Navarretia, Clarkia, and Hesperolinon spp., because they flower and set seed at the same time as the exotic grasses. An arson-caused wildfire burned a mosaic of grasslands around the McLaughlin reserve in October 1999. Serpentine grasslands were more species-rich at the 1 m2 scale than nonserpentine grasslands before the fire, but species richness increased more in response to fire in the nonserpentine than in the serpentine grasslands. The richness of both native and exotic species increased in response to fire on both soils. However, like cattle grazing, fire mainly increased the richness of native species on serpentine soils and of exotic species on nonserpentine soils. Like moderate grazing, the main effect of an autumn fire may be to reduce thatch and thus to benefit low-statured forbs (Harrison et al. 2003). Two years after the fire, richness returned to approximately the same levels as before the fire.

Harvester Ants and Other Insects Seed-harvesting ants (Messor andrei) may be abundant in serpentine grasslands where they are favored by the relatively thin canopy and the sunny soil surface. At Jasper Ridge, Hobbs and Mooney (1985) found an average of one ant mound per 100 m2 (an area 10 m square), with feeding paths 10 – 12 m long. In ant exclusion experiments, they found increases in the abundances of Microseris douglassi and Agoseris heterophylla, two of the forbs whose seeds are selectively harvested by the ants. However, Brown and Human (1997) repeated the experiment and found no effects of ant exclusion on plant composition, even though mounds did tend to contain more grasses and fewer forbs than nonmound areas. Ant species diversity and relative abundances differed slightly between serpentine and nonserpentine grasslands at the McLaughlin Reserve, and these differences appeared to be more strongly linked to the soil itself than to plant community differences (Boulton et al. 2005). The soil microfauna of serpentine grasslands at the McLaughlin Reserve was affected by the activity of Messor andrei. Bacteria, fungi, nematodes, and microarthropods were more abundant in the soils immediately surrounding harvester ant nests than in other nearby soils. Experiments showed this was caused by the food material brought by the ants into their underground nests, which

SERPENTINE GRASSLANDS

151

led to higher soil nutrient levels (Boulton et al. 2003; Boulton and Amberman 2006). Wolf and Thorp (unpublished) found little difference between the bee communities of adjacent serpentine and nonserpentine grasslands at McLaughlin. The authors are not aware of any other studies to date that compare the abundance, identity, or importance of pollinators in serpentine versus nonserpentine grasslands.

Gopher Disturbance Fine-scale spatial patterning in species composition has been of longstanding interest to the researchers at the Jasper Ridge preserve, and early studies implicated an interaction between plant competition and the disturbance caused by the moundbuilding activity of gophers. Hobbs and Mooney (1985) showed that serpentine grassland plants differed both in their ability to colonize mounds and in their growth and survival on mounds. Mound soil was not as productive for Bromus or Plantago as other soil, partly because of lower P (Koide et al. 1987). Early results from a gopher exclusion experiment demonstrated that the mean turnover rate of soil by gophers was once every 3–5 years, although there was considerable spatial clumping of disturbance. Gopher disturbance increased the local abundance of several natives (e.g., Lotus wrangelianus) and exotics (e.g., Bromus hordeaceus) that were good colonizers of mounds, while adversely affecting bulb plants (e.g., Brodiaea spp.) and other native annuals (e.g., Lasthenia californica). Rainfall was also important; gopher mounds tended to be colonized by Bromus in wet years and by Lotus in dry years (Hobbs and Mooney 1991). Inspired by the empirical results, a series of modeling studies treated Jasper Ridge serpentine grasslands as an example of theory on the disturbance-mediated coexistence of competitors (e.g., Moloney et al. 1992; Wu and Levin 1994; Moloney and Levin 1996). These studies combined spatially explicit simulations and analytic modeling with field-measured competition, dispersal, survival, and fecundity parameters to determine how the properties of disturbance affected species abundances and patterning. For example, Moloney and Levin (1996) found that the relative abundances of Bromus hordeaceus, Calycadenia multiglandulosa, and Plantago erecta were sensitive to the frequency but not to the sizes of gopher disturbances. Later studies used remote sensing and spatial statistical analyses to gain a deeper understanding of spatial patterning in this system (Lobo et al 1998; Overton and Levin 2003). These studies indicated that the physical heterogeneity of the environment was a dominant influence. In particular, deeper soils had more perennial grasses, higher productivity, and more gopher mounds than shallower soils did, and soil depth itself had a natural periodic pattern in its spatial structure. Since soil depth was correlated with gopher disturbance, and both were correlated with abundances of common grassland species, it became less clear that gophers were the cause of spatial patterning.

152

ECOLOGICAL INTERACTIONS

Meanwhile, continuation of the gopher exclusion experiment called some of the earlier results into question (Hobbs and Mooney 1995). The short-term increases in perennials (Elymus, Brodiaea/Dichelostemma) in response to gopher exclusion were not sustained, possibly (in the case of Brodiaea) because of compensatory deer browsing. Also, there was little constancy in the colonization of gopher mounds, with different species (Calycadenia, Microseris, Vulpia, Lasthenia, Lotus) dominating in different years, and only Plantago acting as a consistent colonizer. The earlier correlations of Plantago and Lasthenia abundance with rainfall also broke down. Microseris increased dramatically (from 1–2% to over 70% cover), possibly because harvester ants declined. Hobbs and Mooney (1995) concluded that the earlier models had succeeded in simulating spatial patterns but were unable to predict temporal dynamics. A very different study of spatial patterning was conducted by Green et al. (2003), who used serpentine grassland at McLaughlin to test fractal theories of species distribution. According to this theory, one can predict empirical diversity patterns, such as the species-area relationship and the species-abundance distribution, by making the assumption of self-similarity in species distributions across spatial scales. Self-similarity means that if one divides any given sampling area in half, the proportion of species that are found in only one half may remain constant, regardless of the area (Harte et al. 1999). Serpentine grassland is ideal for testing these ideas because it is possible to find a moderately large number of species within a relatively small area and to count all individuals of these species, since few of them have a clonal growth habit. The results partly supported the theory but also indicated the need for further refinements (Green et al. 2003).

Ecosystem and Belowground Processes The role of mycorrhizae in serpentine grasslands has long interested researchers. Hopkins (1987) found that virtually all of 27 serpentine grassland plant species she examined at Jasper Ridge had vesicular-arbuscular mycorrhizae, and that two mycorrhizal species were dominant. Koide et al. (1988) showed that mycorrhizal removal suppressed plant growth and productivity. However, researchers studying a southeastern Pennsylvanian serpentine grassland showed that mycorrhizal species may either enhance or suppress the growth of the plant species that are their preferred hosts (Castelli and Casper 2003). In addition to mycorrhizae, rhizospheres (the soils immediately surrounding roots) host a diverse and poorly known microbial community that includes bacteria, fungi, and protozoa. Some of these play mutualistic roles in nutrient transformation and uptake, while others are pathogens. Rhizosphere microbial communities, as identified by biochemical markers, show some degree of variation among particular plant species. In the serpentine grasslands at the McLaughlin Reserve, Batten et al. (2006a) showed that the invasive species Centaurea solstitialis (yellow starthistle) and Aegilops triuncialis (barbed goatgrass) altered the rhizosphere

microbial community; additional evidence suggested that these changes may favor the growth of the exotics over that of native species. The Jasper Ridge grasslands have been the site of a carbon dioxide enrichment experiment as discussed in Chapter 19. Hungate et al. (1997b) found that CO2 enrichment in both serpentine and sandstone grasslands led to increased carbon uptake and increased allocation of carbon to roots. However, most of the carbon uptake went to root respiration and exudation, both of which led to rapid cycling back into the atmosphere. Rillig et al. (1999a) compared the effects of elevated CO2 on root length, mycorrhizae, and soil arthropods in serpentine and sandstone grasslands and found that root length increased in serpentine, while the growth of mycorrhizal hyphae increased in sandstone. The relatively low productivity and low carbon storage of Jasper Ridge grasslands has made them useful in the development of predictive modeling of ecosystem-level productivity. Valentini et al. (1995) measured atmospheric carbon exchange using the eddy covariance technique, and found that CO2 uptake was at the low end of reported values for grasslands. They also found that no more than 40% of incoming light is intercepted, which was also low for grasslands, but was explainable by the relatively sparse canopy cover. Using these and other measurements, they calculated an overall energy balance and accurately predicted annual net primary productivity. Diversity and ecosystem function have been another important research theme that has benefited from studies on serpentine. Hooper and Vitousek (1998) established experimental communities with different numbers of seasonal functional groups (i.e., early vs.. late-season species) at a reclaimed serpentine grassland at Coyote Ridge to test the effects of functional diversity on ecosystem-level nutrient dynamics. Serpentine grasslands are useful for this because they retain relatively small pools of organic matter in the biomass and soil and the small stature of the vegetation makes composition easy to manipulate. Increased functional group diversity led to higher relative use of available nutrients, but did not affect leaching losses of nitrogen, possibly because the rate of N loss was strongly affected by seasonality and by microbial immobilization rates (Hooper and Vitousek 1998).

Conservation of Serpentine Grasslands Rare Species Twenty-eight rare and endangered species of San Francisco Bay area serpentine soils are the subjects of a recovery plan by the U.S. Fish and Wildlife Service (1998). They include 13 federally listed plants, six plants of concern, one federally listed animal (a butterfly), and eight animals of concern (a moth and seven harvestmen). Most of the plants occur in grasslands (Clarkia franciscana, Castilleja affinis neglecta, Pentachaeta belliflora, Hesperolinon congestum) or on rocky outcrops within grasslands (Dudleya setchellii,

Streptanthus albidus albidus and peramoenus, S. niger, Lessingia arachnoides, L. micradenia micradenia and glabrata), although several occur in chaparral (Ceanothus ferrisiae, Arctostaphylos bakeri bakeri, Cordylanthus tenuis capillaris), two in wetlands (Cirsium fontinale fontinale and campylon), and one in a special soil habitat known as a serpentine Vertisol (Acanthomintha duttoni). The Bay checkerspot butterfly, Euphydryas editha bayensis (Nymphalidae), is now found on only a few serpentine grasslands in the San Francisco Bay area. This butterfly requires dense stands of its principal larval host plant, Plantago erecta, which is not a serpentine endemic but is seldom abundant on other soils. Its larvae may also feed on Castilleja densiflora. The population biology of this butterfly has been very extensively studied by Paul Ehrlich and colleagues at Stanford; Ehrlich and Hanski (2004) review this classic work. The butterfly is now in drastic decline and has gone locally extinct at Jasper Ridge and other Bay Area sites. Besides outright habitat conversion, the main threat appears to be the increasing prevalence of exotic grasses, especially Lolium multiflorum, in the remaining serpentine grasslands. Opler’s longhorn moth, Adela oplerella (Adelidae), is a lesser-known inhabitant of Bay area serpentine grasslands. It is a small brown moth with very long antennae and weak flying ability. Its larvae feed on Platystemon californicus, which, like Plantago erecta, is a small native forb that is not restricted to serpentine. Yet, like the butterfly, the moth is almost completely confined to serpentine, perhaps because of the higher density of host plants and/or more favorable microclimate provided by these soils. In addition to the San Francisco Peninsula, the moth is known from one location in Sonoma County, two (including the Ring Mountain Preserve) in Marin County, and one in Santa Cruz County (its only known nonserpentine occurrence). The moth appears to be in decline for substantially the same reasons as the butterfly (USFWS 1998). Seven species of harvestmen (daddy longlegs) are endemic to serpentine soils in the San Francisco Bay Area. They include the Marin and Edgewood blind harvestmen, Calicina diminua and C. minor, and the Edgewood, Hom’s, Jung’s, Fairmont, and Tiburon microblind harvestmen, Microcina edgewoodensis, M. homi, M. jungi, M. lumi, and M. tiburona. Most are known from fewer than 10 locations, and the entire genus Microcina is restricted to San Francisco Bay area serpentines. All are roughly one millimeter long as adults. Blind and microblind harvestmen favor microhabitats that are moist, warm, and totally dark. The adults live under large rocks and prey on springtails (primitive soil-dwelling insects). Eggs are laid directly in the soil, and adults are thought to become dormant underground during summer. In addition to habitat loss and degradation, these species are vulnerable to predation by the invasive Argentine ant Linepithema humile (USFWS 1998). The federally endangered Yreka phlox, Phlox hirsuta, deserves a brief mention because one of its four known occurrences is on a grassy hill within the city limits of Yreka,

SERPENTINE GRASSLANDS

153

California. The other three occurrences are in the understory of coniferous woodland west of Yreka. This plant is threatened by its apparently natural rarity and by proposed development on the Yreka city site (USFWS 2004).

Ongoing Invasion and Other Threats The increasing prevalence of Lolium multiflorum in some San Francisco Bay area serpentine grasslands has been attributed to atmospheric nitrogen deposition from automobile exhaust (Weiss 1999 ; Eviner and Firestone, Chapters 8; Dukes and Shaw, Chapter 19). Part of the evidence for this is that Lolium abundance and soil N levels are strongly elevated within a few hundred meters of freeways such as Highways 101 and 280. In turn, this raises the question of whether other serpentine grasslands in less massively urbanized settings are still functioning as relatively secure refuges for native species. In their statewide survey of serpentine sites, Harrison et al. (2006b) concluded that there is little evidence for an overall displacement of natives by exotics. However, some of the most important exceptions to this are in serpentine grasslands, which are more invasible than chaparral, woodlands, or barrens. Some indirect evidence suggests that the lower prevalence of exotics on serpentine is partly a function of lower rates of spread, as opposed to complete inability of such species to grow on serpentine. In the Napa–Lake County region, nonnative species were more common on small serpentine outcrops and within 50 meters of the edges of large outcrops than in the interiors of large outcrops (Harrison et al. 2000). The non-native grass Dactylis glomeratus, used in mine reclamation, spread less extensively from revegetated roadsides into serpentine grasslands than into oak woodlands (Williamson and Harrison 2002). However, the proximity of roads had no effect on the prevalence of non-native species in serpentine grasslands, even though the road effect was strong in nonserpentine grasslands (Gelbard and Harrison 2003). Besides N deposition, several other factors may be undermining the resistance of serpentine grasslands to exotic dominance. One is the continued evolution of serpentine-tolerant ecotypes of exotic species (Proctor and Woodell 1975), which has been documented in Avena fatua and Bromus hordeaceus (Harrison et al. 2001). Another is the continued arrival and spread of exotic species, a few of which may happen to be highly serpentine-tolerant and aggressive, such as Aegilops triuncialis and (to a lesser extent) Taeniatherum caput-medusae. Still another is the question of long-term positive feedbacks, such as the possibility that exotic species may modify the soil or its microbial fauna in ways that gradually make the soils more favorable to them (Batten et al. 2006a). Finally, in serpentine grasslands that already have a substantial complement of exotics, the cessation of livestock grazing and the infrequency of fire may contribute to the loss or decline of native species. Urban development threatens serpentine grasslands just as it threatens all of California’s natural habitats. Agricultural

154

ECOLOGICAL INTERACTIONS

conversion has been relatively less of a threat, but there are instances, especially in Napa and Lake Counties, of attempts to grow wine grapes on serpentine grassland soils after a heavy addition of calcium. Guenoc Winery’s “Serpentine Meadow” Petit Syrah is a case in point. The belief that stressed vines produce the most flavorful grapes has not helped the cause of serpentine grassland conservation. Restoration efforts for serpentine grasslands have taken a variety of approaches. At Edgewood County Park in San Mateo County, burning, mowing, and goat grazing are being tested as ways to suppress the dominance of Lolium multiflorum (Weiss 2003). At Inspiration Point in the San Francisco Presidio, there is active removal of introduced Monterey pine and cypress. At the University of California’s Hopland Field Station in Mendocino County, research has shown that two successive years of late-season burning can substantially reduce the abundance of Aegilops triuncialis (DiTomaso 2000), and this tactic is being deployed at several sites in Napa County. A major difficulty is that serpentine soils may not produce enough biomass to carry the second-year fire. Readdition of natives is equally important as the suppression of exotics. At the Sedgwick Reserve, Seabloom et al. (2003a) showed that native seed addition was more successful than mowing or burning alone in raising the abundances of native species in grasslands, although the best results were of course achieved by combining seeding with burning or mowing.

Summary and Conclusions Some take-home messages from this chapter include: •

Serpentine rocks and soils, derived from the earth’s mantle, are characterized by an unusually low Ca:Mg ratio that makes them inhospitable to most plants.

•

Grasslands comprise a relatively small part (roughly 7%) of the vegetation on serpentine soil in California. Serpentine grasslands are mostly found in the San Francisco Bay Area and the foothills of interior southern California.

•

Serpentine “grasslands” are often dominated by native annual forbs, most of which are not restricted (endemic) to serpentine, but which have become less common in grasslands on more fertile soils due to invasion and habitat conversion.

•

Serpentine grasslands are much less invaded by exotic annual grasses than other Californian grasslands, although they are still much more invaded than other types of serpentine vegetation (chaparral, woodlands, barrens).

•

Within serpentine grasslands, there are more exotic grasses on deeper and more fertile soils, and more native species and more serpentine endemics on shallow rocky soils.

•

Around 20 designated rare, threatened, or endangered plants and arthropods are associated with the serpentine grasslands of the San Francisco Bay Area.

•

Serpentine grasslands are threatened by ongoing invasion (especially by Italian ryegrass, Lolium multiflorum, in coastal areas, and barbed goatgrass, Aegilops triuncialis in interior areas), and grazing and burning may be useful management tools.

Serpentine grasslands are not a completely unique phenomenon. Similar features, such as abundant displays of native annual forbs that are maintained in part by livestock grazing, can still be found on other shallow, rocky, or nutrient-poor substrates, such as volcanic outcrops, vernal pools, and sandy soils (Kruckeberg 2005). However, it is unlikely that such features were typical of the pre-European condition of Californian grasslands on deeper, richer soils. In contrast, the bunchgrass-dominated grasslands that sometimes occur on deeper serpentine soils may be more similar to the pre-European state of grasslands on other deep soils; however, these are also the serpentine grasslands that are most susceptible to invasion and replacement by exotic species. Excellent serpentine bunchgrass stands tend to found in small isolated pockets where partial shade, moisture, and/or protection from disturbance

have helped them to survive. Thus, serpentine grasslands are neither an exception to the “rules” that govern Californian grassland ecology in general, nor a window into the past condition of Californian grasslands as a whole. Keeley (1990) provides an excellent discussion of the possible preEuropean state of Californian grasslands and how it was influenced by varying soils. Nonetheless, serpentine grasslands have considerable interest and importance as one of the most widespread native-dominated, lowland, herbaceous community types in California. They are an important habitat for rare species, especially in the San Francisco Bay area. Some of their particular attributes, such as their relatively low standing biomass and soil nutrient storage, make them ideal locations for many kinds of ecological and ecosystem studies. Careful comparisons between serpentine and nonserpentine grasslands can help ecologists understand how low productivity affects the roles of competition, herbivory, and disturbance in structuring grassland communities. Most of all, they are great places to see spring wildflowers.

Acknowledgments We thank Carla D’Antonio, Nona Chiariello, and Mark Stromberg for helpful comments on earlier versions of this chapter.

SERPENTINE GRASSLANDS

155

THIRTEEN

Competitive Interactions J E F F R EY D. C O R B I N, AN D R EW R. DYE R, AN D E R I C W. S EAB LO O M

The diversity, species composition, and relative abundances of vegetation in California’s grassland ecosystems are the result of a complex interaction between historical factors, abiotic conditions including climate and soil, and biotic interactions. The rangeland literature has frequently discounted the importance of competitive interactions between plants, arguing that many rangelands are not in equilibrium to the extent that would allow competitive interactions to drive species composition (Jackson and Bartolome, Chapter 17). However, a variety of studies, particularly those documenting interactions between perennial species but also interactions between annual species, have demonstrated that competitive interactions play an important role in determining the presence and relative abundances of grassland species. In the following sections, studies that have examined competitive interactions in California grasslands are reviewed, and life history characteristics, including longevity, growth patterns, and productivity, that influence competitive outcomes are considered. For the purposes of this review, competition is defined as the interactions between plants that result in decreased performance (measured by fitness, productivity, or survival) of one or more community constituents. Often, the competitive impacts take relative abundances of each competitor into account, rather than measuring impacts on a per individual basis. Indeed, the competitive superiority of exotic annual grasses in many grasslands is understood only when the large number of individuals per unit area is taken into account. In cases in which results from controlled experiments are examined, competitive outcomes are often measured as decreased performance in the presence of another individual or species as compared to the performance in the absence of other individuals or species. In the cases of observations of patterns in natural landscapes, including the competitive dominance of certain species (i.e., exotic annual grasses) under particular conditions, results of competition are inferred by numerical or functional superiority. Admittedly, this is an

156

indirect measure of competition that does not control for other important factors such as herbivory, propagule limitation, or edaphic conditions. However, numerical or functional superiority is still a useful indicator of competitive dominance in this system as long as certain caveats are recognized. First the major components of grassland communities are presented, including descriptions of their phenology and growth strategies. The knowledge as to the competitive outcomes between each group and how they vary with resource availability or region is also reviewed. In order to understand the mechanisms by which the competitive interactions take place, a description of how various life-history characteristics affect growth and survival and the interactions between species is presented. Finally, the question of how management strategies can be used to alter competitive interactions to favor native species, including grazing regimes and seed addition, is considered.

Components of California Grasslands The large number of species that occur in California’s grassland communities can be subdivided into broad functional groups according to phenology and/or life-history characteristics. Annual grasses, perennial grasses, and annual and perennial forbs are considered separately. The outcomes of competitive interactions between these functional groups vary temporally and spatially and can be strongly influenced by environmental conditions and/or management activities.

Annual Grasses Annual grasses are the most important component of most grassland communities in the state, particularly in inland habitats. Species such as Avena spp., Bromus diandrus, Bromus hordeaceous, and Vulpia spp. essentially dominate all low-elevation interior habitats such as those in the interior valleys and foothills lining both sides of the Great Central Valley (Heady

F I G U R E 13.1. Density of annual plants through the growing season

reported in four studies from California grasslands. Symbols represent the mean of all trials conducted in each study, while error bars represent the range (min–max) reported.

1988). These are largely species of European origin that invaded in the early eighteenth century (for further discussion see D’Antonio et al., Chapter 6). Little is known about the extent to which native annual grasses occurred in grassland habitats prior to the conversion; today native annual grasses such as Vulpia microstachys are much reduced in their distribution and abundance. The majority of annual grass seeds respond quickly to the onset of the fall rains, which typically occur by mid-November. Bartolome (1979) reported near-maximum seedling densities for several species, including Vulpia spp., within 1–2 weeks following rains sufficient to stimulate germination (Figure 13.1). Other annual grasses, such as Aira caryophyllea, exhibited a more conservative strategy and germinated later. After the first 1–2 months, however, the number of germinable seeds remaining in the soil declined to very low numbers. After establishment, annual seedlings grow rapidly as long as conditions remain favorable (Gulman 1979), but productivity slows considerably during the winter months as average temperatures drop and light availability decreases (Table 13.1) (Pitt and Heady 1978; Gulman 1979; Chiariello 1989). During the winter (December–February), when resource demand is low, density-dependent mortality is minimal (Gulmon 1979). In early spring (March–April), temperature and day-length increase and soil moisture levels are favorable for growth, which leads to a period of rapid biomass production (Table 13.1) (Savelle 1977; Pitt and Heady 1978; Gulmon 1979; Chiariello 1989). The dense stands of annual plants experience extensive self-thinning—as high as 50–75% (Eviner and Firestone, Chapter 8)—throughout the remaining vegetative period (Figure 13.1) until plants begin producing reproductive structures. Annual grasses such as A. caryophyllea may complete their life cycle by April 1 while others, such as Aegilops triuncialis or Bromus diandrus, may reach senescence in early June.

1997a; Heady 1988). Today, perennial grasses are relatively rare in inland habitats, though they may dominate coastal habitats (Heady 1988; Peart 1989a; Stromberg et al. 2001). The most widespread of the native perennial grasses is Nassella pulchra (purple needlegrass), which can be found in both the more mesic northern coastal grasslands and the hotter, drier interior and southern grasslands. Remnant coastal prairies may be dominated by such native species as Danthonia californica (California oatgrass), Festuca rubra (red fescue), and Deschampsia caespitosa (tufted hairgrass). Exotic perennial grasses, including Holcus lanatus (velvet grass), Festuca arundinacea (tall fescue), Phalaris aquatica (harding grass), and Dactylis glomerata (orchard grass), dominate many other coastal habitats, particularly ones that have been disturbed and that are not grazed. Phenology of perennial grasses is relatively similar to that of annual grasses with germination in the fall and greatest growth during the winter and early spring. However, while annual species develop from seedling to flowering adult over the course of about 6 months with relatively little variation aside from climatic influences, the length of time to maturity of perennial species can be indefinite (Dyer et al. 1996). Perennial grasses can reach the flowering stage within a single growing season or take several years, depending on a wide variety of biotic, abiotic, and edaphic factors. The most obvious difference between annual and perennial grasses is the presence of green tissue in perennial grasses later into the summer. In coastal habitats, where cooler temperatures, greater cloud cover, and the frequent input of moisture from fog moderates summertime conditions (Azevedo and Morgan 1974; Ingraham and Matthews 1995; Corbin et al. 2005), perennial species are able to maintain biological activity during the summer while most annual species die back.

Annual and Perennial Forbs Herbaceous forb species that are widespread in California grasslands include the native species Eschscholzia californica (California poppy), Calochortus spp. (Mariposa lilies), Hemizonia spp. (tarweeds), and the exotic species Erodium spp. (filaree), Cirsium spp. (thistles), and Centaurea solstitialis (yellow starthistle). This group also includes important nitrogen fixers such as Lupinus spp. (lupines), Trifolium spp. (clovers), and Lotus spp. Forbs make up a significant component of grassland biodiversity (Hayes and Holl 2003a) and are responsible for the wildflower displays for which California grasslands are famous; they are also some of the most widespread and problematic weeds in the state (DiTomaso et al., Chapter 22; Cal-IPC 2006).

Competitive Interactions Perennial Grasses California’s grasslands are assumed to have been largely dominated by perennial grasses prior to European settlement in the nineteenth century (D’Antonio et al., Chapter 6; Hamilton

In the vast majority of grassland habitats in the state, exotic annual grasses dominate communities in terms of both cover and biomass. Important exceptions include coastal prairie grasslands along the central and northern coast,

COMPETITIVE INTERACTIONS

157

TA B L E 13.1 Growth rates of total aboveground biomass at an annual and perennial (Nassella pulchra)-dominated site at Hopland Field Station

Annual Site Sampling period

Perennial Site

Absolute g / day

Relative mg / g / day

Sampling period

Absolute g / day

1970–1971 Oct 21 – Dec 7 Dec 7 – Feb 8 Feb 8 – Mar 6 Mar 6 – Apr 4 Apr 4 – May 5 May 5 – June 2

0.340 0.377 1.077 2.897 4.806 0.857

42.5 13.4 19.1 23.9 14.9

Oct 21–Feb 1

0.549

Feb 1–Mar 1 Mar 1–Apr 1 Apr 1–May 1 May 5–June 2

0.893 2.065 3.867 4.057

1971 – 1972 Nov 13 – Dec 4 Dec 4 – Jan 3 Jan 3 – Mar 2 Mar 2 – Apr 1 May 8–June 5

0.524 0.233 0.814 3.811 0.939

95.3 16.1 14.5 21.5

Nov 13–Jan 16

0.406

Jan 16–Mar 1 Apr 15–May 28

1.000 1.163

NOTE : (Savelle (1977). Absolute growth rate is the increase in biomass per day, while specific growth rate is the increase in biomass per day per average amount of biomass present during the sample period.

where perennial grasses frequently dominate (Heady 1988; Stromberg et al. 2001). Another exception is found in seasons where annual forbs—including Erodium spp. and Trifolium spp.—approach or even exceed the cover and biomass of exotic annual grasses. For the most part, the conditions that give rise to these “forb years” are ones in which temperature or precipitation patterns during germination and early growth are unfavorable to grass growth (Pitt and Heady 1978). Annual and perennial forbs also are the dominant life-forms in edaphic habitat types such as vernal pools or serpentine grassland. It is useful to examine the specific relationships between the three groups in turn.

Annual Grasses vs. Annual and Perennial Forbs Annual grasses dominate cover and biomass of inland grasslands in most years (Talbot et al. 1939; Heady 1958; Bartolome 1979; Heady 1988), though particular climatic conditions give rise to years in which the annual Erodium spp. (filaree) is dominant or when it and the nitrogen fixer Trifolium spp. (clover) are co-dominant (e.g., “grass-clover-filaree years”; D’Antonio et al., Chapter 6; Talbot et al. 1939; Heady 1958; Pitt and Heady 1978). In a study of botanical composition of Hopland Field Station (Mendocino County) grasslands from 1955 through 1973, Pitt and Heady (1978) found that the cover of such annual grasses as A. barbata, B. mollis (hordeaceus), and Vulpia spp. ranged from 24.1% to 82.1%. Cover of Erodium spp. ranged from 3.6% to 48.1%, and cover of Trifolium spp. ranged from 2.5% to 20.1%. At no point was Trifolium dominant in terms of percent cover, though it reached its highest cover in the year that annual grass cover was its lowest. Similarly, the

158

ECOLOGICAL INTERACTIONS

highest annual grass cover coincided with the lowest Erodium cover (Pitt and Heady 1978). Pitt and Heady (1978) found that annual grasses were favored in years when germinating rains begin early (while temperatures were still warm) and precipitation through the autumn and winter was regular. Years in which there were late rains, or an extended winter or spring drought, favored filaree and clover, respectively. Another way of describing these patterns is to say that years with “favorable” rainfall patterns have relatively high grass cover, while years with suboptimal rainfall patterns have relatively high cover of nongrasses. Thus, annual grasses appear to be superior competitors vs. forb species as long as climatic conditions do not limit grass productivity or seedling survival.

Annual Grasses vs. Perennial Grasses I NTE R IOR G RAS S LAN DS

A variety of experiments have demonstrated that exotic annual grasses are able to reduce the growth and survival of native perennial grass individuals and to limit the growth of native grass populations where they exist. The effect is most pronounced in the Central Valley, where competition studies between the native perennial N. pulchra and exotic annual grasses consistently favor the exotics (Dyer and Rice 1997b, 1999; Brown and Rice 2000; Marty 2005). The exotic annual grasses are able both to reduce the establishment of seeds and seedlings of perennial grasses and to reduce growth and survival of mature individuals. In established annual grasslands, whatever perennial seeds may be present often germinate

and emerge relatively well, but seedlings are lost rapidly as the season progresses. Rapid vertical growth by annuals quickly reduces light availability to plants of small stature (Evans and Young 1989; Dyer and Rice 1999), including N. pulchra seedlings. Annual vegetation does not have to be especially tall if densities are great. As the season progresses, soil water is reduced faster by annual vegetation and at higher densities (Reever Morghan et al., Chapter 7; Gordon and Rice 1992; Holmes and Rice 1996; Dyer and Rice 1999), thereby compounding the resource stress. Annual grasses are also able to reduce growth of established N. pulchra individuals (Dyer and Rice 1997b; Hamilton et al. 1999). A. Dyer (2003) followed natural N. pulchra individuals in a Central Valley grassland dominated by annual grasses and found that 26.1% (164 of 629) of mature plants died over a seven-year period. Marty et al. (2005) reported a loss of 3 – 30% of adult N. pulchra individuals over a five-year period across a range of grazing and burning treatments at Beale Air Force Base (Yolo County). Hamilton et al. (1999) found that mature N. pulchra individuals at Hastings Reserve (Monterey County) were more water stressed and produced between 1.6 – 3.5 times fewer seeds per plant in the presence of exotic annual grasses, but there was no net loss of natural N. pulchra over a 25 year period. Differences in responses of N. pulchra populations between the Central Valley locations of A. Dyer (2003) and Marty et al. (2005) vs. the more mesic Hastings Reserve of Hamilton et al. (1999) suggests that N. pulchra populations growing in mesic vs. xeric grasslands may respond differently to annual competitors. The negative effects of exotic annual grasses on all N. pulchra life stages—seed, seedling, and adult—strongly suggest that the exotic annuals have a negative effect on many native perennial populations. Even under the best of circumstances, native perennial grasses tend to be relatively minor components (in terms of cover or biomass) in inland grasslands (Heady 1988; Stromberg and Griffin 1996), and the longterm stability of the perennial populations — increasing, decreasing, or stable—is rarely known. Further research is needed to understand the population dynamics of native species in exotic-dominated grasslands, to identify vulnerable life stages, and to design management strategies that will ensure persistence (Box 13.1). Whether perennial grasses are increasing or decreasing in abundance, or have been excluded altogether, depends on the seasonal timing of competitive stress relative to perennial phenology — particularly whether seedlings are able to develop root systems that are sufficient for the individual to survive the summer drought. Furthermore, grazing and burning treatments may moderate the intensity of competition between annuals and perennials and perhaps allow the perennials to coexist. Positive effects of treatments such as grazing may also work in combination with other factors. For example, Dyer and Rice (1997b) found that grazing reduced the negative effect of herbivory by gophers on N. pulchra. Management strategies such as grazing that may alter

competitive relationships between annual and perennial grasses are considered later in this chapter.

C OASTAL G RAS S LAN DS

In contrast to inland grasslands, competitive interactions in grasslands along the central and northern coasts consistently favor perennial species. Corbin and D’Antonio (2004b), working in a Marin County coastal prairie, found that the presence of native perennial grasses reduced aboveground biomass of exotic annual species by as much as 80%. In contrast, while native perennial grasses, planted as seedlings, experienced a negative effect of the presence of exotic annual grasses in the first year following planting, the negative effects of the annuals decreased when the perennials were 2, 3, and 4 years old. In fact, biomass of 4-year-old native perennial grasses grown with exotic annual grasses was not significantly different from the biomass of natives grown without exotics. Similarly, Seabloom et al. (2003b) found reduced cover and fecundity of annuals in established multispecies perennial grass plots at the Sedgewick Reserve (Santa Barbara County). The persistence of relatively undisturbed remnant coastal prairie grasslands further illustrates the competitive ability of native perennials in coastal habitats and their ability to resist invasion by exotic annuals. In Stromberg et al.’s (2001) survey of coastal prairie grasslands in central California, the exotic grasses that were most abundant—Vulpia myuros, B. hordeaceus, A. caryophyllea, and Briza minor—were relatively small-statured compared to the species, such as Bromus diandrus and Avena spp., that dominate many inland grass communities. Mean species richness (per m2) was also higher in the coastal grasslands (22.6 species) vs. inland grasslands (14.7 species), and the ratio of exotic to native species richness was lower. In sharp contrast to their dominance in most interior and southern grasslands, annual grasses are frequently relatively minor components of the coastal grassland community or are relegated to recently disturbed patches. Peart (1989b, 1989c), studying colonization of intact and disturbed grassland patches at Sea Ranch (Mendocino County), found that annual grasses such as V. myuros persist in mixed annual–perennial patches mainly as colonizing species following small-scale disturbances (e.g., gopher mounds). V. myuros was able to colonize mounds in the first year following disturbance but was excluded by perennial species in the following year (Peart 1989c). Patches where V. myuros dominated were also highly susceptible to invasion by exotic perennial grasses such as Anthoxanthum odoratum and Holcus lanatus. By contrast, V. myuros was unable to invade and persist in undisturbed patches dominated by perennial grasses (Peart 1989b). Climatic conditions in coastal habitats are significantly more mesic than in interior grasslands, owing to the cooler temperatures, greater cloud cover, and the frequent input of moisture from coastal fog (Azevedo and Morgan 1974;

COMPETITIVE INTERACTIONS

159

BOX 13.1 MODELING POPULATION GROWTH OF NASSELLA PULCHRA

Matrix models are useful to estimate the intrinsic rate of increase (␭) of a population, or to assess particular stages that are critical to the demographics of a population (Caswell 1989). We used a simple stage-based matrix model to calculate ␭ of Nassella pulchra based on available estimates of the proportion of individuals that survive the transition from each stage to the next. We included four life stages: seeds, seedlings, reproductive adults, and a senescent, nonreproductive stage (Figure 13.2). The latter stage was included to truncate the lifespan of these potentially long-lived grasses (Hamilton 1997), in lieu of detailed data on the shape of survivorship curves. The effect of this senescent stage should be to decrease estimates of ␭. We set all parameters based on published data (Tables 13.2 and 13.3). We included only situations in which native perennial grass seed was added to intact plant communities — we excluded studies in which the vegetation was disturbed or manipulated to enhance seed survival. The model suggests that the perennial lifestyle imparts a competitive advantage. Perennial grass populations can have positive population growth rates even when seedling survival in the first year is exceptionally low. This result is robust to adult lifespan and suggests that it may be generally true for many species of perennial grasses. While we predict successful establishment in the majority of cases, there were cases in which survival through the first year was insufficient to create positive population growth rates, namely samples in the lowest quartile in Figure 13.3. In these cases, it is likely that the presence of exotic annual competitors precludes seedling establishment.

TA B L E 13.2 Survival During First Year in Experimental Seed Additions in California Grasslands.

Grassland

Percent

Data source

Year

County

Sown species

Species type

type

survival

Robinson et al. 1995

1983

Yolo

Eschscholzia

Perennial forb

Exotic annual

0.14

Perennial forb

Exotic annual

60.83 39.82

californica Seabloom et al. 2003b

1998

Santa Barbara

Lasthenia californica

Seabloom et al. 2003b

1998

Santa Barbara

Plantago erecta

Annual forb

Exotic annual

(Brown and Bugg 2001)

1993

Yolo

Annual forb mix

Annual forb

Exotic perennial

7.20

Gillespie and

2001

Riverside

Erodium

Annual forb

Exotic perennial

3.00

Robinson 1971

1967

Monterey

Nassella pulchra

Perennial grass

Exotic annual

59.00

Brown and Rice 2000

1993

Yolo

Mix of native

Perennial grass

Exotic annual

6.62

Allen 2004

macrophyllum

species Robinson 1971

1966

Monterey

Nassella pulchra

Perennial grass

Native perennial

0.00

Robinson 1971

1967

Monterey

Nassella pulchra

Perennial grass

Native perennial

39.75

Dyer et al. 1996

1989

Yolo

Nassella pulchra

Perennial grass

Native perennial

1.10

Dyer et al. 1996

1990

Yolo

Nassella pulchra

Perennial grass

Native perennial

0.20

Hamilton et al. 1999

1995

Santa Barbara

Nassella pulchra

Perennial grass

Native perennial

15.00

Peart 1980

1980

Sonoma

Mix of native and

Perennial grass

Exotic perennial

0.57

Perennial grass

Exotic annual

2.70

Perennial grass

Exotic annual

0.18

Perennial grass

Native perennial

1.16

exotic grasses Peart 1980

1980

Sonoma

Mix of native and exotic grasses

Peart 1980

1981

Sonoma

Mix of native and exotic grasses

Peart 1980

1980

Sonoma

Mix of native and exotic grasses

NOTE : Seed was added into intact plant communities without further management. We treat seed additions in separate years or plant communities as unique replicates, although they are published in a single study. We do not analyze species-specific responses of species sown in a single mixture to control for among-seedling interactions.

BOX 13.1 (continued)

TA B L E 13.3 Parameter Values Used in Perennial Grass Matrix Model

Parameter

Definition

f31

Fecundity of reproductive adults (seeds per plant)

p12

Probability of seed becoming a

Value

Reference

26-95a

Hamilton et al. 1999

0.001 to 0.3

Values from Table E1 for native

1-year seedling p23

perennial grasses

Probability of 1-year seedling becoming

0.71-0.88b

Shoulders 1994

(1 - p33)

Defined by p33

a reproductive adult (stage 3) Probability of a reproductive adult

p34

becoming senescent (stage 4) p33

Probability of a reproductive adult

0.5 to 0.9

Shoulders 1994,

remaining reproductive (stage 3) into the next year p44

Hamilton et al. 1999

Probability of a senescent adult

0.1

Arbitrarily low to ensure rapid

remaining alive to the next year

mortality

a Hamilton measured seed production of adult N. pulchra plants in control plots that received 330 mm of precipitation and in watered plots that received a total of 610 mm. b Shoulders measured survival of transplanted N. pulchra plugs in a dry (280 mm) and wet (784 mm) year.

p12

2. Seedlings

p23

3. Reprod. Adults

p34

3.5

100 10 5

2.5

3.0

Q3

1.0 5

10 15 20 25 30

Percent Seed Survival (100*P12)

1. Seeds

Q1 Median

P23 = 0.88

2.0

2.5 1.5

2.0

100 10 5

0

f31

F3 = 95

1.5

Q3

3.0

3.5

B

1.0

Population Growth Rate (λ)

Q1 Median

P23 = 0.71

Population Growth Rate (λ)

F3 = 26

A

0

5

10 15 20 25 30

Percent Seed Survival (100*P12)

F I G U R E 13.3. Effects of first year survival (p12), second year

4. Old Adults

p33

p44

F I G U R E 13.2. Schematic of matrix model used to calculate

perennial grass population growth rates. The perennial grass life is represented in four stages: 1. seeds, 2. one year old seedlings, 3. reproductive adults, and senescent adults.

survival (p23), mean adult lifespan (1/(1-p33)), and adult fecundity (f3) on predicted growth rates of perennial grass populations. Mean lifespan is labeled on graph (5, 10, and 100 years). Dashed horizontal line shows the point of no net population growth rate (␭⫽1). The first quartile, median, and third quartile of empirical estimates of p12 are shown as vertical dotted lines. The two panels illustrate predictions for two scenarios: (a) low precipitation (280 – 330 mm) and (b) high precipitation (670 – 780 mm).

It should also be noted that the seedling survival estimates used in this modeling exercise are based largely on experiments using N. pulchra, a species that may have very low seedling survival relative to other native perennial grasses (Seabloom, unpublished data). In addition, the model did not allow reproduction until the third year in order to mimic the slow growth of N. pulchra. In reality, many perennial grasses seed in their first and second year (e.g., Bromus carinatus and Elymus glaucus). Models that include seed reproduction in year two have dramatically higher population growth rates (results not shown). For these reasons, the model simulations shown here are conservative estimates of the potential population growth rates for the native perennial grass community as a whole.

Ingraham and Matthews 1995; Corbin et al. 2005). As a result, the balance between evapotranspiration and water availability is significantly more favorable than in hotter and drier regions such as those inland, and perennial species are able to maintain biological activity during the time that winter annual species die back. Perennial species, including native and non-native grasses, have been shown to take up moisture from coastal fog via shallow roots, thereby moderating the effect of the summer drought (Dawson 1998, Corbin et al. 2005). As distance from the coast increases or where topography blocks the marine influence, evapotranspiration increases and moisture inputs from coastal fog decrease. The importance of these climatic differences for the competitive relationship between annual and perennial grasses is discussed later in the chapter.

Perennial Grasses vs. Perennial Grasses Though the invasion of coastal grasslands by exotic annual grasses is frequently less severe than in inland grasslands, another group of grasses—exotic perennial grasses—have a significantly stronger ability to invade. Greater moisture input and lower evapotranspiration during the summer (Chapter 7; Corbin et al. 2005) may favor perennial species— native and exotic — over annuals. Exotic perennial grasses such as H. lanatus, Festuca arundinacea, Phalaris aquatica, and Dactylis glomerata dominate many coastal grassland habitats, particularly ones that are not grazed (Heady 1988; Peart 1989a; D’Antonio and Corbin, unpublished data). Foin and Hektner (1988) documented the replacement of annual species with exotic perennial grasses at Sea Ranch following the cessation of grazing, a process that has been supported by anecdotal observation in such preserves as Bodega Marine Reserve (Sonoma County) and Pt. Reyes National Seashore (Marin County). In these cases, secondary succession toward dominance by perennial grasses is, apparently, the dominant process in ungrazed habitats (Heady 1988), while grazing appears to prevent the competitive exclusion of annual grasses and forbs (Hayes and Holl 2003a). Exotic perennial grasses are also able to invade communities dominated by native perennial grasses. Thomsen (2005) demonstrated that H. lanatus successfully invaded established stands of B. carinatus and F. rubra. In one multi-year sampling of a relatively undisturbed coastal prairie grassland (Tom’s Point Preserve in Marin County), D’Antonio and Corbin (unpublished data) found that percent cover of exotic perennial grasses increased from 1999 to 2006. The increase in percent cover was greatest in years with abundant late-season rain. The invasion of exotic perennial grasses into coastal prairie grasslands raises concerns that even remnant stands, heretofore resistant to invasion, may face an uncertain future. Based on their capacity to invade natural communities and their impacts on intact communities, F. arundinacea, H. lanatus, and P. aquatica were classified as a “medium” threat in the 2006 California Invasive Plant Council’s (Cal-IPC) Invasive Plant Inventory (Cal-IPC 2006). This ranking is on a par with

162

ECOLOGICAL INTERACTIONS

that of exotic annual grasses such as Avena spp., B. diandrus, and Lolium multiflorum. Of particular concern is the fact that these exotic perennial grasses are likely still expanding and have the potential to impact ecosystems dominated by both exotic annual grasses and native perennial grasses. Future research should examine the potential spread and impacts of these relatively recent invaders.

Perennial Grasses vs. Annual and Perennial Forbs Because of the scenic and biodiversity value of native forbs in coastal prairie grasslands (Heady et al. 1988), there has been increasing interest in management strategies that control annual grass growth and maximize forb diversity (Hayes and Holl 2003a). Principle among these strategies is the use of grazing that can control grass growth and that is capable of excluding annual and perennial forbs (Chapter 17). Foin and Hektner (1986) found that succession following the cessation of grazing led to a twofold decrease in the cover of annual and perennial forbs and a similar increase in perennial grasses (mostly exotic) in two Sea Ranch (Mendocino County) grasslands. Hayes and Holl (2003a), comparing paired grazed and ungrazed sites along the central and northern coast, found that native annual forb species’ richness and cover were over two-fold higher in grazed sites than in ungrazed sites. However, grazing had negative impacts on native perennial forbs. Recent evidence has suggested that perennial grasses, including N. pulchra, are competitive against taprooted forb species such as the invasive weed Centaurea solstitialis (yellow starthistle). Reever Morghan and Rice (2005) reported reduced survival and growth of yellow starthistle as N. pulchra stands matured. By contrast, yellow starthistle is able to effectively invade stands dominated by annual grasses, perhaps because of abundant soil moisture below the rooting zone of annual grasses (Enloe et al. 2004). Native perennial grasses, with their deeper rooting profiles, reduce soil moisture deep in the soil profile to a greater extent than exotic annual grasses (Enloe et al. 2004; Gerlach 2004). Thus, the invasion of California’s Central Valley grasslands by annual grasses may have left deep soil moisture resources untouched and promoted the invasion of these sites by deep-rooting species such as C. solstitialis (Chapter 7).

Specialized Competitive Situations—Edaphic Sites S E R P E NTI N E SOI LS

Serpentine communities are dominated by annual grasses and herbs and tend to be very resistant to invasion. The soil characteristics create physiologically harsh growing conditions but support diverse communities with a relatively high proportion of endemic species (Harrison and Viers, Chapter 12; Kruckeberg 1984). Productivity is much lower than in surrounding soils, and water is likely very limiting and is probably a function of soil texture (Harrison 1999a). Very few

non-native species have invaded these communities, but the exceptions are notable. Harrison (1999a) documented 33 alien species in her sites, and soil nitrate was a significant predictor of invasive diversity. Aegilops triuncialis (barbed goatgrass) is common on these soils throughout northern California and is often a dominant species. Its ability to invade serpentine has been linked to high phenotypic plasticity rather than adaptation to the stressful conditions (McKay et al., unpublished data). Other non-native annual grasses do not readily invade serpentine patches, but B. hordeaceus and A. fatua were more likely to be found in small patches than large ones and this was attributed to higher propagule pressure leading to faster selection for serpentine-tolerance (Harrison et al. 2001). Recently, nutrient inputs from atmospheric deposition have increased nutrient availability, particularly nitrogen, in some serpentine grasslands (Fenn et al. 2003b). As a result, exotic grasses such as B. hordeaceous and Lolium multiflorum (Italian ryegrass) have successfully invaded previously nativedominated serpentine grasslands (Weiss 1999). As atmospheric nitrogen deposition continues across large areas of California, the unique species composition of some serpentine grasslands may shift. VE R NAL P O O LS

Vernal pools are shallow seasonal wetlands fed by rainwater and are perched on impermeable clay soils, often with low nutrient availability and low pH. The highly diverse communities are very resistant to invasion by aliens. Holland and Jain (1988) estimated non-natives at 7% of the species, and a recent survey by Michael Barbour and colleagues confirmed that aliens remain at or below 10%, despite the fact that aliens often comprised 90% of the biomass of surrounding grasslands. Establishment and survival in vernal pools is more likely driven by stress tolerance than by resource competition, because species must germinate in the winter while inundated but mature in hot and dry conditions. Resource competition is likely dominated by water and only for a very short period in late spring, depending on the rainfall patterns of the particular year. Native species mature as the water level recedes, and each species can be characteristic of particular zones around the pool. It is very likely that the combination of fluctuating water resources, edaphic stress, and well-adapted natives may limit non-native invasions in vernal pools except by preadapted species from similar habitats.

Life History Characteristics The components of California grassland communities employ very different strategies — including variation in lifespan, summertime activity, rooting patterns, and allocation to root vs. shoot tissue — to survive the conditions of California’s Mediterranean climate and its annual summer drought. A comparison of the life-history characteristics of annual and

perennial species, with an eye toward understanding how differences between various groups—annual grasses, perennial grasses and forbs—influence species composition in various grassland habitat types is useful.

Longevity As previously noted, annual species germinate quickly with the arrival of sufficient rainfall in the fall and complete their entire life-cycle—germination, growth and reproduction — by the time the rains have tapered off in the spring and summer. By contrast, perennial species invest in root systems that allow them to survive the summer drought and live, in some cases, for decades. The difference in longevity between annuals and perennials means that they have fundamentally different approaches to the early stages of growth. Annual plants, for the most part, germinate and grow quickly, investing mostly in aboveground biomass and reproductive structures compared to roots. Perennials, on the other hand, are frequently more conservative in terms of resource use and have slower growth rates. Reynolds et al. (2001), using seeds grown in growth chambers, found that seeds of exotic annual species germinated as much as 1–2 weeks earlier than native or exotic perennial species. This two-week advantage may provide the annual seedlings a considerable advantage vs. perennial seedlings: Biomass of the native perennials Nassella pulchra and Festuca rubra and the exotic perennial Holcus lanatus was significantly greater when their seeds were allowed to germinate 14 days before seeds of B. diandrus, as compared to treatments in which the perennial and annual seeds were allowed to germinate at the same time (Abraham et al., unpublished data). The rapid growth of winter annuals in late spring exacerbates the resource stress experienced by perennial grasses, especially at the seedling stage. For example, depending on soil characteristics, moisture depletion in the upper 60 centimeters occurs earlier in the season when annuals are present (Gordon and Rice 1992; Brown et al. 1998; Dyer and Rice 1999). The diffuse competitive effect of thousands of annual plants per square meter amplifies resource stress at the habitat scale (Dyer and Rice 1997b). The longer lifespans of perennial species may give them an advantage over annual species once the perennials are mature. Several studies have reported a correlation between bunchgrass size (or age) and resistance to invasion by annuals (Brown et al. 1998; Hamilton et al. 1999; Corbin and D’Antonio 2004b; Reever Morghan and Rice 2005). Corbin and D’Antonio (2004b) argued that the ability of perennial species to hold onto space and resources over the summer gave them the advantage of “incumbency” over newly germinating annual species. This mechanism may be especially important in coastal habitats, where perennial grasses maintain biological activity well into the summer, as compared to inland habitats, and therefore can occupy a substantial proportion of resources as the fall rains begin.

COMPETITIVE INTERACTIONS

163

Summer Activity One of the most obvious differences between the life history strategies of annual and perennial species is the senescence of annual species prior to the onset of the summer drought. In effect, populations of annual species, including the dominant annual grasses, survive the summer drought as seeds in or on the soil. In contrast, many perennial species are metabolically active during the summer drought and often have at least some green tissue. Where summertime inputs of moisture are low or absent (e.g., all but the most maritime conditions along the coasts), perennial species survive the drought by utilizing residual soil moisture left over from the wet season. Perennial grasses and taprooted perennial forbs develop root systems that take up water as deep as 50–100 cm below the soil surface (Holmes and Rice 1996; Enloe et al. 2004). Competitive pressure from annual species during the winter and spring may negatively affect the ability of perennial species to reach belowground resources during the summer (Dyer and Rice 1999; Hamilton et al. 1999). Availability of summer moisture, and the ability to use it, is a critical difference between inland and coastal grasslands. Corbin et al. (2005) found that fog was a significant summer water source in coastal grasslands—25–66% of the water in perennial grass tissues derived from fog as opposed to residual rainwater. The ability to use fog water suggests that coastal grass roots are continuously active in the shallow soil profile, even during the summer (Corbin et al. 2005). This pattern stands in stark contrast to the reliance on deep soil moisture by inland perennial grasses (Dyer and Rice 1999; Brown et al. 1998). Inland grasslands experience occasional summer showers, but there is little empirical evidence on the extent to which perennial grasses make use of such ephemeral resources. Experimental rainfall studies with Agropyron desertorum in the Great Basin found either no new root growth (Ivans et al. 2003) or a three day lag after the rain event (Cui and Caldwell 1997). Summer moisture provides a mechanism to explain why perennial grasses can dominate coastal grasslands but not inland grasslands. Dominance in this case is not necessarily through superior competitive ability for water, but by site preemption via continued growth after annual species have become senescent—i.e., through a different phenology that better allows the capture of the late season water not available in inland grasslands. Unfortunately, in coastal grasslands, the conditions that favor native perennial grasses also favor exotic perennial grass, and therefore summer moisture may decrease invasion resistance to those taxa (Corbin et al. 2005).

Rooting Patterns Annual grasses deploy 90% of their roots in the upper 30 centimeters of the soil profile (Holmes and Rice 1996) and greatly reduce soil water in that zone (Seabloom et al. 2003b). In contrast, perennial grasses begin their life with roots in the topsoil, but large individuals root to 1.5m or more in deep

164

ECOLOGICAL INTERACTIONS

soil (Dyer and Rice 1999; Brown et al. 1998). Therefore, in deep soils perennial grasses can partition soil resources vertically and avoid much competition with annuals in the upper 30cm. In shallow soils, however, annual and perennial grasses compete for the same soil resources (Dyer and Rice 1999; Seabloom et al. 2003b). In perhaps the most extensive study on native grasses to date, Brown et al. (1998) investigated the links between resource partitioning and capture, species composition, and age of stand. By creating a variety of experimental swards with up to seven species of native perennial grass, they were able to look at water use, biomass production, and response to non-native annuals over the course of consecutive growing seasons. N. pulchra individuals in containers were capable of reducing soil water to a significantly greater degree and at greater depths than other native grasses. In field experiments, summer-active native perennial grasses deployed much greater proportions of their roots to deeper soil layers compared to annual grasses. Cool-season perennial grasses tended to have similar root profiles compared to annual grasses. These data support the contention that root establishment is critical if perennial grasses are to survive the seasonal drought. Overall, this series of experiments demonstrated that established perennial grasses can be very competitive for water, but also that the conditions in exotic annual communities limit the ability of perennial grass seedlings to capture resources and grow. These differences in rooting patterns likely influence susceptibility to growing season droughts. Evidence from experimental (Hamilton et al. 1999; Sher et al. 2004) and observational (Pitt and Heady 1978) studies suggest that annual grasses are especially sensitive to periods of drought during the winter and may be poor competitors for water when this resource is scarce. Annual grasses do not develop deep root systems that can reach deeper sources of water (Holmes and Rice 1996) and likely can grow only when abundant moisture is available. By contrast, perennial species develop deeper root systems (Dyer and Rice 1999; Enloe et al. 2004) and are adapted to survive even extended droughts, such as those that occur during the summer months. In an experimental test of the relative impacts of a growingseason drought on annual and perennial grasses, Hamilton et al. (1999) found that a simulated drought (35 days) in December and early January decreased the biomass of exotic annual grasses by more than 25%, while survival and biomass of N. pulchra seedlings were not affected. In a study using two annuals—V. myuros and Erodium laciniatum from the Mediterranean-climate region of Israel—Sher et al. (2004) found that these species were sensitive to increasing lengths of growing-season droughts. Increasing the length of time between rainfall events decreased relative growth rate and survival of both annual species at low (100 mm/season) and high (500 mm/season) water regimes. Yet periods without precipitation are regular features of California’s climate, even during the winter. Null (2006), using climate records from San Francisco, reported that dry spells of at least 8 days (mean

duration ⫽ 19 days) during December and January occurred every year from 1950–2005. These results indicate that competition for water in California resembles the “two-phase resource dynamics” hypothesis (Goldberg and Novoplansky 1997). According to this hypothesis, availability of a resource such as water can be divided into periods when the resource is plentiful (or “pulses”) and periods when it is too scarce for plant growth (“interpulses”). While competition may be most intense during resource pulses in productive environments, interpulses may be the most important environmental factor for plant growth and survival in unproductive environments (see discussion of serpentine and vernal pool communities in previous paragraphs). In Mediterranean climates such as California, distinct pulses and interpulses occur within each growing season as well as between the distinct winter pulse period and the summer interpulse period. Pulses and interpulses may occur several times within a growing season as precipitation may be absent for 1–3 week periods (Null 2006). Thus, competition dynamics in annual grasslands — even during the winter— may be as much a function of the frequency and duration of dry periods as it is the availability of water during periods when water is plentiful.

Aboveground Productivity Perennial grass seedlings and small-statured plants are highly susceptible to light limitation by the dense and taller annual neighborhood. Dyer and Rice (1999) found that light penetration in exotic annual grass – dominated plots at Jepson Prairie (Solano County) was as low as 5% of full sun. Of course, reduced light availability greatly slows vegetative growth, thereby making the perennial grass progressively less able to compete for light. Reduced carbon fixation also reduces absolute allocation to roots. By late spring or early summer, the surviving small perennial-grass individuals are ill-prepared for the summer drought. Thus, competition with annuals for light in early spring reduces the root growth necessary for survival through the annual summer drought. The story is almost exactly reversed when competition between established perennial grasses and annual grasses is observed. Corbin and D’Antonio (2004b) and Seabloom et al. (2003b) concluded that the ability of the perennials to reduce light available for exotic annual grass seedlings reduced exotic productivity and maintained perennial dominance. Similar results were reported for the effect of N. pulchra on the survival and growth of Centaurea solstitialis (Reever Morghan and Rice 2005). Regardless of the location of the grassland, light limitation is a function of productivity and this, of course, is closely linked to water availability and soil quality. Thus, light limitation is likely to vary temporally and spatially. In years when climatic conditions are favorable for exotic annual growth (e.g., “grass years,” sensu Pitt and Heady 1978), or in richer soils that can support greater productivity, light limitation should

be greater and should occur earlier in the growing season. In such systems, space acquisition (e.g., rosette formation) and vertical growth should be favored. In less favorable years or in resource-poorer systems, light is less likely to be a limiting factor because of reduced annual grass productivity.

Nitrogen Use The uptake of inorganic N by vegetation during the growing season keeps pool sizes of KCl-extractable inorganic N (NH4-N and NO3-N) small (Eviner and Firestone, Chapter 8; Jackson et al. 1988; Corbin and D’Antonio, 2004a). Though evidence is scarce, it is likely that elevated N levels favor exotic annual grasses more than other components of grassland communities. N fertilization frequently increases productivity in California grasslands (Harpole et al., Chapter 10) and is capable of shifting species composition in serpentine and non-serpentine grasslands toward exotic annual grasses (Huenneke et al. 1990, Weiss 1999, Seabloom et al. 2003b). However, the extent to which species composition of nonserpentine-derived soils varies along gradients of soil fertility is not clear. In one of the few studies documenting community changes following N addition to nonserpentine grassland, Seabloom et al. (2003b) found that N. pulchra cover decreased over 5 years of fertilization. However, there was no corresponding increase in other community components that would suggest a role for competitive interactions. The invasion of N-fixing shrubs such as Lupinus arboreus into coastal prairie grasslands provides a natural experiment showing the effects of nitrogen on grassland community composition. L. arboreus shrubs, which grow rapidly and produce a dense canopy that shades out native grassland species, are capable of reducing native plant diversity and increasing dominance by annual species. Maron and colleagues (Maron and Connors 1996; Maron and Jefferies 1999) demonstrated that repeated cycles of lupine colonization and death led to a doubling of total soil N, greatly increased N availability, and increased vegetative productivity. These cycles caused a large-scale shift in grassland composition from native perennial to exotic annual species, presumably due to the greater responsiveness of annual species to the elevated N levels than perennial species. Several efforts to favor native perennial species in a restoration context by reducing plant-available N have not altered species diversity or dominance. Addition of a labile carbon source, such as sucrose or sawdust, would be expected to stimulate production by C-limited microbial populations and to reduce plant-available nitrogen. Addition of sucrose and sawdust has been attempted in coastal prairies in an effort to reduce productivity of exotic species and increase native biodiversity (Alpert and Maron 2000; Haubensak 2001; Corbin and D’Antonio 2004a; Corbin et al. 2004). However, even where these efforts have been successful at reducing exotic abundance or productivity (Alpert and Maron 2000), they have not resulted in a change in competitive outcomes between annuals and perennials.

COMPETITIVE INTERACTIONS

165

Relatively few studies have experimentally tested the role of N in competitive interactions in California grasslands (Huenneke et al. 1990). Differential abilities to grow under Nlimited conditions have been shown to play an important role in determining the outcome of competitive outcomes in other grassland systems (Tilman 1988), and it is suspected that the same is true in California systems. To date, however, the ability to predict interactions of competitive outcomes under various N levels is limited. Further research is needed in this area as elevated N inputs from atmospheric N deposition have the potential to increase N availability to grasslands throughout the state (Weiss 1999; Fenn et al. 2003b).

Physical Barriers The accumulation of plant litter has been shown to influence seed germination rates and species composition in a variety of North American grasslands (Heady 1958; Young et al. 1971; Hamrick and Lee 1987; Facelli and Pickett 1991a, b; Foster and Gross 1998; Reynolds et al. 2001). The presence of litter in California grasslands has been shown to have a positive effect on seed germination rates (Young et al. 1971), presumably by increasing moisture retention at the soil surface (Facelli and Pickett 1991b). However, litter accumulation can also form a physical barrier that limits establishment of germinating seeds (Reynolds et al. 2001). Responses to the presence of litter varies by species (Young et al. 1971; Reynolds et al. 2001), and thus the presence and abundance of litter is likely to influence community composition. Management strategies such as prescribed burning and grazing have been widely applied in efforts to reduce negative effects that litter accumulation can have on native grass and forb germination and establishment (Chapter 17; Reiner, Chapter 18). However, fire and grazing can alter environmental conditions or competition in a variety of ways, including reducing litter accumulation but also mortality, productivity, and allocation to roots vs. shoots, and it has not been easy to quantify the precise effects of each treatment type. As a result, the mechanisms by which these treatments may affect native seed germination or establishment have been difficult to quantify.

response to disturbances (Stromberg and Griffin 1996). Recent research has turned to ways of influencing competitive relationships between native and exotic species through active management in an effort to achieve management goals such as increased native biodiversity (Corbin et al. 2004).

Grazing The grazing of cows, sheep, and goats has been widely applied in an effort to influence competitive outcomes and favor one group of species over another (e.g., native vs. exotic species). Huntsinger et al. (Chapter 20) provide a detailed review of the application of grazing in the promotion of native biodiversity. In theory, specifically timed grazing regimes could capitalize on differences in phenology to promote desirable species (Augustine and McNaughton 1998). However, although some experimental studies have demonstrated the effectiveness of grazing in influencing competitive outcomes and promoting native biodiversity (Love 1944; Langstroth 1991; Dyer et al. 1996; Hayes and Holl 2003a; Marty 2005), the benefits are by no means universal (Chapter 20) and are sometimes shortlived (Dyer et al. 1996). Huntsinger et al. (Chapter 20) conclude that in the absence of more carefully designed tests of the impacts of grazing on competitive outcomes between native and exotic species, site-specific factors, including the species pool, land use history, and climate, make generalizations difficult.

Fire Prescribed burning has also been used in an effort to manage grasslands, particularly in the control of exotic species and the restoration of native biodiversity (Reiner, Chapter 18). Fire has the potential to alter competitive interactions because of its effects on standing vegetation, residual litter, seed survival, and seed germination. It can also substantially alter growing conditions by altering nitrogen and light availability (Chapter 18). To date, however, the effects of fire on the competitive interactions between various groups of species (e.g., annual grasses versus perennial grasses) has not been examined.

Seed Addition

Management Strategies to Alter Competitive Outcomes The re-establishment of populations of native species in habitats from which they have been eliminated is of tremendous practical importance. Indeed, it is the focus of the multimillion-dollar restoration efforts taking place in grasslands throughout California (Stromberg et al., Chapter 21). No less important are the theoretical implications to understanding population dynamics and community interactions. The low density or even absences of native species in the vast majority of grasslands (Heady 1988; Stromberg et al. 2001) is the net result of many ecological factors operating across the spatial and temporal scales that govern population sizes including seed dispersal, fecundity, competitive ability, and

166

ECOLOGICAL INTERACTIONS

Any discussion of the difficulty of reestablishing native populations in exotic-dominated habitats must begin with the low supply of seeds reaching appropriate habitats. The paucity of individual native grasses or forbs that produce viable seeds in many grasslands means that seed availability is a primary factor limiting native population growth rates (Baker 1989; Hamilton et al. 1999; Seabloom et al. 2003b). Even in grasslands where native species persist, native seedling recruitment is generally very low (Bartolome and Gemmill 1981). Experimental evidence has shown that artificial increases in seed supply increase the density and cover of native grasses and forbs. Hamilton et al. (1999) found that N. pulchra seedling density was five times higher in plots that received seed addition (5,000 seeds per m2) than in plots that did not.

Seabloom et al. (2003b) found that percent cover of perennial grasses (Bromus carinatus, Elymus glaucus, Nassella cernua, N. pulchra, and Poa secunda) increased in previously disked annual-dominated old-field communities following seed addition (500 seeds per m2). Seed addition studies using native annual forbs have found many species to be seed-limited in both native- and exotic-dominated grasslands (Robinson et al. 1995; Brown and Bugg 2001; Seabloom et al. 2003b; Gillespie and Allen 2004). A single seeding can lead to persistent increases in population density, although success is often adversely affected by the abundance of grass competitors. Assessing the degree of seed limitation in perennial species is complex because the expected lifetime fecundity of establishing individuals must be factored into any estimate of population growth rate. This is experimentally intractable for species, such as perennial grasses, that can live for decades or centuries. In lieu of direct measurements, we constructed a simple matrix model that evaluates long-term population growth rates (␭) from short-term seed addition experiments (Box 13.1; Table 13.2). We based our estimates of seedling survival (p12) on seven studies in which native perennial grasses were seeded into intact grasslands and tracked through their first year (Table 13.2). These studies were conducted in six different years and three counties spanning a twofold rainfall gradient from Santa Barbara County (⬃350 mm) to Yolo County (⬃650 mm). Survival ranged from 0–9% (median 6.6%) and did not differ significantly whether seeds were added to annual- or perennialdominated grasslands (Table 13.2). Our matrix model projections show that perennial grasses with a mean lifespan of at least 5 years should have positive population growth rates in 75% of empirical trials (Box 13.1; Figure 13.3). The two studies with the lowest seedling survival were replicated in subsequent years (Robinson 1971, Dyer et al. 1996). It is interesting to note that in both cases, seedling survival was sufficient for positive population growth rates in one of the years and probably insufficient in the other. In fact, Robinson (1971) measured the second highest (39%) and lowest (0%) survival rates in consecutive years at the same site. This high interannual variability suggests that seedling success is driven more by temporal than spatial variability. There are more likely to be good and bad years for seedling establishment than specific locations where seeds can never establish. This result suggests that multiple reseedings may be necessary to establish perennial grass populations. Population growth rates were sensitive to survival through the first year (p12), confirming the assumption that early survival represents a major bottleneck for perennial grass restoration. Population growth rates increased with adult lifespan (p34), although the effects were much less pronounced than those of survival through the first year (p12). Population growth rates were relatively insensitive to adult fecundity (f31), survival through year 2 (p23), and adult survival (p34). The model did not include density-dependent interactions, including interspecific competition, that are likely to occur after establishment.

Competition with annuals at the seedling stage has also been consistently implicated in the failure of perennial grass establishment. Indeed, once a seed reaches an appropriate habitat and germinates, the seedling must still compete with neighbors for resources before it can successfully establish and survive. In many invaded grasslands, dense stands of exotic annual seedlings restrict the access of native seedlings to light and belowground resources (see previous sections) (Langstroth 1991; Dyer et al. 1996; Hamilton et al. 1999; Brown and Rice 2000). Thus, competitive interactions likely limit establishment of native species in most exotic-dominated grasslands, perhaps in concert with seed limitation. Divittorio et al. (in press) suggested that existence of seed limitation of native species may be controlled by the density of exotic competitors. Using correlations between seed rain (native and exotic) and the establishment of native seedlings on experimental disturbances, they concluded that seed limitation of native species existed only when exotic seed supply and resulting interspecific competition was low. Even though native populations in exotic-dominated grasslands may respond to increases in seed supply because of low abundances of (native) reproductive individuals, the presence of numerous exotic seedlings likely shifts limitation to competition (Divittorio et al. 2007). This is similar to the conclusions of Hamilton et al. (1999) in that recruitment of N. pulchra was seed limited because of reduced local abundance, but competition for water was ultimately a stronger factor limiting the potential for native population growth. It is suspected that the rate of increase of native grass populations in many habitats is co-limited by both seed limitation and seedling competition. Treatments that overcome one limitation without considering the other are likely to be relatively unsuccessful. In contrast, restoration strategies that simultaneously increase the supply of native propagules (Box 13.1) while also reducing the competitiveness of exotic competitors are likely to be most effective (Chapter 21; Corbin et al. 2004). Further investigations that test the interactions between native seed addition and treatments that reduce exotic competitiveness, such as herbicide, grazing, or solarization, will help determine the relationship between factors limiting native establishment.

Summary Competitive interactions play an important role in determining the species composition and relative abundances of plant species in California grasslands. The dominance of exotic annual grasses, particularly in interior regions of the state, suggests that they are superior competitors versus perennial grasses and annual and perennial forbs under a variety of settings. The ability of annual grasses to germinate quickly, and their ability to effectively deplete soil moisture and light available to other types of vegetation, are some of the mechanisms that have been suggested to explain the annuals’ success. Perennial species, including both native and exotic grasses, fare better in

COMPETITIVE INTERACTIONS

167

competition with annual grasses in coastal habitats, where greater moisture availability during the summer drought leads to a longer growing season. Once established, perennial species in these coastal habitats have been shown to be able to significantly reduce growth of exotic annual competitors. Management strategies—including grazing, prescribed fire, and seed addition—have been applied in an effort to influence competitive outcomes in favor of native species. Thus far, no single strategy has proven successful in reducing the competitiveness of exotic species across a range of habitats. Most likely, a mixture of methods that includes the addition of native seeds along with treatments such as grazing or burning

168

ECOLOGICAL INTERACTIONS

that are designed to reduce competitors’ growth and survival will have the greatest likelihood of increasing native establishment and survival. Future research should examine such multiple strategies, while specifically addressing mechanisms that are likely to drive changes in competitive interactions between native and exotic species.

Acknowledgments We thank Janneke Hille Ris Lambers, Natasha Hausmann, and Carla D’Antonio for careful reviews that greatly improved this chapter.

FOURTEEN

Dynamics of Woody Species in the California Grassland C L AU D IA M. TYLE R, D E N N I S C. O D I O N, AN D RAGAN M. CALLAWAY

Though by definition grassland is dominated by herbaceous vegetation, woody species can be common associates within all five of California’s grassland vegetation types (Keeler-Wolf et al., Chapter 3). In addition, grassland, shrubland, and woodland communities often become intermixed where they overlap (Figure 14.1). Interior valley grasslands merge with oak woodland and chaparral in the foothills and with saltbush scrub in the southwestern edges of the San Joaquin Valley (Heady 1977). In the moist northwestern regions of the state, coastal prairie “exists in a continuum” with coastal scrub dominated by Baccharis and Lupinus (Ford and Hayes, 2005), and forms a mosaic with mixed evergreen forest (Sawyer et al. 1977). In the coastal areas of southern California, grassland extends into coastal sage scrub, where Artemisia californica and Salvia species are embedded in a grassland matrix (Mooney 1977). This patchwork of vegetation types is dynamic. Grasslands in California, as in many parts of the world, can replace or be replaced by shrub- or tree-dominated vegetation. The boundaries and relative abundance of grassland and woody vegetation in landscapes can shift over time and space. These patterns are influenced by both physical and biological factors, including fluctuating climate, disturbance, and changes in native and non-native herbivores (Sampson 1944; Biswell 1953; Naveh 1967; Bartolome et al. 1986; Hobbs and Mooney 1986; Callaway and Davis 1993; DeSimone and Zedler 2001). Climate, soils, and geologic substrate exhibit the broadest controls on the distribution of grasslands and co-occurring woody vegetation. Our conceptual model illustrates the general range of occurrence of California grassland and woody plant communities with respect to gradients in soil and moisture characteristics (Figure 14.2). Grasslands tend to occupy fine textured soils, but there is potential for woody vegetation in all but the finest textured soils and where climate and soil lead to the most extreme summer water deficit. On sites

where different community types are possible, what determines whether grassland or woody vegetation will dominate at a particular point in time? In this chapter we first describe the dynamics between grasses and shrub vegetation, with a focus on factors that lead to transitions between grassland and shrublands and the stabilizing feedbacks and reproductive processes that maintain them. We then discuss the dynamics at the grassland– oak savanna interface. We define oak savanna as grassland habitat with widely spaced oak trees, whereas oak woodland is composed of closed- or nearly closed-canopy stands of trees with sparse herbaceous vegetation in the understory. These habitats realistically form a continuum of oak canopy coverage. Here we focus on constraints over the recruitment of oaks into grassland settings, because this is an area of active interest to managers and scientists, and then we describe the ways in which oaks modify understory vegetation and productivity once they are established into grassland.

Factors Affecting Grassland–Shrubland Spatial and Temporal Dynamics Conversion of Shrubland to Grassland At the shrubland – grassland ecotone, expansion of grasses into adjacent shrubland or transition from shrub-dominated vegetation to grassland occurs primarily following disturbance. Under natural conditions, the disturbance responsible for this transition is fire (Callaway and Davis 1993), and this change is temporary, because shrubs recolonize the burned areas over time. However, large-scale, often long-term change from shrubland to grassland vegetation occurs as a result of anthropogenic disturbance, both intentional (e.g., brush clearing and type conversion) and unintentional (e.g., changes in fire frequency and indirect effects of nitrogen deposition).

169

F I G U R E 14.1. Mosaic of annual grassland, oaks, and coastal sage scrub (gray vegetation on slopes) on the University of California’s Sedgwick Reserve, Santa Barbara County. Photograph by C. Tyler.

I NTE NTIONAL C ONVE R S ION

In California, humans have converted substantial areas of shrubland to grassland, a process that began with Native Americans and continues today (Cooper 1922; Sampson 1944; Keeley 2002; Anderson 2005). One of the primary motivators for shrub removal has been to increase herbaceous forage for livestock (Burcham 1955). This goal has been accomplished through burning, herbicide, and brush crushing, cutting, and/or removal. Where coastal sage scrub and Baccharis shrubs are removed, either by occasional burning or brush cutting, this conversion may be relatively shortlived because these species are capable of readily recolonizing their former habitat (Zedler et al. 1983; DeSimone and Zedler 1999, 2001). The rate of transition from grassland back to shrubland may be slowed significantly by livestock grazing (Callaway and Davis 1993). Manipulating fire frequency has resulted in longer-lasting vegetation changes because herbaceous grasses and their seedbanks respond differently to fire regime than do woody species. While grasses can recover following fire and reach reproductive maturity in one growing season, the recovery of shrub vegetation is more gradual. It often takes 10–20 years for chaparral shrub canopies to redevelop (Keeley 2001), and it may take longer for seed banks to establish (Odion and Tyler 2002). Many chaparral shrubs rely on accumulating a persistent seed bank to recover from fire (Parker and Kelly 1989), particularly those that do not resprout (e.g., most species of Arctostaphylos and Ceanothus) and must grow from seed. Germination and high seed mortality associated with fire results in little or no carryover of the seed bank. Thus, new chaparral shrub seedlings may not emerge if a reburn occurs before shrubs begin reproducing and accumulating soil seed banks (Zedler et al. 1983; Haidinger and Keeley 1993). Both sprouting and nonsprouting shrub life history types are sensitive to recurrent fire (Zedler et al. 1983; Haidinger and Keeley 1993). All young seedlings established after fire are typically killed by an early reburn. In addition, resprouting adults (e.g., Adenostoma fasciculatum) often have very high mortality if reburned within 1–2 years (Hedrick

170

ECOLOGICAL INTERACTIONS

F I G U R E 14.2. Conceptual model of the general range of soil texture and water balance supporting grassland and associated woody vegetation that may co-occur with grasslands in California (based on information in Barbour and Major 1977 and Jackson and Bartolome 2002).

1951; Zedler et al. 1983). The case in which evergreen chaparral vegetation has been replaced by annual grassland has commonly been referred to as type conversion, reflecting the persistent nature of the change. Drought-deciduous coastal sage species have a higher capacity for recolonizing burned grassland, because they disperse germinable seed away from parent plants. However, if fire returns frequently enough, type conversion of coastal sage scrub to grassland can also be maintained (Eliason and Allen 1997; Cione et al 2002). Native Americans and range managers have relied upon the relative vulnerability of different California shrublands to type-convert them to grasslands (Sampson 1944; Hedrick 1951; Keeley 2002). Range managers often accomplished this by seeding non-native annual grasses following fire and then deliberately reburning when the cured grass would carry fire.

U N I NTE NTIONAL C ONVE R S ION

Accidental or unintentional conversion of shrubland to grassland is a widespread dynamic generally affecting the most fire-prone areas in California (Keeley 2001). An increase in anthropogenic ignitions is associated with population growth, particularly in the mountains of southern and central California (Keeley et al. 1999), and serves as a catalyst for grassland expansion into chaparral or coastal sage scrub habitat in these areas. This process of type conversion now commonly occurs without deliberate range management practices as a result of increases in and positive feedbacks among anthropogenic ignitions, exotic annual grasses, and frequent fire (Mack and D’Antonio 1998; Keeley 2001, 2006; Brooks et al. 2004). Introduced Mediterranean grasses increasingly escape into burned shrublands where the landscape has become fragmented by annual grassland or mixed habitat–supporting grasses. The introduced grasses can be very successful colonizers (D’Antonio et al., Chapter 6). They provide readily

ignitable fuel, which allows for reburns sooner than the natural shrub vegetation would normally be expected to burn. This process leads to complete type conversion if the burn interval is 1 to 2 years (Zedler et al. 1983; Haidinger and Keeley 1993). If the burn interval is longer, but the shrub canopy has not yet redeveloped, an incomplete type conversion may occur and grasses and shrubs may codominate. The occurrence of another fire that completes the conversion is then more likely (Keeley 2001). With less shrub vegetation in the grass/shrub mix, fire severity may be low enough to allow for much greater survival of grass seeds, which may not survive the high-intensity fire that typifies shrublands (Odion 2000; Keeley 2001, 2006). Thus, reduction in fire intensity is an additional important stabilizing feedback that occurs with frequent fire and conversion toward annual grassland (Keeley 2001, 2006). Cattle grazing in shrublands disperses grass seed, and trampling can open the shrub canopy, reducing fire intensity. Thus, livestock may promote non-native annual grasses at the expense of shrubs, particularly if grazing is compounded with fire (Zedler et al. 1983; Callaway and Davis 1993; Keeley 2001, 2006). Once grassland is established, mechanical damage to shrub seedlings from livestock and their late-season utilization of slightly palatable shrubs (Sampson and Jesperson 1963) can help stabilize and maintain annual grasslands (Keeley 2005). Another factor contributing to type conversion of coastal sage scrub vegetation in southern California is anthropogenic nitrogen deposition. Even in wildlands protected from agriculture and development, coastal sage scrub is being replaced by non-native annual grasses in many areas (Allen et al. 1998). These ecosystems have undergone significant changes in N fertility that are likely contributing to these changes in vegetation composition (Padgett et al. 1999). As has been documented in other systems in the western United States (Fenn et al. 2003a), unusually high N concentrations enhance weed invasions by contributing to the growth of annual grasses in coastal sage scrub (Allen et al. 1998; Padgett et al. 1999). In addition, N deposition can indirectly alter the fire regime by increasing the understory fuel load. This could promote more frequent fire, further contributing to the conversion of coastal sage shrubland to annual grassland.

Establishment of Shrubs into Grassland In direct contrast to non-native annual grass invasion into shrublands, the establishment of native shrubs in grasslands occurs in the absence of fire or grazing, suggesting that the persistence of many grasslands may be disturbance-dependent (Keeley 2005). Large-scale colonization of shrubs into formerly grazed grasslands is common in some parts of the San Francisco Bay area (McBride and Heady 1968; McBride 1974; Edwards 2002; Keeley 2005). In this case, it is believed that the shrubs are recolonizing former shrubland habitat largely as a result of the cessation of livestock grazing; the evidence

suggests that in the past wildfires were not frequent enough in these areas to maintain them as grasslands, though shrub recolonization may have been prevented through cultural use of fire (Keeley 2005).

DI F F E R E NCE S AMONG S H R U B LAN D T Y P E S

Shrub establishment into grasslands generally does not involve chaparral species. The presence of a dormant soil seed bank, stimulated to grow by high-severity fire, is particularly important to the establishment of many chaparral shrubs (Zammit and Zedler 1988, 1994; Parker and Kelly 1989). Even if seed of these chaparral shrubs is dispersed into grassland, it may never receive the appropriate stimuli from heat and combustion byproducts required for germination (Keeley 1991). The chaparral shrub species that do have readily germinable seed dispersed far from parent plants, such as Cercocarpus betuloides or Quercus dumosa, are also known to establish primarily in mature chaparral (Keeley 1999). Instead, shrub colonization generally occurs where grasslands interface with either summer-drought-deciduous coastal sage scrub (e.g., Salvia spp., Artemisia californica), primarily toward the southern part of the state (Callaway and Davis 1993; DeSimone and Zedler 1999, 2001; Cione et al. 2002), or evergreen Baccharis scrub, primarily in the central and northern part of the state (McBride and Heady 1968; Hobbs and Mooney 1986). The mechanisms underlying the establishment, or reestablishment, of both of these shrub vegetations have been investigated, elucidating some important differences. Drought-deciduous coastal sage scrub species are light-demanding and have been found to establish primarily in openings caused by disturbance, especially gopher mounds (DeSimone and Zedler 2001). In contrast, survival of evergreen Baccharis seedlings through the summer drought may depend on late spring root growth, which benefits from late spring rainfall; colonization is correlated more with high-rainfall years, particularly in sites that are more moisture-limited (Williams et al. 1987, Williams and Hobbs 1989). Da Silva and Bartolome (1984) suggested that wetter soils are more suitable habitat for Baccharis, and its colonization may be less limited in relatively moist environments (Williams and Hobbs 1989). Soil disturbance may not be a necessary cofactor for establishment of Baccharis scrub (but see Da Silva and Bartolome 1984). There is a relatively recent phenomenon of invasion of grassland by non-native shrubs in relatively high-rainfall areas (e.g., north coastal prairie). In this case, invasion is by nitrogen-fixing, leguminous shrubs such as gorse (Ulex europaea), lupine (Lupinus arboreus), and, especially, French broom (Genista monspessulana) and Scotch broom (Cytisus scoparius) (Odion and Haubensak 2002; Alexander and D’Antonio 2003b; Keeley 2006; D’Antonio et al., Chapter 6). Grazing disturbance or lack of fire is not required for these exotic species invasions. In fact, fire may help promote invasion by brooms because it results in a flush of germination in grassland adjacent to broom patches (Odion and

DYNAMICS OF WOODY SPECIES

171

Haubensak 2002). However, even in the absence of fire, broom invasion occurs as a gradually advancing front or series of small foci. Despite important differences among the shrubs that frequently colonize California grasslands and the conditions that facilitate them, there are a number of traits in common among these shrubs. Many produce large numbers of seeds that are capable of at least limited dispersal into grasslands (DeSimone and Zedler 2001; Odion and Haubensak 2002; Alexander and D’Antonio 2003b). For example, seed production by three coastal sage scrub species was much greater than found for chaparral shrubs such as Ceanothus greggii (Zammit and Zedler 1993), and species of manzanita, Arctostaphylos (Keeley 1977; Tyler et al. 1998). In addition, although predation may be high under shrubs, seed predators do not appear to play an important role in limiting shrub colonization in grasslands (DeSimone and Zedler 2001). Seed germination in these colonizing shrubs may be increased by fire or other disturbance factors but can also occur under a variety of commonly occurring environmental conditions (Hobbs and Mooney 1986; Keeley 1991; Odion 2000). For example, Artemisia californica has light-stimulated germination (Keeley 1991; Eliason and Allen 1997), and French broom produces considerable seed that germinates readily under field conditions (Parker and Kersnar 1989; Odion and Haubensak 2002).

Stabilizing Feedbacks Maintaining Shrublands or Grasslands Although fire and grazing are the major forces causing transitions between grasslands and shrublands, they can also act to stabilize either state. As described earlier, grasslands can be maintained with frequent fire and/or intense grazing, and conversely, shrubland can be stabilized with livestock exclusion (Callaway and Davis 1993; Keeley 2001, 2006). Other mechanisms also provide stabilizing feedbacks. Once shrubs establish in grasslands, they change site conditions, which may then favor woody species over grassland herbs. Herbivores, such as brush rabbits, take shelter under shrubs and can reduce establishment and growth of herbaceous seedlings in the immediate vicinity of adult shrubs and in the ecotone (Connell 1954; Bartholemew 1970; Hobbs and Mooney 1986; Tyler 1995, 1996; DeSimone and Zedler 2001). In addition, volatile allelopathic compounds produced by aromatic species affect soils and may inhibit seed germination around shrubs (Muller 1966; Halligan 1973, 1976). Similarly, nonvolatile allelopathic chemicals wash into soils from the evergreen leaves of chaparral shrubs such as Adenostoma fasciculatum (Christensen and Muller 1975). Low light levels may also inhibit grass growth under shrubs (Keeley 1999). The distinctive bare zone often observed next to these shrubs, particularly the sages, has been attributed to one or more of these mechanisms (Muller 1966; Halligan 1973), although their relative importance has been debated (reviewed by Halsey 2004). Thus, in addition to infrequent fire, the maintenance of the shrub canopy

172

ECOLOGICAL INTERACTIONS

provides shrubland-stabilizing feedbacks (Callaway and Davis 1993; Keeley 2001, 2006). Exotic shrubs also alter soil conditions that initially stabilize exotic shrubland but may eventually favor exotic herbs. Brooms and bush lupine are nitrogen fixers, and they have been found to considerably enhance soil nitrogen levels (Maron and Connors 1996; Haubensak et al. 2004; Haubensak and Parker 2004). Interestingly, however, Haubensak and Parker (2004) observed decreased growth of a bioassay herb grown in soils that had been invaded by Scotch broom. They suggest that Scotch broom may enhance soil nutrient status, but that allelopathic compounds produced by the shrubs are responsible for its net negative effect on the growth of some herbaceous species. However, the full implications of soil changes under various brooms, bush lupine, and other N-fixing species are not yet understood. Many N-fixing shrubs are short-lived or are the targets of removal efforts. They appear to promote establishment of weedy species after their death or removal, because these species can exploit nutrient increases more readily by means of their rapid growth potential (Maron and Connors 1996). Commonly, these species are exotic annual (Maron and Connors 1996) or perennial grasses (Haubensak et al. 2004; Haubensak and Parker 2004). Thus, a consequence of invasion by exotic shrubs is that site conditions may change to be more persistently favorable to other exotic species when shrubs are removed or die (Haubensak et al. 2004), thereby maintaining a matrix of both woody and herbaceous non-native species. Gopher disturbance interacts with other factors to alternately stabilize or destabilize grasslands. Seabloom and Richards (2003) found that in a matrix of native perennial species and non-native annual grasses, pocket gophers preferentially forage in areas dominated by the annuals, and that gopher foraging activity increases the abundance of the annual grassland forbs. Thus, gophers may help contribute to the maintenance of annual-dominated patches within a grassland or grass/shrub matrix. On the other hand, in the absence of livestock grazing, shrub establishment can be facilitated by gopher disturbances if coastal sage scrub species are within dispersal distance (DeSimone and Zedler 1999, 2001). In addition, gopher disturbance has been found to be greater in ungrazed areas than in grazed areas (Stromberg and Griffin 1996). The possibility that gopher activity initially increases following the removal of cattle may contribute to the relatively rapid colonization of shrubs into areas newly released from livestock grazing — a common observation (McBride and Heady1968; Keeley 2005). Therefore, gopher disturbance may act to either stabilize grassland or facilitate transition to shrubland, depending on other factors such as livestock grazing, fire, and proximity to coastal sage scrub.

Summary In many cases, transitions from shrublands to grasslands have resulted from deliberate management actions, usually

involving frequent fire, livestock grazing, and brush removal. In addition, humans have unintentionally type-converted substantial areas of shrubland to grassland by altering fire regimes in former shrublands through increased ignitions or increased abundance of grass fuels. Colonization of shrubs (generally nonchaparral species) into grassland habitats, and the return of shrubs to former shrubland habitats, are also commonplace, often resulting from an absence of deliberate management (Wells 1962; McBride and Heady 1968; Hobbs and Mooney 1986; Keeley 2005). When shrubs establish in grassland, they can significantly alter the site by providing shelter for herbivores, changing soil conditions, and altering the fire regime. These and other disturbance-related mechanisms may stabilize grass-dominated and shrub-dominated vegetation.

Factors Affecting Grassland: Oak Savanna Dynamics We focus our discussion of tree–grassland interactions on oak savanna for several reasons. First, oak-dominated communities are the most extensive of the hardwood associations that interface with coastal and valley grassland in California. Savanna trees are mainly blue oak (Quercus douglasii), valley oak (Q. lobata), coast live oak (Q agrifolia), interior live oak (Q. wizlezenii), and occasionally Oregon white oak (Q. garryana) in the northern part of the state. Blue oak, valley oak, and coastal oak woodland and savanna combined cover more than 18,000 square kilometers of the state, with blue oak-foothill pine woodland adding another 14,400 square kilometers (Davis et al. 1998). All of these oak species provide important wildlife habitat. Another reason to direct our attention to oaks is that the vast majority of oak habitat is either adjacent to grassland or intertwined with patches of grassland species in the oak understories and in the gaps between trees. Within savanna settings many adult trees are dying and not being replaced. Hence there is concern over the future of oaks within grassland settings. Finally, oaks have been well studied, and there is much information about interactions between oaks and herbaceous species. Undoubtedly, other tree associations interact with grasslands where their communities border one another. For example, in moist coastal regions of northern California, Douglas fir (Pseudotsuga menziesii) is reported to be expanding into grassland adjacent to mixed evergreen forest (Kennedy and Diaz 2005; Kennedy and Sousa 2006), probably because of reduced fire frequency over the past century (Arno and Gruell 1986). However, this phenomenon is not widespread, and the interactions between these tree and grassland communities have not yet been well described in the literature. The interface of grassland and savanna differs from that of grassland and shrubland in several general ways. With the exception of large-scale anthropogenic disturbances such as tree clearing, the dynamics and transitions between grassland and savanna occur more gradually, reflecting the slower recruitment rates of trees compared to shrubs. In addition,

while the boundaries or ecotones separating grassland from shrubland can often be quite distinct, comparable boundaries between grassland and savanna can be more difficult to identify. Savanna is generally defined as grassland containing widely spaced trees, but from the perspective of the herbaceous species it may be difficult to delineate where savanna begins and grassland ends, since the understory is usually dominated by herbaceous forbs and grasses. Even oak woodlands sometimes have grassy understories, and transitions from open grassland to savanna to an almost woodland setting can occur very quickly where boundaries are difficult to delineate. Oaks exert strong controls on understory vegetation composition and productivity, which in turn create distinctive patches of herbaceous vegetation, in contrast to the surrounding open grasslands. Therefore, the occurrence of oaks greatly influences spatial patterns in the herbaceous understory. However, even when climate and other physical conditions are suitable, recruitment processes of oaks into open grassland environments may be extremely limited in space and time as a result of many natural and anthropogenic factors operating independently or in concert (reviewed by Tyler et al. 2006).

Factors Limiting Recruitment of Oaks into Open Grassland AC OR N P RODUCTION, P R E DATION, AN D DI S P E R SAL

Acorn crop varies annually and among individual trees (Griffin 1976; Koenig et al. 1994; Sork et al. 2002). In some years few acorns develop, and these may be taken from the tree by woodpeckers or jays or from the ground by other foraging birds and mammals (Griffin 1976). Koenig et al. (2002) reported that in years with low acorn production, up to 100% of valley oak acorn crop may be removed while still on the tree. Significant establishment of oak seedlings is undoubtedly limited to years with heavy acorn crops; in “mast” years, thousands to tens of thousands of mature acorns may be produced by a single adult oak (Johnson et al. 2002). However, in general the vast majority of acorns are consumed by acorn predators, including cattle, pigs, deer, rodents, pigeons, crows, woodpeckers, magpies, and jays (Griffin 1976; Rossi 1979; Griffin 1980a; Hibberd 1985; Borchert et al. 1989; Tietje et al. 1991). Many of those acorns that remain on the soil surface succumb to heat or desiccation. Dispersal of acorns into grassland or shrubland would be rare without the intervention of animals. Acorns drop on the ground directly under the tree canopy or near the canopy edge, except for those that roll downhill or, in riparian oak woodlands, are carried downstream. As observed by naturalist and ornithologist Joseph Grinnell, expansion of oaks, especially upslope, or into new habitats must rely almost entirely on their bird and mammal associates (Grinnell 1936). Although many animals function only as acorn predators, some, notably western scrub jays (Aphelocoma coerulescens) and yellow-billed magpies (Pica nuttalli) serve as

DYNAMICS OF WOODY SPECIES

173

vital agents of dispersal because they scatter-hoard, and thus essentially plant, acorns (Griffin 1971, 1976; Carmen 1988). The acorns that are shallowly buried in the soil or under leaf litter by these birds have a much greater chance of surviving and germinating than those left on the soil surface (Griffin 1971; Borchert et al. 1989). A single western scrub jay may cache up to 5,000 acorns in a season (Carmen 1988) but relocate and consume only a portion of these. Interestingly, distribution and diversification of jays and oaks are highly correlated. The areas of the world most noted for high variety of oaks are also those of high jay diversity, and several characteristics of both jays and oaks appear to suggest a coevolved, symbiotic relationship (Bossema 1979, Keator 1998).

S HO OT P R E DATION, H E R B IVORY, AN D B R OWS I NG

Other important factors limiting recruitment of oaks in general, as well as their expansion into grassland, are shoot predation, herbivory, and browsing. Burrowing animals such as pocket gophers and voles, which consume plant roots and young shoots, cause significant mortality to oak seedlings and young saplings (Griffin 1971, 1976; Adams et al. 1987, 1997; Davis et al. 1991; McCreary and Tecklin 1997; Bernhardt and Swiecki 1997; Tyler et al. 2002). In acorn planting experiments Davis et al. (1991) found that the proportion of blue oak seedlings surviving three years was reduced by half (22% vs. 44%) for seedlings not protected from gophers compared to those protected. Even higher rates of damage have been reported for valley oak seedlings; gophers were responsible for up to 90% of the mortality of planted seedlings (Adams and Weitkamp 1992). Tyler et al. (2002) investigated factors limiting seedling establishment of valley oak and coast live oak and found that maximum rates of seedling emergence were 71% in valley oak and 85% in coast live oak in locations protected from birds and mammals. In locations open to all seed and seedling predators, maximum emergence rates were significantly lower: 30% in valley oak and 32% in coast live oak. Bark girdling by rodents is another source of mortality for oak seedlings and young saplings (Davis et al. 1991). Browsing by deer (White 1966; Griffin 1971) and defoliation by insects (McCreary and Tecklin 1994) further limit the survival and growth of young oaks. Some oaks, particularly blue oaks, may remain shrublike, hedged by repeated browsing, for decades. Harvey (1989) found blue oak “saplings” that were less than 4 feet (1.2 meters) high ranging in age up to 100 years old. Some oak species may be able to withstand herbivory better than others. Moderate clipping treatments that simulated defoliation by insects and light herbivore browsing were found to cause greater mortality in coast live oaks than in blue oak seedlings, suggesting that coast live oak seedlings may be especially vulnerable to such attacks (Muick 1995). Livestock, present in much of the state’s grasslands and woodlands, similarly browse young oak seedlings and

174

ECOLOGICAL INTERACTIONS

saplings present in grassland, with additional direct (e.g., trampling) and indirect (e.g., soil compaction) effects. Ranchers have long used “animal impact” to prevent the invasion of shrubby vegetation into rangeland. In fact, cattle grazing has been implicated as being one of the main factors responsible for poor oak recruitment in rangelands. However, the relative importance of browsing by livestock vs. natural herbivores such as deer in limiting natural oak recruitment remains unclear and may be site- or region-specific, as studies have reported conflicting findings. Hall et al. (1992) conducted planting experiments and found that establishment of blue oak seedlings in controls, which excluded cattle but allowed access to deer and other native grazers, was no different from plots grazed by livestock, suggesting that wildlife had an equivalent effect on oak recruitment. Patterns of natural oak establishment in relation to grazing have been mixed. Livestock grazing has been reported to be negatively correlated with the presence of blue oak seedlings (Standiford et al. 1997) and saplings (Swiecki et al. 1997). In contrast, Muick and Bartolome (1987) found that statewide patterns of oak recruitment could not be explained by the presence or absence of cattle grazing. In addition, the removal of livestock has not resulted in increased levels of oak recruitment at some sites, even after many decades (e.g., blue oaks, White 1966; valley oaks, Callaway 1992). Timing of grazing may influence the outcome of livestock on oak establishment and survival (Hall et al. 1992). A meta-analysis across six studies found that cattle grazing effects on oak seedlings were neutral or positive when grazing occurred earlier in the year but neutral or negative when grazing was only in spring and summer (E. S. Zavaleta and K. B. Hulvey, in preparation). Seedlings in open grassland may be especially vulnerable to browsers. Throughout the summer drought period after grassland annuals have senesced, oak seedlings and saplings may be the only green vegetation still available to mammals within many grassland and savanna habitats and thus may be subject to severe attack. In investigating the impact of season of grazing in affecting establishment of blue oak seedlings, Hall et al. (1992) found that seedlings exposed to spring and summer grazing had significantly more damage and lower survivorship than those exposed only to winter grazing. This pattern of stronger browsing impacts in the summer months is probably also true for other herbivores including deer, rabbits, and mice. The strong associations of natural oak seedlings with shrub canopies reported for coast live oak (McBride 1974, Callaway and D’Antonio 1991) and blue oak (Callaway 1992) are likely a result of protection provided by shrubs from shoot herbivory. Callaway and D’Antonio (1991) surveyed naturally occurring coast live oak seedlings and found that the vast majority were under shrubs and that these were significantly less browsed than oak seedlings in surrounding open grassland. In field experiments Callaway and D’Antonio (1991) determined that acorns planted under shrubs had high emergence and survival rates (up to 55% survival to year 2), whereas no seedlings survived from acorns planted in the

open. Herbivory by deer was the suspected cause of mortality of nearly 40% of those dying in the open, because seedlings were removed with no evidence of soil disturbance. Shrubs did not provide protection from acorn predators for blue oaks in similar studies (Callaway 1992); emergence was lower for acorns planted under shrubs, because of excavation by gophers there. However, as just mentioned, percent survival of emerged shoots was significantly higher under shrubs than in the open, and the majority of mortality in the open was due to herbivory above the soil surface, most likely by deer. In addition to providing safe-sites by protecting from deer, shrubs also provide shade. In experiments conducted in a dry year with only 50% of average rainfall, the combination of these facilitative mechanisms, as provided by caging and shading, led to highest survival in blue oak seedlings (Callaway 1992). However, shade provided by shrubs may be a detriment to establishment of oak species that require high light levels, such as valley oak (Callaway 1992). Rock outcrops may similarly serve as natural safe sites for oak recruitment (Snow 1972), by providing shade and providing some protection from acorn predators and herbivores. Establishment of oak seedlings in grassland may be higher in the vicinity of shrubs, not only because of the protection from browsers but also because birds such as jays and magpies may be more likely to scatter-hoard acorns there than in open grassland (Bossema 1979). Thus, establishment of some oak species, such as Q. agrifolia, into adjacent grassland may be more likely to occur following invasion by shrubs. Callaway and Davis (1998) found that over a 50-year period, recruitment of Q. agrifolia in a central coastal California site was relatively high in coastal sage scrub vegetation and chaparral and very low in grassland.

C OM P ETITION W ITH G RAS S E S

One of the hypotheses commonly proposed to explain the low rates of oak seedling establishment observed in California is that grasses, particularly the pervasive exotic annual species, are fierce competitors for limited resources such as water. Although the mechanisms of interaction between oak seedlings and herbaceous plants were not discerned in his study, Griffin (1971) found that oak seedling survival was significantly higher when grown without competition from grasses. In savanna plots fenced to exclude large herbivores, 31% of valley and coast live oak acorns, and 88% of blue oak acorns produced seedlings that survived 3 years within the areas cleared of the herbaceous layer, whereas no seedlings emerged within the uncleared section of the plot. There is evidence that introduced annual grasses may limit oak seedling establishment more than native perennial species would. Seedling emergence, growth, and survival of blue oak and valley oak were significantly reduced when grown with exotic annuals as neighbors versus grown with no neighbors (Gordon et al. 1989; Danielsen 1990; Gordon and Rice 1993). In addition, oak seedlings grown with exotic annual grasses exhibited reduced emergence and growth

rates compared to those grown with native perennials as neighbors (Danielsen 1990; Welker et al. 1991; Gordon and Rice 2000). The negative effects of annual grasses were attributed to their reducing soil moisture more rapidly than perennial neighbors did. The roots of annuals are shallower than those of perennials and are denser within the topsoil layers (Holmes and Rice 1996; Hamilton 1997b). It has been hypothesized that the exotic annual grasses deplete near-surface soil water early in the growing season, leaving emerging oak seedlings less water; in contrast, oak seedlings growing among native perennials have access to soil moisture long into the summer. If competition with annual grasses were the main factor limiting oak recruitment into grasslands, we might expect to find higher rates of establishment into native bunchgrass stands adjacent to oak woodlands. To our knowledge this has not been observed. It is likely that the relationship between acorn dispersal, germination, and seedling establishment in relation to grasses is complex and that it varies among sites and with the composition of the herbaceous species. For example, among the exotic annuals some species, such as Bromus diandrus, which can form thick monocultures following removal of cattle grazing, may be more limiting to oak recruitment than are other annual species.

F R EQU E NT F I R E

Human-caused fire has been a force shaping oak woodlands and savannas for centuries. Native Americans used fire to improve plant growth for themselves, facilitate acorn gathering, attract game animals, and reduce fire hazard (Biswell 1989). Jepson (1910) and Cooper (1922) speculated that the open savannas observed and described by early settlers were a result of frequent burning by Native Americans. However, Griffin (1977) points out that there is little information available on the extent and frequency of burning by Native Americans, with the exception of accounts related to the coastal areas and parts of the Central Valley. In the early American settlement period, fires set by ranchers and miners were a frequent occurrence (Griffin 1977; McClaran and Bartolome 1989b; McCreary 2004). Stephens (1997) reported that the fire return interval in the oak-pine forests of the Sierra foothills was 8 years during the period between 1850 and 1950. Although this research was based on fire scars, which may be caused by lightning, the use of fire by ranchers undoubtedly contributed to an increase in frequency of burning (McCreary 2004). Natural, or lightning, fires in oak savannas were probably much less frequent, and on the order of many decades (Griffin 1977). Although burning has been cited by some as facilitating oak recruitment, most evidence suggests the contrary. Although adults and established saplings of some oak species are capable of resprouting following low-intensity fire (Griffin 1980), frequent or severe fire kills seedlings and may cause sapling mortality, in part by prolonging the period during which young oaks are susceptible to subsequent fires and other

DYNAMICS OF WOODY SPECIES

175

damaging agents (Swiecki and Bernhardt 2002). Swiecki et al. (1997) reported that frequent fire was negatively associated with blue oak sapling recruitment. Additional support for the hypothesis that fire reduces oak recruitment is provided by Mensing (1998), who examined pollen deposits in Santa Barbara County and found that lack of fire correlated significantly with an increase in coast live oak in the last 100 years. S U M MARY

Many physical and biotic factors limit establishment of oaks into grasslands. Multiple hurdles must be overcome before a single oak sapling will appear in an open savanna. Following a mast year, in which predators do not consume all available acorns, a viable acorn must be planted or otherwise find itself in a suitable microclimate to survive desiccation and germinate. A good rain year and escape from root and shoot predation will result in an emerged seedling. That seedling must withstand competition from neighboring grasses or shrubs, endure summer drought, and survive repeated herbivory and browsing for years, perhaps decades. Any one of these factors can prevent the establishment of seedlings and the subsequent transition to sapling and tree.

Effects of Oaks on Grasslands Once established, oaks, like many other shrubs and trees, have strong effects on the grassland matrix in which they grow. Oaks alter understory species composition, community productivity, soil fertility, soil structure, and soil moisture. These effects can be manifest as competition, in which other species are suppressed by oak canopies, or as facilitation, in which other species benefit from oak canopies. Variation from competition to facilitation can be caused by environmental differences, with the general trend appearing to be competitive effects when abiotic conditions are stressful, and neutral to facilitative effects when conditions are relatively benign (Ratliff et al. 1991; McClaran and Bartolome 1989a; Callaway and Davis 1998). However, the competitive and facilitative effects of oaks on other species also appear to be determined by complex factors such as tree age, root architecture, soil conditions, and the composition of the community they are interacting with (Callaway et al. 1991; Scholes and Archer 1997; Rice and Nagy 2000).

C OM P LE X I NTE RACTION S AF F E CTI NG U N DE R STORY P RODUCTIVIT Y

The importance of these complex factors is well illustrated by the historical management of Quercus douglasii woodlands. In the 1950s, 1960s, and 1970s, range managers in California advocated clearing of Q. douglasii because it was suspected that the tree reduced forage for livestock (Johnson et al. 1959; Murphy and Crampton 1964; Murphy and Berry 1973). Thousands of acres were affected by hand cutting, bulldozing, and aerial spraying of Q. douglasii (Leonard 1956;

176

ECOLOGICAL INTERACTIONS

F I G U R E 14.3. Annual productivity of understory plants in savanna woodland under Quercus douglasii trees with dense surface roots (negative trees), without dense surface roots (positive trees), and in the open grassland matrix surrounding the trees. Data are for 0.125 m2 quadrats sampled along transects aligned due north from the trunks. Error bars are 2 SE. Redrawn from Callaway et al. (1991) with permission from Ecology.

Murphy and Crampton 1964; Murphy and Berry 1973). However, in the late 1970s V. L. Holland demonstrated that many Q. douglasii trees and stands did not suppress understory productivity, and in fact some canopies were associated with a tremendous increase in understory biomass (Holland 1980). This facilitative effect on productivity under the canopies was attributed to much higher levels of soil nutrients (Holland and Morton 1980), a hypothesis supported in later experiments and descriptive studies conducted by others (Kay 1987; Callaway et al. 1991; Dahlgren et al. 1997). Variation among sites in the effects of Q. douglasii has been reported in other studies (McClaran and Bartolome 1989a; Ratliff et al. 1991; Connor and Willoughby 1997), and even within a single site among neighboring trees (Callaway et al. 1991). In the latter study, contrasting positive and negative effects of Q. douglasii were found to occur within very local spatial scales, with adjacent trees exhibiting opposite effects (Figure 14.3). Callaway et al. (1991) collected soil from under trees eliciting these different effects and found that soil nutrients were much higher under all trees. However, fine root biomass near the soil surface was 5 times greater under trees with competitive effects (“negative” trees) than under trees with facilitative effects (“positive”) trees. Furthermore, the predawn xylem pressure potentials of “negative” trees were much lower at the end of the long, dry summer than those of “positive” trees, indicating that the latter utilized a more permanent soil water source. Root exclosures reduced the competitive effects of Q. douglasii roots under “negative” trees, but not under “positive” trees. In sum, greenhouse bioassay experiments and cross transplants of soil in the field indicate that while both “positive” and “negative” trees enrich the nutrient content of the soils beneath them (a positive effect), dense surface roots of “negative” trees result in their net negative competitive effects. The idea that facilitative soil fertility and competitive root interference mechanisms operate simultaneously is supported

30

Percent difference in productivity (1987/1988)

by a re-examination of Kay’s (1987) large-scale removals of Q. douglasii, and 15-year post-removal surveys. Kay found that productivity under intact Q. douglasii canopies was consistently lower than in surrounding open grassland, suggesting competitive effects. However, removing the oaks generally raised productivity to levels significantly higher than in the open, suggesting that understory soils had high potential fertility. By the end of the experiment the higher-productivity at sites where oaks had been removed had disappeared, perhaps because nutrients previously deposited by the oaks had been exhausted. The positive effect of Quercus douglasii canopies on understory grass productivity has been attributed primarily to the way that canopy throughfall and litterfall increase soil nutrients near the trees (Holland 1980; Holland and Morton 1980; Callaway et al. 1991; Dahlgren et al. 1997). More recently, however, Ishikawa and Bledsoe (2000) observed gradual increases in soil water potential at night and rapid decreases during the day in soils under Q. douglasii trees. These diurnal fluctuations in water potential are indicative of hydraulic lift. Hydraulic lift is discernable only in relatively dry soils; when soils are wet, their high water potential relative to the tree roots does not allow water to passively move into the soil. Hydraulic lift may play a role in the general facilitative effect of Q. douglasii trees, but diurnal patterns do not develop until later in the summer, when the annual grasses have senesced. However, water released by Q. douglasii roots could delay the rate of soil water depletion and increase the growing season for the annual understory species. Competitive and facilitative effects of oak canopies on the understory vary to some degree with abiotic conditions. For example, Ratliff et al. (1991) measured herbaceous productivity under oaks and in open grassland for 8 years and in many different habitats. Though on average Q. douglasii increased productivity across all times and sites, in mesic swales, where annual productivity was the highest, the effect of Q. douglasii was negative, decreasing productivity. Similarly, McClaran and Bartolome (1989a) found that the effect of Quercus douglasii on understory productivity (relative to open grassland) varied along a rainfall gradient at five sites; Q. douglasii canopy effects were primarily neutral at more xeric sites and negative at mesic sites. They found no evidence for positive effects. Connor and Willoughby (1997) also found that the effect of Q. douglasii on understory productivity tended to vary among years that differed in precipitation, though the variation was not significant over a 5-year period. Overall, results from several studies suggest that the effects of oak canopy on the understory is positive (facilitation) where productivity is lower (more xeric upland habitats), and oak canopy effects are negative (competition) where productivity is higher. In addition, the impact of rainfall on understory productivity varies with site productivity. A reanalysis of McClaran and Bartolome’s results showed that productivity in the fertile soil under the canopies increased ⬃45% in an above-average rainfall year, whereas in the open, productivity increased only ⬃25%. Likewise, a reanalysis of Callaway et al. (1991) and unpublished data

25

20

15

10

5 6

7

8

9

10

11

12

13

Total soil nitrogen (mg/g soil) F I G U R E 14.4. Relationship between total soil nitrogen beneath Quercus douglasii trees and the proportional difference in understory productivity in a wet year (1987) and a dry year (1988). Data are means of understory productivity and total soil nitrogen (0 – 10 cm) in the subcanopy soils for each of 12 Quercus douglasii trees (n ⫽ 5 per tree). Percent difference ⫽ ([(subcanopy ⫺ open)/open in 1987] – [(subcanopy ⫺ open)/open in 1988]) ⫻ 100. R2 ⫽ 0.61, P ⬍ 0.05.

(R. M. Callaway) indicates that Quercus douglasii trees with higher soil nitrogen show the highest productivity response to increased rainfall (Figure 14.4). OAK E F F E CTS ON U N DE R STORY S P ECI E S C OM P OS ITION

Most studies of oak effects on understory composition have been conducted on either the deciduous oak, Q. douglasii, or on the live oak, Q. agrifolia (Holland 1973; Parker and Muller 1982; McClaran and Bartolome 1989a; Callaway et al. 1991; Marañón and Bartolome 1993; Rice and Nagy 2000). To our knowledge a comparative study of the effects of these species on composition in the same location has not been conduced. Ordination analyses of herbaceous communities show considerable differences in community composition between understory and open (Figure 14.5) in Q. douglassii savanna. These differences appear to be produced primarily by the much higher levels of nutrients and shade under oak canopies, which in turn can mediate competitive interactions. For example, Rice and Nagy (2000) demonstrated that the annual grass species Bromus diandrus, which dominates under Q. douglasii canopies, required these high nutrient soils to outcompete its congener B. hordeaceus. Away from the oak canopies it was not able to dominate and instead coexisted with B. hordeaceus. The latter is a weak competitor in high-nutrient environments and tends to dominate more infertile soils. Thus B. diandrus’s close association with oak canopies is due in part to its ability to outcompete other species in the high-nutrient environment under oaks, but also in part to its inability to compete well in the lower-nutrients soils of the open grassland. By contrast, its congener B. hordeaceus is outcompeted by B. diandrus in the rich environment under oaks but can coexist with it in the more nutrient-poor environment outside the oak canopy.

DYNAMICS OF WOODY SPECIES

177

Mean annual temperature (C)

–10 Grassland Shrubland Woodland

Taiga

0

10

Temperate forest

20 Savanna Thorn forest Thorn scrub

Tropical dry forest

Tropical rain forest

30 0

100

200

300

400

Mean annual precipitation (cm) F I G U R E 14.5. Ordination of vegetation samples (0.125 m2 quadrats)

collected under the canopies of Quercus douglasii (solid circles) and in the open grassland (open circles) of Hastings Natural History Reserve, Monterey County, in central California. Large symbols show the means and 95% confidence intervals for each group of samples. (R. M. Callaway, unpublished data.)

The effects of the evergreen oak, Q. agrifolia, on understory community composition appear to be very similar to those of the deciduous Q. douglasii. However, the effects of these two species arise in part from different mechanisms. The canopies of both species increase soil nutrients, but they shade the understory to greatly different degrees. Quercus douglasii does not shade the understory much during the winter growing season of the annual species that tend to dominate, whereas Q. agrifolia creates deep shade that can limit light-demanding species year-round. As a consequence, species that dominate its understory must be tolerant of low light. Indeed, Parker and Muller (1982) demonstrated that Avena fatua could take advantage of the rich soil from under Q. agrifolia, but only when it was transplanted to open grassland. By contrast, B. diandrus was able to dominate under Q. agrifolia because it tolerates shade as well as responding strongly to high soil nutrients. In a similar study, Marañón and Bartolome (1993) reciprocally transplanted soil blocks between Q. agrifolia understories and open grassland during the autumn before germinating rains. During the first year, species composition in the transplanted blocks resembled that in the grassland from which the blocks had been extracted. By the second year, however, the deep shade under Q. agrifolia caused high seedling mortality of the grasses from the open, including B. hordeaceus and Lolium multiflorum. Marañón and Bartolome also transplanted soil blocks from the open and understory environments to a greenhouse, laid them on top of greenhouse soils unaffected by oaks, and created their own artificial shade to isolate the importance of canopy shading. The least shade-affected species to emerge from the seedbank was B. diandrus, which dominated understory blocks whether shaded or not. Avena fatua, which is abundant in open grassland, was strongly reduced by the dense shade.

178

ECOLOGICAL INTERACTIONS

F I G U R E 14.6. World biome types in relation to precipitation and temperature. Redrawn from Whittaker 1975.

We have observed that in pastures released from livestock grazing, the oak understory often supports dense populations of thistles such as Carduus pycnocepahala and Silybum marianum, particularly near the drip line of individual tree canopies (the authors, personal observation). These species are considered undesirable and are listed as invasive pest plants (see Chapters 6 and www.cal-ipc.org). Marañón and Bartolome (1993) demonstrated that C. pycnocepahala indeed was shade-tolerant and nutrient-responsive. The conditions under which oak canopies promote populations of these species require further exploration. S U M MARY

Oaks have strong effects, both competitive and facilitative, on understory productivity and grassland community composition and thereby alter patterns of landscape biological diversity. The effects that oaks have on grasslands appear to be caused by a complex combination of resource competition, shade, hydraulic lift, and deposition of nutrients at the soil surface. Furthermore, these effects vary both between individual trees and, apparently, with different oak species, tree age, access to ground water, and tree position on the landscape. California grasslands would be far less heterogeneous in the absence of the complex effects of oak canopies.

Conclusions Although climate and geological substrate control the potential vegetation at a given site, there are large portions of California’s natural landscapes that are capable of supporting several strikingly different communities: grasslands, shrublands, oak savannas, and even oak woodlands. In the well-known illustration of biomes/world ecosystems shown in relation to temperature and precipitation (Figure 14.6), these California plant communities would fall into the climate zone “in which either grassland, or one of the types

dominated by woody plants may form the prevailing vegetation” (Whittaker 1975). Bond (2005) suggests that in these areas of the global biome ordination, the dominant forces controlling the structure and composition of the community are fire and large mammal herbivores, both of which act as “consumers of vegetation.” Several examples described in this chapter support the hypothesis that these ecosystems are consumer-controlled. With sufficient time and lack of fire and grazing, the successional trajectory of many grasslands is toward shrubland or savanna. This is observed in the rapid colonization of Baccharis scrub in moist Northern California sites where cattle have been excluded, and in the slow establishment of oaks into unburned grasslands in drier southern California savannas. Once they have established in grassland sites, shrubs and trees alter the environment, affecting composition and abundance of the grassland community and, in some cases, facilitating continued colonization by woody species. Transitions back to grassland are dependent on forces such as livestock grazing, tree cutting, and frequent fire. Managers of grasslands in California face numerous challenges, and one is addressing the almost inevitable colonization of woody species into unburned or ungrazed grassland habitat. The call to action will be clear in some cases, such as eradicating invasive non-native broom species. However, there will also be situations that require making value judgments about what vegetation is preferred, rather than what vegetation is natural at a given site. Is the change from grassland to coyote bush shrubland following the removal of livestock grazing an “invasion,” the establishment of native trees into grassland an “encroachment,” or is this “succession”? How much woody vegetation is acceptable, given that acreage gained by shrubs and trees is essentially acreage lost by grassland? There is generally little or no information available to describe how dynamics between grass and woody vegetation

differed prior to human alterations of native vegetation and disturbance regimes in grassland systems. However, at least over the past century, the presence of woody plants in or adjacent to grasslands has been treated as a nuisance, as if grasslands were a static ecosystem distinct from communities with woody vegetation. In California many thousands of acres of chaparral have been converted to annual grassland or semi-grassland, and close to one million acres of oak savanna and woodlands were cleared between 1945 and 1975 alone, by state-sponsored range “improvement” programs to create more grassland (Bolsinger 1988). In this intriguing grassland–shrubland–woodland biome (Whittaker 1975), where areas of similar climate and substrate can support any one of these vegetation-types, land managers bear a great responsibility. If we accept the notion that fire and large-mammal grazing are the main forces controlling these systems, deliberate and thoughtful management actions will be required. The use of fire as a management tool has become more difficult as natural landscapes have become fragmented and closer to residential areas. With the absence of natural megafauna, grazing means livestock management. In these systems, where grassland and woody vegetation can co-occur, conscious decisions will determine the composition and structure of these communities over space and time. In our shrinking natural landscapes, the management and preservation of grasslands, shrublands, and oak woodlands, including their associated fauna, may be best accomplished by considering them simultaneously and with a long-term view.

Acknowledgments We thank Carla D’Antonio for suport above and beyond the call of editorial duty. In addition, we thank and acknowledge Nicole Molinari, Katherine Suding, and Erica Zavaleta, for their helpful suggestions and comments on this chapter.

DYNAMICS OF WOODY SPECIES

179

FIFTEEN

Ecology of Native Animals in California Grasslands PAU LA M. S C H I F F MAN

We found a whole country to bee a warren of a strange kinde of Conies . . . And besides the multitude of a strange kinde of Conies by farre exceeding them (the deer) in number; their heads and bodies in which they resemble other Conies; are but small; his tayle like the tayle of a Rat, exceedingly long; and his feet like the pawes of a Want or Moale; under his chinne, on either side he hath a bagge, into which he gathereth his meate when he hath filled his belly abroade that he may with it either feed his young or feed himselfe when he lists not to travaile from his burrough. SIR FRANCIS DRAKE

This 1579 account is among the first ecologically meaningful written records of grassland wildlife in California. Sir Francis Drake was a British sea captain, explorer, and pirate, not a naturalist, and the unusual rabbit-like creatures (“conies”) that he described were most likely ground squirrels (Spermophilus beecheyi), pocket gophers (Thomomys bottae), and kangaroo rats (Dipodomys spp.). It is clear that he was impressed by the large, widespread populations of these animals, their capacity for burrowing, and, in the case of gophers and kangaroo rats, their ability to transport harvested plant foods within cheek pouches. Although brief and somewhat imprecise, Drake’s observations identified several processes that are central to animal ecology in California grasslands: soil disturbance, seed dispersal, granivory, and herbivory. This chapter addresses these ecological processes from a historical perspective as well as within the context of our modern-day understanding of grasslands in California. Emphasis is given to the ecological relations of small mammals, particularly burrowing rodents, because a large amount of natural history information on them exists and they have been studied rather intensively by scientists for several decades. Other types of native organisms, including some insects, are also discussed. In addition, special attention is paid to interactions between native animals and invasive non-native plants, because these plants dominate today’s grassland vegetation and are, therefore, habitat components that influence many aspects of animal ecology. Moreover, interactions with animals appear to have been, at least in part, responsible for producing and then maintaining some non-native plants in California’s grassland ecosystems.

180

( IN

JACOBSEN

1918: 130)

Finally, with the hope of stimulating future research efforts, attempts have been made throughout this review to direct attention to ecological topics that are currently inadequately understood but could be rewarding to study.

Soil Disturbance Burrowing Rodents Animal burrowing is one of the most common forms of natural disturbance in grasslands and other ecosystems worldwide (Hole 1981; Tilman 1983; Loucks et al. 1985; Hobbs and Huenneke 1992; Hansell 1993; Huntly and Reichman 1994; Butler 1995; Cox et al. 1995; Whitford and Kay 1999; Reichman and Seabloom 2002). In California, pioneering anthropologist Stephen Powers reported in 1877 that the community of Native People located in the southern San Joaquin Valley near Kern Lake had called themselves “Po-hal’-lin-Tin’leh”—a name that translated to “squirrel holes” and referred to the great number of ground squirrel excavations in that region (Powers 1976). Burrowing rodents, particularly pocket gophers and ground squirrels, were also the grassland animals most frequently commented upon by early European and American explorers, settlers, and naturalists. These observers wrote vivid accounts of enormous rodents populations and described how entire landscapes were riddled with their excavations (Table 15.1). Although the biological accuracy of these reports varied, it is clear that small mammals were important historical constituents of California’s grasslands and that the tremendous amount of soil disturbance that they produced had to have been of considerable ecological importance.

F I G U R E 15.1. (a) Slope riddled with pocket gopher mounds (Thomomys bottae). The grassland biomass was recently burned away, making the

mounds readily visible. (b) Close-up view of pocket gopher burrow mounds with a 1-meter stick. Photographs by P. Schiffman.

Recently, however, archeologists have suggested that the remarkable abundance of wildlife reported in California’s early history was not the typical, long-term ecological condition (Broughton 1994, 2002; Preston 2002a; Raab and Jones 2004). They discuss evidence indicating that, beginning in the late Holocene, the hunting pressure exerted by the growing population of Native People significantly limited populations of wildlife species. This suppression persisted until fast-spreading disease epidemics associated with European contact devastated the Native People (Preston 2002b). It is proposed that the high level of disease-caused human mortality released native wildlife populations from top-down control and allowed them to expand to the extraordinary sizes described in early historical writings (Preston 2002a). This assertion reflects a record of archeological evidence of large game animals (e.g., pronghorn antelope, Antilocapra ameriana; elk, Cervus elaphus; mule deer, Odocoileus hemionus) and predators (e.g., coyote, Canis latrans; gray fox, Urocyon cinereoargenteus; badger, Taxidea taxus). Although bones of small mammals, including gophers and ground squirrels, have been found in large numbers at Sacramento Valley archeological sites, these remains have not been included in analyses (Broughton 1994, 2002), because fossorial animals can disrupt chronological strata of archeological materials and make it difficult to determine when they were deposited (Erlandson 1984). Therefore, it is uncertain whether pre-European contact population patterns of small grassland mammals corresponded with those of larger species (Broughton 1994). However, early records indicate that, by the 1770s, burrowing rodents were extremely abundant in the central coast range and San Joaquin Valley (Coues 1900; Bolton 1926, 1930). Whether the presence of large numbers of burrowing rodents was a natural, long-term ecological condition in California or a side effect of European colonization, it is quite evident that these species have been important grassland constituents in recent centuries. Grinnell (1923) estimated

that 1 billion burrowing rodents lived within California’s borders, and Clements and Shelford (1939: 290) referred to ground squirrels as “the most characteristic species” of California’s grasslands. More recently, Lidicker (1989) explained that ground squirrels and pocket gophers continue to be two of the most ecologically important vertebrates in California grasslands and that their capacities to disturb soils are considerable. His review of the literature indicated that ground squirrels occur in grasslands in densities of 4.2 – 45.2 individuals per hectare and gophers in densities of 26.6–100.8 individuals per hectare. Families of ground squirrels usually occupy large burrow systems with multiple entrances that are surrounded by piles of excavated soil extracted from tunnels that are mostly of 1–2 meters depth (Lidicker 1989). Individual gophers produce systems of tunnels at depths of 15.3–55.9 centimeters and averaging 5 meters in length (Lidicker 1989) with surface heaps of disturbed soil deposited at entrances (Figure 15.1). Although the amount of excavation is greatest when soils are relatively moist, these animals dig year-round (Romañach et al. 2005). It has been estimated that gophers in California grasslands turn over most of the surface soil in their habitat areas every 3–15 years (Hobbs and Mooney 1995; Kneitel 1997). A detailed catalog of California’s grassland vertebrates (Lidicker 1989) included, in addition to ground squirrels and gophers, many other species that are soil disturbers. For example, excavations by kangaroo rats (Dipodomys spp.) are extensive in some grasslands (Grinnell 1932). At Carrizo Plain National Monument, giant kangaroo rat (D. ingens) burrow “precincts” (Grinnell 1932) polka-dot vast tracts of land (Figure 15.2). This species, which is endemic to the western San Joaquin Valley region and is an endangered species, produces burrow systems that are relatively static fixtures in the landscape and can be occupied by succeeding generations of individuals. Although other burrowing rodent species, such as gophers, produce spatially shifting mosaics of soil disturbance over time, Seabloom and Richards (2003) showed that ecological feedbacks between vegetation and burrowing,

E C O L O G Y O F N AT I V E A N I M A L S

181

TA B L E 15.1 Historical Accounts of the Abundance and Effects of Burrowing Rodents in California Grasslands

Pedro Fages (1769, SW California; Priestley 1937: 12) “In this territory there are to be seen, besides a number of other land animals, deer, antelope, conies, hares without number, wildcats, wolves, some bears, coyotes, and squirrels of three kinds.”

Francisco Palóu (1773, Valley in Central Coast Range; Bolton 1926: 226) “They [a group of Native People] also practice hunting rabbits and squirrels, in which the valley abounds.”

Pedro Font (1776, NW San Joaquin Valley; Bolton 1966: 410) “[W]e came to some bare hills which, because they were mined by ground squirrels we called the Lomas de las Tuzas.”

Francisco Garcés (1776, S San Joaquin Valley; Coues 1900: 301) “[A] level plain much undermined by the tusas of which there are infinite numbers in all the plains . . . ; we fell down, the mule and myself, and several times I was in danger of the same, because of the insecurity of the ground. In the fall I lost the compass needle, and did not think of returning to search for it, because it made me afraid to see a land so dry and dangerous to travel.”

Hugo Reid (1832, Los Angeles Plain; Dakin 1939: 4) “Squirrels, rabbits, and gophers were continually scurrying down into their holes, out of harm’s way. Indeed, these tiny animals had so honeycombed the surface of the ground as to make it dangerous to ride anywhere off the roadway faster than at a walk.”

George H. Derby (1850, S San Joaquin Valley; Browning 1991: 85, 86) “The soil was not only of the most wretched description, dry, powdery and decomposed, but was everywhere burrowed by gophers, a small animal resembling a common house-rat, which I had never seen before of whitish grey color, short round body, and very strong bony head. These animals are innumerable; though what they subsist upon I cannot conceive, for there was little or no vegetation. There holes and burrows, into which a horse sink to his knees at almost every step, render their travelling difficult and dangerous”; “reptiles . . . and which with the gophers and ground rats are the only denizens of this unpleasant and uninhabited spot.”

Andrew J. Grayson (1853, S San Joaquin Valley; Grayson 1920: 106) “These lands were literally perforated by gophers, moles and other underground inhabitants.”

TA B L E 15.1 ( C O N T I N U E D ) Historical Accounts of the Abundance and Effects of Burrowing Rodents in California Grasslands

Harris Newmark (1853, Los Angeles Plain; Newmark and Newmark 1916: 24) “Soon after leaving San Pedro, we passed thousands of ground squirrels, and never having seen anything of the kind before, I took them for ordinary rats.” Harris Newmark (1857, Los Angeles Plain; Newmark and Newmark 1916: 215–216) “[T]here were millions of ground-squirrels all over this country that shared with other animals the ups and downs of the season. When there was plenty of rain, these squirrels fattened and multiplied.” William H. Brewer (1862, NW San Joaquin Valley; Brewer 1966: 283) “[O]ften for miles we see nothing living but ourselves, except birds, reptiles, and ground squirrels.” Titus Fey Cronise (1868, many valleys; Cronise 1868: 443) “The Grey Ground Squirrels (Spermophilus Beecheyi . . .) are so numerous and destructive in all parts of the valleys that are not annually inundated, as to be one of the most serious pests. . . . They are the size of a half-grown cat, and have a long, bushy tail, like the tree squirrel; but do not ascend trees, except occassionally for food, making their dwelling in the ground, which in many places is full of their burrows for miles together.” John Muir (1869, NE San Joaquin Valley; Wolfe 1938: 27) “The ground squirrels having found easy tunneling in a soft stratum of one of the hills of Twenty-Hill Hollow, have bored it round and round.” John Muir (1869, San Joaquin Valley; Muir 1894: 342) “Hares and spermophiles showed themselves in considerable numbers.” Alice Eastwood (1902, Carrizo Plain; Wilson 1955: 81) “Their chief concern was to be on the alert for squirrel holes that might trip the horses.” Joseph Grinnell (1923, San Joaquin Valley; Grinnell 1923: 142, 149) “There are pocket gophers in abundance. . . at Fresno”; “On wild land and elsewhere the burrowing rodent is one of the necessary factors in the system of natural well-being.” Harold Child Bryant (1929, many valleys; Bryant 1929: 61–62) “Two mammals are typical of open stretches, the jack-rabbit and the ground squirrel. . . . Ground squirrels usually live in colonies, their burrows often being connected for a considerable distance. . . . Burrowing rodents are the original cultivators and aerators of the soil.”

densities of these rodents exceeded 203,300 individuals per hectare [82,280 individuals per acre] and suggested that they were the direct consequence of habitat loss and regional efforts to eradicate native predators.

Other Animal Excavators and Affiliated Species

F I G U R E 15.2. Thousands of circular giant kangaroo rat (Dipodomys

ingens) burrow “precincts” scattered across the grassland landscape at Carrizo Plain National Monument. In the spring, vegetation on mounds tends to be taller, greener, and different in composition compared to the surrounding grassland matrix. The line at the bottom is a dirt road. Photograph by P. Schiffman.

herbivory, and territoriality can produce similar patchy spatial patterns. The microtopographic relief of California grasslands reflects animal excavation (Black and Montgomery 1991; Seabloom et al. 2000; Gabet 2000; Reichman and Seabloom 2002) and indicates that prolonged periods of fossorial activity can have very dramatic effects. It has been proposed that the contours and spacing of the “hog-wallow” or “mima mound” terrain that sometimes characterizes grasslands with vernal pools is a consequence of the territoriality and tunneling behaviors of burrowing animals (Cox 1984, 1990; Cox and Allen 1987, Cox and Scheffer 1991). Over time in some edaphic conditions, burrowing redistributes soil in centripetal patterns such that fields of dome-shaped hummocks surrounded by shallow depressions develop. Although this sort of mounding has also been attributed to pluvial and seismic processes (Preston 1981; Berg 1990a, b), the high densities of burrowing animals in California grasslands and their capacity for habitat engineering (Reichman and Seabloom 2002) suggest that they are, at least in part, responsible for shaping this unusual terrain. Other soil disturbing rodent species, including meadow voles (Microtus californicus) and moles (Scapanus latimanus), also occur in California grasslands. Although they may be capable of significant levels of disturbance (Rice 1987), the ecological impacts of ground squirrels, gophers, and kangaroo rats are usually much greater. Still, vole populations, in particular, can grow to large size, and when this happens, their ecological effects can be significant (Batzli and Pitelka 1970; Cockburn and Lidicker 1983; Rice 1987; Noy-Meir 1988; Howe and Brown 2000, 2001). In a very extreme case, Hall (1927) documented meadow vole and non-native house mouse (Mus musculus) population explosions near Buena Vista Lake in the San Joaquin Valley. He estimated that

184

ECOLOGICAL INTERACTIONS

The California grassland fauna is characterized by mammalian predators that rely heavily on burrowing rodents, and several of these predators, including badgers, coyotes, and San Joaquin kit foxes (Vulpes macrotis mutica) and gray foxes (Urocyon cineroargenteus) also produce significant soil excavations of their own (Platt 1975; Lidicker 1989; Briden et al. 1992; Cypher et al. 1996). In addition, many raptors prey upon burrowing rodents. In fact, Cronise (1868) referred to the ferruginous hawk (Buteo regalis), a quintessential grassland raptor (Zeiner et al. 1990), as the “California squirrel hawk” in his encyclopedic description of California’s natural features. The burrowing owl (Athene cunicularia), white-tailed kite (Elanus leucurus), rough-legged hawk (Buteo lagopus), northern harrier (Circus cyaneus), and golden eagle (Aquila chrysaetos) are among the birds of prey that frequently consume grassland rodents (Ehrlich et al. 1988; Lidicker 1989; Conroy and Chesemore 1992). Snakes are also common sightings in or near rodent holes, and one of the most common grassland species, the gopher snake (Pituophis melanoleucus), is itself capable of excavating soil (Carpenter 1982). The marked dependency of so many important California grassland predators upon burrowing rodents reflects the foundational role that burrowers play in the grassland food web (Schiffman 2000). A discussion of soil disturbance in California grasslands would not be complete without mention of grizzly bears (Ursus arctos). In some ways, their ecological effects were like those of much smaller burrowing animals. Grizzlies used their enormous claws to dig up very large areas of soil in search of buried foods such as rodents, roots, bulbs, fungi, and insect grubs. Burrowing rodents, in particular, were important prey for grizzlies. Storer and Tevis (1996: 56 – 57) noted that in California “the land was well stocked with rodents” and “only the badger could vie with the bear in digging rodents from the soil.” Historically, grizzly bears had been extremely common in California’s grasslands until eradication efforts forced them into the relative, though short-lived, safety of dense chaparral cover on steep slopes (Grinnell 1938). In 1841, an observer noted that grizzlies were “an almost hourly sight” in the Sacramento Valley and that “it was not uncommon to see 30 or 40 a day” (Storer and Tevis 1996: 24). Many other early observers recorded similarly remarkable descriptions of grizzly bear abundances. Unfortunately, the predator suppression movement of the nineteenth and twentieth centuries decimated the grizzly population before the species’s ecological role in California’s ecosystems could be studied (Storer and Tevis 1996; Mattson and Merrill 2002). So, although grizzlies do still exist elsewhere in North America, the climatic and floristic differences

between California’s grasslands and other North American ecosystems are considerable. It is difficult to identify meaningful functional similarities between the grizzly’s ecology in other regions and its former ecology in California. Nonetheless its multifaceted role as a generalist carnivore, herbivore, and soil disturber strongly suggests that the grizzly was a keystone species (Tardiff and Stanford 1998; Krebs 2001). The extirpation of an animal of such extraordinary ecological importance means that now we can never truly understand historical grassland ecosystem dynamics in California. It also means that genuine restoration of the vegetation to preEuropean contact conditions is now impossible. It has been suggested that the disturbances of feral pigs (Sus scrofa) may, in some ways, substitute for those of California’s extirpated grizzly bears (Kotanen 1995; Cushman et al. 2004). In 1860 zoologist Janos Xántus observed that the California grizzly “sometimes amuses himself digging, like pigs, and sometimes during a moon-light night, he will dig up many acres of lands [so] that not one [blade of] grass is to be found on it” (Storer and Tevis 1996: 61). Although feral pigs disturb large patches of ground and consume a broad diversity of foods, including some of the same things eaten by grizzlies (see Cushman, Chapter 12), pigs and grizzly bears are still very different kinds of animals. Therefore, it is probable that only in a very general sense do these two species function as ecological equivalents in California grasslands.

Secondary Burrow Users A broad diversity of mostly nonburrowing grassland animals use the holes and/or bare soil produced by rodent excavators. These secondary burrow users range from the burrowing owl (Athene cunicularia), San Joaquin antelope ground squirrel (Ammospermophilus nelsoni), and rabbits to the tiger salamander (Ambystoma californiense) and many reptiles (Orr 1940; Zeiner et al. 1988; Lidicker 1989; Nicolai 1992; Knopf 1996). Among invertebrate burrow users, tarantulas (Aphonopelma spp.) frequently occupy grassland rodent holes (Gabel 1972), and some beetles (e.g., Geomysaprinus and Eremosaprinus spp.) are burrow habitat specialists. In addition, on cool mornings Kern primrose sphinx moths (Euproserpinus euterpe) have been known to bask on mounds of rodent-excavated soil (Tuskes and Emmel 1981). Grasshoppers (Melanoplus spp.) also seek out bare gopher mound soils as oviposition sites. Nymphs of these grasshoppers survive best in open microsites, and the presence of gopher mounds in an ecosystem probably results in larger populations of grasshoppers than would otherwise exist (Huntly and Inouye 1988). Some secondary burrow users (e.g., San Joaquin antelope squirrel; tiger salamander; blunt-nosed leopard lizard, Gambelia silus; Kern primrose sphinx moth) are officially protected as threatened or endangered species. Understanding the ecological relationships between the burrowers, the microsites they produce, and these burrow-using species may be of critical conservation importance.

F I G U R E 15.3. A giant kangaroo rat (Dipodomys ingens) burrow precinct

covered with disturbance adapted non-native Erodium cicutarium and native Amsinckia tessellata (foliage green in color). The less disturbed intervening area is dominated by native Lasthenia californica (flowers yellow in color). Photograph by P. Schiffman.

Burrowing Effects on Vegetation Disturbed microsites produced by burrowing mammals differ physically from less recently disturbed microsites located just meters away (Grinnell 1923). Excavation disrupts soil structure, and hence these patches of less consolidated soil have lower bulk densities (Canals et al. 2003). In addition, soil temperatures are typically several degrees warmer than the surrounding areas because they are less shaded or otherwise insulated (Kotanen 1997a; Canals et al. 2003). Nutrient levels of recently excavated soils are sometimes markedly different (Koide et al. 1987; Kotanen 1997a; Hoopes and Hall 2002; Canals et al. 2003) and often, but not always, have higher levels of NH4
and NO3. These patterns are reflected by greater amounts of plant growth and reproduction on burrow mounds than in less disturbed areas (Hobbs and Mooney 1985; Canals et al. 2003). Over time, burrowing animals maintain spatially complex grassland soil and plant landscapes as new disturbances are continually created and vegetation develops on old excavations (Cox et al. 1995). Interesting vegetation composition patterns are observable at the microsite scale when burrow mounds are compared to adjacent, less disturbed areas (Hobbs and Mooney 1985; Koide et al. 1987; Peart 1989c; D’Antonio 1993; Schiffman 1994; Hobbs and Mooney 1995; Hoopes and Hall 2002; Canals et al. 2003). Frequently, burrow mounds support greater abundances and diversities of non-native plants, especially ruderal annuals, than less disturbed microsites (Hobbs and Mooney 1985, 1995; Schiffman 1994; Cox et al. 1995). Grinnell (1932) described how areas disturbed by giant kangaroo rats had tremendous covers of tall non-native Erodium while native Lepidium and Lasthenia plants dominated the less disturbed intervening areas (Figure 15.3). Still, some native annual forbs, including Hesperevax [Evax] sparsiflora, Lotus subpinnatus, Amsinckia tessellata, and Tropidocarpum gracile, do actually occur in large amounts on burrows

E C O L O G Y O F N AT I V E A N I M A L S

185

(Schiffman 1994; Hobbs and Mooney 1995). It seems probable that these native disturbance-adapted plants were much more common before California’s grasslands were invaded by competitive non-native ruderal annuals. Interestingly, Hoopes and Hall (2002) suggest that gopher mounds serve as refugia on which the native perennial grass Sporobolis airoides can escape competition from non-native annual grasses. Somewhat similarly, two non-native perennials, Anthoxanthum odoratum and Carprobrotus edulis, also do best on gopher mounds (Peart 1989c; D’Antonio 1993). In fact, much like S. airoides, C. edulis would not be able to occupy grasslands without gopher disturbances (D’Antonio 1993). The effects of burrowing are also observable at much larger spatial scales. Studies have shown that the vegetations of California grassland sites with gophers, ground squirrels, or other burrowing animals are compositionally distinct (Hobbs et al. 1988; Hobbs and Mooney 1991, 1995; Knops et al. 1995; Robinson et al. 1995; Stromberg and Griffin 1996). Two general features, in particular, tend to characterize these grasslands. First, they have reduced abundances of perennials such as bunchgrasses and geophytes (Hobbs and Mooney 1995; Stromberg and Griffin 1996); second, they have elevated levels of invasive non-native species, particularly annual grasses (Knops et al. 1995; Stromberg and Griffin 1996). Small burrowing rodents’ control of so many important aspects of grassland ecology make them ecosystem engineers (Reichman and Seabloom 2002).

Granivory and Seed Dispersal Native Plant Seed Granivory and seed dispersal have not yet been extensively studied in California grasslands (Kotanen 1996). Seed-eating mammals (e.g., kangaroo rats, pocket mice, meadow voles), birds (e.g., many sparrows; horned lark, Eremophila alpestris; mourning dove, Zenaida macroura), and harvester ants (e.g., Messor and Pogonomyrmex spp.) are abundant in this ecosystem and are capable of both consuming and dispersing many seeds (Batzli and Pitelka 1970; Beattie 1989; Lidicker 1989; Schiffman 1994; Boulton et al. 2005; Espeland et al. 2005). In years of average or above-average rainfall, grassland plants can produce enormous volumes of seed. Yet some native species are strongly seed-limited (Seabloom et al. 2003a), and granivores are likely, in part, responsible for this. They are also often vectors, if only inadvertently, for numerous seeds (Price and Jenkins 1986). Many native seeds have morphological features, particularly barbs and awns, that would be expected to promote dispersal by adhering to animal fur, feathers, or skin. Other plant species have adaptations for wind dispersal (wings, pappus, or plumes) or lack obvious morphological features for animal dispersal, but they may also collectively serve as a significant food source for granivores. Spatial patterns of plants frequently reflect the choices and behaviors of seed-eating animals. For example, Hobbs (1985) found that foraging harvester ants (Messor [Vermessor] andrei)

186

ECOLOGICAL INTERACTIONS

preferentially collected the seeds of the annual grassland forb Microseris douglasii and limited that species’ abundance. In addition, ant nest sites supported more Calycadenia multiglandulosa and Layia platyglossa as well as less Lasthenia californica, Plantago erecta, and Castilleja [Orthocarpus] densiflora than the surrounding adjacent vegetation (Hobbs 1985). However, harvester ants are especially vulnerable to the negative effects of habitat fragmentation and invasive Argentine ants (Linepithema humile), and this means that their control of species distributions is tenuous in grassland fragments, particularly those in close proximity to human development (Suarez et al. 1998). Despite the abundance of potential seed dispersers, the patchiness of grassland plants suggests that the vast majority of native seeds disperse only relatively short distances. Moreover, animals sometimes indirectly influence seed dispersal and seedling establishment by creating mosaics of heterogeneous microsites. Hobbs and Mooney (1985) found that seed rain from grassland species onto gopher mounds was generally lower than in undisturbed areas. This was especially the case for Lasthenia californica. Seed rain of taller species, however, was greatest on burrow mounds because seeds released from inflorescences of greater heights dispersed beyond the dense canopy of undisturbed vegetation.

Non-native Plant Seed Native seed-eating animals are opportunistic, and their diets have expanded to include the invasive non-native species, particularly the annual grasses, that now dominate California’s grasslands (Shaw 1934; Hawbecker 1951; Batzli and Pitelka 1970; Borchert and Jain 1978; Hobbs 1985; Schiffman 1994). Native animals may have been responsible for dispersing the seeds of some of the first invasive plants into California’s grasslands (Mensing and Byrne 1999). The presence of seeds of invasive plants, including Erodium cicutarium, Avena, Rumex crispus, and Medicago polymorpha, in the mud used to make early adobe bricks (Hendry 1931; Spira and Wagner 1983; Mensing and Byrne 1998, 1999) indicates that these species dispersed to California and became established during the period of exploration prior to the 1769 founding of the first European settlement in Alta California. Mensing and Byrne (1999) suggest that dispersal by native animals from Baja California, where European plants had previously become entrenched, to Alta California is one of the ways in which the invasions could have been initiated. In California’s grasslands, the patches of disturbed soil produced by burrowing animals probably acted as “nascent foci” (Moody and Mack 1988) for invasive non-native plant species. Moody and Mack (1988) showed that large numbers of relatively small, circular foci can promote the very efficient spread of invasive plant propagules across landscapes. Rodent burrows have been demonstrated to promote grassland invasibility (Robinson et al. 1995), and studies indicate that disturbed soil microsites allow for the establishment of weedy non-native plants (Smith 1970; Hobbs and Mooney 1985;

F I G U R E 15.4. Soon after European contact, widespread soil

disturbances and seed dispersal by burrowing grassland mammals may have facilitated the rapid spread and establishment of some invasive, non-native, ruderal plant species. During the invisible lag phase, small, localized populations expanded into large widespread populations. Model adapted from Kowarik (1995).

Peart 1989c; D’Antonio 1993; Schiffman 1994; Cox et al. 1995; Knops et al. 1995; Stromberg and Griffin 1996). This is particularly true for invasive annual plants, both grasses and forbs, derived from domesticated crops (e.g., Avena fatua) or with recent evolutionary histories that included prolonged exposure to agricultural soil cultivation (Hobbs and Huenneke 1992; Seligman 1996; Zohary and Hopf 2000). It is not difficult to imagine that, early in California’s history, the enormous numbers of rodent burrows (see Table 15.1) enabled some opportunist ruderal plant invaders to hopscotch rapidly across broad expanses of grassland terrain (Hobbs and Hobbs 1987; Wu and Levin 1994). In addition, the chronic nature of these disturbances meant that newly disturbed microsites were continually made available. This may have been enough to promote the rapid spread of many invasive non-native species. The fact that California’s grasslands were invaded rapidly and that the process apparently happened in a manner that escaped notice of early residents (see Chapter 4), suggest that the mechanisms that facilitated this invasion did not involve factors that would have been perceived as extraordinary. Soil disturbances produced by burrowing mammals were common and widespread grassland elements, and their ability to promote the invasion process would not have attracted the attention of early observers, who were, if anything, attuned to the tremendous scale of disturbance rather than changing abundances of small plants. In addition, invasions are often characterized by lag periods during which time non-native populations spread surreptitiously across regions. In the lag phase, invasive populations are still of relatively small size, and their spread is often undetected or simply viewed as ecologically insignificant (Kowarik 1995). It seems likely that in California grasslands, animals facilitated the invisible lag phase spread of some invasive ruderal plant species by creating

F I G U R E 15.5. Non-native annual grass, red brome (Bromus madritensis), seed heads piled in a temporary “haystack” cache on a giant kangaroo rat (Dipodomys ingens) burrow “precinct.” A few days after making a haystack, the kangaroo rat transfers most of the seeds into the burrow for longer-term underground storage. Length of tape is 1 meter. Photograph by P. Schiffman.

microsites for establishment and physically dispersing seeds (Figure 15.4). The propagules of many invasive, non-native plant species, including common annual grasses and Erodium spp., also have barbs, awns, and other morphological features that readily make them adhere readily. These structures likely facilitated their dispersal by native animals and livestock (Manzano and Malo 2006). In addition, grassland animals purposely transport seeds. Kangaroo rats harvest and cache enormous numbers of seeds (Figure 15.5), and Cox et al. (1995) reported that pocket gophers produce surface-access caches of large numbers of Erodium seeds. Giant kangaroo rats apparently preferentially select nonnative seeds because they are, on average, larger than seeds produced by native grassland species (Schiffman 1994). Some of these non-native seeds escape predation and subsequently germinate and grow in the disturbed soils where they were temporarily cached. These behaviors are, at least in part, responsible for the high levels of non-native species richness and cover that are sometimes found on rodent-disturbed soils (Schiffman 1994). Herbivorous animals also often inadvertently consume and disperse viable seeds. Zedler and Black (1992) showed that rabbits (Sylvilagus audubonii and S. bachmani) and hares (Lepus californicus) disperse many grassland seeds, including non-native species, after consuming them while feeding on foliage. Similarly, grazing animals also disperse seeds. Malo and Suárez (1995) estimated that a single grazing cow is capable of consuming and transporting 300,000 viable Mediterranean seeds per day. The establishment of widespread livestock grazing was coincident with the spread of invasive non-native plants in California’s grassland landscapes. It seems likely that both native animals and livestock were very effective dispersers of non-native ruderal plant

E C O L O G Y O F N AT I V E A N I M A L S

187

seeds in California’s burrow-riddled grasslands (Figures 15.1 and 15.2).

Herbivory Small Mammals In addition to being disturbers of soil and consumers and/or dispersers of seeds, many small grassland mammals are also herbivores (Orr 1940; Fitch and Bentley 1949; Hawbecker 1951; Huntly and Reichman 1994; Olff and Ritchie 1998). In Mediterranean-climate ecosystems, fossorial rodents are the major herbivores (Cox et al. 1995). Fitch and Bentley (1949) estimated that each year ground squirrels, gophers, and kangaroo rats collectively reduced the grassland biomass (dry weight) at the San Joaquin Experimental Range by an average of 512 kilograms per hectare (457 pounds per acre) and that this quantity was equivalent to at least one-third of a typical year’s total production. This was almost certainly an underestimate because it did not include the potentially significant effects of rabbits, mice, and other small herbivores (Fitch and Bentley 1949). In addition, grasslands in more mesic locations, with greater annual productivities and larger populations of small mammals, would be expected to experience greater biomass losses than those documented by Fitch and Bentley (1949). Precipitation and soil fertility are often good predictors of the effects of herbivory on grassland diversity at regional as well as local scales (Olff and Ritchie 1998). Species diversities of dry or infertile grasslands are generally more negatively impacted by herbivory than more mesic and fertile grasslands (Olff and Ritchie 1998). In California, this means that plant diversity in coastal prairies would be expected to be much more positively affected by herbivory than would the diversities of serpentine grasslands of the Coast Ranges and the arid inland grasslands of the Central Valley. Herbivory and its ecological implications have not yet been adequately studied in California grasslands, and therefore there are many uncertainties. However, given that populations of rodents were historically very large, it is likely that the ecological effects of their herbivory were far-reaching (Cockburn and Lidicker 1983; Noy-Meir 1988; Olff and Ritchie 1998; Seabloom and Reichman 2001; Howe et al. 2002). A simulation model constructed by Seabloom and Richards (2003) indicated that gopher territoriality and preferences for feeding on annual plants can result in the development of multiple stable equilibria—compositionally different small plant communities that persist for prolonged periods of time. In addition, fluctuations in herbivorous rodent population sizes have been observed to affect the structure and composition of coastal prairie and inland valley grassland vegetations (Batzli and Pitelka 1970), and the effects of grassland herbivores can extend into nongrassland ecosystems. For example, in the late 1800s and early 1900s, predator suppression caused native rodent populations to boom, and their habitat usage expanded into agricultural fields. As a result, state and federal government agencies and

188

ECOLOGICAL INTERACTIONS

civic organizations encouraged the mass slaughter of California ground squirrels and other rodent species because they were destroying economically valuable crops. Antiherbivore activities became common and included an 1877 “killing bee” in Los Angeles County and a women’s “squirrel association” in Manteca that gave prizes and monetary awards to children who turned in the tails of dead ground squirrels in 1903 (Jacobsen 1918). Although grassland herbivores can be opportunistic, they do have preferences for particular plant foods (Fitch and Bentley 1949). Gophers, for instance, generally prefer to eat annuals, forbs, and geophytes (Huntly and Inouye 1988; Hunt 1992; Huntly and Reichman 1994; Seabloom and Richards 2003). However, the California grassland environments that gophers occupy today are dominated by invasive, non-native plants. Non-native taxa such as Bromus, Avena, Erodium, and Hypochaeris now constitute most of the typical gopher’s diet (Hobbs and Mooney 1995; Hunt 1992). Nonnative plants, particularly annuals, have also become significant components of the diets of other small mammalian herbivores including meadow voles, ground squirrels, rabbits, and hares (Orr 1940; Fitch and Bentley 1949; Batzli and Pitelka 1970; Cockburn and Lidicker 1983; Zedler and Black 1992). Even kangaroo rats, which are primarily granivores, sometimes eat the foliage of both native and non-native grassland plants (Grinnell 1932). Fitch and Bentley (1949: 311) noted that kangaroo rats (Dipodomys heermannii) seem “rather indiscriminate in use of green vegetation in winter” and that much of this plant material was the common nonnative, Erodium botrys. Kangaroo rats also often clip down plants growing on their burrow mounds (Williams and Kilburn 1991). In fact, much of the total amount of the biomass removed by small grassland herbivores is actually associated with behaviors that are not directly connected to feeding (Fitch and Bentley 1949). These activities, which include trampling of vegetation in runways and near burrows, clipping nesting materials, cutting and discarding herbage, and covering-up of plants with excavated soil, can considerably reduce a grassland’s total biomass. In the case of Dipodomys ingens, clipped areas can constitute up to 32% of the terrain (Schiffman 1994) in the dry season. At Carrizo Plain National Monument, such relatively cleared areas can be important habitat for mountain plovers (Charadrius montanus), blunt-nosed leopard lizards, and other grassland species that are apparently not well adapted to environments with heavy accumulations of nonnative grass biomass (Knowles et al. 1982; Williams et al. 1992; Knopf 1998).

Grasshoppers Crawley (1989) has suggested that mammalian herbivores generally have greater impacts on plants than do insect herbivores. However, he refers to grasshoppers as “honorary vertebrates,” because, as relatively large, mobile, polyphagous insects, they are capable of significantly reducing plant

densities. In California grasslands, grasshoppers and other insects are a particularly important class of herbivores. Nearly 200 grasshopper species occur in California (Joern 1989), and more than 50% of them are endemics. In addition, almost half of California’s grasshoppers are in the subfamily Melanoplinae (Joern 1989). Among these, Melanoplus devastator is a massing species that can be especially damaging to vegetation (Strohecker et al. 1968). It feeds on forbs as well as grasses and the effects of a population boom in 1957 and 1958 were reported to have been quite spectacular (Strohecker et al. 1968). Despite their considerable biological diversity, almost nothing is known about the population dynamics, preferred plant food species, and other ecological relationships of California’s grasshoppers and other herbivorous grassland insects (Joern 1989). Insects in other North American grasslands have been studied, however, and it should be possible to extend some of these findings to California. Joern (1989) used this sort of extrapolation to propose that, historically, reciprocal effects between California grasshoppers and the plants they eat may have helped to propel invasive nonnative plants to successfully displace native plants and come to dominate California’s grasslands— an event that punster Joern called the “coup de grass.” Very large grasshopper populations would have been necessary for such a massive ecosystem change. Grasshopper population sizes have been indirectly linked to weather conditions (Strohecker et al. 1968; Rodell 1977; Belkovsky and Joern 1995; Belkovsky and Slade 1995). In years when the temperature and moisture regime promotes abundant plant growth, populations of grasshoppers can become quite large. In California grasslands, where rainfall amounts can vary considerably from year to year and from place to place, grasshopper population sizes would also be expected to vary greatly. Differing historical observations about grasshoppers and their effects in California probably reflect such weather-driven population variations. For example, according to Jacobsen (1918: 133), “in 1829, it was found that throughout California where missions had been established both squirrels and grasshoppers caused considerable damage.” In contrast, Cronise (1868) noted that grasshoppers and crickets had at times been destructive but that their effects were not extensive. Belkovsky and Joern (1995) indicated that grasshopper populations can be strongly influenced by predators in habitats with good-quality plant food resources. In a study of grasshopper ecology conducted in the palouse prairie of Montana, predation on grasshoppers by birds, including some species that also occur in California grasslands (western meadowlarks, Sturnella neglecta; grasshopper sparrows, Ammodrammus savannarum; and western kingbirds, Tyrannus verticalis), was observed to reduce the abundances of largeand small-bodied grasshopper taxa more so than medium sized ones (Belkovsky and Slade 1993). This, in turn, might affect the populations of food plants fed upon by grasshoppers of different sizes. Although the species-level details may

differ, it is to be expected that bird predation also differentially affects California grasshopper species and possibly the grassland plant populations that they eat. In another Montana study, grasshopper herbivory levels were found to influence rates of nitrogen cycling by changing the quality of leaf litter and rates of decomposition. Ultimately, this impacted plant production and grassland species composition (Belkovsky and Slade 2000). It is likely that, in California, grasshoppers also affect decomposition and nutrient cycling rates. However, these relationships may not be as strong in arid California grasslands, where leaf litter (mulch) from non-native grasses accumulates and can cover the ground, undecomposed, for years.

Other Insects California grasslands are also habitat for a wide diversity of additional herbivorous insect taxa. However, as with grasshoppers, the life histories of these insects are very poorly documented. The non-native beet leafhopper (Circulifer tenellus [Eutettix tenellus]) is one species that was studied in some depth in the 1930s and 1940s because, as the vector of curly top disease, it was a costly agricultural pest in the San Joaquin Valley. Detailed studies showed that several annual grassland forbs are important hosts for this herbivorous insect (Piemeisel and Lawson 1937, Lawson and Piemeisel 1943). As with grasshoppers, beet leafhopper populations respond to how moisture availability affects plants. In the rainy season, they feed on winter annuals in grasslands and other native vegetations. However, when summer drying begins, they migrate to agricultural crops (especially beets, tomatoes, spinach, and cucurbits) and weedy summer annuals that tend to be common in grasslands that are heavily grazed by livestock (e.g., Atriplex, Amaranthus, Chenopodium, Salsola, and Malva). When crops are harvested and summer annuals are dry, they shift back to grasslands, where they eat whatever is green until winter annuals reappear (Piemeisel and Lawson 1937, Lawson and Piemeisel 1943). Complicated interactions with annual grassland plants are also important for the Kern primrose sphinx moth. Its caterpillars are apparently specialists that feed on the native annual forb Cammisonia campestris (Tuskes and Emmel 1981). Historically, this moth occurred in much of southern California, but in 1980 it was known from just one population in the Walker Basin of Kern County. Between 1982 and 2002 no individuals had been seen anywhere, and then in 2002 a population was discovered at Carrizo Plain National Monument. The rarity of the Kern primrose sphinx moth appears to be linked to the abundance of non-native species in California grasslands, and a common non-native plant species may act as a reproductive trap. Tuskes and Emmel (1981) report repeatedly observing females ovipositing on the extremely common non-native annual Erodium cicutarium instead of the much less common native Camissonia plants. However, larvae reared on E. cicutarium did not feed and died after three days.

E C O L O G Y O F N AT I V E A N I M A L S

189

If beet leafhoppers and Kern primrose sphinx moths are in any way typical of California’s herbivorous grassland insects, a large proportion of the other species should be expected to also have complex ecological relationships with native and non-native grassland plants as well as other organisms. Research into this large and desperately understudied general topic would almost certainly yield fascinating results with value for the conservation and management of California grassland ecosystems.

Summary As Sir Francis Drake so aptly noted over 400 years ago, California’s native grassland animals are involved in a variety of interesting ecological processes. Small burrowing mammals, in particular, are very abundant and have a tremendous array of important interspecific relationships.

190

ECOLOGICAL INTERACTIONS

The ecological relations of California’s burrowing rodents have been studied rather intensively in recent decades, and we now know a great deal about how their disturbances affect grassland vegetation. Their roles as seed predators, seed dispersers, and herbivores are less well understood. Still, it is clear that the capacity of these animals to accommodate non-native plants (and vice versa) is truly remarkable. Because of this, it is likely that, beginning soon after California’s first European contact, soil-disturbing rodents played instrumental roles in promoting the invasion and maintenance of some of the non-native plants that dominate grasslands. Because a good-sized body of ecological knowledge now exists for California’s small grassland mammals, perhaps it is time to broaden the focus of scientific inquiry to explicitly address the ecological roles of grasshoppers and other common, but possibly less charismatic, grassland organisms.

SIXTEEN

History and Ecology of Feral Pig Invasions in California Grasslands J. HALL C U S H MAN

Disturbance is a critically important factor influencing the structure of ecological systems and is considered integral to the maintenance of species diversity (Connell 1978; Grime 1979b; Huston 1979; Sousa 1984; Pickett and White 1985). However, both natural and human-caused disturbances are also widely recognized to facilitate the spread of exotic species (Mooney and Drake 1986; Drake et al. 1989; Hobbs and Huenneke 1992; D’Antonio and Vitousek 1992; D’Antonio et al. 1999; Sakai et al. 2001). In addition, exotic species themselves can greatly alter disturbance regime characteristics in their novel environments, either by enhancing or suppressing existing regimes or by introducing new forms of disturbance (Mack and D’Antonio 1998). Thus, disturbance in human-altered landscapes can pose significant challenges, particularly to vegetation management. Biotic disturbance agents are important components of most ecosystems throughout the world and frequently modify local soil characteristics, plant community structure, and the dominance of exotic species (Hobbs and Huenneke 1992; Mack and D’Antonio 1998; Cushman et al. 2004; Tierney and Cushman 2006). Disturbances by native mammals — such as ground squirrels, kangaroo rats, moles, shrews, pocket gophers, prairie dogs, badgers, bears, and bison — create small-scale soil disturbances through burrowing, excavating, and wallowing that greatly alter the landscapes they inhabit (Platt 1975; Huntly and Inouye 1988; Whicker and Detling 1988; Hobbs et al. 1988; Lidicker 1989; Martinsen et al. 1990; Tardiff and Stanford 1998; Schiffman, Chapter 15). A growing number of studies document that such disturbances can facilitate invasion by exotic plant taxa, which rapidly colonize these openings (Rice 1987; Hobbs et al. 1988; Peart 1989c; Hobbs and Mooney 1991; D’Antonio 1993; McIntyre and Lavorel 1994). Domesticated mammals have been introduced throughout the world, and many of these taxa have established feral

populations, substantially altering prevailing disturbance regimes, and adversely impacted population, community, and ecosystem processes (Cox 1999; D’Antonio et al. 1999). Especially problematic have been feral goats (Capra hircus) and sheep (Ovis aires) on temperate and tropical islands (Coblentz 1978; Van Vuren and Coblentz 1987), feral horses (Equus caballus) and burros (E. asinus) in arid habitats of the western United States (Cox 1999; Beever and Brussard 2000; Beever et al. 2003), and feral pigs (Sus scrofa) on all continents except Antarctica, as well as many oceanic islands (Mayer and Brisbin 1991). The latter species greatly increases levels of disturbance by overturning extensive areas of vegetation and associated soil while foraging, thereby creating a mosaic of disturbance intensities and ages (Bratton 1975; Barrett 1978, 1993; Kotanen 1994, 1995; Cushman et al. 2004; Tierney and Cushman 2006). Feral pigs are a prominent feature of grasslands in California and represent a challenge to management efforts in these landscapes. As touched on throughout this volume (especially in D’Antonio et al., Chapter 6), grasslands in California have undergone an extraordinary transformation in composition over the past two hundred years and are now dominated by a wide range of exotic plant species (Heady 1988; Heady et al. 1988; 1992). Thus, feral pigs are by no means the first invaders to colonize grasslands in California, but it is nevertheless critical to understand their impacts on these highly altered landscapes, particularly if restoration of native grassland composition is a desired goal. In this chapter, I will explore the ecology and management of feral pig populations in the grasslands of California. Specifically, I will discuss (1) the history, taxonomy, geography, and basic ecology of feral pigs; (2) the response of native and exotic plant taxa from different functional groups to pig disturbances; (3) the impacts of pig disturbances on ecosystem processes; and (4) efforts to control or

191

eradicate pig populations and manage grasslands in the face of invasion by these mammals. Although the focus will be on grasslands in California, studies from other regions or habitat types that are relevant to the discussion will also be considered.

Overview of Feral Pigs and Wild Boar Native to Eurasia and North Africa, domestic pigs (Sus scrofa domesticus, Suidae) were brought to California by the Spaniards in the 1500s and were often allowed to forage freely in oak woodlands surrounding their settlements (Mayer and Brisbin 1991). As a result, by the late 1700s at least, many viable populations of pigs had become established in coastal areas, where settlements were most commonly located (Barrett 1978). In addition, Eurasian wild boar (Sus scrofa scrofa) were subsequently introduced for hunting into Monterey County in 1925 and other counties in the 1950s. These two subspecies readily interbreed, and populations in California are now a mixture of feral domestic pigs, wild boar, and various hybrids and backcrosses (Mayer and Brisbin 1991). Wild pig populations in the state were restricted to a few coastal counties until the 1950s, but they expanded to 33 of California’s 58 counties by the early 1980s. More recently, an analysis of hunting records by Waithman et al. (1999) found that the distribution of feral pigs increased from 10 coastal counties in the early 1960s to 49 of the 58 counties by 1996 (Figure 16.1). Until recently, pigs also occurred on four of the Channel Islands (Mayer and Brisbin 1991). The California Department of Fish and Game (CDFG) formally designated pigs a game mammal in 1957 (Tietje and Barrett 1993; Updike and Waithman 1996), and this status offered them partial protection from eradication and control efforts during an early stage in their invasion. In 1992, individuals were required to purchase license tags to hunt feral pigs in the wild, and revenue generated by CDFG from these sales—approximately 30,000 pigs are killed by hunters each year—are used to support research on the ecology and management of feral pigs (Updike and Waithman 1996). Numerous factors undoubtedly contribute to the tremendous success of pigs outside of their native range (Barrett 1978; Mayer and Brisbin 1991; Waithman et al. 1999). First, humans have been rearing domesticated pigs for 8,000 – 9,000 years and have introduced them to many regions globally, where they frequently escaped or were intentionally released (Mayer and Brisbin 1991). Thus, there have been repeated introductions of this mammal throughout the world. Second, pigs are able to thrive in a wide range of habitats. For example, in California, they are common in oak woodlands, evergreen and mixed evergreenhardwood forests and both grasslands and chaparral adjacent to forests. Third, pigs are efficient and very effective generalist foragers, moving rapidly among patches to capitalize on seasonal variation in resource availability.

192

ECOLOGICAL INTERACTIONS

F I G U R E 16.1. Estimated range and relative abundances of feral

pigs in counties throughout California (Waithman et al. 1999). Black areas correspond to regions with high to very high densities; gray areas correspond to regions with moderate densities; and hatched areas correspond to regions with low densities (pigs are absent from white areas). Reprinted from the Journal of Wildlife Management.

These opportunistic omnivores have an exceedingly diverse and variable diet that includes bulbs and roots, acorns, aboveground foliage of grasses and forbs, fungi, invertebrates, small vertebrates and carrion (Barrett 1978). Fourth, pigs have a high reproductive capacity, with each female producing 10 – 12 offspring per year, which enables populations to increase rapidly in size and promotes range expansion. Fifth, pigs commonly travel through the landscapes in groups (or sounders) of varying sizes. Groups are usually composed of females and their young, with males living on their own once they reach maturity and joining groups only for mating or to visit localized food or water sources. Such group behavior may increase their foraging success and/or reduce predation rates. Lastly, pigs learn quickly to avoid contact with humans and thus may be particularly elusive to hunters (Barrett et al. 1988). Pigs exhibit distinctive foraging activity—often referred to as “rooting” or “grubbing”—that consists of excavating the soil to a depth of 5 – 15 cm (Kotanen 1994, 1995). The disturbed vegetation and associated soil may remain in place, or pigs may push this material to the side, thereby covering adjacent vegetation. As is widely reported in the literature, pigs can disturb large areas of vegetation and soil where they forage (Kotanen 1995; Sweitzer and Van Vuren 2002; Cushman et al. 2004), and Hone (2002) found that the

area of land disturbed by pigs increased significantly with their density.

Responses of Grassland Vegetation to Pig Disturbances Extensive published research has focused on the ecology, behavior, and management of pig populations in California (e.g., Barrett 1978; Waithman et al. 1999; Choquenot and Ruscoe 2003), but considerably less work has explored the impacts of pigs on plant, animal, or fungal populations or communities. Pigs can influence biota both directly and indirectly. Directly, feeding by pigs can adversely affect the survival and reproductive success of target species. The soil disturbances caused by pigs while foraging also directly kill or adversely affect a variety of species. Indirectly, these soil disturbances create openings in otherwise space-limited grasslands, which in turn affect resource availability, create opportunities for colonization, alter the dynamics of species interactions, and modify the composition and structure of communities. The relative importance of direct versus indirect effects of pigs would be difficult to determine. However, in most situations, the greatest effect of pigs probably comes from their soil disturbances, and much less from direct consumption of food species. Studies in Australia, New Zealand, Hawaii, and the southeastern United States have suggested that foraging disturbances by pigs are associated with reduced dominance of native plants and increased abundance of exotic taxa (Bratton 1975; Challies 1975; Spatz and Mueller-Dombois 1975; Stone 1985; Aplet et al. 1991). In contrast, D’Antonio et al. (1999) have suggested that the effects of pig disturbances on grasslands in California may be less dramatic. The most detailed research on the responses of plants in grasslands to feral pig disturbances has come from two sets of studies in northern California. In the first study, Kotanen (1994, 1995, 1996, 1997a, b, 2004) addressed the effects of actual and simulated pig disturbances on grasslands at the University of California’s Angelo Coast Range Reserve in Mendocino County, 240 km north of San Francisco. A second set of studies (Cushman et al. 2004; Tierney and Cushman 2006) assessed the impacts of pig disturbances on grasslands at Salt Point State Park in northwest Sonoma County, 120 km north of San Francisco. These two detailed studies provide an opportunity to examine how consistent results are between sites because they are in close proximity to each other (only 120 km apart), and both involve coastal grasslands that occur in areas with high pig abundance (see Figure 16.1). However, there are numerous differences between these studies that complicate comparisons. First, although both sites are considered coastal grasslands, Salt Point is on the coast and subject to the ocean’s moderating influence, whereas Angelo is farther inland and much hotter. Second, the two sites are different floristically, with Angelo dominated by native perennial bunchgrasses and exotic annual grasses, whereas Salt Point has an abundance of both exotic

and native perennial grasses as well as exotic annual grasses. Third, Kotanen’s research at Angelo was conducted primarily during 1990 to 1993 near the end of a six-year drought, whereas our work at Salt Point occurred from 1996 to 2000 when precipitation levels were more substantial. And fourth, the two sets of studies did not always quantify the same response variables. Kotanen (1995) conducted a comparative study at the Angelo Reserve to assess how plant species richness associated with pig-disturbed areas differed from that in undisturbed areas. He found that pig disturbances initially reduced species richness during the first year after disturbance, and then richness rebounded to levels similar to or sometimes greater than those in undisturbed plots. Surprisingly, he also found that native and exotic richness responded similarly to pig disturbance. This result was due largely to the presence of many species of disturbance-dependent native annual dicots, which responded favorably to pig rooting. In another study, Kotanen (1997a) conducted an experiment to evaluate the responses of vegetation to simulated pig disturbances of different types and found that plant species richness either recovered three years following disturbance or was reduced, depending on the type of disturbance treatment applied. He also found that exotic annual grasses increased in abundance during the first three years following simulated disturbances but had decreased to low levels after 10 years (Kotanen 2004). Cushman et al. (2004) conducted comparative and experimental studies at the Salt Point site to quantify the effects of pig disturbances on species richness and aboveground biomass of native and exotic plant taxa. The comparative study revealed that species richness and aboveground biomass of exotic plants increased significantly with increasing amounts of pig disturbance, whereas no such relationships emerged for native taxa. A four-year exclosure experiment showed that pig disturbances significantly decreased residual dry matter and increased plant species richness, especially for exotic taxa. Other studies have also found that small-scale soil disturbances by mammals can facilitate invasion by exotic plants, which possess life history characteristics that enable them to colonize and become established in these openings (Rice 1987; Hobbs et al. 1988; Peart 1989c; Hobbs and Mooney 1991; D’Antonio 1993; McIntyre and Lavorel 1994; Kotanen 1995, 1997a). Results from Cushman et al.’s (2004) exclosure study for aboveground biomass were complex and varied greatly with plant functional group (annuals vs. perennials; grasses vs. forbs), geographic origin (native vs. exotic), and grassland type. The latter variable refers to two distinctive vegetation types that occurred extensively at the Salt Point site: a lowgrowing grassland vegetation dominated by annual grasses and forbs (the short-patch type) that was interspersed among areas of taller vegetation dominated by large perennial bunchgrasses (the tall-patch type). From a management perspective, two of the most encouraging results were that biomass of native perennial grasses (primarily Deschampsia cespitosa and Danthonia californica) was unaffected by pig disturbances in either patch type, and that biomass of exotic perennial

H I S T O R Y A N D E C O L O G Y O F F E R A L P I G I N VA S I O N S

193

grasses (primarily Anthoxanthum odoratum) was reduced significantly by disturbance in tall patches where they dominate (there was no effect in short patches). Cushman et al. observed that pigs rarely overturned established native bunchgrasses and hypothesized that this was due to the large size and deep roots of these individuals. At another nearby site, Peart (1989c) also found that pocket gophers tended to avoid areas dominated by bunchgrasses, and Kotanen (1995) found that native bunchgrasses were not adversely affected by pig disturbances. By contrast, individual exotic bunchgrasses at Salt Point were small in size, and Cushman et al. hypothesized that they lacked sufficient anchoring from root tissue to resist pig rooting. The exclosure experiment conducted by Cushman et al. (2004) also revealed a number of outcomes that were negative from a management perspective. Pig disturbances increased aboveground biomass of exotic annual grasses (Aira caryophyllea, Cynosurus echinatus, Vulpia myuros, and Briza minor) in the short patch type where they dominate. These results were similar to the numerical increases found by Kotanen (1995) at the Angelo site. However undesired, such results are not surprising, given that exotic annual grasses are well known to respond positively to disturbance (see Hobbs and Huenneke 1992) and undoubtedly have a well-developed seed bank at the Salt Point site. The biomass of exotic forbs was also increased greatly by pig disturbance in tall patches. Like exotic annual grasses, these species possess life history characteristics that allow them to rapidly colonize the openings in grasslands that pigs create. Tierney and Cushman (2006) used both comparative and manipulative approaches to assess how native and exotic vegetation at the Salt Point site changed through time following pig disturbances. The investigators quantified successional changes by comparing pig disturbances of varying ages (2, 14, 26
, and 60
months old) and found that species richness of native plants increased slowly but steadily through time following disturbances, whereas richness of exotic species rebounded much more rapidly. Percent cover of native perennial grasses also increased steadily through time after pig disturbance, whereas the cover of exotic perennial grasses, annual grasses, and forbs initially increased rapidly after disturbance and then remained the same or subsided slightly with time. The cover of native forbs and bulbs either increased weakly through time following disturbance or did not change substantially. These results showed that native and exotic plants from different functional groups varied greatly in how they recovered from pig disturbance, with exotics generally able to colonize rapidly and persist in disturbances, whereas natives slowly but steadily rebounded following pig disturbance. Two additional studies on the impacts of feral pigs in California warrant discussion. First, Peart (1994) conducted experiments on Santa Cruz Island in southern California and found that pig foraging and trampling reduced the abundance and survival of woody seedlings in grasslands and oak woodlands. She also found that the cover of native plants increased as a group when protected from pig disturbance,

194

ECOLOGICAL INTERACTIONS

whereas the cover of exotics increased when exposed to disturbance. Second, Sweitzer and Van Vuren (2002) conducted comparative and experimental studies that assessed the effects of feral pigs on grasslands and oak woodlands at four sites in northern California: Austin Creek State Recreational Area in western Sonoma County; Sugarloaf Ridge State Park in eastern Sonoma County; McCormick Sanctuary in eastern Sonoma County; and Henry Coe State Park in Santa Clara County. Their comparative studies indicated that increasing amounts of pig rooting were associated with reduced aboveground plant biomass in grasslands. An exclosure study also showed that pig disturbances had no effect on the cover of native and exotic plant species at one grassland site (Austin Creek), whereas they increased the cover of exotics and had no effect on natives at another site (Henry Coe) (Sweitzer and Van Vuren, unpublished report). In addition, Sweitzer and Van Vuren found that pigs reduced the abundance of acorns on the ground, and increasing density of pigs was associated with reduced abundance of oak seedlings. To conclude, the studies summarized in this section clearly indicate that feral pig populations can have pronounced impacts on the composition of grasslands in California. The emerging picture is that feral pigs promote the invasion of exotic plants in these habitats. However, the responses of plant communities to pig disturbance are context dependent and highly variable. The composition of disturbed plant communities plays a critical role in determining the overall response. In particular, the impact of pigs will be influenced greatly by the size and richness of native and exotic seed banks, the relative dominance of exotic annual grasses and native disturbance-responsive annuals, and the degree of coastal influence that a site experiences. Despite this complexity, the findings of Kotanen (2004) and Tierney and Cushman (2006) are encouraging because they suggest that native plant species can rebound considerably if protected from further soil disturbances. Thus, removing or reducing the size of pig populations — if feasible (see following discussions)—should be an effective technique for restoring native perennial grasses in coastal grasslands. Unfortunately, we suspect that this resiliency may be specific to coastal grasslands, where the long summer drought is tempered by maritime influences. The responses of more interior grasslands may differ greatly, with native plant groups being less able to recover from disturbances under the more arid conditions that prevail in these regions. The common exotic annual grasses in interior grasslands are different from those found in coastal systems and may be more aggressive competitors with native species. If such exotics are promoted by pig disturbance, they may lead to a greater suppressing effect on native species.

Impact of Pig Disturbances on Ecosystem Processes Pig disturbances have the potential to indirectly influence a number of important ecosystem-level processes. They may

increase soil mixing and alter rates of decomposition, mineralization, and nutrient retention. For example, Singer et al. (1984) found that intensive pig disturbances in a southeastern U.S. deciduous forest was associated with significantly greater nitrate, ammonium, and potassium pools in the soil. In a mesic forest on the island of Hawaii, Vitousek (1986) also found that net nitrogen mineralization was greater in pigdisturbed soils than in areas protected from disturbance for at least 14 years. In contrast to these results, Moody and Jones (2000) found no correlation between pig disturbances and changes in soil pH, moisture, nitrogen pools, or total carbon for oak savannas on Santa Cruz Island, off the coast of southern California. What kinds of impacts have been detected for the two grassland sites in northern California discussed earlier? At Angelo Reserve, Kotanen (1997a) found that simulated pig disturbance did not significantly affect nitrate pools, and results for ammonium were somewhat equivocal: Levels were higher than controls for one type of simulated disturbance and lower for another. At Salt Point State Park, Cushman et al. (2004) found no evidence that pig disturbances affected nitrogen mineralization rates or soil moisture availability. At the same site, Tierney and Cushman (2006) also found that ammonium and nitrate pools and mineralization rates in the soil did not change greatly through time following pig disturbance, nor was organic matter content or particle size affected. These results suggest that soil nutrients in northern California grasslands may not be greatly affected by pig disturbances. Numerous factors could explain the absence of effects. First, these grasslands may simply be able to withstand—or recover rapidly from—soil disturbances without experiencing significant changes in nitrogen availability. This might occur if nutrients do not leach rapidly from the system but instead collect in the many small depressions that pigs create. Another possibility is that grasslands have been intensively disturbed by pigs for many years and this has caused changes in soil characteristics prior to when the measurements occurred. Regardless of which factors are responsible, it seems that variation in soil characteristics cannot explain the many vegetation changes that investigators have detected. Instead, these results are more consistent with the hypothesis that changes in the plant community are caused by space clearing by pigs, which provides greater opportunities for colonization and reduced intensity of competition.

Population Control of Feral Pigs Soil disturbances by feral pigs can have strong effects on grassland vegetation in California and lead to the further degradation of this already altered ecosystem. However, at least in coastal grasslands, the encouraging news is that native vegetation appears to rebound following pig disturbance, provided that additional disturbance does not occur. Thus, reducing or eliminating pigs and their impacts is key for the effective management of coastal grasslands, as well as

more inland systems. However, once established, pig populations are difficult to eradicate, because of their high reproductive rate, their tolerance of diverse ecological conditions, and their ability to avoid contact with humans (Barrett et al. 1988; Waithman et al. 1999). In addition, although reductions in population size can be achieved, they are usually temporary; thus, control efforts must be viewed as continual activities. Multiple techniques have been used to reduce the size of feral pig populations. Ground-based hunting by professionals has been an effective control method in some areas (Cruz et al. 2005), and sport hunters are the most widespread control method in California (R. H. Barrett, personal communication). Aerial-based hunting from helicopters can be effective in some areas, especially in open landscapes (Long 1993; Saunders 1993; Choquenot et al. 1999). Hunting with teams of dogs has also been effective after pig densities are reduced by other forms of control (Sterner and Barrett 1991; Caley and Ottley 1995). Snares can be an effective, albeit controversial, method and have been especially successful in extremely remote areas if other, larger native mammals are absent, as in Hawaii (Anderson and Stone 1993). Poisoned baits have also been used with success in Australia (Twigg et al. 2005), although they can be problematic if native species consume them as well. Whatever control method is used, success can be increased greatly when pig populations occur within fenced areas and their ability to leave the area is reduced or eliminated. Fencing is also critically important for preventing pigs from recolonizing areas that have been previously cleared of these animals (Barrett et al. 1988). Complete eradication of feral pigs is usually considered possible only on islands or within fenced areas. For example, Cruz et al. (2005) reviewed the successful efforts to eradicate pigs from Santiago Island in the Galápagos Archipelago in Ecuador. During a 30-year period, over 18,000 pigs were removed from the island using a mixture of ground-based hunting and poisoning. The success of this eradication effort — which is the largest to date — was attributed to a sustained effort that used a combination of techniques and an intensive monitoring program. Another example comes from southern California, where intensive efforts have been mounted over multiple decades to eradicate feral pigs from four of the Channel Islands: San Clemente, Santa Catalina, Santa Cruz, and Santa Rosa. Pigs were eradicated successfully from San Clemente Island in the 1980s and from Santa Rosa Island in the 1990s (Long 1993). In contrast, the challenge has been more formidable for Santa Cruz and Santa Catalina Islands because of their larger size and more complex vegetation and terrain. In 1989, a study was initiated to assess the feasibility of eradicating pigs from Santa Cruz Island. Trapping, ground-based hunting, and hunting with dogs were all used to successfully eradicate pigs from a 2,250 hectare exclosure (Sterner and Barrett 1991). In 2005, The Nature Conservancy and the National Park Service initiated an effort to eradicate pigs from the entire island (L. A. Vermeer, personal communication). To facilitate this

H I S T O R Y A N D E C O L O G Y O F F E R A L P I G I N VA S I O N S

195

program, they divided the island into five zones using over 27 miles of pig fencing. A professional eradication firm from New Zealand was contracted to systematically remove pigs from each of the five zones. The eradication firm used a variety of techniques, phased over the course of the removal effort, including aerial hunting from a helicopter, use of large, humane corral traps, ground hunting in formation with tracking dogs, and radio-collared sentinel pigs (Morrison et al. 2007). On Santa Catalina Island, an intensive effort succeeded in eradicating pigs from a 38 square-kilometer fenced area between 1995 and 1997, but pigs still remain in others parts of the island (Garcelon et al. 1993).

Conclusions Managing pig-invaded ecosystems is extremely complex because the soil disturbances that pigs create have a mixture of positive and negative effects on different components of a community. This situation highlights the increasingly common challenges that resource managers worldwide must face: They must simultaneously contend with multiple invaders that interact with each other and native taxa in complex and often unpredictable ways. Also, native taxa are heterogeneous in their responses to disturbances or management actions. Different functional or life form groups usually vary greatly in their responses to disturbance, and even members of the same functional group can behave differently as well. Such complexity and variability are not unique to feral pigs or grasslands in California and is discussed in other chapters; for example, variable responses of plant species to livestock

196

ECOLOGICAL INTERACTIONS

grazing are described in Chapters 17 (Jackson and Bartolome) and 20 (Huntsinger et al.), and species-specific responses to climate-change manipulations are discussed in Chapter 19 (Dukes and Shaw). Such variability becomes most apparent when a community-level approach is taken that focuses on the responses of taxa from different functional groups and geographical origins. For understandable logistical reasons, ecologists and resource managers often shy away from such studies because they are labor-intensive and the data they generate are complex and often challenging to analyze statistically. Yet, to effectively and sustainably manage natural landscapes, it is necessary to embrace rigorous, science-based approaches that evaluate community-level effects of invasions and the responses to subsequent management actions.

Acknowledgments I thank the editors for inviting me to participate in the exciting volume. Rick Sweitzer kindly provided details about his research on feral pigs, and Lotus Vermeer provided background on the pig-eradication effort on Santa Cruz Island. Comments from Reg Barrett, Carla D’Antonio, Peter Kotanen, Meredith Thomsen and Lotus Vermeer greatly improved this manuscript. I am indebted to Mark Stromberg for assistance with the bibliography and to Brian Anaker for help with Figure 16.1. My research has been generously supported by grants from the California Department of Fish and Game, California Department of Parks and Recreation, Sonoma State University, Environmental Defense Fund, and the National Science Foundation (DEB-9981663).

SEVENTEEN

Grazing Ecology of California Grasslands RAN DALL D. JAC KS O N AN D JAM E S W. BARTO LO M E

Grazing is an ecosystem process broadly defined as feeding on herbaceous plants, algae, fungi, or phytoplankton (Begon et al. 1996). The present review is constrained to the effects of grazing by large mammals on herbaceous plants and selected animals in California’s grasslands, savannas, and associated herbaceous riparian areas. A brief history of grazing in California grasslands illustrates that while livestock grazing is a relatively recent phenomenon, California grasslands have always been grazed by large and small animals, albeit in many different ways. The grazing process can be described as comprising three phenomena: defoliation, trampling, and nutrient redistribution. However, research results explicitly linking these processes to plant responses at the individual, population, and community levels of ecological organization are lacking for California grasslands. Hence, the present discussion relies largely on phenomenological research generated by range scientists studying productivity and composition responses to grazing management. This work has shown that California annual grassland productivity can be effectively managed for livestock production via manipulation of grazing intensity (Bartolome et al. 1980), but that species composition is more or less entrained by intra and interannual weather ( Jackson and Bartolome 2002). Because of this, these annual grasslands have been characterized as nonequilibrium systems, in which plant-plant and plant-animal interactions are of minimal importance relative to abiotic constraints (Wiens 1984). In contrast, equilibrium systems are those in which biotic interactions such as competition and herbivory are key drivers of plant community structure. Some evidence exists that perennial dominated herbaceous communities in California behave in a more or less equilibrium manner ( Jackson and Allen-Diaz 2006). The discussion is further expanded to grazing effects on native and non-native, invasive plant species and the grassy understory of savanna trees. California grasslands are

components of a broader landscape mosaic that includes tree canopies and riparian corridors. Therefore, grazing effects on oak savanna understory vegetation, which is usually dominated by annual grasses and forbs, are discussed here, but only where the presence of the canopy is known to modify disturbance dynamics observed in open grasslands or where no information exists for open grasslands. Similarly, grazing effects on the herbaceous component of wetlands and riparian areas are discussed, whereas grazing impacts on shrubs and woody vegetation are not. Finally, grazing effects on California grassland wildlife are briefly discussed, and ways are noted in which grazing by nondomestic large mammals such as elk differs from grazing by livestock.

Brief History of Grazing in California Grasslands Grasses appear in the North American fossil record during the Eocene (45–55 million years ago), somewhat coincident with the advent of high-crowned teeth in mammals ( Janis et al. 2002; Stromberg 2002). Grasses have evolved habits (prostrate growth) and structures (basal meristems, awned spikelets, and silica deposits in cell walls) to avoid or tolerate aboveground tissue loss (Briske 1991), and they typically respond to defoliation with elevated relative growth rates (Ferraro and Oesterheld 2002). Hence, the removal of plant tissue (e.g., disturbance) is a fundamental process in the grassland biome (Knapp et al. 1999; Woodward et al. 2004; Bond et al. 2005). The earliest California grass fossils date to the Pliocene (5.4 – 2.4 million years ago) (Axelrod 1944), so grazing animals have been more or less a part of California grassland, savanna, and woodland ecosystems for millennia (Edwards 1992). Before Eurasian contact in 1769 and the establishment of widespread cattle, sheep, and horse grazing, as well as market hunting (Burcham 1957; Burcham 1975), large herds of pronghorn antelope and tule elk grazed California grasslands (Edwards 1996). Descriptions of prehistoric

197

California grassland vegetation are presented in Bartolome et al. (2007), Edwards (1992), and other chapters in this volume (Wigand et al., Chapter 4; D’Antonio et al., Chapter 6). These accounts all paint a picture of dynamic systems in which grazing, in addition to other disturbances such as fire (Anderson, Chapter 5; Reiner, Chapter 18), are fundamental evolutionary and ecological processes. The composition of herbaceous vegetation in California prior to Eurasian contact is unknown. Many believe native perennial grasses, particularly the bunchgrass Nassella pulchra, once were much more abundant (Clements 1934; Beetle 1947; White 1967; Heady 1977). Hamilton (1997a) has rather convincingly argued against overuse of this paradigm, citing overextrapolation of Clements’s climax community concept (Clements 1936) and the dogma that has derived from it. Hamilton suggested that native annuals or shrubs were once dominant, especially in drier parts of the grassland. Holstein (2001) argued that the rhizomatous perennial grass Leymus triticoides dominated the preagricultural Central Valley floor. However, his analysis partially relied on the relict method for which he and others criticize Clements. The relevance of this discussion to grazing ecology of California grasslands is that dramatic increases in livestock grazing intensity in the late nineteenth century are often implicated as one of the main drivers of the shift from a perennial grassland flora to one dominated by annuals (Burcham 1975). Burcham (1957) reviewed the history of human colonization of California and concomitant livestock introductions. The use of the California grasslands for domestic livestock production began with Spanish colonization in 1769 and establishment of missions along the coast, but significant livestock grazing began around 1773 (Bartolome et al. 2007). Widespread grazing expanded inland beginning in 1824 when land was granted for vast cattle ranches. Livestock data for the nineteenth and early twentieth centuries showed that cattle and sheep densities were highly variable from county to county and year to year, but averaged around about 4 million head (Ewing et al. 1988). Since 1970, cattle numbers have remained constant fairly consistent at about 5 million head (USDA-NASS 2006). Sheep densities began a steep decline in 1960, when their numbers went from about 2.5 million head, which had been their 100 year average, to less than 0.5 million head by 2004 (USDA-NASS 2006). Present-day classification of California grasslands falls along geographical boundaries with two main subtypes: perennial grass – dominated Coastal Prairie and annual grass – dominated Valley Grassland (Bartolome et al. 2007). Recently, a third subtype, Coast Range Grassland, was shown to be floristically distinct, exhibiting a relatively even distribution of perennial and annual grasses ( Jackson and Bartolome 2002). However, little specific information is available for this association, so the present discussion relies largely on information generated from the first two broad grasslands categories, both of which grade into areas of increasing oak (Quercus spp.) cover (Allen-Diaz et al. 1997).

198

ECOLOGICAL INTERACTIONS

The Grazing Process and Its Management Grazing affects grasslands directly by removing leaf area via defoliation, compacting soils via trampling, and altering biogeochemical cycles by redistributing nutrients in time and space (Heitschmidt and Stuth 1991; Heady and Child 1994). Impact from each of these phenomena can be manifested at various scales including the genetic, individual, population, community, and ecosystem levels of ecological organization. Furthermore, the relative importance of these factors may be dependent on site- and time-specific variables that are naturally dynamic; for instance, an understanding of grazing effects for tallgrass prairie is not likely transferable to annual grassland. In fact, grassland responses to grazing between two climatically distinct regions of California grassland may be more different than responses between California coastal prairie and Midwestern tallgrass prairie, which are both dominated by perennial grasses. Hence, making generalizations about grazing responses across California grasslands is tenuous at best (Bartolome 1989). When considering the effects of large herbivores on grasslands, it is useful to classify by the type of animal, as well as the intensity, timing, frequency, and duration of grazing (Heady and Child 1994). Common grazing animals in present day California grasslands are cattle and sheep (livestock), as well as elk, deer, gophers, voles, moles, and grasshoppers and other arthropods. The grazing intensity of a particular animal is defined as the proportion of forage (phytomass available for grazing or browsing) removed. It may be measured in a variety of ways, but for grasses it is usually determined by estimating the amount of phytomass before and after a grazing period. Because grazing intensity is strongly linked to defoliation responses by the plant and its ability to compensate for loss of herbage, this parameter is a very important aspect of grazing management (Bartolome 1993). The timing, frequency, and duration of grazing also are important in determining the impact of grazing on target plant species (Heady and Child 1994). These factors often interact with animal preference and plant palatability and thus can be keys to developing grazing prescriptions and understanding the effects of different kinds of grazers (Sampson 1952). Nonetheless, overall there are few quantitative studies comparing the effects of different types of grazers in California grasslands.

Defoliation The most direct effect of an herbivore on a plant is to selectively reduce leaf area, which results in a short-term reduction in carbon gain (Del-Val and Crawley 2005). How the plant responds to reduced photosynthetic capacity largely dictates how the individual will fare. Hence, defoliation intensity, environmental stress, grazing history, genetic potential, and biotic constraints (such as competition and life history stage) interact to determine the plant’s fate. Lemaire and Chapman (1996) cite two general responses to defoliation: (1) short-term redistribution of carbon and nitrogen within the plant and (2) long-term morphogenetic response in which the shape of the plants in a

community changes over time. Upon defoliation, grasses draw on carbohydrate and nutrient reserves stored in nonphotosynthetic structures such as root crowns, roots, and rhizomes, but the magnitude of this reallocation of internal plant resources is related to the remaining leaf area. If some photosynthetic tissue remains after defoliation, regrowth is primarily from carbon gain via photosynthesis (Richards and Caldwell 1985). However, root:shoot ratios of perennial grasses decline as a result of defoliation (Ferraro and Oesterheld 2002) because carbon gained from photosynthesis is mainly allocated aboveground to rebuild photosynthetic apparatus (Turner et al. 1993). Such changes in allocation and reallocation are thought to translate to population-level effects in perennial grasslands because these responses are species-specific (Dyer et al. 1993; Damhoureyeh and Hartnett 2002). Whole-plant responses of several species of annual grasses to variation in the frequency and timing of defoliation were evaluated in a series of pot experiments in the early 1960s (Savelle and Heady 1970). Generally, these studies showed that shorter-stature species such as Bromus madritensis and Vulpia bromoides were able to grow more rapidly and reproduce better under higher-frequency and later-season clipping in pots than were taller species like Bromus diandrus and Avena barbata. Validation of these pot studies in situ has not occurred. In annual grasslands, population-level responses to grazing intensity gradients are hypothesized to arise from soil seedbank alteration from multiple years of grazing pressure (Pitt and Heady 1979; Rosiere 1987; Heady et al. 1992), but these effects are difficult to separate given the overriding influence of intra- and interannual weather fluctuations on the plant community (Bartolome 1979). Morphogenetic responses are common in perennial grasslands elsewhere (Holland et al. 1992) but have not been documented in California grasslands. Shorter-stature annual plants benefit from defoliation events relative to taller-stature annual grasses (Savelle and Heady 1970), and therefore their progeny should preferentially propagate. This should allow them to persist within grasslands where taller-stature species would otherwise be competitive dominants. Holland et al. (1992) showed that perennial grasses from an area with a history of grazing produced greater biomass after experimental defoliation than conspecifics from areas without a grazing history in Colorado shortgrass steppe. These short- and long-term responses are thought to contribute to compensatory growth—the ability of a plant to regrow at faster rates post-defoliation such that total biomass at season’s end is equal to that of an undefoliated conspecific (McNaughton 1979). This phenomenon has been reported for several systems (Williamson et al. 1989; Hik and Jefferies 1990), but Ferraro and Oesterheld’s (2002) meta-analysis showed that it was not a general response in individual plants. Field studies in which plants are growing in competition with each other further cloud the issue. Leriche et al. (2003) modeled compensatory growth showing that resource availability modulated the sign of response; that is, whether plants under- or overcompensated for defoliation depended

on the resource status of the system. These phenomena have not been demonstrated in annual grasslands or for native perennial grasses in California. Determining the effects of defoliation on the structure and productivity of the plant community depends on the spatial and temporal scale examined (Bartolome 1989). In one study of sheep in annual grassland, the selective nature of grazing was confirmed but was not subsequently manifested in changes in plant abundance (Bartolome and McClaran 1992). The inability to reliably scale up research to the community and landscape levels of organization has severely compromised prediction of defoliation and grazing effects in California grasslands (Bartolome 1993; Hayes and Holl 2003b).

Trampling Ungulates physically alter soil structure because their rather substantial mass, as much as 1,000 kg per animal, is carried by relatively small hooves. The usual effect is compaction, which is quantified as bulk density (Pietola et al. 2005); however, compaction responses can be negligible. Increases in soil compaction result in reduced infiltration rates, which in turn increase surface runoff and erosion (Daniel et al. 2002). Trampling effects in California grasslands are not uniform because livestock preferentially use areas near shade and water sources (Tate et al. 2003). That said, two independent studies found that light to moderate livestock grazing in and around riparian areas of oak savanna did not significantly alter the morphology of streambanks (Allen-Diaz et al. 1998; George et al. 2002). However, it has been observed frequently that heavy grazing can reduce vegetation cover and decrease the slope of streambanks, resulting in bank erosion and degraded aquatic habitat (Larsen et al. 1998). Research on trampling effects in California grasslands has mainly occurred at the San Joaquin Experimental Range (SJER) in Fresno County (for a review see Menke 1989). Tate et al. (2004), working in the valley grassland/oak woodlands at SJER, found greater compaction in areas with moderate to heavy grazing intensities compared to historically ungrazed areas, indicating undesirable effects on soil physical properties. D’Antonio and Tyler (unpublished) found significantly increased bulk density within grazed compared to ungrazed (past 10 years) paddocks in Valley grasslands at Sedgwick reserve in Santa Barbara County with effects showing 10% more compaction in some grazed plots. Grazing was at moderate intensity and occurred only in the wet season.

Nutrient Redistribution Biogeochemical cycles are altered by grazing because herbivores mineralize organic matter and return it to the environment in solid, liquid, and gaseous forms (Allen et al. 1996; Hack-Ten-Broeke and Van Der Putten 1997; Oenema et al. 1997; Carran and Theobald 2000; Luo et al. 2000, Di and Cameron 2002; Anger et al. 2003). In general, grazing in grasslands accelerates carbon and nutrient cycling by effectively bypassing the microbial decomposition pathway (Ritchie

GRAZING ECOLOGY OF CALIFORNIA GRASSLANDS

199

F I G U R E 17.1. Typical herbaceous production for California annual

grassland. The magnitude of response varies positively with rainfall. Grazing has some influence on live biomass at peak standing crop (PSC) by altering the amount of dead biomass or residual dry matter (RDM) at autumn germination (see Figure 17.2).

et al. 1998; Singer and Schoenecker 2003). This acceleration happens in a spatially heterogenous manner because livestock use some areas preferentially and because their excreta is deposited in patches that are a small fraction of the grazed landscape (Tate et al. 2000; Tate et al. 2003). Nitrogen quickly cycles within annual-dominated ecosystems (Woodmansee 1978; Jones and Woodmansee 1979; Woodmansee and Duncan 1980; Schimel et al. 1989; Davidson et al. 1990), where plant species possess low nutrient use efficiencies and high litter qualities irrespective of defoliation (Savelle 1977; Eviner 2004). In perennial grasslands of the Great Plains and the Upper Midwest, accelerated nutrient cycling as a result of livestock grazing is credited for stimulating net primary productivity (Frank and McNaughton 1993; Frank et al. 1994; Paine et al. 1999). However, grazing effects on nutrient dynamics in California annual grassland have not been observed (Davidson et al. 1993; Dahlgren et al. 1997; Herman et al. 2003).

Grazing Effects on Grassland Productivity Factors at many spatial and temporal scales interact to control herbaceous productivity in California’s annual grasslands (Bartolome 1989). Aboveground biomass at late spring seed set varies interannually as a function of the timing and amount of precipitation and temperature (Talbot et al. 1939; Bentley and Talbot 1948; Heady 1958; George et al. 1988), and edaphic and topographic characteristics ( Jackson et al. 1988; Callaway et al. 1991). A typical yearly production curve for annual grassland includes the onset of autumn germination with the first rains over 2.5 centimeters occurring within a one week period (Figure 17.1). Slow winter growth progresses as temperatures decline, followed by rapid spring growth as soil temperatures increase concurrent with adequate soil moisture (Chiariello 1989). Peak standing crop of the herbaceous vegetation generally occurs between April 1 and May 15, followed by the death of the annual plants. Standing dead biomass slowly decomposes as summer drought slows microbial activity until the ensuing autumn rains stimulate decomposition concurrent with annual plant germination (Jackson et al. 1988).

200

ECOLOGICAL INTERACTIONS

The first concerted ecological research studies in the annual grassland were directed toward understanding the forage base for livestock grazing (Sampson 1917; Bentley and Talbot 1948; Sampson et al. 1951; Biswell 1956). Much of this work also established the primacy of location and weather as factors controlling herbaceous production and composition (Talbot et al. 1939; Heady 1958). Later studies and management practice have shown that the effects of grazing are related to the abundance of litter or residual dry matter (RDM, e.g., the senesced plant phytomass) remaining at the time of autumn germination (Hedrick 1948; Heady 1956, 1965; Bartolome et al. 1980; Bartolome et al. 2002). To maintain long term livestock production in California annual grassland, range managers must cope with the vagaries of California climate. Bartolome et al. (1980) demonstrated that within a range of RDM levels representative of typical grazing intensities, RDM had a positive relationship to peak standing crop of the ensuing year (Figure 17.2). This relationship was roughly general along a rainfall gradient from southern San Joaquin Valley to the northern Coast Ranges and has been validated for open annual grassland at a Sierran foothills site (Betts 2003). This relationship was weakest at sites with less than 20 cm total annual rainfall. Mechanisms for this relationship have not been elucidated but are believed to be a combination of favorable light, space, nutrient, and water modifications by moderate levels of RDM (Xiong and Nilsson 1997, 1999). The high-RDM extreme (i.e., no defoliation), which was not tested by Bartolome et al. (1980), inhibited production on a Sierran foothill site (Bartolome et al., in press). Belowground biomass responses to RDM were estimated by Betts (2003), who found that any aboveground defoliation treatment (i.e., 50, 100, or 150 g RDM⭈m⫺2) in open annual grassland increased the ensuing year’s root:shoot biomass ratios (⬃0.8) compared to undefoliated controls (⬃0.5), which averaged 500 g RDM⭈m⫺2. This root:shoot response is opposite to that typically found in perennial grasslands in Continental climates, where defoliation reduces root production (Turner et al. 1993; Johnson and Matchett 2001; Ferraro and Oesterheld 2002). To our knowledge, no other studies of belowground biomass responses to defoliation in California grasslands exist, reflecting past emphasis on production of available forage for livestock. The relatively recent emphasis on carbon sequestration in grassland soils to mitigate the accumulation of atmospheric greenhouse gases will likely stimulate belowground productivity research (Conant et al. 2001; Follett et al. 2001; Conant and Paustian 2002; Shaw et al. 2002; Dukes et al. 2005).

Grassland Plant Community Responses to Grazing Species Composition In the 1950s and 1960s, Harold Heady at the University of California at Berkeley conducted a series of experiments showing that fall RDM dramatically influenced productivity and species composition in a high-rainfall (89 cm⭈y⫺1)

over time, while mowing uniformly defoliates in pulse-like events. For the purpose of this study, plant communities from plots under a range of RDM treatments located along the latitudinal, hence rainfall, gradient (Figure 17.2) were first classified. Second, the classification and regression tree (CART) analysis was used to examine the amount of deviance explained in interannual transitions amongst these communities. Approximately 60% of the deviance was attributable to location and weather patterns, less than5% of the deviance was explained by the RDM gradient, and about 35% of the deviance was unexplained. In these sites, composition seems to be entrained by annual weather patterns that appear to render overall community manipulation via RDM manipulations futile. However, plant diversity (Meyer and Schiffman 1999) or a single species, such as a native perennial grass (Hatch et al. 1999; Bartolome et al. 2004) or invasive plant species (Thomsen et al. 1993; Betts 2003), may still be managed by manipulating the timing, intensity, frequency, or duration of grazing as will be discussed below. RDM-based management prescriptions, combined with manipulations of timing and intensity of grazing, will need to be made and monitored on a site- and time-specific basis.

1 2 3 4 5 6 7 8 9

6

Herbage production (kg ha –1 )

5000

4000

2 1 3

Plant Species Richness 3000

4 5

2000

7

1000

8 9

0 280 560 840 1120 Residual dry matter (kg ha -1 ) F I G U R E 17.2. Herbaceous production response to residual dry matter manipulations at nine sites along a N-S rainfall gradient (modified from Bartolome et al. 1980).

Mendocino County annual grassland site (Heady 1956, 1958, 1965). With no RDM at the time of germination, Aira carophyllea, Baeria chrysostoma, Hypochaeris glabra, and Triphysaria eriantha dominated the vegetation on the site. When plant residue was left on the ground, the relative cover of Bromus hordeaceus increased from 1–37% in 3 years; when all RDM was removed each year, the cover of B. hordeaceus remained at less than 2%. However, recent analysis of the species composition component of the Bartolome et al. (1980) dataset, which spanned nine sites and five years, indicated that while some species may respond to RDM manipulations in some years, overall plant community composition is relatively insensitive to these changes ( Jackson and Bartolome 2002). It is important to note that all of these studies used mowing rather than grazing to achieve RDM levels, which may bias results because grazing is a selective process that is distributed

Working in coast live oak savannas, Marañon and Bartolome (1994) showed plant species richness was higher under intermediate aboveground biomass levels compared to either lower or higher levels. These results follow the model of Grime (1979b), which was discussed by Marañon and Garcia (1997), that hypothesizes a unimodal distribution of plant species richness along a productivity gradient. Marañon and Bartolome (1994) listed biomass levels for maximizing species diversity and California annual grassland/oak woodlands at 35–57 g RDM⭈m⫺2. Conversely, Bartolome et al. (2004) found no relationship between grazing intensity, measured as stocking rate, and diversity indices in perennial grass stands in the Coast Range. Fehmi and Bartolome (2002) showed that a possible tradeoff between livestock and rodent herbivory exists in these grasslands. In their study, rodents appeared to preferentially locate burrows (therefore disturb vegetation) in areas where livestock were excluded, i.e., high-cover sites. This agrees with observational surveys by Stromberg and Griffin (1996), who demonstrated higher gopher activity in long ungrazed compared to grazed pastures in Monterey County. Hence, any response of plant species diversity to livestock grazing regime changes may be canceled by rodent activity. Harrison et al. (2003) examined effects of grazing and burning on richness in annual grasslands on serpentine and nonserpentine soils in the Coast Range north of San Francisco Bay. Grazing increased native plant diversity on the less productive serpentine soils but not on the more productive nonserpentine soils. The authors of this study point out, however, that grazing and burning effects were weak relative to extrinsic factors such as soil type. These results were validated by Safford and Harrison (2001), who also showed

GRAZING ECOLOGY OF CALIFORNIA GRASSLANDS

201

that higher plant diversity in grazed serpentine grasslands compared to serpentine roadside verges, which are typically ungrazed areas located between a road and a fenceline. A follow-up survey of 92 sites on Valley Grassland of the northern Coast Range found that the presence or absence of grazing interacted with soil type and aspect in its effect on native and exotic plant species richness (Gelbard and Harrison 2003). Grazing was associated with lower native grass species diversity on nonserpentine soils, but it did not affect the richness of native forb species. Conversely, less fertile serpentine soils supported greater native forb richness with grazing but showed no effect of grazing for native grass richness. In the wetter coastal prairie grasslands, a study of grazed and ungrazed sides of fencelines (Hayes and Holl 2003b) revealed that grazing was associated with a higher diversity of native annual forbs but also a higher richness of exotic annual species. Native perennial forbs showed decreased richness with grazing while exotic forbs increased with grazing. This study did not evaluate the role of other factors such as soil type or aspect.

Grazing Effects on Particular Taxa and Functional Groups Native Grasses and Forbs Most studies of grazing effects on native perennial grasses have emphasized the role of grazing in the demise of the putative dominant of the pre-Eurasian contact vegetation, Nassella pulchra. However, searching for edaphic and geographic correlates with N. pulchra, Bartolome and Gemmill (1981) rejected the notion that this species represents relictual dominance and hypothesized that it likely is a disturbance-adapted species that finds refuge in places where light is less limiting than belowground resources. In support of this, Dyer (2002) found that N. pulchra seed from grazed and/or burned individuals germinated and survived at higher rates than seed from undisturbed plants. Irrespective of its relictual status, many researchers have sought to determine management techniques, namely, combinations of grazing, burning, or grazing removal (Menke 1992; Hatch et al. 1999; Bartolome et al. 2004; Marty et al. 2005), that may enhance its abundance. Some of these studies indicated that N. pulchra was not especially tolerant of defoliation (Dennis 1989; Huntsinger et al. 1996; Marty et al. 2005). However, Dyer and Rice (1999) and Hamilton et al. (1999) demonstrated that N. pulchra is susceptible to competition from non-native species, and when growing among non-native annual species, it benefits from grazing because diffuse competition from these annuals is reduced (Dyer and Rice 1997b; Malmstrom et al. 2006). Bartolome et al. (2004), working in Coast Range Grassland, found a positive response of Nassella pulchra and N. lepida to a spring grazing treatment but only after the cessation of the treatment. These results indicate some positive residual effect of their seasonal grazing treatments that was magnified by removal of the disturbance. Many of the studies of Nassella spp. responses to grazing stress the importance of

202

ECOLOGICAL INTERACTIONS

climate variability in influencing their results and Nassella dynamics in general (Dyer et al. 1996, Merenlender et al. 2001; Bartolome et al. 2004; Marty et al. 2005). California grasslands dominated by other perennial grass species, which are found mainly along the coast (i.e., Coastal Prairie), have received considerable attention recently (Stromberg and Griffin 1996; Hayes and Holl 2003b; Corbin et al. 2005). Working in three coastal prairie sites, Hayes and Holl (2003a) found that a gradient of defoliation intensity (clipping) reduced exotic grasses but had no effect on native perennial grasses, including the common coastal prairie grass Danthonia californica. This species was present in the previously discussed Bartolome et al. (2004) study plots, and it increased with removal of livestock grazing. These results conflict with those of Hatch et al. (1999) and Biswell (1956), who found grazing removal decreased cover of D. californica. Likewise, in their extensive coastal fenceline survey Hayes and Holl (2003b) found that ungrazed areas had lower cover of D. californica than grazed areas, suggesting that this species benefits from grazing in at least some settings. Several recent studies have evaluated the effects of grazing or clipping on native forbs. In a three-year clipping and mulch-removal study in the grassland of Carrizo Plain National Monument, San Luis Obispo County, Kimball and Schiffman (2003) found that clipping one to three times per growing season significantly reduced cover and species richness of native annual forbs and the native annual grass Vulpia microstachys. In contrast, non-native grass and forb cover and species richness were largely unaffected by the simulated grazing. They speculated that Mediterranean grassland species, exposed to livestock grazing for many centuries, have adaptations such as compensatory growth that enable them to recover from grazing, while native species generally lack these mechanisms. In contrast to these results, grazing has been shown to favor native annual forbs in several other recent studies from other parts of California (Safford and Harrison 2001; Gelbard and Harrison 2003; Hayes and Holl 2003b; Marty 2005). These varied and conflicting results indicate the importance of separating grazing effects on individuals, populations, and communities as well as the species-specific nature of grazing responses. If target perennial grasses are not already present on a site, no amount of grazing management or removal will encourage their abundance (Merenlender et al. 2001). This may also be true of some native annual forb species, although seed bank limitation has not been tested for most forbs in California. If present on a site, native perennial grasses and native annual forbs have been shown to respond favorably to some disturbance treatments as previously discussed. The particular combination of treatments depends on geography, soils, interspecific competition, and possibly weather patterns.

Non-native Species Little direct evidence exists that grazing promotes invasion of undesirable non-native plant species in California

grasslands, though the nineteenth century invasion of exotic annuals that now dominate occurred at the same time that livestock densities dramatically increased (Burcham 1957; Heady et al. 1992; Belsky and Gelbard 2000). Grazers may alter the competitive balance of natural communities by altering nutrient cycles and creating gaps for colonization— in effect, creating an alternate ecosystem more conducive to the invader (D’Antonio et al. 1999). But Levine et al.’s (2004) meta-analysis showed that herbivory by native herbivores generally reduced colonization and spread of invasive species, although most of their examples were from nongrassland communities dominated by native species. Of course, this ignores the fact that grazers, particularly large mammals, are excellent dispersal agents for plant species, whether invasive or not. Theory notwithstanding, little or no evidence linking grazing intensity to current plant invasion exists for California grasslands. Yellow starthistle (Centaurea solstitialis) is a summer annual whose cover is expanding at an alarming rate on California rangelands. Gerlach and Rice (2003) determined that C. solstitialis colonized disturbed open patches more readily than its less invasive congeners. While they did not explicitly examine grazing or defoliation, they inferred that grazing intensities that exposed bare ground would facilitate invasion and spread of this noxious species. Nonetheless, C. solstialis was less sensitive to competition from grasses than the congeners they also tested. Holmes and Rice (1996) and others (Reever Morghan et al., Chapter 7) imply that winter annual grass dominance facilitates C. solstitialis invasion by creating an untapped pool of soil water that C. solstitialis can exploit with its summer active phenology. Whether grazing promoted the nineteenth-century invasion of these annual grasses (which then promoted C. solistialis) is a question that cannot be decisively answered. Although C. solstitialis is relatively unpalatable and indeed is toxic to horses, Thomsen et al. (1993) showed that properly timed late spring/early summer intensive grazing by cattle or goats can be used as a tool to reduce cover and reproductive capacity of this species (Huntsinger et al., Chapter 20; Stromberg et al., Chapter 21; DiTomaso et al., Chapter 22). Betts (2003) found that the invasive annual barbed goatgrass (Aegilops triuncialis) was more likely to germinate and reach maturity in undisturbed plots than in plots with reduced RDM levels. In addition, A. triuncialis plants in ungrazed areas were more robust and produced more seeds than those goatgrass plants in grazed areas. Seedheads in grazed areas experienced significant predation by granivorous rodents, and the high amount of RDM in the ungrazed areas also may have created a more favorable microenvironment for A. triuncialis. Although grazing may not be able to control A. triuncialis directly (goatgrass is unpalatable to livestock), grazed areas may be under lower threat of goatgrass establishment than ungrazed areas. Medusahead (Taeniatherum caput-medusae) is another annual grass that is a significant pest in grazed grasslands of California because of its low palatibility and nutritional

quality for livestock (Young 1992; Betts 2003). This grass has been estimated to reduce grazing capacity by 40–75% (Major et al. 1960). Although sheep will not eat silage containing medusahead (Bovey et al. 1961), they will eat young medusahead plants (Lusk et al. 1961) such that moderate-intensity, early-season grazing may reduce medusahead cover (Cooper 1960). Results from the studies previously cited illustrate the importance of grassland management based upon scientifically produced information that is applicable to appropriate sites, scales, and taxa. Grazing managers control animal type, intensity, frequency, and duration of grazing; therefore, grazing effects should not be assessed as an “either-or” proposition.

Grazing Effects in Grassland Landscapes Grazing effects on California grasslands have been treated here as though the grasslands are very discrete, uniform units. However, these communities grade more or less strongly into oak savanna and woodland (Allen-Diaz et al. 1997). Riparian zones are nested within these landscapes, forming corridors and patches of mainly perennial vegetation. Virtually all of California’s drinking water passes through annual grassland and oak woodland ecosystems (Tate et al. 1999), which are predominantly managed with grazing (McClaran and Bartolome 1985; Standiford and Tinnin 1996). Suburban development within these rangelands places increasing pressure on management agencies to reduce wildfire fuel loads (Fried and Huntsinger 1998). On the steep terrain of California’s annual grassland/oak woodland watersheds, grazing and prescribed fire are the most feasible vegetation management options (Stephens and Ruth 2005). Hence, a watershed-level approach to understanding how grazing affects water quantity and quality is critical to managing these landscapes. The importance of the watershed level of organization notwithstanding, at this time only a single controlled and replicated paired-watershed experiment is under way to understand grazing and burning effects on water quantity and quality as well as vegetation dynamics at this scale (Dahlgren et al. 2001). Recent work showed that nitrate pulses in streams draining large Sierra Nevada watersheds were mainly the result of sedimentary rock weathering (Holloway et al. 1998), which demonstrates the importance of carefully separating management effects from “background” contributions to water quality (Tate et al. 1999). Experimental work at this scale will require long-term (decadal) datasets in which temporal and spatial variability can be separated from management signals (sensu Lewis et al. 2000; Jackson and Allen-Diaz 2006).

Grazing Dynamics Modulated by the Presence of Oak Canopy Dahlgren et al. (1997) described soils beneath oak canopy as “islands of fertility” because of greater carbon, nitrogen, and phosphorus stocks compared to adjacent open grassland sites. The patchiness of oak woodland canopy may be enhanced by the ability of oaks to garner water and nutrients

GRAZING ECOLOGY OF CALIFORNIA GRASSLANDS

203

from beyond the canopy perimeter, from the open grassland spaces between them and their neighbors, and then preferentially returning leaf litter below the existing canopy. If this model holds, it would constitute a redistribution of ecosystem resources within the savanna landscape (sensu Schlesinger and Pilmanis 1998; Cross and Schlesinger 1999; Huenneke et al. 2002). However, this process would constitute a positive feedback that is theoretically unsustainable in the long term, depressing herbaceous production in the open (assuming no nitrogen fixation or deposition in the open), but this effect has not been observed. An untested hypothesis is that herbivores provide a check on this effect by harvesting herbaceous resources from beneath the canopy and redistributing them in a more homogeneous way across the landscape. Herbaceous understory production is primarily controlled by interannual weather variability; however, several workers have demonstrated that tree canopy cover exerts a strong but variable influence on peak standing biomass, depending on regional location, tree density, and tree type (Frost and McDougald 1989; Callaway et al. 1991; Ratliff et al. 1991; Bartolome et al. 1994; Tyler et al., Chapter 14). Relative to open grasslands, canopy cover inhibits herbaceous production in areas of California receiving less than 50 centimeters annual precipitation (McClaran and Bartolome 1989a). The inverse relationship generally holds for drier portions of the state, where canopy cover attenuates drought stress. However, Callaway et al. (1991) showed experimentally that within-site variation in the effect of an oak canopy was related to shallow, fine root abundance of oaks (Tyler et al., Chapter 14). High oak fineroot biomass suppressed herb production, whereas low oak fine-root biomass promoted herb biomass exceeding that measured in adjacent open grassland. From a livestock production perspective, Frost et al. (1991) found that an increase in tissue and litter quality per unit biomass under oak canopy in drier regions more than compensated for any reduction in herbage mass from the canopy cover. Nutrient concentrations differed as a result of species composition differences rather than some change in individual plant nutrient use efficiencies.

Herbaceous Riparian Zones Nested within Grasslands High intensity grazing can negatively affect water quality, plant biodiversity, productivity, wildlife habitat, wildlife species biodiversity, and nutrient cycling in riparian areas in regions with Continental-type climates (Kauffman et al. 1983a, b; Kauffman and Krueger 1984; Fleischner 1994; Clary 1995, 1999). However, extrapolation of these results to Mediterranean-type regions should be made very cautiously (Larsen et al. 1998; Gasith and Resh 1999). Effects of moderate to light grazing on ecosystems of these regions tend to be overwhelmed by larger-scale environmental fluctuations. Nested within annual grasslands are riparian zones where relatively little scientific study about grazing effects has occurred in California. Riparian zones provide many important ecosystem services (Naiman and Decamps 1997; Sabater et al. 2000), which may be modified by livestock grazing or

204

ECOLOGICAL INTERACTIONS

its cessation. Much of the water exiting grassland watersheds passes through the highly productive plant communities that are situated at the terrestrial-aquatic interface (Huang 1997; Tate et al. 1999; Lewis et al. 2000). Wetland vegetation in these zones, typically cattail (Typha spp.), sedges, rushes, and perennial grasses, may act as nutrient filters (primarily nitrate) as waters emerge at the soil surface. Jackson et al. (2006a) determined with a paired-plot grazing removal experiment that nitrate concentrations in surface waters where grazing was removed for only two years were as much as five times greater than grazed counterparts, these concentrations far exceeded the U.S. EPA maximum standard for surface waters of 10 ppm NO3-N (Fan et al. 1987). They speculated that reduced herbaceous production that resulted from litter accumulation under grazing removal led to reduced plant nitrogen demand and hence greater throughput of nitrate to the downstream aquatic ecosystem. Spring-fed wetlands and first-order riparian areas are often the only sources of surface water in grasslands and are especially heavily utilized by grazing animals. However, light to moderate autumn/winter grazing had little effect on Sierra Nevada foothill spring-fed vegetation after 6 years of treatment (Allen-Diaz and Jackson 2000). Continued monitoring of these systems under experimental treatments showed that by years seven through ten, moderate grazing reduced herbaceous cover, light grazing had minimal effect, and grazing removal significantly increased cover (Jackson and AllenDiaz 2006). Furthermore, the long-term results from this study demonstrated that the riparian creeks emanating from the more marshy wetlands, where spring waters emerge, display dynamics (i.e., equilibrium community dynamics) that were fundamentally different from the nonequilibrium response exhibited in their upslope marshy spring counterparts (Figure 17.3). Equilibrium dynamics were investigated by Marty (2005) for vernal pools in the Valley Grassland, whose endemic, diverse, largely annual flora is deleteriously affected by the removal of livestock grazing. In this work, three years of grazing removal resulted in higher cover of exotic annual grasses, and lower cover and richness of native species, in vernal pool edges compared to grazed pool edges. Species in the pool bottoms did not respond to grazing regime manipulations. Further, species richness of aquatic invertebrates declined when livestock grazing was removed from pools. Marty also found that the upland grassland species between the pools responded to grazing similarly to the pool edge species: Native species richness and cover increased with grazing. The grazing regime producing the strongest response was continuous grazing (contrasted with wet-season and dry-season grazing). Marty argued that these pool assemblages evolved in the presence of large herds of tule elk, which possibly explains their current positive responses to grazing. Livestock production grazing management decisions are usually made at the landscape level based on the matrix vegetation (annual grassland in this case), but nested ecosystems, such as herb-dominated vernal pools, spring-fed wetlands,

F I G U R E 17.3. Means by year of DCA site scores (an index of plant community composition), which were normalized to 1992 pretreatment values for headwater springs and their resultant creeks. Mixed effects ANOVA showed DCA site scores were significantly different amongst all three grazing treatments at creeks, but not springs, indicating alternative, relatively stable equilibria at the former and nonequilibrium at the latter (from Jackson and Allen-Diaz 2006).

and creeks, may respond differentially, requiring a more nuanced approach that includes site-specific information relevant to management goals.

Wildlife Grazing and Grazing Effects on Wildlife Most literature treating grazing in California focuses on the productivity and composition of annual grasslands under livestock grazing, because this is the dominant land use of California rangelands (Huntsinger et al. 1997). However, a growing appreciation and understanding of wildlife grazers, such as elk, deer, and small mammals, is apparent in the literature (Lidicker 1989; Hobbs and Mooney 1995; Fehmi and Bartolome 2002; Eviner and Chapin 2003b, 2005). Schiffman (Chapter 15) provides a thorough review of small-mammal herbivory effects on the structure and function of California grasslands. Animals are known to graze selectively—they choose certain plants over others—but the impacts of selection have not been well-linked to observed changes in the plant community or productivity (Bartolome 1993). Of the modern large grassland herbivores, cattle, horses, and tule elk tend to prefer grasses (McCullough 1969; Heady and Child 1994), while sheep (Bartolome and McClaran 1992) and deer (Gogan and Barrett 1995) prefer forbs. Antelope, which were formerly widespread in the grassland but are now highly localized, are opportunistic and prefer grasses, forbs, or shrubs depending on the season (Yoakum and O’Gara 2000). Elk also change their diet seasonally, preferring forbs in the

spring and summer and grasses in the fall and winter (Gogan and Barrett 1995). Although native grazers and livestock differ in dietary preference, there is considerable overlap (Elliott and Barrett 1985). Recent work by Johnson and Cushman ( Johnson and Cushman 2007) demonstrated that removal of tule elk grazing from Coastal Prairie dramatically decreased annual plant cover while increasing some perennial grasses (e.g., the invasive Holcus lanatus) and having no effect on others. These results point to the importance of selective defoliation in perennial grasslands, which is likely to alter competitive interactions among functional groups. However, when only one functional group is present, as is the case with many annual grasslands, grazing effects on composition are difficult to detect. Given the importance of RDM at autumn germination on plant productivity and the nominal or highly variable effect of grazing on community structure, there is no reason to expect that wildlife grazing would impart any different effect on annual grasslands than livestock grazing unless their patterns of seasonal use are strongly different. Wildlife’s effect on productivity should be mediated by grazing intensity, as with livestock. The effects of grazing on several key wildlife taxa are listed in Barry et al. (2006). This publication attempts to decompose the interacting effects of grazing animal type, intensity, and season on the following: California ground squirrel, burrowing owl, kit fox, Bay checkerspot butterfly, Smith’s blue butterfly, steelhead trout, California quail, and tiger salamander. The numerous and varied interactions that make such an approach necessary preclude their exposition here. This and similar publications should prove useful for managers struggling with multiple, often conflicting, management goals and objectives. Posting and updating of these efforts on the Internet should allow them to be used opportunistically and adaptively — two key characteristics for management of highly variable nonequilibrium systems (Westoby et al. 1989).

Summary Grazing in Californian grasslands has occurred for millennia, but domestic livestock have grazed for only about 250 years. The introduction of domesticated livestock corresponded with a massive conversion from some more or less unknown past flora to one dominated by annual grasses and forbs native to the Mediterranean region. Grazing research in California was initiated early in the twentieth century. The main response variables during the majority of the century were forage production and species composition. This research resulted in a useful management model for livestock production, manipulation of residual dry matter (RDM), because a simple linear relationship between RDM and ensuing year’s production was roughly generalizable across the state’s annual grasslands. Although many have speculated on the mechanisms underlying this relationship — e.g., favorable modification of temperature, moisture, soil protection,

GRAZING ECOLOGY OF CALIFORNIA GRASSLANDS

205

nitrogen availability–the mechanisms have not been uncovered experimentally. Little literature exists that assesses grazing effects on California grassland nitrogen cycling, decomposition, or belowground productivity. Early range managers understood that intra and interannual floristic variability was dominated by weather patterns. Nonetheless, many sought to understand species composition in annual grasslands as a function of grazing management, because the assumption of equilibrium community dynamics, which were prevalent in the Great Plains and the Midwest (Clements 1936; Dyksterhuis 1949), dictated that plant-herbivore dynamics exert primary control on the plant community. The slow realization that nonequilibrium dynamics prevail in California and other arid and semiarid grasslands indicates that empirical approaches to understanding plant community responses to grazing are necessary, because they are data-driven and flexible and allow contingent effects of weather, geography, and history to be assessed (Allen-Diaz and Bartolome 1998). Similarly, site- and time specific data are beginning to show that grazing can be

206

ECOLOGICAL INTERACTIONS

effectively applied as a tool for native grass enhancement, invasive species control, fuel load reduction, and habitat management. Maintenance or restoration of ecosystem services in California grasslands demands a watershed-level understanding of grazing effects, which is sorely lacking at this time. Grazing managers in California’s Mediterranean climate should not expect to realize narrowly defined outcomes, especially when considering species composition. Instead, a range of possible results should be anticipated. If anything is apparent about California’s nonequilibrium grasslands, it is that the response of the system to grazing depends on when and where the response is observed.

Acknowledgments Thanks to Mitch McClaran, Rich Reiner, and Carla D’Antonio for comments that improved this manuscript and to Carla for her careful editing.

EIGHTEEN

Fire in California Grasslands R I C HAR D J. R E I N E R

Grasslands are characterized by the plant and animal species that inhabit them and also by periodic natural disturbances (Axelrod 1985; Jacobs et al. 1999; Sauer 1950; Wells 1962). Herbivory, burrowing by rodents, flooding, wind, and fire all affect the species composition, structure, and distribution of grasslands (Vogl 1974). Fire is often a major force, and understanding its role has important implications for grassland conservation, restoration, and management. This chapter provides some basic concepts about grassland fire and explores the role of fire in determining where grasslands occur, the impact of fire on species composition, and the ways in which fire can be used as a management and restoration tool. The focus primary is on California annual grasslands, which occur within the Mediterranean climate zone. Remnant native perennial grasses and stands composed of mixed annuals and native perennials are also considered.

because of their high surface-to-volume ratio. Unless there are woody or wet materials, the charcoal phase of grassland fires is short.

Fire Regimes The role that fire plays in a natural community is typically termed the fire regime. It is, in a sense, the community’s natural fire history. In grasslands, the fire regime is typically described by four factors: fire intensity, fire frequency, size and pattern, and seasonal timing. Each factor influences which species flourish in particular grasslands. The fire regime is influenced by characteristics of the vegetation, such as its vertical and horizontal distribution, floristic composition, and the accumulation of biomass. It is also influenced by physical site factors such as topography, soil type, and weather. Figure 18.1 illustrates how vegetation and physical site factors interact and influence the ultimate fire regime (Riba and Terradas 1987; Whelan 1995).

Fire in Grasslands Grasslands burn by releasing the energy stored in photosynthesis. The more energy that is stored in plant biomass, the more heat that can be released. Fires can occur only when an ignition source heats the fuel to the point of combustion. When the fuel is burning, oxygen is consumed and the byproducts are carbon dioxide, water vapor, and small amounts of other minerals and compounds bound in gases and ash. Three stages of combustion have been described as a fire advances. First, fuels in front of an advancing fire are preheated; second, they combust at their “kindling temperature”; and third, they continue with glowing combustion, in which the remaining material burns as a charcoal (Whelan 1995). The preheating phase is particularly relevant to grassland fires, because an advancing fire can quickly dry grass fuels

Grassland Fire Behavior Once ignited, three factors most affect fire behavior: fuel, weather, and topography (Biswell 1989). The effect of grassland fuels on fire behavior depends on moisture content, the ratio of dead to live fuel, the fineness (diameter) of the fuel, fuel continuity, and fuel loading. The moisture content of grassland fuels is an especially important factor in understanding fire behavior because grass fuels react quickly to changing temperature and humidity. This characteristic is due mostly to their high surface-to-volume ratio. The aspects of weather that most influence fire behavior are temperature, relative humidity, and wind velocity and direction. In general, higher air temperatures result in greater ease of ignition, rate of spread, and fire intensity. Humidity is related to temperature, and it affects fire intensity in grasslands

207

F I G U R E 18.2. A back fire at Vina Plains Preserve, Tehama County,

moves slowly into the wind. This ignition pattern applies maximum heat to the ground. Photograph by Rich Reiner.

F I G U R E 18.1. Schematic diagram illustrating interactions among fire

regime, vegetation, and physical factors. Modified from Riba and Terradas (1987), Whelan (1995) with the permission of Cambridge University Press.

mostly though its effect on fuel moisture. Dead annual grass will track the relative humidity closely as it changes throughout the day. Grass fires typically burn with moderate intensity if the humidity is between 40% and 60% and the temperature is below 90
F. Wind speed and direction are important factors in predicting fire behavior. Winds accelerate the delivery of oxygen to the fire, increase spread, and influence the direction of spread. Topography also affects wind patterns and directs heat upward. A fire burning on level ground can double its rate of spread when it moves onto a 25% slope (Biswell 1989). Aspect can also affect fire behavior, with south slopes having higher temperatures and lower humidity. On the other hand, north and east slopes may sometimes exhibit more intense fire behavior as a result of higher biomass production on shaded slopes (Wright and Bailey 1982).

Ignition Pattern The behavior, spread of fire, and the resulting fire effects can also be influenced by ignition pattern, the relationship of the fire to the wind, the topography, and the presence of other fires. There are three general ignition patterns, each with unique characteristics: back fires, head fires, and flank fires. Back fires burn downslope or against a steady wind (Figure 18.2). They burn slowly and with moderate intensity. Because they cause a longer residence time than other ignitions patterns, back fires are able to direct more heat into the ground, thus killing seeds and some perennial plants. Head fires move with the wind or upslope. They are the most intense fire type. They move fast and are more difficult

208

ECOLOGICAL INTERACTIONS

to control. Energy stored in the fuel is released rapidly upward, and the fire moves quickly over the ground. It is a paradox, however, that head fires are often called “hot.” Their rapid movement actually reduces heat penetrating into the ground compared to slower-moving back fires. Head fires are often able to burn grazed grasslands with low fuel loads, where a back fire might not spread. Flank fires are lit into the wind. They are sometimes used in prescribed burning to widen fire breaks that run toward the wind. The fire behavior of flank fires is intermediate between that of back and head fires.

Fire Effects on Grassland Soils and Nutrients The effect of fire on soil nutrients and stability is related directly to fire intensity (Biswell 1989, Whelan 1995). Most studies in grasslands, however, show that surprisingly little heat is transported more than a few centimeters into the soil. Soils are good insulators, and most heat from fires is directed upward. Temperatures of greater than 100
C below 3 centimeters are rare for a grass fire (Whelan 1995). Fires in annual grasslands often result in a short-term increase in available nitrogen and phosphorus (Menke and Rice 1981). Other nutrients may also become more available when they are mineralized by fire or when ash raises the soil pH (Whelan 1995). These effects, however, are usually present for only a single season. Increased available nitrogen may stimulate weed production and have negative effects on species diversity by favoring select, fast-growing forbs such as Erodium sp. (Foster and Gross 1998). If a fire is extremely hot, nitrogen may be volatized, resulting in an overall loss to the ecosystem. Repeated hot fires can reduce the overall pool of nutrients available to plants (Wan et al. 2001). Soil carbon may also increase after very hot fires as a result of the death of plant roots. This increase in carbon could further reduce the availability of nitrogen for plants by releasing microbes from carbon limitation and

causing them to immobilize more nitrogen (Seastedt et al. 1991).

Fire in California Grasslands Fire almost certainly played an important part in California’s pre-European grasslands, helping structure a mosaic of grasslands among other natural communities such as woodland, savanna, and chaparral (Callaway and Davis 1993; Mensing 1998). Lightning strikes and Native Americans are believed to have been the primary ignition sources (Blackburn and Anderson 1993; Greenlee and Langenheim 1990), and it is likely that fire frequency increased upon the arrival of people to California about 12,000 years ago (Parsons 1981). The use of fire is documented for almost every Native American group in California (Anderson 2005, Aschmann 1977, Blackburn and Anderson 1993), and it appears that Native American fires were very frequent in grassland and oak savanna settings (Anderson, Chapter 5). Some sites may have been burned every year (Aschmann 1977). Reynolds (1959) found evidence that 35 tribes set fires to increase the abundance of plants used for food, fiber, and medicine as well as to open up brushy areas to improve hunting for deer, rabbits, and quail (Biswell et al. 1952; Lewis 1973). Fire was also used to create plant growth suitable for basketry (Dalrymple 2000). Early European accounts of California, such as the journal of the 1769 Portola Expedition, include numerous descriptions of recently burned grasslands in the southern Coast Range (Barry 1972; Anderson 2005). The role that lightning ignitions played in shaping contemporary California grassland distribution is debated. Keeley (2002) suggests that lightning strike fires alone could not explain the shrub/grassland patterns now found in the Coastal Range. He suspects that current patterns mostly resulted from Native American burning, followed by the expansion of “rangeland improvement” burning when the California Division of Forestry began to issue burn permits in the 1940s. In contrast, Biswell (1989) emphasizes the role of lightning ignitions in maintaining grasslands and cites a record of over 1,200 lightning strike fires in lower-elevation oak woodlands in a single season. The National Interagency Fire Agency in Boise, Idaho, records lightning-caused fires and predicts lightning strike probability on a daily basis for all of California (Latham and Schlieter 1989; WFAS 2005). Their records clearly show that frequency of lightning ignition is region-specific. The elimination of Native American ignitions, fire suppression, reduction in “range improvement” burns, and intensive grazing of fine fuels have greatly reduced fire frequency across California grasslands over the last century. Greenlee and Langenheim (1990) estimated that the fire frequency in the Central Coastal Range before 1880 was once every 1–5 years, whereas since then burning has occurred every 20–30 years. In this region, as well as the mid-elevation Sierra Nevada, the area of grassland has decreased at sites where the edaphic climax community without fire becomes

dense forest, oak woodland, or shrublands (Bakeman and Nimlos 1985; Biswell 1956; Mensing 1998; Van Auken 2000). In contrast, fire frequency has increased in some annual grasslands found near urban areas. Accidental ignitions increase as cities and new roads push into ranch land (Westmann 1976). Higher fire frequency promotes the conversion of shrublands to annual grasslands, which further promotes frequent fires (Keeley 2001; Tyler et al., Chapter 14). In some areas, the loss of shrub communities is a conservation concern. For example, an increase in ignition sources along roads in coastal southern California, in combination with the promotion of the grass growth caused by nitrogen deposition from pollution, results in widespread loss of coastal sage scrub in this region (Allen et al. 1998; Zedler et al. 1983). The seasons in which fires occur are controlled by climate and ignition source. California’s seasonal precipitation pattern, along with the life cycle of annual grasses, produces conditions in which most grasslands are capable of burning from May to October. Low-elevation grasslands away from the coast dry out early because of the late winter/early spring senescence of European annual grasses, whereas coastal and high-elevation grasslands may not be capable of burning until July. Fuel load, weather, topography, the proximity to ignition sources, and the effectiveness of suppression attempts determine the actual probability of grasslands burning during the dry season. Little is known about the size and pattern of historical grassland burns. It is suspected that the preinvasion grasslands that were subjected to regular burning by Native Americans were more spatially patchy than are grasslands today (Keeley 2002). Fires burning though bunchgrasses, occasional shrubs, and low-growing forbs burn cooler, patchier, and potentially with lower spread rates than typical invaded annual grasslands with continuous fine fuels. Attempts to describe and map the departure of fire regimes from what is believed to be historic or “natural” is a common method of identifying and setting land management priorities across much of the United States. The Fire Regime Condition Class system is an interagency, Web-based tool for determining the degree of departure from what is thought to be the historic conditions for vegetation composition, fuel loads, and fire regime (FRCC 2005). In California, however, use of this tool is complicated by the unknown history of fire, the large climatic and topographic gradients over which grasslands occur in the state, and the mosaic nature of the vegetation. Attempts to increase native species cover in invaded annual grasslands by re-establishing a “historic fire regime” should therefore be viewed with caution, because we do not know what plant species were actually dominant in the pre-European grasslands (see Keeler-Wolf et al., Chapter 3; Schiffman, Chapter 4; and D’Antonio et al., Chapter 6), and fire may increase the abundance of non-native plant species that now dominate much of the region.

Fire Regime and the Invasion of Introduced Plants The dominance of introduced annual grasses in California grasslands strongly influences all of the components of the

FIRE IN CALIFORNIA GRASSLANDS

209

fire regime, including the severity of fires as defined by flame length, heat output, and residence time. An increase in fire severity promoted by the presence of introduced annual grasses may have also played a role in the invasion process itself. Carried by Spanish explorers, non-native plants began to arrive in California in the 1600s (Bossard et al. 2000). Newly arrived annual grass species found a favorable climate similar to their origin in the Mediterranean Basin. These grasses were tolerant to grazing, drought, and fire. In addition, they produced large amounts of seeds which were protected in the soil during the dry season. These adaptations, and a native California flora that offered few species that could compete with them, likely encouraged their spread (Heady 1977; and see Chapters 4, 6). This invasion may also have been encouraged by a long-term warming and drying trend that has occurred in California since the Pleistocene (Raven and Axelrod 1978) as well as a major drought in the late 1800s, the expansion of crop agriculture, and extreme overgrazing during the Gold Rush (see Chapters 4 and 6). Once annual grasses became established, their continuous cover, combustible nature, and thatch-building character may have promoted hotter and less patchy fires, assuming the grasses were not intensively grazed. This would have created conditions that selected against native perennial grasses and for non-native annual grasses. This hypothesis is supported by experimental burns that indicate that hot fires can damage native bunchgrasses such as Nassella pulchra (Marty et al. 2005). Other supporting evidence is the general gradient of annual grass invasion from dry, low-elevation grasslands to wetter mountain habitats observed in California (Keeley 2001). Native perennial grasses are now most common on coastal terraces and in montane settings where cooler fires might be expected. It is unfortunate that we do not have better records that document the invasion and conversion and the role played by fire.

Effects of Fire on Grassland Plants The Individual Plant The persistence of an individual plant after a fire is determined by its anatomical, physiological, and life history characteristics. Plant phenology and susceptibility of meristems and seeds to fire are important determinants of a plant’s tolerance to fire. In addition to direct mortality, plants are also affected by postfire conditions (Whelan 1995). Season of burn, fire residence time, pattern of fuel consumption, and depth to which heat penetrates the soil all affect plant mortality during fire and the conditions the plant will face after the fire (Miller 2000). Although it is tempting to classify certain plants as “fire adapted,” the effect of fire on plants can vary from fire to fire and within a single fire. Direct mortality of plants is related to how much heat their tissues receive and how long the heat is present. If fuels are dry enough to generate intense heat, even green plants with actively growing meristems and high moisture content

210

ECOLOGICAL INTERACTIONS

can be affected. The rapid expansion of heated water in growing cells can kill plants otherwise not directly consumed by the fire (Wright and Bailey 1982). Annual grasses that disseminate their seed early in the fire season, and perennial species that keep growing points low and protected by duff and soil, are less affected by fire. Intense fires, however, can damage seeds lying on the surface as well as kill the crowns of bunchgrasses or shrubs. A significant body of work examines the effects of fire on needlegrass (Nassella sp.), California’s most widespread native perennial grass genus (Ahmed 1983; Garcia-Crespo 1983; Fossum 1990; Langstroth 1991; Dyer 1993; Dyer et al. 1996; Hatch et al. 1999; Fehmi and Bartolome 2003; Marty et al. 2005). Overall, these studies do not yield a consistent picture of Nassella responses to fire, even though it is often seen along roads and rail right-of-ways that frequently burn, and it can be observed greening up and flowering just days after spring fires. Several studies involving seeding and germination of N. pulchra after fires show increased seedling densities in burned compared to unburned plots (Ahmed 1983; Garcia-Crespo 1983; Dyer et al. 1996), although results were not statistically significant in Garcia-Crespo, and in the latter study survival everywhere was low because of drought in the years following fire. In studies evaluating fire effects on adult Nassella plants, most showed no or negative effects of fire on cover, frequency, or density of Nassella in the first year after fire (Langstroth 1991; Pollak and Kan 1998; Hatch et al. 1999), but some measured significant increases in the second postfire year (Langstroth 1991; Klinger and Messer 2001). Also, some studies show increased seed production or seed weight from burned compared to unburned adult Nassella plants (e.g. Ahmed 1983; Fossum 1990; Langstroth 1991; Dyer 2002), but this was not uniformly the case even within a single study. Fire effects on other native perennial grasses in California have also been mixed. DiTomaso et al. (2001) saw an increase from 1% to 10% of the native perennial Hordeum brachyantherum after conducting fires to control annual goatgrass, but H. brachyanthemum has not been otherwise studied. D’Antonio et al. (2002) reviewed effects of fire on the coastal prairie bunchgrass Danthonia californica and concluded that effects tend to be negative (Arguello 1994) or negligible (Hatch et al. 1999). Variation in the impact of fire on perennial grasses may be due to variation in fire intensity. High-intensity fires, for example, may negatively affect native bunchgrass survival, although this relationship has not been systematically evaluated. Marty et al. (2005) studied the effects of fire and grazing on N. pulchra at a site in the foothills of the Central Valley that had not been burned or intensively grazed for many years. They found that mortality of bunchgrasses was 10% higher in burned versus unburned plots, and even though seedling recruitment was increased by fire, this did not offset adult plant mortality, and densities were still lower in burned than unburned plots after two years. These results may have been influenced by thatch accumulation in their plots, which,

prior to this study, had not been burned for several decades and had been only lightly grazed. Also, the experimental burns were conducted using back fires, which tend to increase the duration of heat exposure by individual plants. It is important to note that interpreting studies regarding bunchgrasses and fire is made difficult by the fact that fire not only affects the mortality of established plants and the regeneration of new seedlings, but also can cause mature plants to fragment into what may appear to be separate individuals. For this reason, studies that examine only density data should be viewed with caution. Fires can also reduce annual forbs by directly killing the plant or their seeds before they disperse. Perhaps the best studied introduced annual plant in California Grasslands is Centaurea solstitialis, or yellow starthistle (DiTomaso et al. 1999; and see DiTomaso et al., Chapter 22). The seeds of this species remain viable in the soil for 3 or more years; thus, multiple years of burning are needed for control. The best time for burning is in the early summer following the seed dispersal of desirable plants, but prior to viable seed production or dispersal in yellow starthistle (DiTomaso and Johnson 2006). Fire has rarely been used to control perennial and biennial forbs, and few published examples exist from anywhere within the United States. Single burn events often do not reduce such plants, suggesting that multiple years of burning would be necessary for control (DiTomaso and Johnson 2006). Fire is commonly used to control woody species in California grasslands (Alexander and D’Antonio 2003b; Tyler et al., Chapter 14; DiTomaso et al., Chapter 22), but control is effective only if woody species cannot regenerate after fire or if fires are frequent enough to eventually kill resprouting adult plants or their seedlings before they have reached reproductive size.

with high temperature tolerances by having cells that are in a dormant and dehydrated stage during the fire season (Whelan 1995). Other seeds have protective coatings that resist burning and require heat scarification for germination. Species in the genera Lotus, Lupinus, Astragalus, and Trifolium all showed higher germination after fire in California grassland (Parsons and Stohlgren 1989). Unfortunately, many weed species in California grassland also have fire-stimulated seed germination. For example, fire stimulates seed germination of broom species in California grassland (Alexander and D’Antonio 2003b). The resulting seedling crop can then be treated with another control method (DiTomaso et al., Chapter 22). Postfire environmental conditions may increase the germination of seeds with dormancy. Rice (1985), for example, demonstrated that removal of thatch by fire increased the summer diurnal temperate range that seeds were exposed to, thereby increasing seed germination rates for the European annual forbs Erodium botrys and E. brachycarpum in a California grassland. A common adaptation that may aid plants in surviving fire is to produce seeds that are able to quickly move into protected sites in the soil. On clay soils with cracks, heavy smooth seeds fall to the ground and are quickly buried. Plants such as Nassella sp., Avena spp., and Erodium sp. produce seeds with mechanical appendages that screw individual seeds into the soil, helping to bury them. Caching insects and burrowing rodents such as Botta’s pocket gopher (Thomomys bottae) also bury seeds (Kneitel 1997). Some plants will “hedge their bets” by using long-distance wind dispersal via winged or tufted seeds, which can move seeds out of a burn area. Long-distance dispersal also transports seeds into recently burned areas (Miller 2000).

Seeds

Grassland Plant Community Composition

Plants protect their seeds from fire either by physical adaptations or by dispersing seeds widely. Seeds are most vulnerable to fire when they are still held above the ground where temperatures during fire are the highest. This fact can sometimes be used as a strategy for controlling weed species if a fire can be applied when seeds are still elevated (DiTomaso and Johnson 2006; Pollak and Kan 1998). In general, plants with annual life cycles and prolific seed production are favored in grasslands that burn often. High numbers of seeds increase the odds that some will, by chance, make it to a fire-safe microsite. Annuals that then germinate and grow in the postfire environment often have prolific seed production as a result of nutrient-rich conditions created by fire. Perennial plants may also respond to fire by producing more and larger seeds the year following a fire. Dyer (2002) found that Nassella pulchra seeds from burned areas were larger and had higher germinability than seeds from unburned plots. He also found previously that survival of N. pulchra seedlings increased in burned plots (Dyer 1993). Plants can also produce seeds with specific adaptations that aid in fire resistance. For example, many plants produce seeds

Fires most often shift community composition of annual grasslands and shrublands away from woody species and toward forbs and geophytes (Graham 1956; Hansen 1986; Heady 1956, 1972). This shift primarily is due to direct mortality of woody species in the fire and to indirect improvement of the growing conditions for forbs and geophytes by the reduction of thatch. Thatch is the dead and decaying material, mostly of grasses, accumulated from growth in previous years. On the positive side, thatch helps recycle nutrients, retard erosion, improve water infiltration, and shades the soil (McNaughton 1968). In the absence of grazing and fire, however, thatch accumulates and eventually reduces diversity through inhibiting germination or shading out forb seedlings (Bentley and Fenner 1958; Heady 1956). Grasslands with years of accumulated thatch tend to be dominated by tall, introduced annual grasses such as ripgut (Bromus diandrus) and wild oats (Avena sp.) (Heady 1958), two strongly thatch-producing grasses. Fires immediately remove thatch, expose the mineral soil, increase light passage to the ground, and provide warmer and more fluctuating ground temperatures for germinating forbs (Baskin and Baskin 1998; Gill 1977; Stone 1951). For this

FIRE IN CALIFORNIA GRASSLANDS

211

FIGURE 18.4. Average pre- and postburn cover of exotic annual grasses

(p  .051), native annual grasses (p  .02), exotic annual forbs (p  .001), and native annual forbs (p  .003) from 13 spring burns in Tehama County, California, during 1996–2003. Adapted from Pollock (2006).

F I G U R E 18.3. The right-hand side of this Tehama County grassland

was burned the previous spring. The left side was excluded from the fire. The forbs on the right are mostly goldfields (Lasthenia spp.) The left side is dominated by the invasive grass medusahead (Taeniatherum caput-medusae). Photograph by Rich Reiner.

reason, it is common to observe abundant wildflower displays in the spring following prescribed fire in California (Figure 18.3). Fires can also destroy the seeds of annual grasses, thus reducing their dominance the following growing season. Figure 18.4 shows the average change in the foliar cover of vegetation guilds the spring following 13 burns in Tehama County, California (Pollock 2006; Reiner et al., in press). Notice the increase in both native and exotic forbs and the decrease in non-native annual grasses. The frequency with which grasslands are burned also affects plant species composition. In most studies, forbs continue to increase and annual grasses decrease with multiple consecutive burns (Delmas 1999; Parsons and Stohlgren 1989; D’Antonio et al. 2002). For example, three consecutive years of prescribed burning in a Sonoma County grassland increased native forb cover in spring by nearly 400%, boosting the native proportion of total forb cover from 17% to 67% (DiTomaso et al. 1999). Hansen (1986), working in the San Joaquin Valley, found that the non-native annual grass Hordeum leporinum declined after a single burn and continued to decline after repeat burning. Likewise, DiTomaso et al. (2001) and Betts (2003) found that repeated burning was necessary to reduce cover of barb goatgrass. Although it is clear that frequent fires can reduce annual grass abundance, most studies show that annual grasses quickly rebound to prefire abundance within 2 – 4 years after rest from burning (D’Antonio et al. 2002). Season of burn also affects postfire plant composition, because individual species vulnerability to fire changes by the season. For example, annual grasses that mature in early to midsummer are most vulnerable to spring and early summer fires (Eller 1994; Meyer and Schiffman 1999; DiTomaso and Johnson 2006). Fires that occur before seed maturation or dispersal usually have the greatest negative effect on these species. These species include some weedy invasive plants such as medusahead (Taeniatherum caput-medusae), red brome (Bromus madritensis ssp.), Japanese brome (Bromus japonicus),

212

ECOLOGICAL INTERACTIONS

and barb goatgrass (Aegilops triuncialis). Native perennial grasses such as Nassella pulchra also appear to be sensitive to season of burning, with the largest increases from burning in June when compared to fall burns (D’Antonio et al. 2002). Season of burn also affects forb abundance in a grassland community. Species of filaree (Erodium spp.), most of which are introduced, are the most common “fire followers” in California annual grasslands. In a study by Meyer and Schiffman (1999), filaree increased after spring burns but less so after fall burns. These investigators also found that spring burning increased abundance of native forbs compared to fall or winter burns. Likewise, Parsons and Stohlgren (1989) demonstrated an increase in the native legume Trifolium microcephalum after three consecutive spring burns. In their meta-analysis of burn studies in California grasslands, D’Antonio et al. (2002) further showed beneficial effects of spring compared to fall burning: average postfire exotic forb cover was higher in fall than spring burns. Nonetheless, Parsons and Stohlgren found that the native forb Orthocarpus attenuatus responded best to fall burns. These results underscore the idea that responses to fire are ultimately speciesspecific. Intensity of grassland fire also affects species composition. High-intensity fires tend to negatively affect perennial species such as bunchgrasses and shrubs more than annual species that have dispersed seed by the height of the fire season. Grasslands that are grazed by livestock have less available fuel in the fire season and thus burn cooler. In fact, grazing appears to negate the negative effect of fire on both Nassella pulchra and Danthonia californica, perhaps by reducing fuel density around the bunchgrass crowns (D’Antonio et al. 2002). In addition to grazing, fire intensity can be manipulated by altering the season of a burn. Also, fires conducted in the early morning, when humidity is often high, will produce less heat. In some situations deferring grazing for a season or more may be necessary on low-productivity grasslands if an intense fire is desired (DiTomaso and Johnson 2006; Hansen 1986).

Fire Effects on Grassland Animals The effects of fire on animals can be either direct or indirect. Direct effects are generally negative, resulting from injury or death from the heat of the fire. Indirect effects include

alterations to habitat or food sources and can be either positive or negative. Most animal-fire studies depict at least a temporary “reorganization” of animal communities after fires (Bendell 1974, Smith 2000). In most grasslands, however, animal populations return to prefire conditions within 3 years.

Birds Like other animals, grassland birds respond both to the fire itself and the changes in habitat structure it produces. Even though birds are rarely killed in fires, ground-nesting species can lose eggs and young when fires occur during the nesting season. Some birds, however, actually seek out fires. Swallows and other insect eaters such as Swainson’s hawks are attracted to the smoke and feed along the fire’s edges as potential prey flee. The response of bird populations to changes in habitat character is highly species- and site-specific. Species that respond to shorter grasslands, such as horned lark (Eremophila alpestris), burrowing owls (Athene cunicularia), and mountain plovers (Charadrius montanus), generally favor grasslands that are kept low by either grazing or fire. The actual results seem to depend on the burn’s timing, its intensity, unburned patch sizes, site vegetation, and the grassland bird species present in the area. For example, mourning doves have been shown to experience both positive (Bock and Bock 1992; Johnson 1997) and negative (Zimmerman 1997) effects by fire at different sites. Also, grasshopper sparrows have been found to experience positive ( Johnson 1997), negative (Bock and Bock 1992; Vickery et al. 1999; Zimmerman 1997), and no significant (Rohrbaugh et al. 1999) response to fire. A study conducted in a mixed oak woodland and grassland in coastal-central California found no shortterm (2 years) effect to breeding birds after a low-intensity prescribed fire (Vreeland and Tietje 2002). Waterfowl such as mallards and cinnamon teal are important yet inconspicuous nesting species that use valley grasslands. No data are available for California, but studies in mixed-grass prairie show that duck nesting is significantly lower in areas that were burned in the spring than in areas that were burned in the fall (Ward 1968). The California Partners in Flight Grassland Bird Conservation Plan (CPIF 2000) discourages prescribed burning during the breeding season for all ground-nesting birds. This can pose a challenge for grassland managers trying to control exotic plants with early spring burns.

Mammals The effects of burning on grassland mammal species are most often minor and short-term. Deer mice (Peromyscus sp.) and voles (Microtus sp.) are the most studied species, because they are key prey species in trophic chains. In most cases deer mice and voles escape the direct effects of fire by seeking refuge in burrows (Erwin and Stasiak 1979). They often leave the blackened area for adjoining habitat but are the first to recolonize grassland after the new growth (Lawrence 1966). Small

mammal populations will sometimes increase after fires because of the abundance of seed-producing annual plants. Vreeland and Tietje (2002), however, found that low-intensity prescribed burning had no effect on the populations of nine species of small mammals in Central California oak woodland. Fires that make nitrogen available to plants can increase the protein content of forage for herbivores, and it is common for livestock and deer to move to an area the winter or spring following a burn (Smith 2000). Biswell (1967) found that burning in chaparral increased deer use in the Central Coast from 30 to 131 animals per square mile. This study also demonstrated higher ovulation rates in does from the burn area. Zavon (1982), working at the Hopland Field Station, showed that sheep increased consumption of forbs and had 33% higher lamb gains after a fire. Although nutritional quality may improve, the overall biomass of forage is sometimes reduced by fire. This reduction is due to the reduction of annual grasses and a shift toward low-growing forbs, which often produce less total biomass (Bentley and Fenner 1958; Hervey 1949).

Insects and Other Invertebrates Insects such as ants, grasshoppers, and butterflies, as well as other invertebrates, are important contributors to ecological processes such as pollination and seed dispersal. As with other animals, fire causes both direct losses of individuals and longer-term indirect effects on habitat (York 2000). Underwood (2004), working in California blue oak savanna, studied the effects of spring and fall prescribed fires on grounddwelling invertebrates. She found that prescribed burning had relatively minor and short-term effects on most of the invertebrates she sampled. There were, however, significant temporary effects on several groups. Seed-harvesting ants decreased after fall fires, and omnivore ants increased after spring fires. Hunting spiders decreased after all burns, whereas funnel web spiders increased. It is likely that these effects diminish in subsequent years after a fire. A 30-year study in midwestern oak savanna also showed that fire did not significantly alter arthropod diversity (Siemann et al. 1997). Kerstyn and Stiling (1999) found an increase in grasshoppers after multiple years of prescribed burning in the Florida Sandhills and attributed it to increases in desirable forbs. Other studies have shown that fires can influence the spatial distribution of grasshoppers, and a reduction in herbivory toward the interior of burns has been recorded (Knight and Holt 2004). Butterflies are of interest to conservationists because of their role in plant pollination and, in some cases, their rarity. When large, intact habitats burn, butterflies tend to do well. Fleishman (2000), working in the Great Basin, demonstrated that butterfly species richness and composition did not differ significantly between burn units and controls. However, when habitat is fragmented and the butterfly species is rare, large fires could have significant detrimental effects. Schultz

FIRE IN CALIFORNIA GRASSLANDS

213

and Crone (1998) point out that designing strategies to manage habitat for species habitats may involve tradeoffs that include negative short-term impacts to achieve positive long-term results. Working with life history data on Fender’s blue butterfly (Icaricia icarioides fenderi), they constructed a model to predict the most efficient burn interval and size. They suggest that burning one-third of the available habitat each year would maximize population growth for this species.

Other Species There is very little information about fire and other grassland animal taxa in California. Only a single study looked at fire effect on grassland reptiles, and the results were not definitive (Vreeland and Tietje 2002). For species-specific information, Internet-based databases such as the Fire Effects Information System are valuable (http://www.fire.org).

Fire as a Grasslands Restoration Tool Because fire is an important controlling factor in structuring and maintaining grasslands, it is often considered as part of a restoration strategy (Meyer 1997; see Stromberg et al., Chapter 21). The working hypothesis in most restoration burning is that properly timed fire can be used to directly kill the seeds of the dominant introduced annual grasses and specific weeds and, in turn, reduce competition for native species (Fossum 1990; George et al. 1992; Menke 1992; Meyer and Schiffman 1999). It has been demonstrated that fire can be successfully used to suppress specific exotic species (see DiTomaso et al., Chapter 22). Fire has been shown to effectively control species such as medusahead (Taeniatherum caput-medusae) (Pollak and Kan 1998; Betts 2003), goatgrass (DiTomaso et al. 2001; Betts 2003), and yellow starthistle (Centaurea solstitialis) (DiTomaso et al. 1999), and some studies show remarkable increases in native forbs after prescribed fire (Meyer and Schiffman 1999). DiTomaso et al. (1999) found that native forbs, particularly in the family Fabaceae (peas) and Geraniaceae (geranium), increased after 3 years of burning. They also saw large (250% to 2,300%) increases in native species such as Linanthus bicolor, Minuartia californica, Lotus wrangelianus, Lupinus nanus, and Trifolium gracilentum. At a site in Tehama County, Mitchelson (1993) found significantly higher reproductive output of the geophyte Zigadenius fremontii and the annual forb Sidalcea calycosa after fires (Hunter 1986). Betts (2003) evaluated the effects of burning in goatgrassand medusahead-infested grasslands in Yuba County. Her study defined the desirable postfire community state as having a low abundance of goatgrass, medusahead, and ripgut brome (Bromus diandrus) and a higher abundance of more palatable species such as B. hordeaceus, Avena, Trifolium spp., and Erodium. She found that burning greatly increased the probability that plots would transition to more desirable states even though these “more desirable states” contained

214

ECOLOGICAL INTERACTIONS

few native species. Nonetheless they were more diverse in both forb and grass composition than unburned plots. Regardless of site specific restoration successes, it is debated whether repeated fires generically restore overall native species diversity and cover (D’Antonio et al. 2002). In fact, some studies report that consecutive fires actually reduce species diversity (Delmas 1999). Most likely, plant community composition before burning plays an important role in determining whether fires help promote native diversity. For example, Harrison et al. (2003) studied the effects of fire on both invaded annual grasslands and mostly native serpentine grasslands in the Northern Coast Range. They found that fire increased both native and exotic species richness across sites. However, it increased native species richness more than exotic species richness at sites dominated by natives (serpentine soils), whereas it increased richness of exotic species more than that of native species in the heavily invaded (nonserpentine) grasslands. They concluded, “The rich tended to get richer and the poor poorer in terms of native and exotic species.” Therefore, the presence of exotic species and the pool of natives that could possibly respond to the fire in a grassland area should be considered before using fire as a tool in restoration and management. D’Antonio et al. (2002), in their meta-analysis, which quantitatively measured the effects of fire in California annual grassland across 19 studies, concluded that “Although fire does not result in a straightforward increase in native vegetation or consistent decreases in exotic cover, elements of the native vegetation can benefit.” Their analysis showed that fire generally had a positive, although small, effect on total cover of native vegetation (forbs and grasses combined) and a small negative effect on the total cover of exotic species. This result, however, was partially due to exotic annual grasses showing a greater decrease than the increase shown by exotic forbs. Their analysis also showed that burn frequency and the presence of grazing affected the outcome of burning. Native forbs benefited the most from annual burning but not from the combination of annual burning and grazing. Grazing, however, did help sustain the positive benefit for native forbs into the third year after burning. Although most species that benefit from burning are desirable plants, in some cases invasive and undesirable plants, particularly forbs, can increase following fires. For example, burns conducted in Sequoia National Park achieved control of invasive annual grasses, but the exotic annual forb Malta starthistle (Centaurea melitensis, also known as tocalote), which was not found in the unburned control, increased significantly in the burned area (Parsons and DeBenedetti 1984). Similarly, black mustard (Brassica nigra) increased to almost 100% cover after a burn designed to reduce annual grass in the Santa Monica National Recreation area (DiTomaso and Johnson 2006). Managing for low available plant nutrients in the soil might be incorporated in a restoration strategy. It is known that repeated hot burns can reduce the overall pool of available nutrients, and there is some evidence that native plants

do best where nitrogen is limited (Weiss 1999; Wan et al. 2001). In a 12-year experimental study of nitrogen deposition on Minnesota grasslands, plots dominated by native grasses shifted to a lower-diversity community as nitrogen was added (Wedin and Tilman 1996). Grasslands with high nitrogen retention and carbon storage rates were the most vulnerable to species losses. Likewise, there is evidence for some California grasslands that N enrichment leads to an increase in exotic grass cover and a decline in native species cover (Huenneke and Mooney 1998a; Maron and Connors 1996; Weiss 1999). Fire might be a way to reverse the impacts of N enrichment. Haubensak et al. (2004) found a 40% decrease in soil N with repeated fire used for control of French broom in Marin County grasslands.

Managing California Grasslands with Prescribed Fire The use of prescribed fire is becoming a preferred restoration and management technique for grasslands worldwide (Figure 18.5). It is attractive because it emulates a natural disturbance, and it can be applied at scales relevant to conservation (Baker 1994). Prescribed fires in California are conducted to reduce thatch, to control non-native plants such as medusahead (Taeniatherum caput-medusae) or yellow starthistle (Centaurea solstitialis), and to improve rangelands or grassy parklands by reducing invasion by woody species such as Scotch broom (Cytisus scoparius) and French broom (Genista monspessulana) (Alexander and D’Antonio 2003a, b). In theory, these goals can be obtained while at the same time promoting native, fire-adapted plant species. In practice, most prescribed grassland burns are focused solely on control of non-native species. Reseeding with native species after fire is rare. Alexander and D’Antonio (2003a), for example, surveyed grassland sites burned for control of broom species in Marin County. No follow-up seeding occurred in any site, and native species were generally not enhanced by burning. Prescribed burns are conducted by government land management agencies that maintain staff trained in fire safety, logistics, and ecology. On private lands within State Responsibility Areas (SRA), the California Department of Forestry and Fire Protection is the lead agency and offers services to landowners through their cost-share “Vegetation Management Program.” By involving an agency in a prescribed burn, landowners can obtain expert advice and receive help implementing the fire. Also, the liability for damage caused by an accidental escape will usually not fall to the landowner if an agency is involved.

Planning Fires in Grasslands Fire planning is most often conducted at two spatial scales. At a large scale, “Fire Management Plans” describe how an entire landscape should be managed for both wildfire and prescribed burns. Goals are set at the watershed level or larger, and a

FIGURE 18.5. Prescribed fire can be an effective restoration tool in annual grasslands. This photo shows the establishment of a “black line” fire break. The person in the foreground is igniting a strip of grass while the person in the background extinguishes it. Photograph by Tom Griggs.

common fire management vision is created for multiple managers and landowners. The plans include long-term goals for vegetation, wildlife, biodiversity, water quality, and air quality. They are also used to organize information on wildfire, fire suppression, access roads, natural firebreaks, and the location of water supplies. Fire Management Plans should include information on how fire is expected to influence future vegetation on the site. There are various approaches to doing this. Filling in the boxes of a conceptual “line and box” model, such as illustrated in Figure 18.1, can be a useful tool to help organize information about how fire affects the landscape. They can be constructed from local land managers’ fire experience and from published studies. If sufficient information exists, a more complex model may be useful. Computer-based modeling frameworks such as The Vegetation Dynamics Development Tool (VDDT), developed by ESSA Technologies, are useful for examining the role of succession, disturbances, and management actions in vegetation change (Merzenich et al. 1999). Models such as this allow users to create and test hypotheses of vegetation dynamics, simulating them at the landscape level. Whatever approach is taken, models should be viewed as a set of hypothetical relationships regarding the vegetation community and fire. Each burn should be used as an opportunity to test the model and, if necessary, modify it as new information is collected. The “Fire Unit Plan,” or Incident Action Plan, is a logistics plan that lays out the course of action for a specific prescribed fire. It includes a list of the required fire staffing, the line of command, required equipment, permits, fire line construction details, ignition pattern, fuel and weather prescription, smoke management, accidental escape, and “mop up” logistics. When burning is conducted under state programs such as the California Department of Forestry and Fire Protection’s Vegetation Management Program, a California Environmental Quality Act evaluation must also be undertaken. The Fire Unit Plan also includes the specific ecological and operational outcome objectives for the fire as well as how

FIRE IN CALIFORNIA GRASSLANDS

215

each will be evaluated. An example of an ecological objective might be “medusahead grass should be reduced by at least 65% with no loss of blue oak seedlings.” An operational objective might include evaluating the use of new equipment, or training new staff. There should also be an accompanying description of how each objective will be monitored. Wright and Bailey (1982) describe in more detail considerations and techniques for successfully carrying out prescribed burns.

Four Assumptions Used to Justify Prescribed Burning in California Grasslands •

Because fire is a historic component of grasslands it is assumed that reintroduction of fire will promote native species richness.

•

Reduction of thatch will provide better growing conditions for native species, especially native grasses.

•

Burning in the late spring will allow early-seeding natives to set seed but kill late-season exotic species before seeds mature.

•

Burning will remove unwanted woody perennial species and promote natives.

Developing a Fire Prescription The conditions under which a fire can be ignited and controlled safely and achieve the resource objectives is called the “fire prescription.” Most often the prescription lists the range of fuel moisture content, air temperature, humidity, wind speed and direction, ignition pattern, and season under which a fire can be conducted. Those fire conditions that will best achieve ecological objectives may not be the same as those allowing safe containment of the fire. Hence, ecological objectives may have to be modified to meet safety and smoke (see following discussion) concerns. All of the aforementioned factors work together to affect fire behavior. The reliable prediction of fire behavior has become increasingly important as prescribed fire has increased as a management tool. BEHAVE and BEHAVE-Plus are widely distributed fire behavior predictive models, developed by the USDA Forest Service (http://www.fire.org). They allow planners to predict fire rate of spread, flame length, and fire line intensity (rate of heat release) using one of several generalized fuel models (Andrews 1986). California annual grasslands are generally covered within fuel model 1 or 3 depending on the height of the grass (Anderson 1982).

Smoke Management Burning in California presents major challenges for preserving good air quality, making smoke management a critical component of any prescribed fire. In some air basins, such as the South Coast and the Central Valley, meeting air quality requirements can become a significant hurdle when considering fire as a management tool. A typical grass fire, even when burning hot, produces 15 pounds of particulate matter per ton of fuel (7.5 kg per metric ton; USDA 1985). Approximately 90% of these particles are less than 10 micrometers in diameter, can be inhaled, and thus produce a significant heath risk for many people (Breysse 1984). Most prescribed fires require air quality permits issued by either the county agricultural commissioner’s office or the local air quality agency. The key to successful smoke management lies in the land manager’s ability to work within existing regulations and to conduct fires using techniques to minimize smoke impact. Smoke can be minimized using several techniques. First is to encourage burning of dry fuels under conditions at which they burn completely without smoldering. When combustion is complete, two products — carbon dioxide and

216

ECOLOGICAL INTERACTIONS

F I G U R E 18.6. Common assumptions used to justify a prescribed

grassland fire to promote native species. Adapted from Christian (2003) and Reiner et al. (2006).

water — make up over 90% of emissions. As combustion efficiency decreases with wetter fuels and cooler burns, less carbon is converted to carbon dioxide, airborne particulate matter increases, and more carbon moves into pollutants such as carbon monoxide, hydrocarbons, nitrogen oxide, and sulfur oxides (Breysse 1984). Besides ensuring full combustion, fire managers need to work closely with meteorologists to conduct fires under weather conditions that both direct smoke away from populated areas and disperse smoke quickly. Burning immediately after a frontal weather system has passed can help control smoke. Depending on the type and severity, weather fronts can produce predictable winds, as well as cause unstable air to break down inversions that trap smoke close to the ground. In the Central Valley, a passing front often leaves a several-day window during which a smoke plume can be dispersed high into the atmosphere.

Monitoring Fire Effects Despite the growing use of fire as a method of restoring and managing grasslands, the effectiveness of this approach in suppressing exotics and restoring native species is far from clear. Improving the monitoring of prescribed fires presents an opportunity to advance our understanding of this potential restoration tool. Better data will also help validate basic assumptions that underlay many California prescribed-fire programs (Christian 2003; Reiner et al. 2006) (Figure 18.6). One or more of these assumptions should be tested in almost any restoration burning conducted in California grasslands. A monitoring program should help managers understand the validity of their assumptions, as well as the other factors that influence the outcome of a prescribed fire. The challenge is to design a sampling protocol that is both statistically powerful and affordable. Conducting a true experiment with real statistical replication will always yield the clearest results with the highest probability to explain complicating factors.

Unfortunately, the high cost of setting up replicates makes this approach mostly impractical for grassland managers. Monitoring designs, however, need not be conducted as experiments in order to inform decision making and the adaptive management processes. The most important ingredient for a nonexperimental monitoring approach to succeed is for the design to include sampling of an unburned control area (Elzinga et al. 2001). The control and treatment areas must be sampled before the burn to ensure that there are no significant differences in the vegetation. Establishing a valid control is needed to unravel the interacting environmental and human-caused factors that affect the outcome of most burns.

Summary Most grassland ecologists and managers in California agree that fire is, or should be, an important if not essential component of California grasslands. The heat of fire induces direct effects on plants and animals and immediately changes the species composition and physical structure of the habitat. There are also indirect postfire effects that alter competition between species and habitat for native species. Yet using fire to manipulate composition in California grasslands is still an inexact science. Although our understanding of the historical vegetation of California is far from complete, it is clear that the grassland species composition and fire regimes we observe in California today are not representative of the historical grassland

community. Both the ignition sources and fire frequencies and the plant and animal species involved have changed. Human population growth, as well as the introduction of highly flammable, yet fire-tolerant, exotic grasses, have influenced the grassland community we observe today. Prescribed fire can be a powerful tool for managing grasslands. However, consideration of its use must recognize that the results will likely be highly site-, species-, and season-specific. It is clear that prescribed fire can be used to reduce some weed species, yet it is less clear under what conditions fire can be used to restore native species richness or abundance to degraded grasslands. Managers contemplating prescribed fire as a management tool should begin on a small scale, using an experimental approach, and carefully monitor the results before taking the activity to larger scales. Prescribed fire is only one management treatment available to land managers, and it can be used in combination with other treatments such as livestock grazing and herbicides. Strategies that combine management tools will likely be the most effective and will certainly become even more important as the urban–wildland interface continues to expand and smoke and safety concerns limit use of fire.

Acknowledgments I would like to express my gratitude to Carla D’Antonio, Scott Collins, Erin Espeland, Perry Grissom, Rebecca Herring Reiner, Peter Hujik, Jaymee Marty, Guy McPherson, and Kipp Morrill for advice, critical reviews, and editing of this manuscript.

FIRE IN CALIFORNIA GRASSLANDS

217

NINETEEN

Responses to Changing Atmosphere and Climate J E F F R EY S. D U K E S AN D M. R E B E C CA S HAW

As previous chapters have illustrated, humans have radically transformed the extent and species composition of California’s grasslands. We continue to change the environment that those grasslands experience in both direct and indirect ways. Many of the current environmental changes are related to fossil fuel burning; this activity, along with cement manufacturing and changing land use patterns in the tropics, has led to a ⬃35% increase in the concentration of carbon dioxide (CO2) in the atmosphere over the past 200 years (Houghton et al. 2001). Fossil fuel burning also creates reactive nitrogen compounds such as nitrous oxide (N2O). These nitrogenous compounds are released into the atmosphere, where they join similar molecules released by agricultural activities. Eventually, much of the nitrogen in these compounds returns to the land surface as some form of nitrogen deposition, often fertilizing natural ecosystems. Carbon dioxide and nitrous oxide, along with methane and other gases released by human activities, trap heat. As atmospheric concentrations of these gases increase, climate changes around the world (Houghton et al. 2001). The climate in California is expected to warm substantially by the end of the century; the amount of warming will depend largely on human actions between now and then (Hayhoe et al. 2004). Each of these human-caused changes—the increase in CO2 concentration, the added nitrogen inputs, and the changes in climate — will affect California grasslands. In this chapter, we explore how the grasslands will respond (and have been responding) to these perturbations. We also review the limited evidence for how these changes may affect the prevalence of invasive species in California. Other chapters examine how California grasslands are affected by changes in land use (see Eviner and Firestone, Chapter 8; Jackson et al., Chapter 9; Harrison and Viers, Chapter 12; and DiTomaso et al., Chapter 22), and how species may evolve in response to environmental changes (see Rice and Espeland, Chapter 11).

218

We focus primarily on responses relevant to management concerns. Responses of grassland production and composition are relevant to the well-being of wildlife and livestock as well as the preservation of rare plant and animal species. In the context of climate change, the amount of carbon stored in ecosystems is increasingly relevant to land managers and politicians. This is because, at the same time humans are releasing heat-trapping gases, terrestrial ecosystems are taking them up. Current estimates suggest that ecosystems on land take up and store one-fifth to one-quarter of the carbon dioxide released by human actions (the oceans take up a similar amount; Houghton et al. 2001, Field and Raupach 2004). This natural carbon sequestration slows the increase of atmospheric CO2 concentrations and thus slows climate change. To understand how rapidly climate may change over the coming centuries, it is important to know the extent to which natural ecosystems will continue to be able to sequester the carbon dioxide released by human activities under future environmental conditions. California grasslands have proven to be useful ecosystems in which to examine this question. To better understand the question of carbon storage, a great deal of research has examined responses of nutrient cycles, microbial communities, decomposition, and other variables. Here, we can include only some of the highlights of this research.

Predicted Environmental Changes in California Predicted Changes in Temperature and Precipitation Like many other terrestrial habitats, California grasslands face an uncertain future climate. Although climate has changed repeatedly over past millennia, for a variety of reasons (Houghton et al. 2001), anticipated human-driven changes are likely to be unusually fast and large. Many of California’s ecosystems are likely to be vulnerable to future

climatic change, because their current ranges are limited and their potential ranges are bounded by the coast or other topographical features (Snyder et al. 2002). To better understand the degree to which California’s climate may change in the near future, climate modelers apply greenhouse gas emissions scenarios that encompass a range of energy futures to the suite of general circulation models (GCMs; Houghton et al. 2001). Most commonly, modelers will apply relatively “low” and “high” emissions scenarios to a few GCMs chosen for their particular physical climate sensitivities. To effectively project these global climate changes to local and regional scales where environmental impacts can be evaluated, GCM outputs derived at the grid scales of 2.5⬚ ⫻ 3.25⬚ are applied to the region of interest using one of many downscaling methods that provide increased spatial resolution (Houghton et al. 2001; Pan et al. 2001; Leung et al. 2003; Salathe 2003; Wood et al. 2004). There are significant uncertainties at each step of such an analysis, and these uncertainties are reflected in the wide range of predicted temperature and precipitation changes for California. In this century, the average annual statewide temperature is projected to rise 1.7 – 3.0 ⬚C (3.0 – 5.4 ⬚F) under low emissions, or 3.8 – 5.8 ⬚C (6.8 – 10.4 ⬚F) with higher emissions (Hayhoe et al. 2004; Cayan et al. 2006). There is no clear trend among the precipitation projections. Statewide, the projections for change in annual average precipitation range from a decrease of 157 mm to an increase of 38 mm (Hayhoe et al. 2004; Cayan et al. 2006), with significant variation in projections among GCMs, scenarios, and downscaling methods (Pan et al. 2001; Salathe 2003; Wood et al. 2004). Regional climate studies indicate that, on average, California’s ecosystems may experience substantially warmer and wetter winters, slightly warmer summers, and an enhanced El Niño/Southern Oscillation (ENSO) in the next 100 years (Field et al. 1999; Gutowski et al. 2000; Cayan et al. 2006), but there is no evidence to suggest that the seasonal, Mediterranean climate will change (Cayan et al. 2006). That is, the winters will remain relatively wet and cool and the summers will remain hot and dry, but the temporal and spatial pattern of precipitation within the winter season, the form of the precipitation (rain or snow), the frequency and severity of extreme events (storms and drought), and the length of the growing season will likely change (Hayhoe et al. 2004). Downscaled temperature projections consistently project a pattern of lesser warming along the southwest coast and increasing warming to the north and northeast (Figure 19.1; Hayhoe et al. 2004). These changes in temperature and precipitation will drive changes in species interactions, species ranges, community assemblages and ranges, and ecosystem processes. It is not possible to predict the full suite of changes in California grasslands, but we can seek clues from a wide range of experimental and modeling approaches to better understand the implications of climate change for California grasslands.

F I G U R E 19.1. Projected winter (DJF: December through February) and summer (JJA: June through August) temperature change (⬚C) for 2070–2099, relative to 1961–1990, for a 1/8⬚ grid, based on statistical downscaling from GCM outputs. Statewide, projections of winter temperature increases for the end of the century are 2.2–3 ⬚C and 2.3 –4 ⬚C for the PCM and HadCM3 models, respectively. End-ofcentury summer temperatures are projected to increase by 2.2 – 4 ⬚C and 4.6–8.3 ⬚C based on the PCM and HadCM3 models, respectively. These projections were generated by Hayhoe et al. (2004), and are based on two scenarios for future societal structures (SRES B1 and A1fi, which project low and high dependence on fossil fuel, respectively). Figure from Hayhoe et al. (2004).

Predicted Changes in Fire Regime Wildfires have helped to shape California’s ecosystems over evolutionary time and are important in many ecosystems today for species regeneration. Fire occurs at somewhat regular return intervals, determining the type and structure of the vegetation, with consequences for all other species (Franklin et al. 2001). Historically, ground fires are thought to have occurred every 15 to 20 years in red fir forests, 4 to 20 years in mixed conifer forests, 6 to 8 years in coastal redwood forests, and 5 to 10 years in grasslands and woodlands (Brown and Swetnam 1994; Skinner and Chang 1996; Swetnam et al. 1998), although Anderson (Chapter 5) and Reiner (Chapter 18) argue that Native Americans may have burned some California grasslands every one to three years. In more recent history, a practice of fire suppression and intensive livestock grazing has radically decreased the fire return frequencies in some grassland regions, allowing shrubs to increasingly colonize and dominate. A number of important factors determine fire frequencies in grasslands, including climate, vegetation type, and extreme weather events such as drought or heavy winds (see Reiner, Chapter 18). Reconstructed fire history shows that fire frequency is highly dependent on climate (Grissino-Mayer and Swetnam 2000; Carcaillet et al. 2001). Consequently, fire models project an increase in both ignition rates and fire spread with the warmer temperatures, lower humidity, higher winds, and drier fuels that are expected under future climate scenarios (Torn et al. 1998; Fried et al. 2004). If precipitation remains high during some years, fire risk during

R E S P O N S E S T O C H A N G I N G C L I M AT E

219

dry years is likely to increase as a result of increased plant productivity in previous years. Torn et al. (1998) projected that climate change will greatly increase the number of wildfires that escape containment in regions of California that have large amounts of grass or brush fuels. Since the 1850s, fire patterns in California have been altered by human suppression, climate change, and land use changes. The history of fire suppression and changes in land use will play an important role in the character of the fires in a future climate. Even in the absence of climate change, fire management in some parts of California has led to larger fires that are more likely to occur in extreme weather (Minnich and Chou 1997; but see Keeley et al. 1999). In addition, any increase in the frequency of Santa Ana wind conditions, which contribute to many of the large fires in California coastal ecosystems (Moritz 2003), combined with warmer, drier summers, could escalate economic and environmental loss to wildfires.

Predicted Changes in Nitrogen Deposition Throughout California, high levels of ammonia and nitrogen oxides are emitted to the atmosphere by nitrogen fertilizer use and feedlots in agricultural areas, and internal combustion engines in urban areas. Although extremely patchy, nitrogen deposition rates in parts of California are among the highest in the United States (Fenn et al. 2003b), with up to 45 kilograms per hectare per year (kg ha⫺1 yr⫺1) in Southern California (Bytnerowicz and Fenn 1996; Padgett et al. 1999), 16 kg ha⫺1 yr⫺1 in Northern California (Blanchard and Tonneson 1993), and up to 90 kg ha⫺1 yr⫺1 in areas of extensive fog exposure (Fenn et al. 2003b). In southern California about 80% of the nitrogen enters ecosystems via dry deposition during the long, dry summers (Bytnerowicz and Fenn 1996), and in the north 75% enters as wet deposition (Blanchard and Tonneson 1993). Because nitrogen can limit primary productivity in grassland ecosystems in California (see Harpole et al., Chapter 10), nitrogen deposition is likely to alter basic ecosystem functions as discussed in this chapter. Under one future scenario, increasing nitrogen emissions are likely to drive a 250% rise in global nitrogen deposition over the current century (Lamarque et al. 2005). A recent estimate suggests that about 30% of California’s annual grassland area currently receives at least 5 kg-N ha⫺1 yr⫺1, and less than 10% receives 7 kg-N ha⫺1 yr⫺1 or more (Figure 19.2; Weiss 2006).

Potential of Invasive Species to Interact with Other Environmental Changes Species native to areas outside the state have been purposely and inadvertently imported to California and are now a permanent feature of California’s ecosystems (see D’Antonio et al., Chapter 6). Non-native, invasive species are commonplace and have replaced native species as dominants in many of California’s grassland ecosystems (Mooney and Hobbs 2000). These species can threaten native species by

220

ECOLOGICAL INTERACTIONS

F I G U R E 19.2. Total annual nitrogen deposition over areas of annual grassland in California. Nitrogen deposition is shown at 36 km resolution. Areas of annual grassland are shown in black. Note that local hotspots and sharp gradients in N deposition rates cannot be resolved at the 36 km scale. Figure from Weiss (2006).

competing with them for critical resources, preying on them, hybridizing with them, or making the environment less suitable for them. Where climate change occurs at rates that undermine the ability of native species to persist, disperse, or become reestablished elsewhere, non-native species may be favored. Changes in climate could either facilitate or suppress grassland invasions, depending on the type and degree of climate change and the extent to which non-native invasive species are adapted to the new climate regime (Dukes and Mooney 1999; Mooney and Hobbs 2000). Thus, climate change could further facilitate biological invasions and decrease biodiversity in a state where over 350 species are already federally listed as endangered or threatened (CDFG 2005). Ecosystem processes, already impacted by climate change, will also be affected by the presence and expansion of nonnative species. In some ecosystems, ecosystem processes are likely to be more strongly affected by non-native species than by climate change. Some species that have already dramatically altered ecosystem processes in California grassland have the potential to further alter the systems under future environmental changes. For example, yellow starthistle (Centaurea solstitialis) populations have rapidly expanded in California grasslands since 1958 (Maddox and Mayfield 1985), increasing soil water consumption in invaded grasslands (Enloe et al. 2004; Gerlach 2004). Yellow starthistle success was positively correlated with late spring water availability in microcosm studies (Dukes 2001a), suggesting that changes in precipitation

regimes might influence the species, and there is evidence that the species could benefit from increasing CO2 concentrations (Dukes 2002a; Ziska 2003; Dukes et al., unpublished data).

Responses and Feedbacks to Global Environmental Changes Two main questions have motivated recent global change research in California grasslands. First, will the functioning of the system change in ways that feed back to affect atmospheric composition? Stated another way, will grasslands sequester more (or less) carbon, thus influencing climate change? Second, will the functioning of the system change in a way that affects the abundance and composition of different species? California grasslands will not respond uniformly to environmental changes, because the grasslands themselves are not uniform. California’s formerly extensive valley grasslands differ in species composition from coastal and coast range grasslands, and both of these communities bear little resemblance to the communities found on serpentine soils. Even within these community types, there is substantial variation from site to site (Harrison 1999b). During the 1990s a seven-year experiment at Jasper Ridge Biological Preserve (JRBP), near Woodside in the San Francisco Bay Area, studied CO2 responses of neighboring grasslands on serpentine and more fertile sandstone soils. These grasslands differ strongly in productivity and composition, with native perennial grasses and native forbs comprising a higher fraction of the biomass and cover in the serpentine grassland and the sandstone grassland resembling a more typical, drier coast range or “valley grassland” dominated by non-native annual grasses (McNaughton 1968; Heady 1988). It is not surprising, then, that the two grasslands can differ in their response to elevated CO2 (Field et al. 1996; Fredeen et al. 1997; Rillig et al. 1999a); these responses are further described in subsequent sections. To date, the majority of California grassland research relating to global environmental changes has taken place in what can be generally called the valley grassland type (see Keeler-Wolf et al., Chapter 3). Understanding how the grasslands will respond to increased nitrogen (N) deposition is perhaps simpler to address than changes in temperature and atmospheric CO2. Motivated by the desire to increase forage yields, researchers have conducted dozens of fertilization trials in California grassland. These provide clear insight on grassland responses to additional N inputs.

Responses to Nitrogen Deposition Studies conducted over several decades in a wide variety of sites and soil types have shown that nitrogen (N) can be the major limiting nutrient for plant growth in California grasslands (e.g., Jones 1974; Woodmansee and Duncan 1980; Jackson et al. 1988; Huenneke et al. 1990, Hull and

F I G U R E 19.3. Proportional responses of NPP (measured as root ⫹ shoot biomass), shoot biomass, root biomass, and root-to-shoot ratio to the four global change treatments in the JRGCE. Each line represents the response over time to a single global change factor, and each data point represents the sum of eight elevated treatment averages divided by the sum of eight ambient treatment averages. Elevated CO2, C (gray dashed line and filled diamonds); increased temperature, T (thin solid black line); increased rainfall, R (thick black dashed line and triangles); nitrate deposition, N (thick solid black line). Figure from Dukes et al. (2005).

Mooney 1990). Recent studies at JRBP have shown an increase in biomass production with 7 g m⫺2 y⫺1 supplemental nitrate deposition in a valley grassland (Shaw et al. 2002; Dukes et al. 2005). Total and aboveground biomass responded to N deposition in four out of five years (Figure 19.3); total biomass increased by 21 – 42%, aboveground by 36 – 61%). The response was largely driven by the two dominant species of grass, Avena barbata and Bromus hordeaceous. Forb production did not respond to increased N deposition (Zavaleta et al. 2003b). Many studies have documented decreases in grassland diversity with increasing N inputs (Grime 1973; Tilman 1993); in California this pattern is driven primarily by decreasing forb diversity and a rise in dominance of one or two species of non-native grasses (Hobbs et al. 1988; Huenneke et al. 1990; Brooks 2003, Zavaleta et al. 2003a, b). In many cases, added N inputs also alter species abundances in the grasslands. In serpentine grasslands, Huenneke et al. (1990) found that supplemental nitrogen addition increased exotic grasses at the expense of native forbs. Weiss (1999) suggests that anthropogenic N deposition has had the same effect on serpentine grasslands in the San Francisco Bay Area that were previously N-limited. He argues that, by favoring non-native grasses over native forbs, N deposition is reducing the vivid displays of California wildflowers that typically occur on serpentine soils and decreasing habitat quality for the Bay checkerspot butterfly (Euphydryas editha bayensis; a “threatened” subspecies under the U.S. Endangered Species Act), which depends on particular host plant species that are small forbs. Supplemental nitrate deposition also favored grasses over forbs in valley grassland at JRBP (Zavaleta et al. 2003b). In coastal grassland increased soil nitrogen availability under dead Lupinus arboreus shrubs

R E S P O N S E S T O C H A N G I N G C L I M AT E

221

favored non-native annual grasses over native perennial grasses and forbs (Maron and Connors 1996). This effect could be mimicked by adding nitrogen to these soils (Maron and Jefferies 1999) and suggests that nitrogen deposition might eventually do the same.

Responses to Precipitation Change Although climate models do not yet converge in their projections for California’s future rainfall patterns, past research has provided some insight on how the state’s grasslands respond to different precipitation regimes. This research is reviewed here, as well as by Reever Morghan et al. (Chapter 7). For decades, researchers have attempted to correlate forage yield (i.e., shoot growth) in California grasslands with the timing and amount of annual precipitation. Their work suggests that there is regional variation in the times in which grasslands are most responsive to precipitation. At the Hopland Field Station (HFS) in northwestern California, Murphy (1970) found that early-season precipitation (i.e., prior to November 21) explained 49% of the variability in shoot growth, but this explanatory power weakened to 34% with additional years of data (George et al. 1989b). On the San Joaquin Experimental Range (SJER) in central California, early-season precipitation explained only 8% of this variability, and combined data from the three most predictive months could only explain 38% of variability (Duncan and Woodmansee 1975). Shoot growth in this grassland was better explained (r2 ⫽ 0.72) by a multiple regression that included several environmental factors, including spring precipitation (George et al. 1989b). Pitt and Heady (1978) used precipitation in October, November, March, and April, along with temperature, to predict 90% of variability in shoot growth at HFS. These authors also used precipitation and temperature to predict growth of different species and groups of species in this grassland. Species responded individualistically in this study. Responses of single species to rainfall in JRBP’s valley-type grassland were also variable, but grouping species into broad functional types allowed Zavaleta et al. (2003b) to detect patterns of response. A 50% increase in each precipitation event and a three-week extension to the rainy season increased the abundance, diversity, and production of forbs, as well as the diversity of grasses. Over five years, this treatment did not affect total production of the grassland, because increased shoot growth (⫹10%) was offset by a decrease in root growth (⫺15%; Dukes et al. 2005). A similar treatment applied by (Suttle et al. 2007) to a coastal grassland in Mendocino County achieved similar results, with moderate increases in shoot growth of plants during some years. A late spring watering treatment in this experiment achieved more dramatic results, more than doubling aboveground biomass in most years. Increased spring precipitation (or shorter summer droughts) can also increase establishment and reproduction of native perennial grasses (Jackson and Roy 1986; Suttle et al. in press). In some locations, wetter springs

222

ECOLOGICAL INTERACTIONS

might trigger the conversion of grassland to shrubland by increasing establishment of native woody species such as Baccharis pilularis (Williams and Hobbs 1989). In competition experiments between blue oak seedlings (Quercus douglasii) and annual grassland species, growth and survivorship of blue oak were greatest in treatments where soil moisture was highest, suggesting that wetter springs could also favor encroachment of oak woodland into grasslands (Gordon and Rice 1993). Precipitation patterns may affect the relative abundance of non-native species in some grasslands. For instance, in the Mendocino County grassland mentioned in the previous paragraph, Thomsen et al. (2006b) found that increasing late spring water availability led to increased survival of seedlings of Holcus lanatus. This non-native species is a highly aggressive perennial grass and is considered a threat to native species diversity in California grasslands (Cal-IPC 2006). In a more interior site, by contrast, Hamilton et al. (1999) found that supplemental water benefited the native perennial grass Nassella pulchra because it ameliorated some of the negative effects of competition with annual grass neighbors. Hobbs and Mooney (1991) attributed increased populations of the invasive annual grass Bromus hordeaceus to successive years of high rainfall in serpentine grasslands at JRBP. However, correlative studies such as this are difficult to interpret with confidence, and additional data (Hobbs and Mooney 1995) made this conclusion less clear. Populations of non-native plant species such as B. hordeaceus likely played a role in the recent extinction of the Bay checkerspot butterfly (Euphydryas editha bayensis) at JRBP by reducing its host plant densities. However, McLaughlin et al. (2002) also strongly linked this extinction to a post-1971 increase in the variability of annual precipitation. Taken together, these studies suggest that precipitation effects on California’s grasslands vary by region, with wet fall or spring seasons increasing shoot growth in the northwest and a wet spring maximizing shoot growth in central regions. An increase in late spring precipitation, or a decrease in the length of the summer drought, would favor perennial species (herbaceous and woody) and would likely promote invasive non-native species in some grasslands. Composition and production of the grasslands are relatively insensitive to increases in precipitation during the wet months of winter. In a Kansas prairie, increasing variability of precipitation has been shown to decrease shoot growth and soil CO2 flux while increasing plant diversity (Knapp et al. 2002). In California, consequences for plants of an increase in the variability of precipitation have not been studied, but one might expect to see similar responses. Increases in variability appear to have already contributed to the extinction of at least one butterfly population in California.

Responses to Warming Warming will likely change production, phenology, and the composition of California grassland, but our ability to quantify these responses is still somewhat basic. From first

principles, one would expect that warming during winter months would generally increase production of the grasslands. Net primary production (NPP) typically increases with temperature across ecosystems (Schuur 2003), and soil warming studies in a variety of ecosystems have typically found that warming accelerates mineralization rates in soils, leading to increased nitrogen availability and therefore plant production (Rustad et al. 2001). Observational evidence from California grasslands supports this line of reasoning. Degreedays are commonly used to predict shoot production, and this metric is more accurate than accumulated days in some areas, for example, SJER but not HFS (George et al. 1988). In developing a model to predict shoot growth at HFS, Pitt and Heady (1978) used mean minimum temperatures in November through February. As discussed, precipitation is also important to grassland production, and degree-days alone are not always useful in explaining variability of forage production (George et al. 1989b). Despite the supportive observational evidence, research from the Jasper Ridge Global Change Experiment (JRGCE, discussed subsequently) suggests that small amounts of warming have little effect on plant production in valley-type grassland (Dukes et al. 2005). In the JRGCE, a ⬃1 ⬚C warming did not detectably affect plant growth (Figure 19.3). Warming during the wet winter months accelerated the flowering (Figure 19.3; Cleland et al. 2006) and senescence (Zavaleta et al. 2003c) of many grassland species. Because warming caused the dominant annual grasses to senesce earlier, transpiration of the warmed grassland decreased in late spring, and the grassland retained more soil moisture through at least the early summer. This additional water (about 1% volumetrically; Zavaleta et al. 2003c) might benefit species that depend on spring and summer moisture, such as late-season forbs and woody species. By accelerating plant phenology, warming will affect the animal species that depend on plants. Working in serpentine grasslands, Hellmann (2002) found that warming accelerated senescence of the host plants of Bay checkerspot butterfly larvae (Euphydryas editha bayensis), causing the larvae to switch to an alternative host species earlier in the season. Consequences of such host shifts are not yet understood.

Responses to Rising Atmospheric CO2 Concentrations LEAF-LEVE L R E S P ON S E S

A rise in CO2 increases net rates of leaf photosynthesis in many plants, including native and non-native C3 grasses and forbs in the California grasslands (Jackson et al. 1995). Under an approximate doubling of CO2 relative to current levels, assimilation rates may increase by as much as 70 to 132% depending on the environmental conditions and plant growth form (Jackson et al. 1995; Dukes 2002a). Long-term exposure to elevated CO2 can result in a range of responses in grasslands, including the maintenance of increased rates of photosynthesis (Huxman and Smith 2001), the down-regulation of

photosynthesis (a homeostatic adjustment that aligns wholeplant processes and carbon gain rates to within some bounds important for coordinated function; Oechel et al. 1995), or nonsignificant responses (Jackson et al. 1995). In the case where increased photosynthesis under elevated CO2 occurs initially, the plant may later down-regulate its photosynthetic capacity, resulting in a decrease or leveling off of the CO2 fertilization effect (Tissue et al. 1993), often characterized by a decline in essential nutrients in the leaf that leads to the reduction of photosynthetic capacity. Elevated CO2 can result in greater leaf-level photosynthesis, therefore, during periods of high water and nutrient availability. However, in California grasslands, this CO2 fertilization effect on leaf-level photosynthesis does not consistently lead to an increase in plant growth (see subsequent discussion). The behavior of stomata on plant leaf surfaces simultaneously regulates carbon dioxide uptake and water loss from the leaf to the atmosphere. The term stomatal conductance refers to the rate at which gas molecules can enter or exit the leaf. When stomata close down, stomatal conductance decreases, meaning that gas molecules, including water vapor, exchange more slowly. Increases in atmospheric CO2 concentrations cause stomata to close somewhat (Linsbauer 1917), slowing transpiration (Ketellapper 1963). The role of CO2 in regulating stomatal conductance has been well studied, with C3 herbaceous species showing, on average, a 40% decrease in conductance in response to a doubling of ambient CO2 concentration (Morison 1985). In California grasslands, elevated CO2 led to a 45% decrease in stomatal conductance in midseason (Jackson et al. 1994; Oechel et al. 1995). Continuous, long-term exposure to elevated CO2 is unlikely to dampen this response, based on studies of Mediterranean grassland species growing near natural CO2 springs, where plants have been exposed to an elevated CO2 environment for many generations (Betterini et al. 1998). Thus, rising atmospheric CO2 allows plants in Mediterraneantype ecosystems to meet the growth demand for carbon with less water lost through transpiration. This leads to an increase in water-use efficiency (WUE; the mass of carbon fixed per mass of water transpired), and less drawdown of soil moisture (Jackson et al. 1994). The direct CO2-induced responses of decreases in stomatal conductance—increases in leaf water potential and increases in WUE — can enhance production and fecundity of individual species in the California grassland (Jackson et al. 1994) and reduce plant community evapotranspiration (Field et al. 1997; Fredeen et al. 1997; Lund 2001). This reduction in plant community water use causes soils to be wetter in the spring relative to soils in ambient CO2 (Fredeen et al. 1997; Zavaleta et al. 2003c; Morgan et al. 2004). The additional soil moisture is particularly important at the end of the growing season, when moisture is limiting to plant growth, effectively lengthening the growing season (Fredeen et al. 1997; Lund 2001; Zavaleta et al. 2003c). This indirect response to elevated CO2 can potentially accelerate nitrogen cycling (Hungate et al. 1997a) and increase plant

R E S P O N S E S T O C H A N G I N G C L I M AT E

223

production in water-limited ecosystems such as California grasslands (Chiariello and Field 1996; Volk 2000; see subsequent discussion).

availability in the late spring is potentially more important for increasing total plant productivity in Mediterraneanclimate grasslands than is the direct effect of elevated CO2. This possibility deserves further study.

B IOMAS S AN D ALLO CATION CHANG E S M IC ROB IAL C OM M U N IT Y R E S P ON S E S

Increased carbon fixation at the leaf level does not always lead to an increase in the biomass production of individual species, functional groups, or the plant community (Morgan et al. 2004). In addition, from existing analyses of the data, growth responses to CO2 are not consistent from species to species, within functional groups, across years, or across grassland types (Poorter and Navas 2003). In California grasslands, there have been a number of efforts designed to address the impact of elevated CO2 on grassland species and community production responses in both sandstone and serpentine systems using microcosms (Field et al. 1996) and in situ experiments (Hungate et al. 1997a; Shaw et al. 2002). The great majority of this work has been conducted at JRBP. In microcosms (essentially large pots of grassland soil maintained outdoors in field conditions), single- and multiplespecies experiments consistently showed increases in aboveground biomass with elevated CO2 (Chiariello and Field 1996; Verville 2000; Joel et al. 2001; Dukes 2002a), but the species-specific and functional responses varied in each study (Joel et al. 2001). In microcosms simulating nutrient-poor serpentine grasslands, where aboveground production is 50% of that in the sandstone community in ambient conditions (approximately 800 vs. 400 g/m2; S. S. Thayer, unpublished data), the response of NPP was less than in microcosms using more fertile sandstone-derived substrate (18% vs. 54% increase in NPP, respectively; Joel et al. 2001), but responses of both were consistently positive. In contrast to microcosm experiments, in situ experiments in sandstone and serpentine grasslands have shown inconsistent responses from year to year. Field experiments have shown small (Field et al. 1996; Shaw et al. 2002) or flat (Dukes et al. 2005; see also Hungate et al. 1997a) responses of plant growth to elevated CO2, indicating that indirect feedbacks play an important role in determining grassland production responses to elevated CO2. In a year when CO2 increased total plant production, root growth decreased sharply with CO2 enrichment (Shaw et al. 2002). Zavaleta et al. (2003b) found no aboveground production response after three years of exposure to elevated CO2. There is some indication that summer-flowering species respond more strongly to elevated CO2 than the grassland dominants. Unlike the dominant annual grasses, these species require soil moisture in late spring and summer and may take advantage of the soil moisture savings seen during the spring as a result of reduced evapotranspiration of the dominants (Chiariello and Field 1996; Field et al. 1996; Moore and Field 2006). Shrubs such as Baccharis pilularis, because they are active into the summer, may also respond to water savings generated by elevated CO2 (Zavaleta 2006). This indirect effect of enhanced water

224

ECOLOGICAL INTERACTIONS

Elevated CO2 alters microbial communities in California grasslands, with the most dramatic effects on fungal communities. Supplemental CO2 significantly increased active fungal biomass early in the growing season on both sandstone and serpentine soils at JRBP (Hungate et al. 2000). Infection of roots by arbuscular-mycorrhizal (AM) fungi increased on both substrates under elevated CO2, and hyphal lengths of these fungi increased on sandstone substrates (Rillig et al. 1999a). The increases in AM fungi could have been driven by increases in carbon fixation and root carbon exudation by plants under elevated CO2. The increase in AM fungi may have led to observed increases in soil aggregation, which could lead to increased carbon retention in soils (Rillig et al. 1999b). AM fungi produce glomalin, which is thought to be important in maintaining soil aggregate stability (Tisdall and Oades 1982). In contrast, artificial climate warming increased AM fungi, but decreased glomalin concentration and, therefore, soil aggregate stability and carbon retention (Rillig et al. 2002). R E S P ON S E S OF P LANT C OM M U N IT Y C OM P OS ITION, AB U N DANCE AN D DIVE R S IT Y

Changes in individual plant performance should affect competitive hierarchies and the representation of different species in terrestrial communities at elevated CO2 (Bazzaz 1990). This theory is based on the assumptions that elevated CO2 will change total biomass production and that changes in species-specific production will scale directly to changes in fecundity. This assumption is supported by the strong, species-specific effects of elevated CO2 seen in grasslands (Jackson et al. 1994; Chiariello and Field 1996; Verville 2000; Joel et al. 2001; Zavaleta et al. 2003b). In a one year sandstone microcosm experiment, Joel et al. (2001) found increases in the number of individuals of Avena barbata and Lotus wrangelianus, and concurrent increases in production of those species relative to others, but no change in the allocation to reproductive structures. How seed production and the growth potential of future offspring are influenced by elevated CO2 is not well known and has been addressed in only one grassland study with one species, Avena barbata, where seed production increased by 30 % (Jackson et al. 1994). Evidence from multiple year studies suggests that elevated CO2 may not cause drastic shifts in dominance among the species typical of California’s valley grasslands. In a three year in situ experiment on the sandstone grassland, Zavaleta et al. (2003b) found a small and inconsistent growth response of Avena barbata to elevated CO2, but little or no response from other species. At the community level,

elevated CO2 produced no overall change in functional group production, a decreased abundance of forbs, and a decrease in overall species richness, but analysis across multiple years showed no directional change. The observed and predicted rise in CO2 may foster nonnative species invasions, especially in conditions where plant growth rates are CO2-dependent (Dukes and Mooney 1999; Dukes 2000; Weltzin et al. 2003). Recent studies suggest that native and non-native species sometimes differ in their growth responses to elevated CO2, and that some nonnative species could be more competitive in a more CO2-rich atmosphere (Dukes 2002a). However, results to date are merely suggestive and will likely depend on the particular composition of a site. Future species composition, abundance, and diversity of California grasslands are likely to be determined by the suite of global changes, the novel environment that results as well as the movement of native and non-native species.

Responses to a Complex Future: The Jasper Ridge Global Change Experiment Over the current century, changes in the composition of the atmosphere and in the climate are expected to occur simultaneously. How can we predict the future composition and performance of any ecosystem when so many factors are changing at once? It is tempting to extrapolate from studies that have looked at responses to a single variable such as CO2. However, there are many reasons to suspect that some effects of these factors may not be additive. For instance, will effects of water savings from elevated CO2 disappear if precipitation increases? Globally, very few experiments have tested the responses of natural ecosystems to more than one global change at a time. Grasslands, because of the high density and small stature of individual plants, are some of the more tractable systems for examining such questions. In California grassland, many such questions have been addressed by one of the oldest multifactor studies: the Jasper Ridge Global Change Experiment (JRGCE), which began operating in 1998. In its original design, the JRGCE exposed plots of valleytype grassland at JRBP to a factorial combination of warming, elevated CO2, increased precipitation, and increased nitrogen deposition (Figure 19.4). Circular, 2-m diameter plots experienced either ambient conditions, ⬃1 ⬚C of warming from infrared heat lamps, CO2 enhancement of ⬃300 ppm via a Free-Air CO2 Enrichment (FACE) system, or warming and CO2 enrichment together. Each plot was divided into four quadrants, which received ambient or increased precipitation and nitrogen deposition. Some results from this multifactor experiment are described in the following sections.

erated the onset of flowering of grassland species, precipitation had no clear effect. Elevated CO2 and supplemental nitrate deposition both accelerated flowering of forbs but delayed flowering of grasses (Figure 19.5). Since the grasses generally flower earlier than the forbs, these two treatments led to a more synchronous flowering of all grassland species, with possible consequences for the timing of resource uptake in this ecosystem. Phenological responses to warming and CO2 were additive, as were the late spring water savings due to these treatments (discussed above in previous sections; Zavaleta et al. 2003c). Shaw et al. (2002) described responses of the grassland’s peak biomass (above- plus belowground) to the global change treatments after three years of exposure, focusing primarily on the CO2 response. During the 2000 – 2001 growing season warming, precipitation, and nitrogen all tended to stimulate biomass production, individually or in combination. Plant growth responses to some treatment combinations were approximately additive. For example, the plots that were warmed increased NPP by 18% relative to all those at ambient temperature and the N deposition plots increased NPP by 34% overall, but the combination of warming and N deposition increased NPP by 62%. Other treatment combinations showed similar patterns. The interaction of CO2 with other manipulations was the most complex. CO2 enrichment increased aboveground biomass when all other factors were at ambient levels, an observation typical of grassland responses to CO2 elsewhere (Nowak et al. 2004). However, elevated CO2 significantly suppressed production responses to elevated levels of precipitation, warming, and nitrate deposition (Figure 19.6). Much of the suppression occurred in the root biomass, which declined by 22% in response to CO2 across all treatment combinations. Although this strong interaction appeared in the 2001 season, interactions involving CO2 or warming were less common in other years. Dukes et al. (2005) found only one interaction

C OM M U N IT Y R E S P ON S E S TO M U LTI P LE G LOBAL CHANG E S

Most global change treatments affected the phenology of grassland species, with occasionally surprising consequences (Cleland et al., 2006). Although warming consistently accel-

F I G U R E 19.4. One of 32 experimental plots in the Jasper Ridge Global Change Experiment.

R E S P O N S E S T O C H A N G I N G C L I M AT E

225

FIGURE 19.6. Response of NPP (root ⫹ shoot biomass) to F I G U R E 19.5. Shift in date of flowering onset caused by the four

simulated global changes, for nine common species observed in the JRGCE. The effect shown for each species is the difference in mean first flowering date for all plots, where each aspect of environmental change is elevated (N ⫽ 64) versus ambient (N ⫽ 64), over three years of observation. Figure from Cleland et al. 2006.

in plant growth responses over the first five years of the experiment (a four-way interaction in the 1999 –2000 growing season). Beyond their effects on total production, some global change treatments affected the relative abundance of species or groups of species. Zavaleta et al. (2003b) characterized these responses over the first three years of the study. The treatment combination that featured warming, elevated CO2, and increased precipitation caused the greatest change in relative abundance of the functional groups: a 50% increase in forb abundance. Despite this, forb richness tended to decrease in plots with elevated CO2 or supplemental N. Much of the response of grassland composition to global change may be driven by altered patterns of herbivory. Peters et al. (2006) found that changes in gastropod feeding preferences under the various global changes could frequently account for ⬎50% of the changes in grassland composition observed in the JRGCE. E C O SYSTE M R E S P ON S E S TO M U LTI P LE G LOBAL CHANG E S

Systems that do not increase biomass production under future conditions are unlikely to store additional carbon and therefore unlikely to slow climate change. Thus, it is important to know why elevated CO2 did not increase plant growth in this system. The answer may provide clues to whether other systems will respond similarly. Some evidence suggests that phosphorus (P) limitation may account for the lack of CO2 response by the plant community. Cleland (2005) found reduced P concentrations in plants growing under elevated CO2 or increased N deposition. Under some treatment combinations, elevated CO2 even reduced total plant P uptake. The mechanism behind this pattern is still unclear. Phosphorus has been shown to be limiting to production in some California grassland ecosystems (Chapter 10).

226

ECOLOGICAL INTERACTIONS

environmental change treatments in the JRGCE during the 2000–2001 growing season. Bars show percent changes in NPP for treatments or pooled groups of treatments, relative to a corresponding reference. For each treatment combination, the reference is defined as all ambient CO2 plots in which the variable(s) that defines each treatment combination are also at ambient. Abbreviations: amb, ambient; C, elevated CO2; N, nitrate deposition; T, increased temperature; R, increased rainfall. For treatments T, N, and R, n ⫽ 24 plots. For treatments TR, NR, and TN, n ⫽ 12. For treatments C and TNR, n ⫽ 6. For example, the open bar for the increased temperature pair (T) is calculated using all treatments with increased temperature but not elevated CO2 (n ⫽ 24). The gray bar in the same pair is calculated using all treatments with increased temperature and elevated CO2 (n ⫽ 24). Figure adapted from Shaw et al. (2002).

In the summer of 2003, an accidental fire burned through part of the JRGCE, changing the experimental design somewhat and providing the opportunity to compare the global change responses of burned and unburned patches of grassland. An analysis by Henry et al. (2006) found a negative response of production to elevated CO2 in unburned grassland, but no response in burned areas. Plant growth responded more positively to experimental nitrate deposition in burned compared to unburned areas. The fire may have released P limitation by making phosphorus available in ash. This possibility is supported by N:P ratios of annual grass shoots, which were lower in burned compared to unburned plots. Although the burn may have released nutrients, the burn alone did not affect plant production in the grassland, suggesting that increases in CO2 and nitrate availability may somehow trigger P limitation. Because the fire simultaneously added ash and changed the microclimate in burned plots, Henry and colleagues could not conclusively link P availability to the observed CO2 and N responses. The role of P in the grassland is being further explored, and researchers are examining whether changes in herbivory, allocation, phenology, or other factors may prevent a positive CO2 response. Because species differ in their chemical makeup, shifts in community composition can affect litter quality. Henry et al. (2005) found that global change treatments affected lignin and nitrogen concentrations in litter but that some of these direct effects were counteracted by the shifts in community composition. Direct effects included CO2-

induced increases in lignin (which were attenuated by warming in grasses and increased precipitation in forbs) and increases in litter nitrogen concentration in response to nitrogen addition. This latter effect was dampened by increased precipitation. Overall, these changes in litter quality did not markedly affect decomposition rates over the period measured. Litter decomposition and many other ecosystem functions could be altered by shifts in microbial community structure. Horz et al. (2004) observed shifts in the abundance and community composition of ammonia-oxidizing bacteria in response to some global change treatments. In this case, all responses depended on the level of other factors. Nitrogen deposition affected community structure, but only at ambient temperature and precipitation, and elevated CO2 affected bacterial abundance, but only under increased precipitation.

Modeling Grassland Responses It is not possible to assess the full extent of ecosystem responses to global environmental changes through experimentation alone (Aber et al. 2001). To better understand how these changes might alter the distribution and extent of grassland ecosystems in California, researchers use ecosystem models that incorporate information from field experiments. Dynamic global vegetation models (DGVM) (Cramer et al. 2001) use plant physiological and ecological traits to categorize vegetation into plant functional types (PFTs) and establish life history characteristics for each type. The models parameterize the physiological processes and climatesensitivity associated with each PFT based on experimental results. The models also consider competitive interactions that take place when two different PFTs occupy the same physical space, also based on experimental results. The newest generation of DGVMs incorporates the complexity of ecosystem dynamics into the analysis of responses to a changing climate (Lenihan et al. 2003). There are important limitations of DGVM exercises. Because of the focus on plant functional types, these models do not incorporate species migration patterns or dispersal capabilities. Consequently, these models have limited value in projecting ecosystem distributions, given that species will move as individuals in response to climate change and not as ecosystems or plant functional types. In addition, they simulate only potential or natural vegetation, not existing vegetation, and they do not adequately integrate the impacts of other global changes such as land use change (fragmentation and human barriers to dispersal), nitrogen deposition, or invasive species. The majority of California grasslands are under some form of management (e.g., grazing), which itself will likely change as socioeconomic considerations change. Also, as discussed earlier, many grassland sites are experiencing N deposition. It is

unclear how climate change will interact with these other changes, and the models as yet do not address this question. Despite these caveats, modeling exercises have provided interesting insights into possible responses of California vegetation to climatic changes. Lenihan et al. (2006) conducted a modeling exercise to understand biome shifts in California under a changing climate. Their objective was to “dynamically simulate the response of vegetation distribution, carbon and fire to three scenarios of future climate change for California using the MAPSS-CENTURY (MC1) dynamic vegetation model,” driven by climate output from two GCMs (Bachelet et al. 2001; Lenihan et al. 2003, 2006). Valley grasslands, southern coastal grassland, and desert grassland were modeled as a single biome type (no land was occupied by agriculture or urban areas in the simulations). Grassland total percent cover increased by ⬎65% under all future climate scenarios (Figure 19.7). Grasslands and shrublands were both initially favored by the increase in moisture under future climate scenarios, but increases in grass biomass produced more fine fuel, which promoted a higher frequency of fire, resulting in the expansion of grasslands (Lenihan et al. 2006). All scenarios showed percent cover declines of ⬎60% for alpine/subalpine forest, ⬎30% for shrubland, and ⬎20% for mixed evergreen woodland. Only grasslands and mixed evergreen forest increased in total percent cover. Other modeling efforts, using a statistical climate envelope model, also support the probability of large shifts from woody vegetation to grasslands (Kueppers et al. 2005). Although none of these simulations and exercises can be interpreted as predictions of the future, they provide important insight into direct and indirect feedbacks that will influence biome distribution under a changing climate. In the study by Lenihan et al. (2006), fire plays a critical role in major biome shifts, either by slowing encroachment of shrubland into grasslands under high-precipitation scenarios or by increasing the rate of transition of woody biomes to grasslands under low-precipitation scenarios. Field studies in the Central Coast of California support this finding; there, woody communities are intolerant to high fire frequency and are replaced by grassland under such conditions (Callaway and Davis 1993; Keeley 2002; Tyler et al., Chapter 14; Reiner, Chapter 18). Changes in fire regime and the relative dominance of woody and grass PFTs will have consequences for climate impacts on total ecosystem carbon storage (Lenihan et al. 2006). Under a warmer, wetter climate, increases in productivity could lead to increases in ecosystem carbon storage but this process could be undermined by an increase in fire frequency. Under a warmer, drier climate, decreases in productivity could be accompanied by decreases in decomposition, which would cause fine fuels to persisfor longer. In addition, the increase in percent cover of grasslands, which are better adapted to frequent fire than are most woody vegetation types, could decrease carbon storage in the short term by

R E S P O N S E S T O C H A N G I N G C L I M AT E

227

Alpine/Subalpine Forest Conifer Forest Mixed Evergreen Forest

decreasing the amount of carbon stored in plant biomass. In the long-term (⬎100 years), an increase in the percent cover of grassland could either increase total carbon storage (plant and soil) by increasing carbon inputs to soil (Jackson et al. 2002; Golubiewski 2006), or decrease total carbon storage by reducing plant carbon pools (Hibbard et al. 2003).

Mixed Evergreen Woodland Grassland Shrubland Arid Lands California Boundary

Historical (1961 – 1999)

Summary Several experiments and dozens of researchers have sought to better understand how California grasslands will respond to changes in the atmosphere and climate (Table 19.1). Biomass production in these grasslands is not highly responsive to increases in atmospheric CO2, small increases in temperature, or increases in winter precipitation. Rather, changes in the length of the rainy season would be more likely to affect production, as would increases in nitrogen deposition. Species compositional changes may be more dramatic than changes in production but are also generally more difficult to predict because of the specificity of responses and the high diversity of species in California grasslands. Studies to date have largely focused on changes that might occur within the current vegetation type. Nonnative species and native woody species, however, may have the opportunity to enter and spread under new climate and atmospheric regimes, which might lead to greater changes in the functioning of the ecosystem. Studies examining this possibility are now underway at JRBP (C. Field, personal communication). Model simulations suggest that changes in fire regime could strongly promote or constrain any shifts in vegetation types.

Acknowledgments We thank Nona Chiariello and Carla D’Antonio for their insightful and constructive comments on earlier versions of this chapter. We thank Stuart Weiss for editing Figure 2 for this publication, and John Wells and Jim Lenihan for translating Figure 7 into black and white. Future (2070 – 2099)

FIGURE 19.7. Simulated historical (1961–1990) and future (2070 – 2099) vegetation class distributions from the MC1 dynamic vegetation model of Lenihan et al. (2006). The vegetation class shown in each cell occupied that cell most frequently during the relevant time period. Future distribution is based on the GFDL-B1 climate scenario. Figure from Lenihan et al. (2006).

228

ECOLOGICAL INTERACTIONS

High Medium

Species composition Species diversity Low Low High Medium Low

Medium Medium

Root growth Flowering onset

Shoot growth Root growth Flowering onset Species composition Species diversity

High

Shoot growth

Medium

Species diversity

Medium

Species diversity Medium Medium Low High

Medium

Species composition

Shoot growth Root growth Flowering onset Species composition

Low Medium Medium

Sensitivity to global change factor

Shoot growth Root growth Flowering onset

Response variable

Medium Very high Medium High High Medium Medium Medium High Medium Low

⫹ ⫹ Forbs earlier, grasses later Favors grasses ⫺forbs ⫹ 0/⫹ All species earlier Favors some forbs 0

High Medium High High, but see note

Medium

Medium

High Medium High

Confidence in predicted responseb

⫹ ⫺ 0 In winter, favors forbs; in spring, favors perennials ⫹forbs

0/⫹/⫺ ⫺ Forbs earlier, grasses later Favors lateseason species ⫺forbs

Direction of response

Dependent on timing

Dependent on timing Dependent on timing

Varies by year

Notes

3 3 6 8 9

9

8, see also Chapter 10

3, 10, 11, 12, see also Chapter 10 3 6

9

3, 4, 13 3 6 4, 8, 13, 14, 15

9

1, 7, 8

1, 2, 3 2, 3, 5 6

Related referencesc

bConfidence

is a subjective assessment of the likelihood of the environmental change occurring in the indicated direction prior to 2100. These assessments are informed by several documents cited in the text. in predicted response is a subjective measure but is based on the number of studies that have examined a given response, the magnitude of the treatment applied in these studies, and the consistency of response across these studies. More studies, greater departures from ambient conditions, and greater consistency among results lead to higher confidence. cKey to references: 1: Field et al. 1996; 2: Shaw et al. 2002; 3: Dukes et al. 2005; 4: Suttle and Thomsen, in press (note that this is a coastal range grassland); 5: Higgins et al. 2002; 6: Cleland et al., 2006 7: Chiariello and Field 1996; 8: Zavaleta et al. 2003b; 9: Zavaleta et al. 2003a; 10: Jones 1974; 11: Woodmansee and Duncan 1980; 12: Jackson et al. 1988; 13: Seabloom et al. 2003b; 14: Jackson and Roy 1986; 15: Gordon and Rice 1993.

aThis

Warming (high)

Nitrogen deposition (medium; spatially variable)

Increased precipitation (low; spatially and temporally variable)

Elevated CO2 (very high)

Global change prediction (confidence in predictiona)

TA B L E 19.1 Some Expected Responses of California Valley-Type Grasslands to Global Changes

P O LI CY AN D MANAG E M E NT

TWENTY

Grazing Management on California’s Mediterranean Grasslands LYN N H U NTS I N G E R, JAM E S W. BARTO LO M E, AN D CA R LA M. D’ANTO N I O

Grazing management uses ungulates to achieve desired ecological, social, and economic outcomes. It is the least costly and in some ways the most flexible tool for managing vegetation on California grasslands. Grazing can be used to manipulate plant community structure, decrease fuel loads, control invasive weeds, create wildlife habitat, and enhance species diversity (Marty 2005; Pyke and Marty 2005). Grazing is an important natural process in grassland ecosystems (McNaughton et al. 1989; Milchunas and Lauenroth 1993; Perevolotsky and Seligman 1998), and a number of studies have recorded higher biodiversity in grazed compared to ungrazed systems (Noy-Meir et al. 1989; Harrison 1999b; Hayes and Holl 2003a). Like almost any other human-directed or influenced activity in grasslands, grazing can also harm vegetation and wildlife if improperly managed (Painter and Belsky 1993; Fleischner 1994; Freilich et al. 2003). The history of livestock grazing and its impacts in California is controversial and complex (D’Antonio et al., Chapter 6; Jackson and Bartolome, Chapter 17). This chapter concentrates on what is known about goals of contemporary grazing programs and the manipulation of vegetation using domestic livestock. Understanding the effects of grazing on vegetation in the California grassland is complicated by a large climatic gradient, pronounced interannual variation in weather, strong variation in topography and land-use history, and regional variation in the species pool. At any one site, the impact of grazing arises out of the interaction of land use history, the current and recent grazing management scheme, the abiotic environment, and the species pool in the local plant community (Heady 1984). The influence of the abiotic environment, including soil type, elevation, precipitation, and temperature, is particularly important in the California Mediterranean environment. Because so much of the grassland is composed of annual species, composition, density, and productivity are highly influenced by the annual pattern and amount of rainfall, with production varying by orders of magnitude among

years. California’s heterogenous soils also add to the heterogeneity of the grassland and response to management actions. For these reasons California grassland scientists have developed methods for managing the grassland that emphasize coping with variation and recognizing the overwhelming role of abiotic factors. Management programs that are based on a conceptual model of long-term, competition-driven vegetation shifts in a relatively consistent environment (e.g., equilibrium-based theory) have limited application in California grasslands, particularly annual-dominated sites. This chapter first highlights the common grazing management goals of the agencies, private landowners, and conservation organizations that manage most of the California grassland. Next, general principles for developing management practices and plans based on the existing knowledge base for grazing and vegetation in California are presented and discussed. The final section of the chapter reviews the knowledge base for implementing specific practices to achieve some common management goals. It is widely acknowledged that management varies widely from place to place on California grasslands, potentially providing abundant examples of sustainable and unsustainable management practices on private and public lands. In this chapter we do not attempt a catalog of historic or current practices, an assessment of their detriments and benefits, or a referendum on different kinds of management. Instead, the focus is on the common goals of grazing management in California Mediterranean grasslands, and what we know about the practices that can contribute to achieving these goals.

Landowners, Managers, and Management Goals Most annual grasslands in California are privately owned (Table 20.1). The United States Forest Service, Bureau of Land Management, National Park Service, and Department of Defense also own a small but significant amount of California

233

TA B L E 20.1 Examples of California Grassland Ownership in 2002 and Grazing Management Goals

Ownership

Annual grasslands 1,000s of hectares (% of woodlands)

Oak woodlandsa 1,000s of hectares (% of woodlands)

Private

3,806 (88%)

1,733 (82%)

Livestock production based on grassland production, lifestyle benefits, family heritage, working with friends and relatives, making a living (Liffmann et al. 2000; Sulak et al. 2002; Larson et al. 2005).

Speculative

Unknown, included in private

Unknown, included in private

Fire hazard reduction, maintaining land in agriculture for tax purposes.

Land Trust

Unknown, included in private

Unknown, included in private

Invasive species control, maintaining agriculture and facilitating land restoration, fire hazard reduction, biodiversity. Preservation of ranches, rancher stewardship (CRT 2005).

181 (4%)

98 (5%)

The use of BLM rangelands for “grazing is supported by the Federal Land Policy and Management Act of 1976 (FLPMA), the Public Rangelands Improvement Act (PRIA) of 1978, and the Taylor Grazing Act (TGA) of 1934, as amended. The TGA requires the BLM to prevent overgrazing and soil deterioration; to provide for orderly use, improvement, and development; and to stabilize the livestock industry that depends upon the public rangelands” (BLM 2005: 90).

United States Bureau of Land Management (BLM)

Grazing management goal

“To improve the health and productivity of the land to support the BLM multiple use mission and to sustain the health, diversity, and productivity of the public lands” (BLM 2005: 4). United States Forest Service (USFS)

90 (2%)

126 (6%)

Sustainable use in concert with multiple uses; maintaining and improving rangeland health, restoring rangeland ecosystem functions; managing the vegetation resources across the landscape to serve a multitude of resource needs (USDA-USFS 2005a). “To promote rural economic development and a quality rural environment” (USDA-USFS 2005b). “To manage, maintain, and improve public rangelands to support all rangeland values in accordance with management objectives and the land use planning process. To prevent economic disruption and harm to the livestock industry” (PRIA).

National Park Service (NPS)

15 (
1%)

17 (
1%)

“To preserve unimpaired the natural and cultural resources and values of the National Park System for the enjoyment, education, and inspiration of this and future generations” (NPS Organic Act of 1916). Pt. Reyes National Seashore: “To acknowledge the historic, cultural, and ethnic diversity of the area, resources such as the cultural landscapes embodied in the historic ranches, will be preserved” (NPS 2003).

TA B L E 20.1 ( C O N T I N U E D ) Examples of California Grassland Ownership in 2002 and Grazing Management Goals

Ownership State and Other Public Lands

Department of Defense

Total

Annual grasslands 1,000s of hectares (% of woodlands)

Oak woodlandsa 1,000s of hectares (% of woodlands)

241 (6%)

125 (6%)

Fire hazard reduction, management of invasive species and vegetation. Utilizing cattle, sheep, and goats as a vegetation management tool to maintain and improve habitat conditions for resident plants and animals and to prevent wildfires. Ongoing research indicates that moderately grazed areas generally display a greater diversity and density of plant and animal life (EBRPD 2005).

Included in “other public”

Included in “other public”

Fire hazard reduction, restoration of native species, and ecosystem management. Camp Pendleton: “An ecosystem based approach to facilitate maximum support for the military training mission while promoting the sustainability of native species and habitat diversity and compliance with laws and regulations” (USMC 2005). Air Force Instruction 32-7064 (USAF 1997) states, “Agricultural outleasing for cropland and grazing can be used to maintain ecologically sound stewardship of public lands. Outleasing can produce a cash flow to sustain the program, enhance other aspects of the natural resources management program, and reduce the maintenance costs of semi-improved lands. However, all agricultural operations must be compatible with the military mission and long-term ecosystem management goals.”

4,333

2,096

California grasslands are 11% of the state; oak woodlands are 5%.

Grazing management goal

NOTE : CDF-FRAP 2003 p. A-27. Because open oak woodlands often support significant grasslands used primarily for grazing, these are included here. The available data do not distinguish open oak woodlands known as savannah, from closed canopy woodlands. They are also for total area owned, not grazed acres. aIncludes Blue, Coastal, Foothill, and Valley Oak hardwood types from the California Wildlife Habitat Relationships System as assessed by CDF-FRAP (2003). Montane types are excluded.

grassland habitat. In addition, state, local, and regional agencies and watershed, utility, and recreation districts often manage and own annual grasslands. A growing phenomenon in California is ownership by land trusts: the amount of land conserved by trusts through purchase, transfer, or easement increased by 147% between 1998 and 2003 (LTA 2004), to more than 600,000 hectares (ha). Of this land, 121,000 ha are protected by conservation easements. There are now at least 173 land trusts in California, more than in any other state (LTA 2004). Data on how much of that are grassland is not available. Although the various ownership groups tend to have different management goals for livestock grazing, maintaining grassland “sustainability” is a common theme. Private landowner goals are usually more influenced by the need to generate income, while agencies and land trusts may be

more focused on goals related to their particular mandates, such as fire hazard reduction, restoration of native species, water quality protection, and protection of endangered species. There is, however, considerable overlap and blending of goals. Agencies and land trusts that use grazing to manage vegetation need to accommodate the goals and needs of those who supply the livestock, just as the livestock provider needs to accommodate the goals of the agency or trust. In the following sections we briefly review goals of different groups.

Private Landowners Livestock producers manage more California grassland than any other kind of owner, and also provide the livestock for

GRAZING MANAGEMENT

235

grazing on millions of hectares of public and private leased land. More than three quarters of grassland parcels larger than 80 hectares are grazed by livestock (Huntsinger et al. 1997, 2007). The low-intensity use typical on many large parcels means that these properties may provide significant wildlife habitat, floristic diversity, watershed, and viewshed. About 42% of oak woodlands are owned by those who produce livestock for sale, and another 10% of owners produce livestock for their own use only. As estimated from recent surveys, an additional 10% of oak woodland landowners graze stock on their property by leasing out their land to ranchers (Huntsinger et al. 2007). County tax assessor data show that many hectares of California oak woodlands and annual grasslands are owned by corporations and investment groups. A significant portion of these are holding land as an investment, anticipating continued rising land values. Maintaining grazing on these properties reduces fire hazard and qualifies the land for tax benefits based on agricultural use. Homeowners’ associations may also use livestock grazing. At Sea Ranch along the Sonoma county coast, sheep and goats are used to control noxious weeds, reduce the frequency and intensity of wildfires, improve wildlife habitat, and enhance biodiversity within the 4,000-acre residential community (Barry 2005). The great majority of livestock producers live on their properties and manage the land themselves. Most have owned their land for a long time: 63% of public land ranchers in one survey reported that their families have owned their ranch for more than 100 years (Sulak and Huntsinger 2002). In 2005, only one quarter of oak woodland ranchers in a statewide survey reported that the majority of household income came from ranching, while 10% reported farming as their major source of income. About 22% cited off-ranch wages as their major income source, and another 22% earned most of their income from other forms of “selfemployment,” including investments, pensions, and so forth (Huntsinger et al. 2007). What ranchers say makes ranching worthwhile is experiencing the lifestyle, raising a family on a ranch, working with livestock, and enjoying the natural environment. This is common throughout the west (Bartlett et al. 1989; Gentner and Tanaka 2002; Rowe et al. 2001; Martin and Jeffries 1966; Sulak and Huntsinger 2002; Torell et al. 2005; Smith and Martin 1972). In a three-county California survey more than 90% of ranchers agreed that ranching “makes them feel closer to the earth” and that it is a “good place for family life,” and about a third reported that they belonged to some sort of “environmental organization” (Liffmann et al. 2000). On the other hand, most consider land appreciation an important, long-term financial asset and have planned retirements and estates accordingly. As a result they strongly defend their right to market their land at a good price. California livestock production is not diverse, with the vast majority of ranchers producing cattle only. In 2004, less than a fifth of oak woodland landowners grazed goats, sheep, or llamas, and most of those also grazed cattle (Huntsinger

236

POLICY AND MANAGEMENT

et al. 2007). The majority of ranchers have a herd of cows that produce calves for market each year. After weaning, calves may be retained by the owner if they have adequate feed, or sold to other ranches or feedlots. About 720,000 beef cows grazed California rangelands in 2005, down from a million in 1985, with half a million to a million weaned calves, known as “stockers,” also using rangeland resources, depending on rainfall, markets, and other factors (NASS 2005). Most stockers are produced by ranchers that also have a cow-calf herd, with less than a tenth of oak woodland cattle producers raising only stockers in 2004 (Huntsinger et al. 2007). In 2005 there were 275,000 ewes in California, the mature female sheep of the kind likely to use rangelands, down from 776,000 in 1985 (NASS 2005). Goats are unreported and uncommon overall. California has significant numbers of dairy cattle, but dairies are rarely grassland-based, except in some northern regions. Young animals may use rangelands. In a 2000 study comparing northern and central California ranchers in three counties, management goals were similar, although grazing practices varied among areas (Table 20.2; Liffman et al. 2000). Though not a statewide survey, the study allows comparison of ranchers in different parts of the state. The impact of geographical differences is apparent. In Tehama County the summer climate is drier and hotter, and high-elevation mountain ranges are nearby, so moving stock to upland or irrigated pastures during the summer is common, as is using rotational grazing practices that are often prescribed by the land management agencies that manage higher-elevation grazing lands. In coastal Alameda and Contra Costa counties, livestock can be maintained on grassland year-round, but, as a result, more ranchers manipulate water sources for the benefit of their stock so that water is available on low-elevation rangeland in the summer (Table 20.2). Putting in a trough can be simply to water livestock, or it can be to divert livestock from natural watering holes to reduce impacts to these areas. Throughout the state, there are divergences in practices that reflect particular geographic and economic conditions, but nonetheless the commonalities, particularly in the most important goals, stand out. Ranchers overall rely on sustainable, productive rangelands, and the goals of respondents reflect this need, as they emphasize protecting soils and increasing forage. The most ubiquitous management goals for the surveyed ranchers are to maintain and increase forage and livestock productivity (Table 20.2). However, improving wildlife habitat, protecting scenic values, and reducing the need for herbicides and pesticides are also important goals. A significant proportion use prescribed burning, herbicides, and other methods to clear brush, control invasive species, and make other vegetation changes (Table 20.2). Most respondents reported that they use some form of controlled grazing to manage their lands and leave forage behind after grazing to protect soils from erosion and to create conditions favorable to the germination of valued species (Table 20.2). Many use seeding and fertilizers

TA B L E 20.2 California Ranchers: Goals and Practices in Three California Counties

% Tehama county ranchers (n  132)a

% Alameda & Contra Costa county ranchers (n  113)

P
.1 (X2)

78 95 94 89 78 41 79 99

69 97 90 91 67 37 68 98

nsb ns ns ns .05 ns .05 ns

34 35 32 18 65 53 12 40 23 65

25 36 28 18 73 37 40 43 15 57

ns ns ns ns ns .02 .00 ns .06 ns

44 39 38 24 70 46 40 87 85

24 26 39 64 90 71 30 86 87

.01 .09 ns .00 .00 .00 ns ns ns

The following are important goals for my ranch:

Improving wildlife habitat Increasing forage Improving soil stability Increasing livestock production Protect scenic values Diversify operation Reduce need for pesticides and herbicides Improve livestock quality Have carried out the following practices in last 5 years:

Prescribed or controlled burn Herbicides Chaining or thinning Browsing by goats or others Controlled grazing Deferred or rest rotation Continuous year-round grazing Seeding rangeland Fencing riparian areas Leaving behind ungrazed forage for soil and seedbank benefits Commercial fertilizers Manure application Electric fencing Developed springs Put in water tanks or troughs Laid water pipe Stabilized streambanks Overall satisfied with range condition Used volunteer labor by neighbors NOTE:

Liffmann et al. 2000. varies slightly by question. bnon-significant (ns). an

to enhance productivity, and grazing systems such as rotational systems (Table 20.2), where cattle are moved from one pasture to another on a regular basis. Public acquisition is one way to protect grasslands from development, but it is costly and controversial (Merenlender et al. 2004). Instead, in recent years efforts have been made by the conservation community to offer incentives to private landowners for maintaining the natural characteristics of

their land as a way to conserve open space and wildlife habitat (Barry and Huntsinger 2002; Huntsinger and Hopkinson 1996). The rancher provides on-site management for these “working landscapes,” and the land remains productive agricultural land. The majority of ranchers voluntarily participate in one land conservation incentive program through the California Land Conservation Act (CLCA) of 1965 (also known as the Williamson Act), which allows them

GRAZING MANAGEMENT

237

to pay property taxes at a rate based on the land’s agricultural value as long as they agree to keep their land in agriculture for ten years into the future (for further discussion of such conservation practices, see Chapter 17). In exchange for much needed cash or tax relief, a small but growing number of ranchers have placed conservation easements on their properties. There are many different kinds of conservation easements, but in general they indicate that the part of the title that allows development of the land has been sold or donated to a land trust by the owner (Sulak et al. 2004; Merenlender et al. 2004). This restriction stays with the land even if ownership eventually changes. Selling an easement can provide capital to the rancher for improvements to the operation, paying taxes, or meeting other personal needs. Because ranches often contain valuable natural resources, they are also subject to natural resources regulation. The majority of ranchers in the three-county survey believed that overregulation is a problem for them, and despite the widespread interest of ranchers in maintaining and improving wildlife habitat, more than 80% of them reported that they felt the Endangered Species Act was a threat to ranching (Liffmann et al. 2000). Most also felt that “society” is at least somewhat hostile to ranching (Liffmann et al. 2000) and that the state does not do a very good job of consulting with those affected by regulations and other policy. Overcoming these perceptions is one barrier to more collaborative management between public entities, conservation groups, and private landowners. There is some public funding available to help ranchers meet some regulatory requirements for wildlife, water quality, and other resource benefits, and in some areas county agencies have helped farmers and ranchers access these funds as part of an overall county planning effort for land conservation (EDF 2007). This helps to build trust and partnerships (Hart 1991). In general, private landowners prefer voluntary and incentive-based programs that rely on the high value ranchers themselves place on the natural amenities of their property. The full cost and effectiveness of voluntary, incentive-based programs, as opposed to regulatory efforts, varies with the situation, but impacts on the attitudes and perceptions that underlie both short- and longterm landowner decision making must be considered when regulatory versus incentive-based options are debated. In short, the most widespread goals for the landowners who own most California grasslands are sustainable grassland and livestock production, economic viability, and benefiting individually or as a family from the ranching lifestyle. Few manage for any of these goals exclusively. All goals should be considered when making policy and management decisions.

Public Lands Twelve percent of annual grasslands are in public ownership (CDF-FRAP 2003). A diverse array of public agencies lease public grassland for grazing, including the Bureau of Land

238

POLICY AND MANAGEMENT

Management (BLM), Department of Defense (DoD), U.S. Forest Service (USFS), water districts, and local and regional parks. Many California ranchers use some public lands for part of the year, including local as well as federal lands. About 35% of ranchers in Alameda and Contra Costa Counties used public land in 2000, most of it under local and regional jurisdictions, while about 30% of ranchers in northern California’s Tehama County also used public land, most of it federal (Liffmann et al. 2000). Studies have consistently found strong competition for the available private grassland, with ranchers in the Sierra Nevada foothills stating that competition was intense (Sulak and Huntsinger 2007). In the San Francisco Bay area region, ranchers are active seekers and users of the available public and private leases (Sulak and Huntsinger 2007). Competition for grazing leases has been augmented by the administrative withdrawal of millions of acres of federal lands from grazing, and the continued decline in grazing permit issuance (CDF-FRAP 2003). Declining public forage supply stresses the industry and potentially contributes to grazing pressure on private lands (Sulak and Huntsinger, 2002). On the west side of the Sierra Nevada, montane forage on USFS and private forest land complements the winter forage produced by California annual grasslands. The USFS had 466 grazing leases in California in 2005 (USDA-USFS 2006). Some of these, however, are connected to eastern-Sierran vegetation types rather than Mediterranean grasslands. In a survey of livestock producers using USFS land in the Sierra Nevada, rotational grazing practices were the norm. Interviews with ranchers in the Sierra foothills found them to be active in developing water sources and controlling brush and invasive species in the annual grasslands of their home ranches. On the adjacent USFS lands, activities tended to center on fencing livestock out of riparian zones and wetlands (Sulak and Huntsinger 2002). BLM leases approximately 162,000 hectares of grasslands in the San Joaquin Valley and adjacent interior Coast Range valleys for grazing. There were 340 permits for cattle and 40 for sheep in 2005 on BLM lands in California (BLM 2006). Much BLM land is in the southern deserts, however, making it difficult to estimate what proportion of permits are connected to or on Mediterranean grasslands. Less than 1% of California grasslands are managed by the National Park System (CDF-FRAP 2003). Pt. Reyes National Seashore is one of the rare NPS parks where land is leased for grazing. Grazing, especially now-rare grass-based dairy production, is considered part of the history of the area, as well as a key to preserving agriculture in the surrounding areas. The private ranches using the “pastoral zone” of the park help support the regional infrastructure that supports beef and dairy production, processing, and marketing in the area (Hart 1991). The DoD and local and regional agencies administer about 5% of California grasslands (CDF-FRAP 2003) (Table 20.1). Grassland sites of the DoD include Fort Hunter Liggett (Monterey County, 160,000 acres grassland habitat), Camp

Pendleton (San Diego County, 186,471 acres with 28,000 acres grazed), Vandenberg Air Force Base (Santa Barbara County, 98,400 acres, 23,500 grazed), Beale Air Force Base (Yuba County, 23,000 acres), Fallbrook Naval Weapons Station (San Diego County, 8,850 acres), Concord Naval Weapons Station (Contra Costa County, 12,800 acres, 5,170 acres grassland), Lemoore Naval Air Station (Kings County, 39,173 acres), and a number of other bases and installations. Grazing for endangered species is included among DoD goals; for example, at Fallbrook Naval Weapons Station and Camp Pendleton, the endangered Stephens kangaroo rat does better in grazed habitats (Kelt et al. 2005; USFWS 1997). Other agencies managing grazing include state, regional, and local parks and water and utility districts. Management objectives vary among these many agencies, together managing about 5% of the grasslands. Some use grazing for fuel reduction or to manage annual grasses for habitat improvement for both native plants and animals. For example, the U.S. Fish and Wildlife Service recently reintroduced cattle grazing to the Warm-Springs Seasonal Wetland Unit in Alameda County to enhance habitat for California tiger salamanders, vernal pool tadpole shrimp, Contra Costa goldfields, and western burrowing owl. The agency found that after 10 years without grazing, habitat for these species had become degraded. The Contra Costa Water District, as another example, uses cattle grazing in the Los Vaqueros Watershed to improve habitat for protected and candidate species such as the red-legged frog, tiger salamander, California kit fox, western burrowing owl, and western pond turtle (CCWD 2005). One problem faced by most public agencies is that livestock grazing on public lands is controversial. Complaints about manure, damaged trails, possible harm to plants and wildlife, and even threatening cows are common, particularly in areas used for recreation. On the other hand, homeowners with properties adjacent to parks can be adamant in their support of grazing for fire hazard reduction (Huntsinger et al. 1995). Water quality issues are also controversial, with concerns about pathogens and sedimentation (Derlet and Carlson 2006). Studies suggest that buffer strips can help mitigate these concerns in grassland systems (Tate et al. 2004, 2006).

Land Trusts The two biggest land trusts in California are The Nature Conservancy (TNC) and The California Rangeland Trust. The Nature Conservancy has goals for grazing including vernal pool and native grass restoration, maintaining good relations with rural communities, and an overall interest in conserving biodiversity. Conservation easements are often held on ranching properties, where the landowner relies on income from livestock production. Management goals are a blend of TNC goals and rancher goals. Grazing also takes place on some TNC-acquired properties. Researchers for TNC have found significant benefits to vernal pool plant and animal species

from grazing (Marty 2005; Pyke and Marty 2005). Also in the Ramona Grasslands of southern California, TNC uses grazing to control invasive species and thatch buildup and to improve habitat for the Stephens kangaroo rat (Eastman 2005). The California Rangeland Trust’s mission is to conserve the open space, natural habitat, and stewardship provided by California’s ranches, using conservation easements. There are also many smaller land trusts with varying objectives and resources for management and monitoring of their easements. The Muir Heritage Trust in Contra Costa County, for example, reintroduced cattle grazing to manage a Contra Costa goldfield population when they saw a significant decline in plant numbers following the removal of cattle grazing. At the Bouverie Preserve, part of Audubon Canyon Ranch in Sonoma County, biologists use grazing to maintain diversity and promote native wildflowers (Barry 2005). Again, however, the controversies common to livestock grazing mean that trusts and those who manage conservation properties are faced with the task of convincing donors and the public that grazing programs contribute to their management goals.

Grazing Management Principles To decide what practices to implement, the manager needs reasonable certainty about the outcome of a management action relative to goals. This means knowing as much as possible about how patterns of grazing will affect grassland composition, structure, and function. The more information available about grazing outcomes under various conditions, the more likely it is that management actions using grazing will achieve desired goals. However, uncertainty and risk are ever-present in California’s unpredictable weather. Grazing management relies on information from ecosystem sciences as well as traditional knowledge and personal experience. Monitoring after actions are implemented increases the information available for making further decisions, especially if similar “control” or untreated areas are established for comparison. Clearly defined management practices, along with monitoring of outcomes and the specific conditions under which particular outcomes were achieved, contribute to the reliability of future predictions of management effects. Controlled experimentation is one of the best ways to improve the information available for management decision making. The resources necessary for establishing large-scale grazing experiments are, however, frequently outside of the purview of managers, and as a result few controlled grazing studies exist from which to quantitatively evaluate management scenarios. By default, on ranch properties today management is largely based on traditional knowledge and experience, including knowledge of the grazing history of the area and the results of past management efforts. Such knowledge has the advantage of being site specific, based on long-term observation, and time-tested. However, observations vary by the perspective of the individual observer, and what looks

GRAZING MANAGEMENT

239

like “good conditions” to one person may look degraded to another. In addition, ecosystems are changing, so what worked in the past may not work in the present or future. Further, laws and regulations for grasslands have changed. Hence, ranchers and managers may have to incorporate new and diverse forms of knowledge into their management. Scientific knowledge attempts to get beyond individual perspectives using quantitative analyses. It allows testing the use of various practices in a way that reduces the chance that results stem from a random effect rather than being actually due to the implementation of a practice or treatment. Drawbacks are that there is seldom the completeness of information desirable for any given site. A risk is also that scientific results, like traditional knowledge, may be overgeneralized by managers who think that results from one site apply more broadly. Experimentation must always be done within clearly specified site conditions. Adaptive management is an approach to management that acknowledges uncertainty and the need to learn about the resource to be managed (Walters 1986, 1993; Walters and Holling 1990). The term adaptive refers to managers learning about systems as they attempt to manage them. This learning is accomplished ideally by structuring management activities as experiments to test hypotheses of system functioning. However, more realistically, it means monitoring the outcomes of management with some “controls,” or unmanaged areas, for comparison, so that the results of management can be distinguished from the results of weather and other forces outside of management control. The results of monitoring are then used to adjust management practices and, as necessary, goals. Eventually, this enables better predictions of management outcomes and increases the likelihood of meeting management targets. Another benefit of adaptive management is that any form of knowledge can be incorporated. The knowledge a rancher or manager has gained over time through experience and observation can be used to establish management practices at the outset. With controls and monitoring, this knowledge can be refined and improved. Adaptive management offers a structured way to evaluate traditional knowledge and its applicability to current goals. Much ecological information that contributes to developing management practices and scenarios is presented in Chapter 17. The highly variable precipitation, soils, temperatures, topography, species pools, and land use histories across California mean that the outcome of management practices will be variable over both space and time. The applicability of scientific studies conducted at just one or two sites to diverse sites is often not well known.

Grazing Prescriptions Livestock can be placed on particular rangelands at particular times in specified numbers for a specified period. Though constrained by what available graziers can provide or accommodate, the manager may also have the opportunity to select

240

POLICY AND MANAGEMENT

animals of a particular species or breed, experience, and specific age class and physiological state: all these choices affect the ultimate impact of the animals on grasslands and natural resources. Prescriptions are put together using what is known about the ecosystem to be grazed and the animal that is doing the grazing. In theory, because plant species differ in phenology (e.g., early- versus late-season annuals or annuals vs. perennials), the timing of grazing could differentially suppress or promote species by mitigating competitive interactions or reducing fecundity of grazed species (Augustine and McNaughton 1998). The frequency and intensity of grazing also influences the rate of live biomass accumulation, affecting the development of plant crowding that might suppress some plants (McNaughton 1968; Noy-Meir et al. 1989). The proportion of plant biomass removed increases with the density of livestock and the number of times plants are grazed. The proportion of biomass removed affects the amount of plant litter at the soil surface, in turn influencing patterns of germination and seedling establishment (Heady 1956; Facelli and Pickett 1991a), a particularly important consideration in annual grasslands, where the grassland renews every year from seed. Grazing livestock also modify physical and chemical properties of soils, with important implications for nutrient cycling, hydrology, and plant composition (Weaver and Rowland 1952; Hobbs 1996; Jones 2000). Grazing can influence species composition directly through herbivore selectivity (i.e., the plants chosen for consumption) and indirectly through the variation among species in tolerance to being grazed (Belsky 1986; Augustine and McNaughton 1998; Kimball and Schiffman 2003). Selectivity is most strongly expressed when forage within the available space is abundant relative to the number of livestock. As forage supplies become more limited, animals consume a broader variety of species. Plant selection, as well as where animals concentrate their activity, is also strongly affected by animal characteristics such as species, breed, and phenological state and by other causes such as topography and water supplies. Using fencing, herding, and domestic livestock, then, the following four grazing factors are widely considered to be important in determining grazing patterns, and can be manipulated by the manager (adapted from Heitschmidt and Stuth 1991): 1. The kinds and classes of animals: Are they young, old, pregnant; what species or breed? 2. The spatial distribution of the grazing animals: Are they fenced and crowded in a small pasture or herd, or will they distribute themselves easily over a large pasture or site? 3. The temporal distribution of grazing: When and how long? 4. The density of animals: How many grazing animals per unit area?

These factors, moderated by the environmental characteristics of a site, determine the intensity, timing, and frequency of the grazing of plants in a grazing area. Specific examples of their application to common management goals, including enhancing native species and managing invasives, are found later in this chapter. For annual grasslands, when managing for resource sustainability, a key management approach is to use these four factors to influence the following year’s germination. By managing grazing to leave particular amounts and patterns of ungrazed plant matter behind at the conclusion of an annual grazing cycle, the manager has the best opportunity, within the confines of weather conditions and other abiotic factors, to influence the next year’s germination and to protect the soil. This is because of the major environmental characteristics that largely control forage productivity and species composition in California annual grasslands, including rainfall, temperature, soil characteristics, and the amount of plant residue that remains ungrazed; ungrazed plant residue is most within the control of the grazing manager. Research has shown that this residue, or residual dry matter (RDM), can protect the soil from erosive forces (Bartolome et al. 2002; Tate et al. 2006) and create seed bank conditions that, depending on the depth and characteristics of the RDM layer, influence the likelihood of germination of different grassland species (Heady 1956; Bartolome 1979). It can also influence soil characteristics by returning organic matter to the soil, and can be an indicator of impacts to soil bulk density (Tate et al. 2004). Residual dry matter also has the advantage of being easy to measure. It can be monitored at specific spots to evaluate levels of use and to decide when livestock should be moved, and it can be mapped to allow for landscape-scale evaluation of animal grazing patterns. Mapping makes it clear where fencing, herding, or distribution of feed and salt should be used to influence grazing patterns. Though of course not accepted by everyone, using RDM for monitoring has become the most widely recommended approach to California annual grassland grazing management. The amount of RDM recommended for protecting soils and forage quality varies with rainfall, slope, soil characteristics, and other factors (Bartolome et al. 1980). Prescriptions are unique to the ecosystem and the goals of the manager. A grazing prescription for California annual grasslands should specify each grazing factor for a particular setting, and in most cases should specify an acceptable range of RDM at the end of grazing. For example, “on this site of 300 acres, 25 mature female cattle will be grazed for three months from January to March, leaving 700 –1,500 lbs per acre of RDM.” The prescription should also include what will be monitored to determine whether the grazing is meeting management goals. In this example, that might be measuring or mapping RDM each year. If there are other goals for the prescription, such as enhancing the proportion of a particular species or reducing an undesirable species, monitoring should be established that will allow the manager to determine whether or not results are being achieved.

For example, in grasslands where a goal might be grazing to increase grassland structural diversity, continuous, yearlong grazing may be appropriate. In another situation, rotational grazing or high-intensity grazing may be used to achieve goals that can include certain levels of RDM, specific timing for grazing, or reduction of targeted undesirable species. Vegetation characteristics themselves influence the potential of a site for grazing use. Some plant species are toxic to some or all herbivores. Some are more nutritious or attractive to grazers at particular times of the year. Others have spines, thick wood, or other features that make them inedible or less preferred during all of part of their life stages. The tendency of a plant to be consumed will also vary by what else grows in its locality: Once the most preferred species are consumed, grazing animals, wild or domestic, will select the next most preferred species. This selectivity must be considered by the manager in predicting grazing outcomes. In the past, grazing prescriptions often had the goal of creating an even utilization of species. Recently, as grazing has begun to be used to mimic “natural disturbance” or to increase species diversity (Noy-Meir et al. 1989; Collins et al. 1998; Perevolotsky and Seligman 1998; Harrison 1999b; Maestas et al. 2003; Marty 2005) and the limited role of competitive interactions in disequilibrium systems such as the California annual grasslands is more understood, using prescriptions that lead to uneven or patchy use, reflected in an uneven distribution of RDM, may be considered useful for various goals (Fuhlendorf and Engle 2001). In addition, in some cases grazing may be used to suppress undesirable vegetation, and the manager may want to achieve different patterns and intensities of grazing than would be typical for the goal of sustainable livestock production alone (Huntsinger 1996).

Grazing Animals Cattle, sheep, goats, and horses have different diet preferences and grazing behavior patterns, and these also vary by breed and even by individual. There is also evidence that experience and “learning” from parents and the herd can influence animal forage choice (Provenza and Balph 1988). Although most livestock have considerable dietary flexibility, depending on the goals of the manager, some animals are more likely to meet particular needs than others. Diet preferences of herbivores represent the priorities of the animal in selecting plants to eat (Table 20.3). So goats, known as the archetypical brush eaters, will consume grass effectively if there is no edible brush available, for dietary variety, or if experience and training have caused them to favor grass. Within the broad generalities of species, various breeds of cattle, sheep, and goats may exhibit differences in behavior and forage preference. Cattle of Asian origin, such as Brahma or zebu cattle, are considered to be more likely to cover greater distances, use less water, and make better use of hilly, hot country than cattle of more common English breeds,

GRAZING MANAGEMENT

241

TA B L E 20.3 Dietary Preferences and Grazing Patterns of Common Domestic Herbivores

Dietary preferences

Grazing patterns

Notes

Beef cattle-brood herd

Grass, forbs

Like lowlands, water

Suitable for open range, develop regular patterns at the landscape scale that is part of herd memory

Beef or dairy cattle, young animals

Grass, forbs

Like lowlands, water

Need to be fenced in, more sensitive to dogs and predators than brood herd

Dairy cattle-mature Grass, forbs (rare on rangelands)

Like lowlands, water; must stay near milking facilities

Not able to spread out or wander as widely as beef cattle; returning to the barn twice a day for milking means that trails can be an erosion concern

Sheep

Forbs, grass (less fiber than goats) Will repeatedly use same bed-grounds, which can lead to overuse of some spots if they are not moved

Can be herded for tighter control; vulnerable to predators and dogs

Goats

Brush, forbs, grass

Like uplands, avoid water

Have extra large liver that allows them to cope with toxic plants; vulnerable to predators and dogs

Horses

Grass, forbs (high-fiber)

Like uplands

Nonruminants, upper teeth allow them to consume very fibrous forage in large amounts

such as the almost ubiquitous Hereford and Angus, which are likely to congregate near water. Though relatively rare in California, grazing two or more species together can result in more complete utilization of diverse vegetation. For example, grazing cattle and goats together will result in both grasses and shrubs being grazed. Because their species of concentration are different, in some vegetation types — for example, in mixed shrub and grass areas—goats can be added to a range grazed by cattle with no need to reduce the number of cattle. In California there are now numerous companies offering “prescribed grazing” to landowners for the purpose of controlling invasive plants, reducing fire hazard, and creating a more aesthetic environment. Often these firms use goats, because their propensity to eat brush aids in fire control and complements the foraging of cattle.

Exclusion Livestock can be excluded from landscape features, from pastures, and from grazing particular plants or small areas through fencing. Likewise, parks and reserves may completely exclude livestock for several years because of public pressure. A common management practice of federal agencies is to fence streams to exclude livestock. This may result in increased vegetation cover, and the development of trees and shrubs, within the protected area. Similar results were obtained by fencing streams in northern California to exclude deer (Opperman and Merenlender 2000).

242

POLICY AND MANAGEMENT

When an area is fenced to preclude grazing in an otherwise grazed area, it is called a “livestock exclosure.” Interpretations of grazing effects from enclosures can be difficult (McCreary 2001). In California annual grasslands, year-round exclusion has been shown to reduce diversity of herbaceous native and exotic plant species, in some cases to the detriment of threatened species that depend on nongrass species (Weiss 1999; Hayes and Holl 2003a; Kelt et al. 2005; Marty 2005; Pyke and Marty 2005). Virtually all of the California grassland has been grazed by livestock at some point; hence there is no “pristine” grassland to serve as a baseline for comparison with grazed sites (Fleischner 1994). Exclosure studies compare the community composition of actively grazed plots to that of plots that had been previously grazed but subsequently protected from livestock for varying numbers of years. In California, many shortcomings limit generalization from existing exclosure studies. First, most studies are unreplicated or do not have appropriate controls. Second, studies measure grazing intensity in inconsistent ways (or not at all), precluding standardization across studies. Third, knowledge of the land use history is imperative to the interpretation of results of exclosure studies, yet it is often not reported. Additional considerations include small spatial scales (Bartolome 1989), temporal issues (initial responses of species to protection from grazing may not represent the long-term response), and issues related to propagule availability for colonization of the protected area. Despite their shortcomings, exclosure studies are one of the few practical

approaches to evaluating long-term grazing impacts or the outcome of releasing sites from grazing (Bock et al. 1993; Fleischner 1994). A common conclusion of exclosure studies is that native plants do not typically become dominant after protection from livestock grazing. One hypothesis to account for this finding is that livestock grazing explains less of the variation in plant distribution than site-specific factors such as land use history, climate, and soils. Stromberg and Griffin (1996) note that many native grassland plants were absent from previously cultivated sites, independent of livestock grazing, and that land cultivation, elevation, soil texture, and aspect all explained patterns of community composition more effectively than grazing (Stromberg and Griffin 1996). Similarly, Harrison (1999b) found that soil type and aspect better accounted for patterns in plant species richness than did grazing. This does not mean that grazing has no impact (see Harrison and Viers, Chapter 12; Jackson and Bartolome Chapter 17), but rather that site-specific factors are extremely important in controlling species composition. The importance of site-specific factors is consistent with the variation in the response of native plants to protection from grazing. This variability underscores the need for studies that employ sufficient replication to permit rigorous analysis and occur over a spatial scale that encompasses environmental gradients.

Grazing Systems Grazing systems are a set of specifications for grazing based on the grazing prescription factors listed in the preceding section. Common systems include “rotational systems” that move grazing animals from one area to another over the year. Some systems also involve “rest,” in which an area is not grazed every year but “rested” one year out of every few. The idea is that plants will have a chance to recover from grazing impacts during the rest period. “Deferment” means that an area is grazed later in the season, after the grasses have set seed, with the idea that this will cause less stress to particular plants. For managing annual grasses as a grazing resource, in terms of their physiology, there is little utility in rest, rotation, or deferment because grasses sprout from the soil in response to rain and the cycle begins anew each year. Research-based evidence has not shown a consistent response to rotational grazing in annual grasslands and is not sufficient to broadly recommend this practice (Bartolome et al. 2002). However, it remains a tool available if rotational patterns of grazing are known to be more beneficial for a particular species or setting. Having more than one pasture also allows animals to be herded in ways that may cause them to consume plants they would not ordinarily consume, to avoid grazing some plants at certain times of year, to keep from overusing an attractive area, to control weeds, or to benefit native perennials. “Holistic Resource Management” is a management system for livestock producers that emphasizes goal setting and mimicking “natural” grazing systems by manipulating herd

grazing patterns, typically using a rotation-based system. Scientific evidence of the benefits of many of the grazing practices typically espoused by this system is lacking for California grasslands, but many ranchers and managers find the goal setting exercises helpful. Because today the major forage species on California’s Mediterranean grasslands are annual, this grassland has been relatively easy to manage solely for livestock production for a long time, although the recent spread of certain undesirable species challenges that management. Also, interest in using grazing for native species restoration and other goals has made it worthwhile to delve into more complex management practices.

Grazing Plans A grazing plan may include grazing prescriptions, specified grazing systems, and a year-round plan for the livestock herd or, from the agency perspective, for the grazing area. It should describe the facilities available, the productive capacity of the land, sources of water, minimizing conflict with other uses, and so forth. It may also describe how management could respond to the many vagaries of grazing California grasslands, including unpredictable drought, fire, noxious weed invasion, and so forth. In the grasslands, plan flexibility needs to somehow accommodate annual variations in weather that can cause magnitudes of difference in forage production.

Grazier Constraints and Needs The needs and constraints of livestock producers influence what kinds of prescriptions, systems, and plans can be implemented on a given site, whether on public or private land. These factors also influence the availability and flexibility of contract grazing. In general, the producer wants security and predictability, e the manager needs to meet specified targets and to be able to adapt quickly to changes in rainfall and other factors. As one management document for a water district states, “the livestock owner/operator must be flexible enough to increase or decrease his herd to meet land management objectives within a very short window of opportunity following notification by the agency” (Nuzum 2005). Livestock producers, for their part, need to maintain their herd year-round or for a certain number of months somehow, and to protect their income they need to be able to control the costs of forage sources. When their animals are taken off grassland, they must resort to feeding with expensive stored forages such as hay or agricultural by products unless irrigated croplands are available. Most livestock producers in California are cow-calf producers, and must maintain a herd of brood cows through the year. They cannot easily alter the numbers of animals in their herds. It is difficult for them to cope with lots of variation in forage availability and unexpected events: to have a successful grazing program, it is useful to have alternatives available

GRAZING MANAGEMENT

243

in case there is an unexpected need to reduce or remove livestock from an area for a period of time. Grass banks are one way to provide some insulation for the unexpected (NGN 2005). Grass banks are rangeland areas that are set aside to provide forage during drought or when improvements or rest is desirable in their usual grazing areas. For example, southern Arizona’s “MalpaiBorderlands Group” provides access to grass on other ranches for local ranchers who agree to place a conservation easement preventing subdivision on their own land. Ranchers use the grass bank during drought or when prescribed burning is being carried out. This is considered a way to protect the traditional ranching community of the area (MBG 2006). The impacts of sporadic grazing on a grass bank flora would need to be assessed for California sites. Producers need grazing areas with certain characteristics during breeding, calving, kidding, or lambing periods and may not be able to match this with lessee needs. Markets, fuel costs, family needs and goals, and a host of other factors may also affect what a livestock producer is able to do in a given year. Producers must also be concerned with the safety of their animals and possible conflicts with other users, transportation costs, water supplies, fences, and work that they may need to put into improvements and maintenance. An increasingly common problem for public agencies is providing and maintaining the fences, troughs, and other infrastructure that livestock producers need, especially when funds are available for acquisition but not for maintenance. Producers often make the improvements in exchange for rent reductions. In general, cow-calf and lamb producers have the greatest need for stability, and have the least flexibility in their operations. Yet their stability also means that they are likely to form a long-term connection with the leaser and land. Contract goat grazers face the continuous challenge of finding projects that fill out their year-round needs, and they must transport animals from site to site. Mobile contract goat grazers also usually need herders on site to protect the goats from dogs and predators and keep them confined; goats are notoriously difficult to fence securely, though technological advances in electric fencing have improved the situation and played a large role in enabling the entire nascent industry.

Example Management Approaches Grazing prescriptions become part of a grazing plan that includes needs for facilities such as fencing and water, potential conflicts with wildlife and recreation and how to cope with them, economic considerations, and so on, depending on whether the area to be grazed is public or private, part of a livestock enterprise, or being leased for grazing. Grazing for modification of vegetation is usually most effective when done as part of an integrated management program that considers use of prescribed fire, chemical, mechanical, and hand treatments as options or components of a management plan. An integrated management plan that follows an adaptive

244

POLICY AND MANAGEMENT

management framework, even if truly experimental treatments are not feasible, is one approach to managing grazing. Here we review what is known about the outcome of grazing practices, including grazing to control yellow starthistle (Centaurea solstitialis, hereafter YST), to reduce brush and control fire hazard, to manage vernal pools, and to restore native grasses.

Grazing Management to Control YST and Noxious Weeds1 Grazing can be used to manage noxious weeds in many rangeland settings (DiTomaso 2000, 2006). Intensive grazing will counteract inherent dietary preferences of livestock, resulting in more equal impacts on all forage species, including weeds, but it can potentially cause damage to native species and soils. Moderate grazing intensity can minimize the impact on native plants and soil disturbance. Multispecies grazing distributes the impact more uniformly among desirable and undesirable plant species; whether or not this is desired should be decided in advance by the manager (Olson 1999; Walker 1994). In the case of YST, a widespread noxious weed in California, grazing alone will not provide long-term eradication, but it can be a valuable tool in an integrated management program (see DiTomaso et al., Chapter 22). The ideal time to graze for noxious species control is when individuals of the target plant are most susceptible to defoliation, or when the impact on the desirable vegetation is minimal. Improperly timed grazing can benefit the wrong species. For example, livestock grazing in late winter or early spring will primarily feed on young grasses with an erect growth form, causing little damage to seedling YST rosettes. This increases the amount of light reaching the rosettes and stimulates their growth. By contrast, intensive grazing in late May and June, when starthistle plants are bolting, can reduce YST growth, height, survivability, canopy size, and seed production (Thomsen et al. 1989, 1990). During this stage, YST can provide palatable, high-protein forage, which can be particularly useful when other annual species have senesced. The plant’s crude protein concentration ranges from 28% at the rosette stage down to 11% at the bud stage and should be sufficient to meet the general maintenance requirements for most grazers. When it is abundant, YST appears to have the ability to sustain animals several weeks beyond annual grass “dry-down.” But in the mid to late summer months, livestock will selectively avoid the now spiny mature plants, allowing ample seed production. If it is not grazed in the bolting stage prior to spine production, YST will not be successfully controlled by livestock grazing. In contrast to cattle and sheep, goats continue to browse YST plants even in the flowering stage (Thomsen et al. 1993). For this reason, goats have become a popular method for

1. Much of this discussion is developed from the Web site “Yellow Starthistle Information” maintained at the University of California (DiTomaso 2006).

controlling small infestations. Goats preferentially consume seeding stems, reducing the spread and perpetuation of weeds by seed. Even though mature seeds sometimes survive passage through the digestive tract, goats usually consume the seeds in an immature stage, so they would not be expected to survive. Allan and Holst (1996) observed that goats reduced the seed bank of thistles. Residual dry matter, or thatch, may influence starthistle success, as shading has been shown to reduce YST seedling survival rates (Roché et al. 1994; Gerlach and Rice 2003). In a report by Weber (1985), it was noted that when Roché delayed spring grazing of wheatgrass, starthistle was controlled because ungrazed, taller wheatgrass plants blocked sunlight from the rosettes of starthistle. Since most defoliated YST plants will recover from one-time grazing, it is necessary to bring the animals back up to four times at about two-week intervals under rotational grazing. Alternately, grazers can be left on site for two to three months under a continuous grazing regimen (Thomsen et al. 1993). But adequate residual dry matter should be left to shade YST seedlings in the following year (DiTomaso et al. 2004). Also, excessive trampling by livestock has also been shown to increase the density of YST (Miller et al. 1998), so using grazing to control YST must be done with caution. Short periods of intensive grazing can result in more uniform and complete utilization of a pasture, including weeds, and minimize the animal’s ability to avoid less palatable species. In such a system pastures are intensively grazed from 3 to 5 days, often with the use of electric fencing. After animals are moved, the grazed area is allowed to recover for at least a month before grazing is repeated. This may or may not be desirable, depending on the ecology of the particular weed. When grazing animals refuse to consume a weed, that behavior can encourage the weed’s growth by eliminating the competition, leading to more rapid infestation. On the other hand, grazing may endanger sensitive nontarget species if they are strongly preferred. Goats are typically browsers and can effectively control certain noxious species. However, when confined they will eat both desirable and undesirable species and may even strip the bark off trees. Livestock can also trample desirable, sensitive species and can spread noxious weeds over a wide range when seeds become attached to hair (DiTomaso 1997).

Grazing for Restoration of Native Grasses and Forbs Despite the position of some environmentalists that livestock grazing is incompatible with native biodiversity preservation, managed grazing practices have been endorsed as a tool for promoting biodiversity in native grassland remnants and for restoration projects in California (e.g., Menke 1982; Edwards 1995, 1996; Reeves and Morris 2001; Hayes and Holl 2003a; Stromberg et al., Chapter 21). This has generated some controversy. Reconciling these views in the context of the California grassland is complicated by the difficulty of separating site, weather, and historic effects from grazing impacts.

Across sites, the interaction between climate and grazing in relation to native plants is likely important but has not been rigorously examined (but see Dyer et al. 1996; Langstroth 1991; Jackson and Bartolome 2002). Also, although studies of grazing in California grassland have gone on for some time (see Table 20.4 for a review), most studies suffer from design flaws, so results cannot always be unambiguously assigned to grazing treatments or cannot be generalized across sites. Systematic investigations of native plant palatability and species responses to grazing have not been carried out since the early studies of Gordon and Sampson (1939), but scattered anecdotal and experimental evidence is available suggesting variability in tolerance to grazing within and among species. For example, based on observations of moist coastal grassland at Sea Ranch (Sonoma County), Elymus glaucus and Calamagrostis nutkaensis (reed grass) are thought to be unpalatable to livestock, resulting in their persistence on some grazed sites (Dwire 1984). A somewhat better studied species is Danthonia californica, which has been observed to be both palatable and tolerant of moderate levels of grazing in central and northern coastal California (Cooper 1960; Hatch et al. 1999; Hayes and Holl 2003a). Several species of native forbs (e.g., Iris spp., Orthocarpus spp., Ranunculus californica, Limnanthes spp., Orcuttia spp., Limnanthes floccosa) may increase under light to moderate levels of grazing (Edwards 1995; Barry 1998; Hayes and Holl 2003a), possibly as a result of the suppression of co-occurring exotic annual grasses. A study of valley grassland species that employed clipping of plants in the greenhouse and the field found that native and alien species had contrasting responses to biomass removal (Kimball and Schiffman 2003). After clipping, density of European grasses was unaffected, and the alien forb Erodium cicutarium increased, whereas native grasses such as Poa secunda decreased in density and growth, presumably because of being intolerant of biomass losses. It has been suggested that some grazing regimes may enhance native biodiversity more than others (WallisDeVries et al. 1998). Many manipulations of grazing intensity have been performed in exotic annual grassland sites that had only negligible native plant cover (e.g., Pitt and Heady 1979; Rosiere 1987; Jackson and Bartolome 2002). In these settings species composition appears to be largely unaffected by livestock grazing (Heady 1977; Jackson and Bartolome 2002). In grasslands composed of mixed exotic annual and native annual species such as those that often occur on serpentine soils (see Chapter 12), or in vernal pool grasslands in the Central Valley, grazing is being used as a tool to promote native annual wildflowers (Weiss 1999; Weiss, personal communication; Marty 2005). In the case of serpentine grasslands, wet-season grazing is the preferred management method for reducing the annual grasses that are increasing in response to anthropogenic N deposition (Weiss 1999). Where grazing is not feasible, mowing is being successfully used (Weiss, personal communication). In mixed annual and perennial grasslands, the outcome of grazing is more variable. Bartolome et al. (1980) examined

GRAZING MANAGEMENT

245

Observational/ release

Grazing manipulation/ release

Grazing manipulation/ release

Harrison 1999b

Hatch et al. 1999

Hatch et al. 1991

Yuba?

San Mateo

Napa and Lake

Sonoma

Observational/ release

Foin and Hektner 1986

Humboldt

Marin

Observational

Cooper 1960

Contra Costa

Observational/ release

Grazing manipulation/ release

Bartolome et al. 2004

Mendocino

Elliott and Wehausen 1974

Observational/ release

Bartolome and Gemmill 1981

Several

Solano

Mulch manipulation

Bartolome et al. 1980

County

Dyer et al. 1996d Grazing manipulation

Study type

Citation

Paddock

Paddock

Pasture

Pasture

Pasture

Paddock

Pasture

Paddock

Pasture

Plot

Scale of treatment applicationa

Stocking rate

Stocking rate

Presence/ absence

Stocking rate (qualitative)

Stocking rate

RDM/stocking rate

Stocking rate?

RDM

Qualitative

RDM

Measure of grazing intensity b

Cattle (nd)

Cattle (15) and sheep (200)

Cattle (nd)

Sheep (nd)

Cattle (1.16–2.63 ha per cow)

Sheep (15)

Cattle (ca. 1,000?)

Cattle

Sheep (high)

Simulated

Grazing agent (density)

Continuous

Continuous

nd

Continuous

nd

Spring/wet vs. summer/dry (2–3)

Winter, summer, or spring and fall (nd)

Spring vs. summer vs. continuous vs. release

nd

Late summer/early fall

Season (duration in days)

1–2

1–2

1–2

1, 3, 5

6 after release

1–4

1

1–6

2, 3, and 23

5

Sample yearsc

TA B L E 20.4 Summary of Studies of the Effects of Livestock Grazing on Native California Grassland Plants

All sites historically grazed

Observations 7 to 13 yr after release from grazing

Ungrazed  6 yr release from grazing

Grazed until 400–500 kg/ha

Crossed with burning after 3 yr of grazing treatments, grazing treatments ceased after 3–4 yr

Comments

Observational/ release, clipping and mulch manipulation

Grazing manipulation

Grazing manipulation

Grazing manipulation

Kimball and Schiffman 2003

Langstroth 1991

Love 1944

Marty 2005

Mendocino

Merenlender et al. 2001

Observational/ release

Yuba

Marty et al. 2005 Grazing manipulation

Sacramento

Sacramento

Solano

San Luis Obispo

San Mateo

Several

Grazing manipulation

Mulch manipulation

Jackson and Bartolome 2002

Contra Costa

Kephart 2001

Grazing manipulation

Jackson, unpublished

Mendocino

Tulare

Biomass/mulch manipulation

Heady 1956

Several (Mendocino south to San Luis Obispo)

Keeley et al. 2003 Observational/ release

Observational/ fenceline

Hayes and Holl 2003a

Pasture

Paddock

Vernal pools

Pasture

Paddock

Pasture and plots

Paddock

Pasture

Plot

Paddock

Plot

Pasture and plot

Goats (500)-yr 1; sheep (40) and goats (7)-yr 2

Cattle and horses

Simulated

Cattle, light to moderate

Simulated

Qualitative

RDM and stocking rate

1 AU/2.4 ha

Stocking rate

RDM/stocking rate

Sheep

Cattle

Cattle

Sheep (264–275 early; 83 late)

Sheep (15)

Presence/absence Cattle and Clipping intensity simulated RDM removal

Stocking rate

Presence/ absence

RDM

RDM

RDM

Presence/absence Cattle

nd

Light vs, heavy winter/spring, vs. continuous vs. release

Dry vs. wet vs continuous vs. release

Spring and summer or fall (20–30)

Spring/wet vs. summer/dry (2–3)

Winter/spring

May (12)

Winter-summer

Late summer/ early fall

Spring (7), summer (7), continuous

Fall

nd

43 years

3

3–4

Same

1–4

2, 3

1–2

1

6

1–3

1–2

(Continued)

Followed permanent plots over 43 years after removal

Treatments crossed with burning

Sampled zones from upland to pool bottoms

Grasses seeded into disked pasture

Grazed until 400–500 kg/ha

Grazing release site unreplicated 11 yr since grazing.

Grazed to bare mineral soil

Long-term grazing

Clipped to fixed height and varying percentages returned

Compared 17–25 grazed/ungrazed pairs in coastal prairie

Observational/ release

Grazing manipulation (no control)

Observational

Observational/ release

Observational/ release

Grazing manipulation

Grazing manipulation

Observational/ release

Micallef 1998

Reeves and Morris 2000

Saenz and Sawyer 1986

Safford and Harrison 2001

Stromberg and Griffin 1996

Thomsen et al. 1993

TNC 2000

White 1967

Monterey

Tehama

Colusa/Yolo

Monterey

Napa and Lake

Humboldt

San Benito

Contra Costa

County

Pasture

Paddock and pasture

Paddock

Pasture

Pasture

Pasture

Pasture

Pasture

Presence/ absence

RDM

RDM

RDM (qualitative)

Presence/ absence

Qualitative

Stocking rate

RDM/stocking rate

Measure of grazing intensityb

Cattle, horses

Cattle

Cattle (18 cow-calf pairs); sheep then goats (20–40)

Cattle (nd)

Cattle

Cattle (nd)

Cattle (nd)

Cattle (nd)

Grazing agent (density)

aplot

NOTE:

1–2

1, 20, 24

Sample yearsc

nd

Rotated November–April

May and 2–3 follow up later in the season (
3 days)

Seasonal vs. continuous (nd)

nd

27

3

3

23

1–2

Early–late (8 mo) 1 vs. late (4 mo)

nd

nd

Season (duration in days)

nd  data not available;  experimental unit gen
10 sq. meters; paddock  experimental unit gen
0.5 acres; pasture  existing unit, generally 0.5 acres; bRDM  residual dry matter at end of treatment application, grazed or clipped until a given RDM is achieved cYear data collected after initiation of treatment (or release from grazing) dIncludes Fossum (1991) data and is same experiment as Langstroth (1991).

Study type

Citation

Scale of treatment applicationa

TA B L E 20.4 ( C O N T I N U E D ) Summary of Studies of the Effects of Livestock Grazing on Native California Grassland Plants

Ungrazed for 27 years prior to study except horse pasture

Cattle reintroduced after 11 years of release from grazing

Ungrazed for 51 years prior to study except horse pasture

No ungrazed control

Monitored increased stocking rate

Comments

the response of native perennial grasses to various levels of RDM manipulation in three Humboldt County coastal sites. Native grass growth was most enhanced by the highest level of RDM employed (1,120 kg/ha), suggesting that a low grazing intensity may be most appropriate for increasing cover of native bunchgrasses. By contrast, Savelle (1977) found that removing mulch from Nassella pulchra tussocks increased seed production, and other experiments suggest that mulch can inhibit N. pulchra seedling establishment (Dyer et al. 1996; Reynolds et al. 2001). In a three-year study testing burning and four different grazing intensities including rotational and continuous grazing, Marty et al. (2005) found no difference due to grazing treatment in N. pulchra mortality, seedling density, or culm production. In general, grazing reduced the height and reproduction of the bunchgrasses, but this did not affect subsequent seedling density, which was driven by weather. Across all plots, including controls, there was a fivefold decrease in bunchgrass density, probably due to declines in rainfall over the study period. The seasonality of grazing has been hypothesized to be a critical determinant of native grass enhancement. “Properly timed” application of livestock grazing theoretically enhances the establishment and growth of native grasses by suppressing competitive annual grasses. Experimental support for this hypothesis has been mixed. In one study, early spring grazing suppressed the faster-germinating exotic annual grasses, thereby reducing the competitive suppression of perennial bunchgrasses or native forbs, whose seeds germinated later than the grasses (Love 1944; Langstroth 1991; Dyer et al. 1996). Seemingly small differences in timing can have big effects. For example, an early study (Love 1944) in valley grassland in Sacramento County compared the establishment of N. pulchra and N. cernua in plots that were grazed by sheep in early April versus late April. Love (1944) suggested that the greater success of the native grasses in the earlygrazed field was due to the removal of the taller annual grasses, resulting in the release of native perennials from competitive suppression. The late April grazing treatment occurred when the wet season ended and the perennial grasses had not developed sufficient root biomass to survive the dry conditions that followed, and they were more likely to be grazed because the annuals were senescent (Love 1944). More recently, a series of related experiments on N. pulchra distribution compared the effects of grazing in the wet and dry seasons at the Jepson Prairie Preserve (Solano County; Fossum 1990; Langstroth 1991; Dyer et al. 1996). Seedling emergence and survival were higher in the wet-season grazing treatment (especially in combination with burning) compared to the dry-season grazing treatment and an ungrazed control (Fossum 1990; Dyer et al. 1996). The putative mechanism behind this result was that litter and annual grass removal increased light levels at the soil surface and enhanced Nassella seedling survival (Langstroth 1991; Dyer et al. 1996). However, neither grazing nor burning had a significant effect on survival when the experimental treatments were repeated a year later during a very dry year,

suggesting that the long-term effects of the treatments was ameliorated by climatic conditions (Dyer et al. 1996). Furthermore, only a tiny fraction of the seedlings that emerged survived after four growing seasons (99.9% mortality), possibly as a result of below-average rainfall (Dyer et al. 1996). For established plants, wet-season grazing decreased the basal diameter of individuals but enhanced reproduction via plant fragmentation, leading to an increased density of N. pulchra plants relative to ungrazed plots. Though wet-season grazing improved the establishment and vegetative reproduction of N. pulchra, plant microsite (i.e., on top of or between mima mounds) exerted stronger effects on plant responses than the grazing treatments (Dyer et al. 1996; Langstroth 1991). Bartolome et al. (2004) found an increase in N. pulchra with spring (wet-season) grazing compared to summer and continuous grazing over several years, but in their study N. pulchra also responded positively to complete release from grazing. Other studies have employed specialized grazing regimes with the goal of increasing the native, perennial component in annual-dominated grasslands, although most studies suffer from a lack of control plots (Table 20.4). For example, Reeves and Morris (2000) found that applying a grazing regime based on holistic management principles — basically a short-duration, high-intensity rotational grazing system (Savory and Butterfield 1999)—to grassland in San Benito County resulted in increased abundance of native perennial bunchgrasses over three years, but their study design is unclear and lacks control plots. Observations from coastal grassland in Humboldt County indicate that a change from a continuous, high-intensity cattle-grazing regime to an intermittent, moderate-intensity grazing rotation corresponded with a 10% increase in Danthonia californica abundance and a 35% decrease in the abundance of the invasive exotic grass Taeniatherum caput-medusae over 3 years (Cooper 1960), but, again, no appropriate controls were available. In another setting without control plots, Kephart (2001) used sheep and goat grazing to reduce the dominance of aggressive annual weeds such as Centaurea solstitialis in mixed grassland in San Mateo County. After two years of short-duration, intensive grazing, native species richness increased and native cover increased slightly. Exotic species richness was unchanged, but cover of exotics also increased substantially. Fenceline and road verge studies have recently been used to infer the impacts of livestock grazing on native species richness in California grasslands. Their usefulness in advising management needs further study. In one study across a wide range of coastal grassland sites, the native perennial grass Danthonia californica appeared to benefit from grazing (mostly continuous grazing), as it was more abundant on the grazed compared to the ungrazed sides of fences (Hayes and Holl 2003a). This is consistent with an experimental study in which it decreased in frequency and cover within grazing exclosures (relative to grazed plots) in a coastal grassland site in San Mateo County (Hatch et al. 1999). In the fenceline study, cover of native annual forbs also appeared to benefit from grazing, but so did the cover and richness of

GRAZING MANAGEMENT

249

exotic annual and perennial forbs. This positive response of exotic forbs to grazing was also seen in a comparison of ungrazed road verges with nearby grazed sites on nonserpentine soils: Safford and Harrison (2001) found that grazed sites had higher exotic species richness than their ungrazed counterparts. In their study, native species were unaffected by grazing. The results of livestock grazing studies suggest that sitespecific factors (e.g., species pool, land use history, soils) and climate exert an influence on the response of native plant species to manipulations of grazing regime. Indeed, a recent extensive survey of grazed and ungrazed sites in northern California suggested a strong role for slope/aspect and soil type in influencing the outcome of grazing (Gelbard and Harrison 2003), with colder slopes showing different responses than warmer slopes and soil type interacting with these slope/aspect factors. In a broad survey of grassland sites in Monterey County, the consistent lack of recovery of native grasses in sites that had been excluded from grazing for decades but that had been tilled for agriculture, suggests the critical importance of land-use history (Stromberg and Griffin 1996). In these cases, no amount of grazing or specialized grazing regime will enhance native perennial grasses if they have been completely eliminated from the site and no nearby seed sources are available. If livestock grazing management and the restoration of native grassland diversity are to be compatible goals, sources of variation in the relationship between grazing regime and native plant abundance need to be better understood to provide site-specific guidelines for the development of grazing prescriptions. More information is needed on the response of entire assemblages, particularly native forbs, to grazing treatments (including cessation). Also, the presence of an oak canopy can greatly affect grassland composition and structure (Callaway et al. 1991; and see Tyler et al., Chapter 14), yet we know very little about the interaction of livestock grazing with the presence or absence of oak canopies (Chapter 17). Because of the variation in response to grazing that is found in the native grassland flora, it is reasonable to assume that no single grazing regime or the complete cessation of grazing will be optimal for all native species. A management plan that varies the timing and intensity of grazing on a landscape scale may better enhance native plant diversity than the uniform application or the uniform elimination of grazing. Properly designed experiments are needed that explicitly examine the response of native species to wellquantified grazing regimes, replicated across a range of sites where site conditions and history are quantified. In many sites, the lack of a seed source for native species will limit responses to grazing, and seeding will need to be combined with grazing to enhance native biodiversity.

Grazing to Control Brush and Fire Hazard Many chaparral species are highly palatable to goats, and these animals have long been used to maintain firebreaks in

250

POLICY AND MANAGEMENT

California or to reduce brush in areas that have been converted from shrubland to grassland. In California grasslands, Himalyan blackberry (Rhubus discolor), poison oak (Toxicodendron diversilobum), coyote brush (Baccharis pilularis), and broom species (Cytisus scoparius and Genista monspessulana) are all commonly invasive in coastal and coast range grasslands and can be managed using grazing. In many areas the use of angora and Spanish goats shows promise in controlling Himalayan blackberry (Daar 1983). Crouchley (1980) mentions that blackberry is readily eaten by goats throughout the year, even when there is an abundant supply of other plants. Sheep, cattle, and horses can also be effective in reducing the spread of Himalayan blackberry (Amor 1974), especially when plants are small. Blackberry was reduced by 73.5%, and more than half of the cover of toyon (Heteromeles arbutifolia), coyote brush, honeysuckle (Lonicera hispidula), herbaceous species, and madrone (Arbutus menziesii) was removed by goats on a fuel break in a study in coastal foothills, breaking the sequence of live fuels horizontally up to 1.5 meters, and reducing the amount of one- and ten-hour dead fuels by 33.2% and 58.3% respectively (Tsiouvaras et al. 1989). Regrowth has also been controlled by grazing sheep and goats in areas where mature plants have been removed. Grazing of poison oak by sheep and goats can be effective for its control. Deer or horses will also graze poison oak when the foliage is young, before the plant flowers (DiTomaso et al. 2001b). Grazing of cattle can prevent the encroachment of poison oak and coyote brush into grasslands in the eastern San Francisco Bay area (Russell and McBride 2003; EBRPD 2005). In this area, goat utilization of poison-oak was 67% in a fuelbreak under controlled grazing, somewhat lower than utilization of toyon, California blackberry, and coyotebrush. Grazing must be maintained over the long term to control shrubs, since in this rainfall region grasslands tend to convert to shrubland without fire or grazing. Brooms such as Scotch broom (Cytissus scoparius) appear to be more resistant to grazing, even by goats. In one study goats had a major impact when broom density was low (4% ground cover) but no impact when broom density was at 10% ground cover. Goats stripped bark from broom stems during winter, reducing broom vigor in pastures. Both sheep and goats removed stem and flowering parts, preventing seed production within browse reach and removing new broom shoots over summer. When broom seeds were fed to goats, 8% of the seeds remained viable following ingestion (Holst et al. 2004), so at least some viable seeds could be transported among sites by goats. The dry grasses left on annual rangelands in summer and fall are fine fuels that promote rapid fire spread. Cattle, sheep, goats, and horses can all be used to reduce fine-fuel loads. The effectiveness of grazing on fire behavior has not at this point been quantified but is inferred from the removal and alteration of fuels. Grazing is an alternative to prescribed fire for fire hazard reduction at urban-wildland interfaces and in other situations where risks of fire escape cannot be tolerated. Goat herders usually charge to graze for vegetation management,

while cattle grazers typically pay for grazing leases. Management is straightforward, emphasizing control of shrub encroachment while protecting other resources.

Grazing Management for Vernal Pool and Small-wetland Biodiversity Although grazing in riparian areas is generally outside the scope of this chapter, vernal pools, small grassland wetlands, and small ephemeral streams are characteristic of and integrate with Mediterranean grasslands in California. Many of these habitats are considered important for native flora and fauna, including threatened species such as the red-legged frog and tiger salamander. Livestock influence vegetation, soils, and water quality in these small wetlands, but the ultimate impacts on species of concern are not well studied in most of these habitats. An adaptive management approach, with careful monitoring of management effects, is recommended. The assumption has often been that livestock grazing, as an “unnatural” influence, is harmful (Barry 1998; Griggs 2000). However, some recent evidence suggests that grazing can be a tool for improving or maintaining biodiversity in annual grassland vernal pools and other grassland aquatic environments. One study argues that grazing could mediate the effects of global climate change in vernal pools by reducing cover of exotic annuals that benefit from warmer temperatures (Pyke and Marty 2005). In an extensive study of grazing impacts on vernal pools in the Central Valley, Marty (2005) found that without any livestock grazing, exotic annual grasses dominated pool edges and reduced the duration of pool inundation because of their rapid use of water. By reducing litter buildup and cover of exotic annual grasses, cattle were effective in increasing the diversity of forbs and other species. Three years of study of numerous pools showed that ungrazed pools had 88% higher cover of annual grasses and 47% lower relative cover of native species than pools grazed for most or all of the year. Without grazing, species richness of native plants declined by 25%. Many of the native vernal pool plants (e.g., Lasthenia spp., Downingia spp., Veronica spp.) are small and require an open environment to successfully germinate and reproduce (Linhart 1988). Cattle grazing may be particularly effective at reducing exotic grass cover, because cattle selectively forage on grasses (Kie and Boroski 1996; Knapp et al. 1999). Removal of sheep grazing has also been reported to result in degradation of vernal pool habitats (Reiner 2001). In addition to affecting vernal pool plant composition, grazing enhanced invertebrate diversity in vernal pools by 28% (Marty 2005). In ungrazed pools, exotic annual grasses evapotranspired large amounts of water, reducing pool inundation period by 50–80% and making it difficult for some vernal pool endemic animal species to complete their life cycle (King et al. 1996; Marty 2005). Thus, Marty (2005) suggested that grazing prescriptions to protect and promote native species diversity should be based on the practices utilized when pools were identified as having species of conservation concern. There was a low-intensity, October to

June grazing regime, and release from this regime via total exclosure reduced the species of special concern (Marty 2005). Under an adaptive management scenario, grazing might be adjusted over time to see whether further understanding of the interaction of grazing practices and weather patterns could be gained by establishing and monitoring plots where modifications could be tested. Small spring-fed wetlands, or seeps, are found in many annual grassland areas. A ten-year study of different levels of grazing in such wetlands showed that plant diversity was not affected by grazing regime in and around the springs, but diversity increased under moderate levels of grazing along the small creeks flowing from the springs. Details of grazing in these springs are discussed in Chapter 17. Grassland soils act as natural nitrogen sources to springs. Removal of livestock grazing from small wetlands resulted in a reduced capacity for the system to take up nitrogen, and hence increased nitrates were found in the springs (Jackson et al. 2006b). Indeed, Jackson et al. (2006b) determined with a paired-plot grazing removal experiment that nitrate concentrations in surface waters where grazing was removed for only 2 years were as much as five times greater than in grazed counterparts, far exceeding U.S. Environmental Protection Agency (EPA) standards for surface waters. The authors of this study hypothesized that the buildup of litter in the absence of grazing reduced herbaceous production, and therefore N sequestration, by the vegetation (Typha, rushes, sedges) in and immediately around these springs. On the other hand, direct inputs of animal excrement into ponds and wetlands have been found to increase levels of ammonia and nitrite in small ponds in intensively grazed areas (Clausnitzer and Huddleston 2002; Knutson et al. 2004). There is widespread concern about the impacts of grazing on riparian areas in the western United States (Belsky 1999), and it has been a difficult topic to study for logistical and experimental reasons. Two studies of grazing around small grassland streams in California grasslands showed little to no impact of a light to moderate grazing regime on channel morphology (Allen-Diaz et al. 1998; George et al. 2002). However, it has been observed that heavy grazing can reduce vegetation cover and decrease the slope of streambanks, resulting in bank erosion and degraded aquatic habitat (Larsen et al. 1998). Finally, stockponds constructed by ranchers are recognized to provide refugia for aquatic species such as red-legged frog and tiger salamander, especially as vernal pools are lost to urbanization and vineyard production (EPA 2003, 2005). Their quality as habitat depends on the inundation period, timing of livestock use, and whether predators such as bullfrogs or introduced fish are present. Allowing or creating a dry period can control many such introduced species.

Manager Information Needs Interviews with 35 grassland managers and scientists across the state reveal three major areas of concern (Chadden et al. 2004). First, grassland managers want more information to

GRAZING MANAGEMENT

251

help them set goals and evaluate progress. For those with restoration goals, finding out all they can about the past distribution and composition of grasslands and grassland species is a common enterprise. Restoring past conditions, including native species communities, is one common goal, though complete restoration to past conditions has never been achieved and is widely accepted as impossible. A countervailing narrative by some managers is that ongoing changes to species and communities are irrevocable. Instead of trying to restore past conditions, they define a goal that is reasonable and desirable for each particular situation. This could be maintaining or enhancing native species diversity, improving conditions for broad-leaved plants needed for a threatened butterfly (Weiss 1999), or increasing forage production for livestock. Another approach is to manage for diversity of species composition and structure in general, with the assumption that this will provide habitat for the most diverse array of plant and animal species. In all cases, management flexibility and monitoring are needed to cope with constant change. Commitment to a long-term analysis is essential. Response to management tends to be slow and often difficult to detect because of radical weather-driven changes in the habitat each year. A second major area of concern of land managers clusters around habitat fragmentation. Increasingly, managers find themselves faced with disconnected pieces of land, subject to urban stresses and the effects of roads and trails. Such effects include incursion by weeds along roadsides and fuelbreaks (Gelbard and Harrison 2003; Merriam et al. 2006). Goats are increasingly used as vegetation management tools in these settings because they appear to be more acceptable to adjacent homeowners and recreationists than other livestock (R. Budzinski, personal communication), but they are an expensive tool. Fragmentation also destroys metapopulations of various wildlife species and increases predation by dogs and feral cats. Finally, managers realize that they need more information that is directly applicable to the sites and species that they need to manage. They know that they are unlikely to have an amount of knowledge that they consider “adequate” any time in the near future. Instead, adaptive management, which makes management itself into an experimental process, is seen as desirable. Unfortunately, the resources to design and carry out the sound experiments called for in an “adaptive management” scenario seem to be only rarely available. Monitoring with some controls is usually a more feasible goal. Recent interviews with ranchers about rangeland conditions and management (Ford and Huntsinger 2006 interview notes) revealed that a major concern is that they are often held accountable for factors outside of their control. General misunderstandings about the California grasslands often result in attempts to use grazing management to influence vegetation characteristics that are controlled by weather and environmental conditions and little influenced by grazing. Interviewed ranchers shared manager concerns about invasive plants, and they support efforts to improve wildlife

252

POLICY AND MANAGEMENT

habitat when such efforts are not highly detrimental or costly to the livestock operation or when the ranchers are compensated for extra conservation measures. In many cases the ranchers believe that in the course of their management they take good care of the land and wildlife already, and protect resource productivity. On public lands, they believe that short and uncertain leases reduce ranchers’ investment in the land and that, without long-term commitments from both lessor and lessee, management is often less than optimal. Ranchers also argue that while each ranch is unique in environmental and human terms, regulations can be too broad. Finally, ranchers are concerned about bureaucratic complications that prevent the implementation of projects that have been worked out face to face by ranchers, wildlife biologists, and conservationists. Sometimes such projects are delayed so long that desired habitat quality is lost through a lack of grazing over time or through the infilling of ponds that need maintenance. Such problems take the enjoyment out of ranching, and make ranchers feel that they are “doing something wrong” by ranching. A growing concern in the ranching community is that with current high land prices and the difficulties and costs of navigating a complex regulatory environment, families are leaving ranching and traditions are being lost. The skyrocketing value of land prohibits newcomers from joining the ranching community.

Conclusions California’s grasslands are managed for diverse goals by public and private landowners. There is great variation in the quality of management and in the resources available for management. However, most managers are interested in maintaining the environmental values of the resource, though opinions about what that means and how to get there vary. Basic principles of livestock grazing management can be used to develop prescriptions for accomplishing manager objectives, but grazing impacts occur within the huge annual variation caused by weather patterns and across a heterogenous landscape of varied soils, topography, and land use history. Most quantitative information shows that California grasslands respond quickly to weather (Reever Morghan et al., Chapter 7; Harpole et al., Chapter 10; Dukes and Shaw, Chapter 19) and relatively slowly to management. Research into the goals and capacities of various kinds of grassland managers is needed to develop programs and practices that fit with their objectives and that will motivate improved management. Demographically, land ownership in the grasslands is changing, with new kinds of owners, more real estate speculators, and more land trusts owning large properties. Yet little is known about the implications of these changes for grassland sustainability and opportunities to enhance values derived from rangelands. Research geared to producing information useful in reducing land fragmentation and habitat loss is needed, as is finding ways to stimulate increased “boundary-crossing” management and protection of habitat. The interconnections between private and public lands via

grazing leases and public manager/stakeholder cooperation toward attainment of mutual goals have not been well assessed. As more and more land is converted to urban and suburban development, the environmental services produced by private rangelands increase in value. Yet methods to value these services have been little researched or applied in California. The most common grazing management approach is to use RDM or “mulch” monitoring as a way of protecting future productivity in annual grasslands (Bartolome et al. 2002). Carefully timed grazing can be used to control invasive species such as yellow starthistle and medusahead. Livestock grazing can also be used to promote native species, including rare ones, in vernal pool settings (Marty 2005). Nonetheless, a clear relationship between livestock grazing and California’s native grassland plants is difficult to establish. Although an extensive literature documents the impact of grazing in annual grasslands dominated by introduced species (e.g., Heady 1956, 1958; Pitt and Heady 1979; Rosiere 1987; Bartolome and McClaran 1992), relatively few studies have quantified the impact of grazing on a range of native plants, either annual or perennial. The existing data show

that the interactions among livestock, exotic plants, and native plants are variable across regions and years. Grazing can benefit some native plant populations, but a positive response is not universal even across locales for any one species, as demonstrated by the variable response of Nassella pulchra to grazing. Grazing can negatively impact some native plant species, but responses, again, are not universal across species or sites. More experimental research is needed to understand the contingencies that regulate species responses to grazing prescriptions. Tests of the effects of specialized grazing systems on range production and vegetation composition on different ecological sites would also improve grassland management. With growing emphasis on management for biodiversity, interactions between livestock and wildlife or livestock and threatened and endangered species need more research. As wildfire becomes more costly and suburban development patterns put more residential areas at risk, considerable public investment in reducing fire hazard is contemplated by public agencies. As a relatively low-cost tool, the effectiveness of grazing management for grassland fuel modification needs further study.

GRAZING MANAGEMENT

253

TWENTY ONE

California Grassland Restoration MAR K R. ST R O M B E R G, CA R LA M. D’ANTO N I O, TR U MAN P. YO U N G, J EAN N E W I R KA, AN D PAU L R. K E P HART

Changes in California’s grasslands associated with human activity vary along a gradient from complete destruction in the worst case to preserved fragments of presumed pristine stands in the best. Yet the most “pristine” California grasslands represent only a small proportion of California’s current grasslands, and even they include some non-native species (Safford and Harrison 2001; Stromberg et al. 2001; Gelbard and Harrison 2003). Some grasslands mostly or entirely devoid of native grasses still harbor relatively rich floras of native herb species (Lulow 2004). More common are stands dominated by a few species of non-native grass, with scattered natives (both grasses and herbs) making up varying degrees of relative cover. The challenge of restoring California grasslands is to develop site-appropriate goals along with prescriptions that match this wide range of grassland conditions. While preserving and managing California grasslands with a higher component of native species may seem like an easier task, it is often those that have been completely destroyed that can be most easily restored, simply because they are less complex and the practitioner can start from a clean slate. This paradoxical quality of grassland restoration and the dilemmas faced by grassland restorationists are the subjects of this chapter. Restoration is the complex set of efforts to reverse or mitigate effects of human activity on the landscape (Packard and Mutel 1996). The Society for Ecological Restoration (SER) has defined ecological restoration as the process of assisting the recovery of an ecosystem that has been degraded, damaged, or destroyed. An ecosystem has recovered— and is restored—when: 1) it contains a characteristic assemblage of the species that occur in the reference ecosystem and that provide appropriate community structure, 2) it consists of indigenous species to the greatest practicable extent, 3) all functional groups necessary for the continued development and/or stability of the restored ecosystem are represented or, if they are not, the

254

missing groups have the potential to colonize by natural means, 4) the physical environment of the restored ecosystem is capable of sustaining reproducing populations of the species necessary for its continued stability, 5) it apparently functions normally for its ecological stage of development, and signs of dysfunction are absent, 6) it is suitably integrated into a larger ecological matrix or landscape, with which it interacts through abiotic and biotic flows and exchanges, 7) potential threats to the health and integrity of the restored ecosystem from the surrounding landscape have been eliminated or reduced as much as possible, 8) it is sufficiently resilient to endure the normal periodic stress events in the local environment that serve to maintain the integrity of the ecosystem and 9) is self-sustaining to the same degree as its reference ecosystem, and has the potential to persist indefinitely under existing environmental conditions (SER 2004).

Restoration of California grasslands, once thought to be nearly impossible (Sampson et al. 1951; Heady 1988) is now under way at many sites, although usually with less ambitious goals than the complete eradication of exotics or complete ecological restoration as defined above. Restoration offers the hope of creating a landscape that is more weed-resistant, maintains its productivity over time and other ecosystem services, and is somewhat tolerant or resilient to a variety of stresses. A broad continuum of effort exists from small-scale landscaping and creation of prairie gardens, to landscape architecture projects focused on native grasses, to largerscale reclamation and full “ecological restoration” as defined above. Reaching the idealistic definition of restoration may be exceedingly difficult in much of California’s grasslands, because of constraints that will be discussed subsequently. Nonetheless, people are working along all parts of the continuum toward restoration of one or more of the attributes described by SER, and the issues discussed here are generally relevant at many levels. The term restoration will be used to refer to all of the efforts along the continuum.

Restoration practitioners, seed producers, academic researchers, consulting biologists, agronomists, ranchers, and landowners have made some significant advances in California grassland restoration. Many were involved in the establishment of the California Native Grasslands Association (CNGA) and the California chapter of the Society for Ecological Restoration (SERCAL), both in 1991. Hundreds of grassland restoration projects have been initiated across California. Initially, most were designed to establish permanent grassland habitats with native perennial grasses as the backbone (Anderson and Anderson 1996). Once established, individual native perennial grasses may survive for hundreds of years (Hamilton et al. 2002), and the basal clumps form the structural basis for a physically more complex habitat. The focus on perennial grasses in California grassland restoration is based on the assumption that by restoring the structural diversity of perennial bunchgrasses, colonization and survival of associated herbs, shrubs, insects, small mammals, and other community members will eventually occur (MacArthur et al. 1966; Bell et al. 1991; Huston 1994; Rosenzwieg 1995; Vickery et al. 2001; Goerrissen 2005). Also, because native perennial grasses are persistent, it is assumed they will provide greater resistance to invasion and resilience to stress than annual species. Whether sites restored to native perennial grasses achieve the ultimate goal of a restored ecosystem, as defined by SER, however, has rarely been evaluated. More recently, there has been greater emphasis on other plant groups, such as native forbs. For example, a few grassland restoration projects have introduced up to 20 forbs (Kephart 2001). Restoration typically involves the selection of a reference ecosystem that is chosen because it is a realistic target for the particular site conditions. Ecologists have a long legacy of historical ecology to determine reference ecosystems (Egan and Howell 2001). However, there are no formal guidelines for what defines a ‘reference ecosystem” or remnant stand of native California grassland. What is considered as viable remnant grassland, and thus a reference ecosystem for restoration, will vary widely across California. For instance, in the dry interior habitats above the Central Valley, bunchgrasses may be rare and the “native” community may have been one largely dominated by forbs and shrubs (Schiffman 1995). By contrast, foggy coastal terraces are often dominated by plants other than grasses (Stromberg et al. 2001; Hayes and Holl 2003a), and total plant diversity may exceed 20 native species/square meter (Stromberg et al. 2001), whereas in interior dry grasslands, one may only find 5 to 10 species/square meter (Harrison 1999b). Drier slopes of the coastal ranges or the central valley foothills may be only co-dominated by grasses (Carlsen et al. 2000). It is not at all clear that the bunchgrass Nassella pulchra (purple needlegrass) dominated in the drier, upland plant communities there (see KeelerWolf et al., Chapter 3; D’Antonio et al., Chapter 6). Portions of California’s central valley were locally inundated seasonally and may have been relatively diverse, supporting alkaline-tolerant plant communities and a variety of grasses other than N. pulchra (Holstein 2001; Lombardo et al. 2007).

By comparison to California, the Midwest has a rich literature on plant community composition, structure, and controlling processes of reference remnants of the tall grass prairies (Clements 1934; Packard and Mutel 1996). Yet even there, the definition of a successful restoration has been elusive and now includes some criteria for the amount and distribution of native plant diversity (Martin et al. 2005). Explicit inclusion of native animals in grassland restoration (Martin et al. 2005) is rare but may be critical in drier grasslands (see Schiffman, Chapter 15). While little data exist with which to build a quantitative classification—or to reconstruct a historic flora—it is clear that remnant California grasslands have great geographic and floristic diversity (see Keeler-Wolf et al. Chapter 3) and high ecological value (Jantz et al., Chapter 23). Sawyer and Keeler-Wolf (2007) describe 25 vegetation series dominated by native perennial grasses in California, with an additional eight series dominated by introduced annual or perennial grasses. Most of the native grassland once thought to have occurred in and around the Great Central Valley and surrounding foothills has largely been converted to agriculture and urban uses (Huenneke and Mooney 1989b). Remnant grasslands provide habitat to many federally listed species (see Jantz et al., Chapter 23), including 48% of California’s listed terrestrial invertebrates, 50% of the listed terrestrial vertebrates, and 82% of the listed vascular plants (HCPB 2006). California’s native-dominated remnant grasslands are clearly of conservation importance—and therefore of interest to restorationists—throughout the state. Ecological restoration is intimately related to the population biology of each of the species being assembled (Montalvo et al. 1997). A successful long-term restoration will include populations large enough to survive in a dynamic landscape and to allow adaptive natural selection (see Rice and Espeland, Chapter 11). Population biology in California grasslands has another unique implication for restoration planning. As California’s grassland composition and relative species composition are largely driven in a particular year by annual rainfall patterns (Reever Morghan et al., Chapter 7; Dukes and Shaw, Chapter 19), and this can be highly variable (Jackson and Bartolome 2002), the use of a relatively pristine “aboriginal” grassland as a model for restoration (White and Walker 1997) with data taken from only one year could be misleading. The assembly of species in a restoration on a particular site represents a complex of decisions in a matrix of constraints.

Restoration Constraints: Legacies of the Past Grassland restoration efforts must take into account the current and desired species composition of a site and effects of past human activity (Baker 1989). Some of the more important legacies to consider in restoration are discussed in the following paragraphs.

Invasive Non-native Species Invasive non-native species represent the single greatest impediment to grassland restoration in California (and

G R A S S L A N D R E S T O R AT I O N

255

throughout the United States west of the Rocky Mountains). How they prevent the return of native perennial grasses is not clear, but introduced grassland species are highly competitive in many circumstances (Corbin et al., Chapter 13). They maintain a very large soil seed bank and can overwhelm native seedlings after fall rains. Some may take advantage of unique associations with soil biota (Reinhart and Callaway 2006) or anthropogenic sources of added nitrogen (Weiss 1999), while others may be allelopathic to natives (Tinnin and Muller 1971, 1972). The replacement of perennial grasses with annual grasses has also increased the deep soil water availability, thus providing a new resource for late-season, taprooted invasive plants such as starthistle (Reever Morghan and Rice 2005). Grasslands in California dominated by nonnative species appear to be new stable states (Seabloom et al. 2003b, Seabloom and Richards 2003); these communities persist until some active restoration including native species are undertaken.

Soils and Land Use Historical land use often involved plowing, including deep disking, and has been associated with the loss of much of California’s perennial, native grasslands on deep, arable soils (Stromberg and Griffin 1996). Initially, such plowing completely eliminates the native perennial bunchgrasses. When formerly farmed fields are abandoned, exotics quickly invade. However, plowing may also create longer-term conditions that favor exotics over natives. Disrupting the relationship between native plants and their complex soil microbial communities often harms efforts to re-establish natives (Perry et al. 1989; Allen et al. 2002; Wardle 2006). Plowing California grassland soils results in dramatic loss of both microbial species diversity and composition (Steenwerth et al. 2002). These effects persist in old fields even if they are not further disturbed for up to 70 years (Jackson et al., Chapter 9). Partial restoration of a microbial community toward that of remnant native grasslands appears to take place in soils of native grass production fields, but these have been intensively managed through irrigation and repeated weed control (Potthoff et al. 2005b). Whether the “annual grassland” microbial communities in rangeland soils with a history of tillage will affect the success of re-establishing natives remains largely unanswered. In addition to plowing, many areas in California were subject to mechanical raking or large-scale vegetation type conversion (Merenlender et al. 2001). Oak savanna edges or edges of abandoned agricultural fields in California (Stromberg and Griffin 1996), unlike the Midwest tallgrass/woodland edges, may be quite stable (Carmel and Flather 2004, 2006), defying the Midwestern concept of ecological succession (Clements 1916). Likewise, dense oak woodlands have been bulldozed in recent times and resist subsequent restoration efforts, remaining instead as weedy grassland (Brooks and Merenlender 2001; Merenlender et al. 2001; and see Tyler et al., Chapter 14). Also, there exists a decades-long recruitment gap in blue oaks

256

POLICY AND MANAGEMENT

and valley oaks in some wooded grasslands (Tyler et al. 2006), which may presage an eventual conversion to pure grassland unless oak restoration efforts are successful. Large-scale grading and soil loss was common in much of the historic land use in California (1840s – 1940s) (Kinney 1996; Heise and Merenlender 2002). The loss of the productive A horizons, exposing the less productive subsoils, was widespread. Some species commonly used in California grassland restoration (e.g., Nassella pulchra) can survive and perform well on these less productive soils (Lombardo et al. 2007) or subsoils (Jaymee Marty, personal communication, 2006), in contrast to establishment efforts on rich, deep soils, where competition with non-native, invasive plants is common. For example, native grasses achieved 75% cover with very few weeds on a two-acre site that had been spread with subsoil from a pond excavation (Maxwell Flat project; Table 21.1). When sites with deep, productive soils do become available for restoration and there is effective weed control in place, a variety of native grasses can quickly establish dense stands that effectively exclude exotics, as illustrated by the Mace site in the city of Davis (Table 21.1).

Viruses Another historical legacy of the invasion of California grasslands is the presence of viral diseases introduced with European agriculture. For example, barley yellow dwarf virus (see D’Antonio et al., Chapter 6) can thrive on the widespread, abundant, non-native Avena fatua and other species and can infect and reduce the survivorship of the nearby perennial native grasses (Malmstrom et al. 2005a, b). Grazing appears to partially counteract the effects of these viral infections in Nassella pulchra (Malmstrom et al. 2006). At the very best, however, the presence of viruses may limit grasslands restoration goals to an equilibrium in which perennial, native grasses persist as low-density populations in a background of non-native annual grasses (Malmstrom et al. 2005b).

Road Construction Roads increase both the spread of invasive species and the loss of native species (Gelbard and Belnap 2003; Gelbard and Harrison 2005). They promote and concentrate seed vectors (vehicles, domestic animals, etc.), and common road maintenance practices (e.g., annual grading) provide the disturbance required for the persistence of many invasive, non-native annuals.

Fire With the arrival of humans in the late Pleistocene, and the development of fire as a tool by the indigenous people (Greenlee and Langenheim 1990), large areas of California were burned regularly, including grass-dominated sites (see Reiner, Chapter 18; Anderson 2005). Fires were set frequently (every 2 – 5 years) by Native Americans to keep grasslands relatively free of shrubs and trees (Margolin 1989) and for

F I G U R E 21.1. (a) Headwaters of Wildcat Canyon, San Pablo Ridge, Contra Costa County, California, 1902.

Bald Peak is on left in the distance, and Grizzly Peak is right of center. This is very typical of the Berkeley hills, where trees were largely absent except along watercourses. From: Lawson and Palache (1902), plate 15. (b) Headwaters of Wildcat Canyon, as in (a), July 2002, with Bald (
Vollmer) Peak on left and Grizzly Peak on right. Tilden Botanic Gardens is on the left, with mowed, irrigated lawn in foreground. Plantations of Monterey pine, eucalyptus, and other conifers, plus other woody vegetation, have replaced former grasslands (Edwards 2002).

many other purposes in grasslands (Anderson 2005). Since European settlement, fire frequency has been considerably reduced in many areas (Greenlee and Langenheim 1990), resulting in conversion of former grasslands to shrublands (McBride and Heady 1968; Edwards 2002). For example, the Berkeley hills and much of the San Francisco Peninsula were open grassland at the time of intensive European settlement (Figure 21.1). After decades without fire, they have converted to woody vegetation. Maintaining a fire return interval that is locally appropriate can be particularly challenging for a grassland restoration but may be critical (see Reiner, Chapter 18).

Grazing Pleistocene California grasslands were one of the more spectacular grazing systems in the world (Wigand, Chapter 4;

Edwards, Chapter 4; Schiffman, Chapter 4). With the arrival of humans in the Pleistocene, there was a precipitous drop in the larger animals (“megafauna”), perhaps due to overhunting by humans (Martin 1974; Alroy 2001). California’s 13 species of large carnivores dropped to one, and the 18 large herbivores were reduced to five (Edwards 1996). With the arrival of hunting-gathering people in the Pleistocene, and the megafauna disappearing, it has been suggested there was some other sudden change (Lambert and Holling 1998) on the landscape scale (Owen-Smith 1987). This loss of megafauna may have been associated with a human-caused increase in fire frequency (Greenlee and Langenheim 1990), which would have landscape-scale effects. With European settlement, fires were suppressed and the last of the large herds of elk and antelope on the grasslands were eliminated and replaced with feral cattle and later domesticated livestock

G R A S S L A N D R E S T O R AT I O N

257

TA B L E 21.1 Selected Native Grass Restoration/Seeding Projects in Northern California 1990–2005

Project name and landowner

Preplanting site treatment

Planting date, method and rate

Bromus carinatus Elymus glaucus Melica californica Nassella lepida Nassella pulchra Poa secunda

Burn 6/2000 Disking 10/2000

Drill seeded 12/2000 19 lbs/acre

Elymus glaucus Elymus multisetus Nassella lepida Nassella pulchra Poa secunda

Pre-fire grazing 3/2002 Burn 5/2002 5/2003 Disking 11/2002 11/2003

Drill seeded 35 acres 11/2002 21 lbs/acre

Burn 6/2000 Disking 10/2000 10/2001 Herbicide 4/2001 11/2001

Drill seeded 11/2002 20 lbs.acre

County

Grassland type

Acres

Species used

Corral Pasture Privatea

Yolo

Foothill grassland Grazed pasture

30

Ranchette 1 Privatea

Yolo

Foothill grassland

50

Back 40 Privatea

Yolo

Foothill grassland

40

Elymus glaucus Elymus multisetus Nassella lepida Nassella pulchra Poa secunda

25 acres 12/2003 21 lbs/acre

Postplanting management

Initial site conditions/problems

Grazing 4/2001–2005 Swathing 6/2002 Herbicide 12/2000 4/2001

Stand condition as of 2006

Lessons learned

Very compacted holding pasture at large cattle ranch. Heavily invaded with medusahead, goatgrass and yellow starthistle with virtually no native grass cover, although natives were presented in shaded riparian zone

Relative cover of native grasses approximately 30% after 5 years. While overall cover seems to have stabilized at the site, the relative cover of individual native grass species shifted with B. carinatus and P. secunda dropping to low levels and N. pulchra and E. glaucus persisting. Initial prescribed fire had a significant negative effect on relative cover of medusahead, goatgrass, and yellow starthistle, but by 2005 relative cover of goatgrass had rebounded to levels that are not significantly different from preproject levels.

Success of project due in part to highly cooperative rancher willing to move cattle in and out of pasture at appropriate times. Goatgrass very difficult to control with methods that work well for medusahead and starthistle (e.g., burning, grazing). Goal at rangeland sites should not be to eradicate non-native forage grasses (e.g., wild oats and soft chess) but to manage site for maximum biodiversity with appropriate grazing. Long-term management necessary.

Grazing Mowing Herbicide

Site dominated by medusahead and yellow starthistle (YST) with annual ryegrass in swale areas. Private ranchette with no commercial grazing made it necessary to import grazing animals. Very hilly; difficult to drill and mow.

Relative cover of native grasses approximately 31% in 2005, 3 years after initial burn and 2 years after seeding. Cover of yellow starthistle (YST) remains low due to treatment with herbicide, but cover of non-native grasses, especially medusahead, on the rise. Annual ryegrass persists in swale areas.

By third year after initial burn, medusahead had rebounded to near preproject levels, indicating that repeat burning may be necessary. Grazing was limited and done with sheep and goats and did not significantly control weeds. Repeated mowing was not sufficient to keep medusahead at bay.

Mowing 4/2003 4/2004 4/2005

Diverse mixed annual grassland with high cover of native and non-native forbs and scattered perennial bunchgrasses. Surrounded on two sides by neighboring ranch dominated by medusahead and goatgrass. High cover of starthistle.

No preplanting data are available, but one-year postplanting native grass cover is nearly 50%. Site is very hilly with individual native grass species alternating dominance depending on slope and aspect.

Site was kept fallow with disking for 2 years following initial burn treatment, which may explain initial stand success. Herbicide was not applied after grass planting to maintain existing populations of beneficial forbs. No resident population of grazing animals, so site was maintained with mowing instead of grazing.

(Continued)

TA B L E

Project name and landowner Maxwell Flat Privatea

21.1 ( C O N T I N U E D ) Preplanting site treatment

Planting date, method and rate

Three different mixes used with varying combinations of: Elymus glaucus Elymus multisetus Koeleria macrantha Melica californica Nassella lepida Nassella pulchra Poa secunda Vulpia microstachys

Burn 5/2003 5/2004 5/2005 Disking 11/2003 11/2004 Herbicide 11/2003 11/2004

Drill seeded 50 acres 11/2003 21 lbs/acre

County

Grassland type

Acres

Species used

Solano

Foothill grassland Grazed pasture

66

16 acres 11/2005 21 lbs/acre

Colusa N USFWSb

Colusa

Wetland?

23

Hordeum brachyantherum Nassella pulchra Elymus glaucus

Burn Sum/1997 Disk 2  Fall 97

Drill seeded 12/97 15 lbs./acre

Llano Seco Tract 1 USFWSb

Butte

Riparian understory

65

Hordeum brachyantherum Nassella pulchra Elymus glaucus Elymus trachycaulis Leymus triticoides

Burn 1999 Disking Herbicide

Drill seeded 30 acres 1/2002 (with fertilizer) 35 acres 11/2002 15 lbs./acre (no fertilizer)

Turtle Bay Discovery Park City of Reddingc

Shasta

Riparian terrace

28

Leymus triticoides Carex barbarae (sedge)

Herbicide (no date given)

Plug planted 28 acres 11/2004
1500 plugs/acre

Postplanting management

Initial Site conditions/problems

Grazing 4/2004 4/2005 spot spraying

Stand condition as of 2006

Lessons learned

Site a large, relatively flat pasture adjacent to riparian areas. Historically grazed and farmed, with high cover of medusahead, goatgrass, and yellow starthistle. Subsoil from pond excavation covered 2 acres. Landowner leases grazing rights to neighboring ranchers, so timing of grazing episodes have to be coordinated among many pastures.

50-acre site planted in fall after 2 sequential years of pre-planting fire shows higher relative cover of native grasses (nearly 50% after 2 years) and lower weeds than 16 acre site also burned twice but at which planting was delayed a year. Rainfall and temperature seem to be a factor. Two-acre site covered in subsoil from pond excavation had very high cover of native grasses (close to 75% relative cover) but almost no native or nonnative forb species.

“Cookbook” approaches to grassland restoration are inadequate. Analyze what is going on and tailor treatment regimes to that. Have patience; areas that start out with low cover of natives may catch up with nearby areas that start out stronger. High rainfall does not guarantee success; long periods of low temperature may hinder germination. Well-timed grazing key to success.

Herbicide 12/1997 4/1998 6/1998 4/1999 3/2001 Mowing Sum 2000 Burn 12/1999 11/2002 Sheep grazing 4-7/2001

Primary weed initially was yellow starthistle (YST) along with wild oats and ripgut brome. Surrounding areas weedy as well, but phenoxy herbicide restrictions in place after April 1.

Very robust stand as of fall 2005.

Could have used more weed control preplanting and in surrounding areas. A larger, later seeding (Colusa S, 49 acres in 2003) in same area using a similar seed mix and methods developed a very poor stand in comparison, possibly due to delay in seeding until January and February.

Mowing 3/2002 Herbicide 2/2003

Site borders Sacramento River with riparian communities on 3 sides. Historically dry farmed. Non-native invasive annual plants dominated the site with small patches of native shrubs and trees until habitat restoration began in 1999. Primary understory weed annual ryegrass with wild oats.

Excellent stand with close to 90% cover of native grasses, which are competing well with non-natives.

Mowing 5/2004

Site is a low riparian terrace along the north bank of the Sacramento River in the City of Redding. Historically used for agriculture, the site was initially infested with vetch and moth mullein.

Good establishment. Carex growing well with drip irrigation.

(Continued)

TA B L E

Project name and landowner

21.1 ( C O N T I N U E D ) Preplanting site treatment

Planting date, method and rate

N. pulchra E. glaucus P. secunda H. brachyantherum

Burned 7/1992

Drill Seeded 11/1992 with Roundup

100

N. pulchra E. glaucus Hordeum brachyanthum Elymus trachycaulis H. californicum P. secunda Melica californica Festuca idahoensis

Burned, Disked 10/1993

Drill seeded 11/1993 with Roundup immediately after drilling.

90

N. pulchra E. glaucus Hordeum brachyanthum H. californicum P. secunda Melica californica Festuca idahoensis

Complex series of restoration trials; spraying or burn in fall.

Drill in fall, starting in 1990. Continued through 2006.

County

Grassland type

Acres

Species used

Citrona Farms Road 26 Privated

Yolo

Foothill grassland

80

Citrona Farms Bottomland Privated

Yolo

Foothill Grassland

Hedgerow Farmsd

Yolo

Foothill Grassland

Postplanting management

Initial site conditions/problems

Broadleaf Herbicide 2/1993, 2/1994, 2/1995, 2/1996. Mowed MarchApril each year, 1993–1997. Winter of 2002, 2003, 2004, grazed sheep.

Stand condition as of 2006

Lessons learned

CRP (Federal Conservation Reserve Program lands, under Farm Act; withdrawn from farming) were fallow for several years, previously dry-farmed. Weeds included YST, Avena sp., B. diandrus, annual rye grass, B. hordeaceus. CRP did not allow grazing.

In 2000–2001, medusahead grass began to show up. N. pulchra surviving but surrounded by medusahead. Grazing enhanced relative abundance of native perennials. Seed sources not known, would have preferred local ecotypes. Hordeum, Nassella, and Elymus had mixed successful stands throughout the site. After 4–5 years, clearly a separation of species relative to terrain; Nassella dominated some areas, Elymus others, and Hordeum dropped out entirely.

They had persistent broad-leaf treatment, and wiped out Brodiea and probably other forbs with persistent 2,4-D treatment. Grazing early in program is advantageous. Need to get rules for CRP changed to allow grazing. More selective herbicides coming into certification for use in 2006 can control YST without as much damage to the native forbs.

Broadleaf Herbicide 2/93, 2/94, 2/95, 2/96 to 2/97, 2/98 Grazed sheep March ’95 through June.

Previously dry land farmed and grazed. Very productive soils. Many, complex treatments and many complex seeding trials.

Excellent stand establishment, managed with sheep and spraying through 1998 then left alone. YST became abundant, especially in alluvial bottomland where E. glaucus and N. pulchra were established. YST and medusa head invaded much of the planting, especially the uplands. Stands did well through June of 2003.

Don’t let YST go to seed; stay after it with spot spraying. If you let it go one year, say from 20/plants/acre, it can become the dominant plant in a year. N. pulchra dominated the hillsides, and in this case, E. glaucus persisted and dominated in alluvial plains. The other species were only occasionally present for unknown reasons.

Various herbicides, mowing, grazing, burning, reseeding, and interseeding forbs.

Previously dry farmed (winter wheat, grain crops). Corning red gravel with shallow durapan in many places (standing water).

Stands are about 50–50; never have a thick stand of grass; 1 bunch/m2 is as good as it gets. In-between is the struggle; Hemizonia, vinegar weeds (Trichostema), and other native forbs are coming in with mowing and burning. Dominant natives are N. pulchra, some areas good stands of P. secunda, patchy Melica, Elymus glaucus restricted to better soils (alluvial, swales, where oaks thrived). Elymus multisetus persists in shallow soils. Constant management for weeds between the established grass plants. Grazing makes the most sense; no smoke, relatively cheap.

Would like to do it all again. Would have used site-specific ecotypes. Continue monitoring and management on all sites. Medusahead has arrived in area, but prescribed fire, some grazing (sheep), and permanent fencing (to move cattle in) has helped. One of the best management options is sickle bar swathing, just before weeds make mature seed. Windrows burn easily and hot later in fall, and this kills seeds. Windrows may shade the established native bunches for one year, but they regrow the next year. Animals will eat windrows of ripgut and wild oats. Good for first-year stands when you can’t get animals in to graze to allow light into small plants. Baling windrows removes much of the seed.

(Continued)

TA B L E

Project name and landowner

21.1 ( C O N T I N U E D ) Preplanting site treatment

Planting date, method and rate

Hordeum brachyantherum Nassella pulchra Elymus glaucus Leymus triticoides

Cover crop of legumes Mowed 4–5/2003 4/2004 Prism 11/03 Herbicide 2/2004

Drill seeded 12/2004 15 lbs/acre

8 properties totaling 257

Hordeum brachyantherum Nassella pulchra Elymus glaucus Leymus triticoides Elymus trachycaulis

Rice crops through fall 2002; fallow through summer 2003; Disked 3  9/2003

Drill seeded (on-site aerial broadcast 11/2003 13.5 to 16 lbs./acre

Coastal Terrace

20

Bromus carinatus (20#) Elymus glaucus (15#) Deschampsia cespitosa (8#) Nassella pulchra (15#)

Disk

Drill seeded, 11/1997 50 lbs./ac.

Coastal Prairie, Inland

200

12 forbs planted with N. pulchra B. carinatus E. glaucus Festuca californica H. brachyantherum Koeleria macrantha

Burn, mow, herbicide, graze (goats, sheep) hand weeding (1996–2000)

No-till drill

County

Grassland type

Acres

Species used

Sunset Ranch The Nature Conservancye

Butte

Riparian

30

Colusa CREP and WRP sites Various privatef

Colusa

Former rice fields

South Ranch, Diablo Canyong

San Luis Obispo

Russian Ridgeh

San Mateo

Postplanting management

Initial site conditions/problems

Herbicide 12/2004 3/2005 5/2005

Stand condition as of 2006

Lessons learned

Riparian site along Sacramento River, initially infested with rye grass and mustard. Cover crops were used preplanting as a smother mulch.

Excellent germination, but too early to determine ultimate stand quality. Ongoing problems with fluevellin, Johnson grass, and Russian thistle. A second larger TNC project (135 acres) on nearby USFWS land at the same time with similar techniques will prove an interesting comparison over time.

Thatch from ryegrass and other weeds caused problems with pretreatment weed control. Cover crops may not have been necessary. Perhaps best to keep the ground barren for 2 years prior to planting.

Disked sum-fall 2003. 2 sites ring rolled. Herbicide on only one site.

Former rice fields, very wet. Silty clay, frequently flooded. Wet conditions precluded use of ground equipment on one site, so seed was aerially applied. A variety of mesic soil weeds present, including smart weed, curly dock, rye grass, rabbit’s foot grass, and clover.

Some sites had excellent initial germination, but prolonged flooding delayed or stunted some sites. Driest sites had the most growth first season.

Communication and coordination with private landowners is key. Success rates among the sites was correlated with how many times and for how long they flooded. Repeat flooding will slow growth rates of natives. Residual vegetation from fallowed fields proved difficult to deal with even with disking; preplanting fire or grazing would have helped.

Grazing continues. Site is mowed, native grass is baled and used for erosion control (and planting) on balance of site.

Former pea fields, abandoned and weedy before planting. Weeds were radish and mustards, decreased with mowing over time.

Current stand is Nassella and Deschampsia, the other species were lost. Brome lasted 3 years, now 10%; Elymus 2%; but cover of Nassella is 80% and Deschampsia at 20%.

Would have gone with different species, sometimes moved to site cattle too early in a year. First 3 years, B. carinatus did very well, with many nesting birds. E. glaucus not native/did not thrive on flats, but higher in scrub. Replace B. carinatus and E. glaucus with Hordeum brachyantherum. Deschampsia did very well, cattle select it first, and eat it to ground; care needed not to overgraze. Project successful without irrigation, herbicides, or fertilizer. Grazing reduced cover of fast-growing annuals. but over several years.

Fire in second and third year, then discontinued.

Abandonded pastures and dryland farm lots on coastal forbland, bald hills between deep ravines supporting forests/shrubs. Primary weed is yellow starthistle. Lowest cost/most effective reduction of YST was with spot spraying of Transline.

Herbicide reduced YST to nearzero. Spectacular flower shows for years after fire and planting. Where drill seeding, native grasses doing well; however, in other locations, YST is returning. Grazing with goats did not reduce YST adequately; adding sheep and increasing herd density resulted in near-eradication of YST rosettes and stems.

Counted on being able to burn to maintain the grasslands, but could not get permits and staff in subsequent years. Now need to find replacement for fire, considering grazing. Without sustained burning, the flowers are less abundant and YST is returning to pretreatment cover levels.

(Continued)

TA B L E

Project name and landowner

21.1 ( C O N T I N U E D ) Preplanting site treatment

Planting date, method and rate

Elymus triticoides Others

Recontoured to swales, breached levee for seasonal flooding. Disked and rolled

Drill seed 11/2000

10

Nassella pulchra

Burn, nonburn, grazed, nongrazed. Seed is collected locally and then grown commercially and harvested as hay.

Fall, 1997 to present. 16 bales/acre
10 lbs seed/acre

Seven projects in area; alluvial terrace Sacramento River

1 mile long, 40–100 feet in width.

Elymus glaucus Elymus triticoides Deschampsia elongata Nassella pulchra Festuca idahoensis Bromus carinatus Lotus purshianus

Restored stream banks, realigned stream, connected it to historic floodplain.

Fall 1997 to 2005

Sandy, coastal grassland

10

Nassella pulchra

Excavated holes for individual plants with auger on Bobcat.

March–April, 2006

County

Grassland type

Acres

Species used

Mace Site City of Davisi

Yolo

Alluvial along Putah Creek

85

Dye Creek Ranch TNCj

Tehama

Volcanic terrace and foothills

Sulphur Creek City of Reddingk

Shasta

Rice Ranch Orcutt, CAl

Santa Barbara

SOURCES : aAudubon California, Stewardship Program Yolo County; contact: Chris Rose. bU.S. Fish and Wildlife Service, Native Grass Working Group, Butte County; contact: Joe Silviera. cRiver Partners, Chico, California; contact: Tom Griggs or Dan Efseaff. dNRCS, Woodland Field Office and Hedgerow Farms; contact: John Anderson. eThe Nature Conservancy, Chico, California; contact: Ryan Luster. fNRCS, Colusa County; contact: Jessica Groves. gPacific Gas and Electric, Diablo Canyon Nuclear Power Plant; contact: Sally Kren. hMidpeninsula Regional Open Space District; contact: Cindy Roessler. iCity of Davis; contact: Mitch Sears. jTNC, Chico; contact: Rich Reiner. kCity of Redding, Sacramento Watersheds Action Group; contact: John McCullah. lRice Ranch, LFR Inc.; contact: Mary Carroll.

Postplanting management

Initial site conditions/problems

Grazed (cattle), mowed, (swathed and baled), spot herbicide.

Stand condition as of 2006

Lessons learned

Tomato fields, managed as clean row crop agriculture. Two management zones; alternating grazing between years. Fires are planned.

Excellent stand establishment, almost entirely creeping wild rye. About 90% cover of E. glaucus. Very productive soils. Drip irrigation put in for shrubs and trees. Tolerating standing water in two of four years, period of inundation; several weeks.

Seeded immediately after farming. Weed seed bank depleted by long-term use in agriculture. Good soils produce excellent stands of creeping wild rye. Deep thatch was a problem; swathing and baling effective solution. Grazing with nearby, cooperating farmers very successful, justified adding permanent fencing.

Baled native hay is spread by hand or with a blower on experimental plots (grazed, ungrazed, burned).

Low-productivity annual grasslands on level or slightly sloping shallow soils. Site is dominated by Bromus, Vulpia, and forbs. Hay is spread on spring burns designed to reduce medusahead grass and improve native grass establishment.

Site is monitored, and data are available at TNC Chico. Cover of planted Nassella approaches that of remnant local stands.

Hay spreading into burned areas is a viable way to reintroduce N. pulchra at sites where seed drilling is not possible. Establishment is best on sites with soils and aspect similar to remnant stands of Nassella found on the ranch.

Hand broadcasted the seed. Covered with native or certified weedfree mulch.

Site historically mined with dredges, leaving huge boulders and cobbles. Did biotechnical erosion control in stream for stream bank stabilization (federally listed salmon).

Excellent stand establishment. YST control is still an issue. Stands are reproducing. No management since establishment.

Used everything from heavy equipment to volunteers to successfully restore native grasses in a site where extensive erosion control, soil stabilization and stream bank restoration were conducted.

Irrigated over summer months

Site dominated by Nassella pulchra, to be converted to housing; plants salvaged, moved to mitigation site on nearby old field (2 acres) and disturbed sandy coastal scrub (2 acrease)

Excellent survivorship, but areas between mature plants dominated by common nonnative invasive weeds. Additional Nassella planting (from seed) planned for balance of mitigation sites.

Salvage on a large scale appears to be successful at early stages of project. More weed control before planting would have been helpful, but no time. Mulch mats helped control weeds around mature plants.

some models for restoration. Individual projects throughout the state also suggest that with intensive management, restoration toward native dominance is possible.

Practical Issues in California Grassland Restoration Establishing Goals and an Implementation and Management Plan

F I G U R E 21.2. View from area near Tassajara road and Carmel Valley

road, to west showing complex mosaic of grasslands, chaparral, and oak woodlands in central California Coast Range. Photograph by Mark Stromberg.

(McCullough 1969). Livestock grazing was intense and yearround, likely contributing to the loss of native perennial grasses and changes in soil structure. Over much of California, grazing pressure today is less than during much of the late 1800s and early 1900s, but land management today is often limited by effects of this intensive recent grazing. The role of grazing in California grassland restoration remains controversial (see Jackson and Bartolome, Chapter 17; Huntsinger et al., Chapter 20), but certain grazing practices can be effective tools in restoration, particularly in the roles of reducing exotic species (Marty 2005) and increasing native annual forbs (Hayes and Holl 2003a). Given these constraints, it may be unrealistic to expect the re-creation of the original California grassland flora across extensive landscapes. Goals might be more appropriately focused on restoring a mosaic of patches of native grassland communities (including the associated animal and plant species) using similar aboriginal grassland communities (Keeler-Wolf et al., Chapter 3) as models, and these interspersed with other vegetation types as in the upper Carmel Valley in Monterey County (Figure 21.2). Native grasses can probably be found in almost any area of Mediterranean California that has not been entirely transformed by agriculture or housing, but patch size, even in very large, relatively undisturbed landscapes, can be highly variable (Taylor and Davilla 1986; Huenneke 1989). The scale of this patchy distribution may be fractal (Green et al. 2003), with areas of native grass populations varying in size from a few individuals to 2 hectares. We still have little evidence of large California landscapes dominated by native California grasses (Hamilton 1997a). Restoration and grassland management plans alike need to take this inherent nature of California grassland’s patchiness into account. This complex historical legacy presents significant challenges to restoring California grasslands. However, the widespread distribution and persistence of native species–dominated grasslands in small patches is encouraging and presents

268

POLICY AND MANAGEMENT

The most important step in considering the restoration of a grassland area is to establish broad goals, discrete objectives, and measurable success criteria for the project. Clear goals will guide the initial plan and the implementation process according to the resources available. Measurable objectives will allow the project to stay on track and assist with adaptive decision making as the project site undergoes change. Typical grassland restoration goals in California include increasing native species diversity and habitat protection, control of invasive non-native species, erosion control or soil stabilization on badly disturbed sites, site water management (water quality or water retention), forage quality improvements (See Box 21.2), or aesthetic improvements. Often restoration goals are associated with legal requirements, such as mitigation. Although many restoration projects are spatially delineated, all else being equal, bigger is better. One of the more general guidelines from ecological theory is that species diversity and persistence are positively correlated with the size of a functional habitat (MacArthur and Wilson 1967). However, small patches, even down to the size of gardens, can harbor an amazing diversity of native plants and insects and add interest and beauty to urban and suburban settings (see Box 21.1). The goals and objectives of the project are the foundation of an implementation and management plan. The management plan should include a timeline for all restoration activities as well as a detailed management strategy for many years after the initial project implementation. Many native grasses require at least three years (Bugg et al. 1997) to establish, requiring more intensive weed control during the establishment phase. An accurate establishment evaluation of Nassella pulchra often requires up to seven years from seed to be readily apparent (personal observation, and J. Anderson, personal communication, 2006). Ongoing management (beyond three years) may be less intensive, but nevertheless critical. Ecological restoration of California’s grasslands is a longterm commitment involving inherent scientific, political, social, and economic uncertainties. Because uncertainty is a normal part of any scientific question or human endeavor, it should not prevent effective ecosystem restoration (Lemons 1996). Instead, it should be recognized as part of the decisionmaking process (Clark and Cragun 1994; Brunner and Clark 1997). Decisions in restoration can often benefit from considerations developed in social and policy sciences (Gobster and Hull 2000). Ethical issues in restoration (e.g., client wishes versus practitioner recommendations) should also be considered (Dickinson et al. 2006).

BOX 21.1 GREEN ROOFS

Living grassland on roofs can provide many ecological services in what is otherwise a relatively hard landscape of concrete sidewalks, impervious streets, and roofs. In Europe, the use of “green roofs” has continued for many centuries. Industry figures suggest that 10% of German roofs are greened. In Zurich, Switzerland, a large water storage tank project, covering a 5-acre (2 ha) site, was built 100 years ago. The builders needed soil to cover the concrete tanks for thermal insulation. The top layer of soil from a nearby pasture was excavated and placed on the roof. Now, the roof is home to 175 plant species, including several that have gone extinct elsewhere (Landolt 2001; Bazilchuk 2006). Between 1989 and 1999, German roofing companies installed nearly 350 million square feet of green roofs, and the rate is increasing (Penn State Center for Green Roof Research, http://hortweb.cas.psu. edu/research/greenroofcenter). In 2001 alone, Germany installed 13.5 million square meters (33,400 acres) of green roof (Grant et al. 2003). The history of green roofs can be traced from the hanging gardens of Babylon to the present (http://www.greenroofs.com/Greenroofs101/history.htm). Living roofs are essentially a thin layer of soil (4–12”, 10–30 cm) over a variety of roofing systems (Figures 21.3, 21.4, 21.5). The layer of soil and plants significantly increases the expected lifetime of the roof surface (membrane, concrete, etc.). Green roofs’ advantages include: •

Providing an esthetically pleasing appearance and a natural landscape for relaxation and nature appreciation.

•

Capture of storm water that otherwise would have to be treated in municipal water treatment plants. At Ford Motor Company’s new River Rouge complex, the green roof of about 10 acres absorbs and transpires up to 4 million gallons of rainwater each year. The savings in decreased costs for treating storm drainage made the roof economically viable.

•

Substantial reduction of the urban heat island effect (EPA 2001) and reduction of the heating and cooling costs for the building. Cooling occurs when the water in the roof’s soil evaporates. The mass of plants, water, and soil insulates the roof in the winter.

•

Substantial reduction of noise in the building, as the soil and plants absorb a great deal of urban noise.

•

Metapopulations of native plants from the local flora that in turn can support native insects, birds, and other animals able to get to the roofs to use them for food or shelter. Desirable plants for living roofs should be:

•

Native in the local ecosystems

•

Long-lived

•

Slow growing

•

Tolerant of summer droughts

•

Tolerant of seasonal rainfall and inundation

•

Tolerant of urban pollutants

•

Tolerant of poor soils

•

Capable of providing seasonal flowers

There are some excellent examples of green roofs in California, many of which are using some native grasses. William McDonough, architect for Ford Motor Company, designed a 69,000-square-foot (0.7 hectare) roof on the Gap corporate headquarters in San Bruno (Figures 21.3, 21.4). In San Francisco, the new California Academy of Sciences building will have a green roof of 250,000 square feet (2.5 hectares). If ecologists would seek out architects in the design phase of new buildings with green roofs or green sloping sides, there is a huge opportunity for designing experiments to build and study grassland ecosystems (Felson and Pickett 2005).

F I G U R E 21.3. Green roof on bath house at Esalen Institute, Big Sur,

California. Photograph by Paul Kephart.

F I G U R E 21.4. Coastal terrace grasses on roof of Gap headquaters,

San Bruno, California. Photograph by Paul Kephart.

F I G U R E 21.5. Coastal terrace grasses on roof at Gap headquarters,

San Bruno, California. Photograph by Paul Kephart.

Implementation should consider all the activities required at and around the site. For example, if a landowner or land manager is going to spray or mow a site anyway, try to coordinate that activity with mowing or spraying to support the restoration activity. Or, if you know a weed-clean cropland will be abandoned to a restoration use, then establish the plants there immediately, before weeds have a chance to arrive (Mace site, Table 21.1). If native grasses are going to be incorporated into a riparian or wetland restoration project, make sure the land managers responsible are aware of the specific cycle of management required for grasses, which may differ from other plants at the site.

Site Survey The importance of an initial site survey cannot be understated. A thorough survey for both the biological and physical characteristics of the site not only will establish benchmark conditions against which future stages of the restoration can be compared, but will determine the overall strategy for the restoration itself. Depending on the condition of the site, the site survey may also reveal an appropriate model ecosystem for the restoration (Clements 1934). Knowledge of similar, nearby sites, particularly as they change over several years with varying rainfall, will greatly enhance the choice of a reference site or help to clarify what sorts of “natural successional processes” might be expected to occur at the site regardless of management activities (e.g., whether shrubs are likely to encroach). In general, there are two main strategies for grassland restoration, and each is driven by the site characteristics (Packard and Mutel 1996). If the site has a considerable population or populations of remnant native grassland species and is not completely overrun with weeds, a “passive” restoration strategy can be less intrusive and focus on management (Hayes and Holl 2003a; Bartolome et al. 2004). On the other hand, if a site has no native species (including the native soil seed bank), or almost none, and is heavily infested with weeds, it would be far more effective to plan an “active” restoration from scratch; for example, an expanse of bare soil as free of weeds (including weed soil seed bank) as is practical. The site survey will also determine whether and which permits will be required by local, state, or federal government agencies. If a wetland or stream course is included in the restoration, the permitting process can be more complex. If the restoration is deemed a significant change in the environment, it may require review under the California Environmental Quality Act (CEQA) (Jantz et al., Chapter 23). A query of the California Natural Diversity Database might reveal the presence of listed species. If federally or state-listed species are present, permits and planning will involve both state (California Fish and Game) and federal (U.S. Fish and Wildlife, Office of Endangered Species) agency involvement. Land managers often hire consultants to conduct site surveys, but it is possible to explore cooperative actions with local watershed groups, local resource agencies and extension

offices, or state and federal agencies (Jantz et al., Chapter 23). These include field offices of the Natural Resources Conservation Service, California Fish and Game, California Department of Transportation, chapters of the California Native Plant Society, and others. For a list of restoration projects in California, consult the Web site of SERCAL or CNGA (Table 21.1). These can provide a start for finding a qualified restoration practitioner to evaluate the site. The initial survey should describe the biota of the site. Any native plants on the site or nearby should be listed and considered clues to restoration potential. Ideally, several samples of the soil should be taken from throughout the site for evaluation of the seed bank, as well as for soil tests to determine the soil properties. In practice, such sampling, particularly for soil seed bank composition, is rarely done. Presence of state or federally listed plants and animal species must be reported, and the areas of their occurrence should be treated as remnants and left undisturbed even if they have many exotic species, unless specific permissions are obtained. If the site has a population of native grasses or other native plants, one must decide whether they are abundant enough to warrant on-site management that will both increase their abundance (see subsequent discussion) and not further degrade the site. Grassland communities may have site-specific protocols; for example, coastal dune grassland restoration presents unique challenges and requires unique methods (Pickart and Sawyer 1998; Pickart and Barbour 2006). Abiotic site characteristics that influence site potential include slope and aspect, soil chemistry, texture, rockiness, drainage, and depth, plus local climate, rainfall, and the probability and duration of flooding. These can be critical factors in selecting appropriate plants for the restoration as well as planting and management techniques (for example, rocky sites may preclude drill seeding and mowing with heavy equipment, leaving managers to rely on manual or innovative seeding methods such as native grass straw and grazing; see Dye Creek Project in Table 21.1). Finally, the evaluation should consider practical constraints. Some sites will only have seasonal access. Some sites can be grazed, sprayed with herbicides, or burned, while others will have strict limits on some of these restoration activities. Some sites require vegetation to be low, either for visibility (such as highway interchanges) or to maximize water flow (Yolo Bypass, Mace Site, Table 21.1).

Site Preparation Except in the cases of conversion of clean agricultural fields to native grassland or simple enhancement of existing grasslands, some kind of site preparation is necessary. The extent of site preparation and the techniques chosen will depend on the restoration goals, the results of the site survey, the type of plant material to be introduced (e.g., seed, rhizomes, plug plants, or hay) and on communication with the landowner and the resources available to do the preparation. For most grassland projects, site preparation involves mostly weed

G R A S S L A N D R E S T O R AT I O N

271

control and seed bed preparation. The most common techniques for initial weed control are burning, disking, mowing, mulching, the use of selective or broad-spectrum herbicides, or a combination of these (see DiTomaso et al., Chapter 22; Table 21.1). Postplanting weed control in small sites or at individual (woody) plantings can also include hand-pulling, spot spraying, hoeing, and so forth. Often, glyphosate can be applied at the time of drill seeding (Hedgerow Farms, Citrona Farms, Table 21.1) to eliminate germinating weeds, as it is not persistent and will selectively control growing plants. It is important to begin weed control at a site as soon as possible; even several years prior to planting of native species, depending on the weed species present and the abundance of the weed seed bank (Mace Site, Table 21.1). Soil treatments as a part of site preparation (increasing/decreasing soil fertility or pH, adding microbial elements, or ripping to enhance infiltration) are at early experimental and observational stages. Soil ripping depth was only weakly, but positively, associated with cover of Nassella pulchra in year two in one experiment (Montalvo et al. 2002). Additional trials of soil treatments are needed. It is very important to time preparation actions to maximize their effectiveness. For example, weed control burns should ideally be timed to occur after the annuals have committed to reproduction (and death) but before the seeds have fully matured and dispersed (Moyes et al. 2005). Burning in the spring and early summer provides the greatest control of the noxious weeds medusahead (Taeniatherum caput-medusae) and yellow starthistle (Centaurea solstitialis), respectively (Hastings and DiTomaso 1996). Spring burning is nearly as effective as the more expensive solarization of soils with plastic sheeting, as measured by Nassella pulchra establishment (Moyes et al. 2005). Although burning in the fall will not be as effective in killing annual weeds, it can provide an excellent seedbed, especially for no-till planting methods in the Central Valley (Anderson and Anderson 1996) (Citrona Farms, Table 21.1). Most annual exotic grasses germinate shortly after the first fall rains (October to December), and their seed bank has little carryover into the next year (Marañon and Bartolome 1989; Rice 1989b). Tillage during these months will often germinate additional seeds in the soil seed bank. Several cycles of tillage following the flush of additional weeds germinating can reduce the annual exotic seed bank (Stromberg et al. 2002). Irrigation in the fall, and subsequent repeated tillage of the germinating annual exotics, before the winter rains is another option. Various herbicides can be used as preemergents (see DiTomaso et al., Chapter 22) to selectively reduce annual weeds, yet allow native grass establishment or growth. Controlling weeds that germinate with the native plant materials is far more difficult, as each strategy for weed control must then consider the continued survival of the natives (see DiTomaso et al., Chapter 22). One of the most effective ways to impede invasive weeds in sites where they have been reduced is to establish dense stands of native species (Carlsen et al. 2000; Reever Morghan

272

POLICY AND MANAGEMENT

and Rice 2005; but see Hamilton et al. 1999). It can reasonably be said for many heavily invaded grassland sites that there can be no effective restoration without weed control, and there can be no effective weed control without restoration. Although soil disturbance should be minimized wherever possible, some grassland restorations occur after construction or other activities have already disturbed the site significantly. In such cases, tillage such as disking, imprinting, or deep ripping along contour lines may be required to reduce soil compaction and maximize seed germination (Montalvo et al. 2002). Controlling erosion at such sites may require additional grading and erosion control measures (ABAG 1995). Where ongoing surface erosion is a problem, bioengineering using native grasses and other plant materials in conjunction with geotextiles, willow wattles, and/or straw (Sulphur Creek, Table 21.1) has proven successful in stabilizing soils (ABAG 1995). One of the functional groups that may need to be restored are those soil fungi intricately associated with the root system and known as arbuscular mycorrhizae, or AM fungi (Snyder 2003). Topsoils serve as reservoirs of AM fungal spores and hyphae, but disturbance (compaction, vegetation removal, physical destruction of hyphae, etc.) almost always reduces the diversity and abundance of AM fungi (Allen and MacMahon 1985; Jasper et al. 1989). There are about 150 species of AM fungi (Morton et al. 1995), forming associations with about 70% of the plants worldwide. Nearly all grasses form mycorrhizal relationships. Virtually any AM fungus can associate with a vascular plant species capable of forming arbuscular mycorrhizae (Allen et al. 1995), but local sources of AM fungi are probably best to use, although commercially produced AM fungal inoculum is available (Snyder 2003). Soil assays could determine the inoculum potential at a site, or one could look for any plants present known to be mycorrhizal. If they are still present and vigorous, there is likely a remnant mycorrhizal community (Snyder 2003). Mycorrhizal fungi can be reintroduced onto a site using a variety of methods (Snyder 2003), including culturing soil to amplify inoculum, translocating inoculated plants with some associated soil, or culturing spores in sterile soil (Sylvia and Williams 1992). Although plants that can form arbuscular mycorrhizal associations perform much better with the fungi, evaluating the improved performance of target species on restoration sites where AM fungi have been inoculated remains an area where more research is needed (Peters 2002). Remnant patches of native plants can be salvaged (Rice Ranch, Table 21.1), particularly from sites where they will be destroyed by a planned disturbance (mining, road cuts, construction). They can then be handled like other plant materials. One of the first grassland restoration projects, on the Curtis Prairie in Wisconsin, used this method (Umbanhowar 1992; Howell and Stearns 1993; Sperry 1994). Sod or bunchgrass clumps with a significant root/soil mass can be propagated in a nursery, layered on the ground, irrigated, or covered with straw or shade cloth until transplanted into the grassland restoration site. Salvaging intact grasses ensures

that site-specific plants are used for restoration and will preserve local soil microorganisms as well as some of the soil seed bank. Plants may also be clonally fragmented and numerous “sibling” plants divided, thereby substantially increasing plant material available.

Seed/Plant Material Selection Plant species chosen for the restoration should come from the palette of species that naturally occur on or near the site or can be (with more risk) reasonably inferred to have been native to the site. Also, they should be matched to the site characteristics (see Rice and Espeland, Chapter 11). A database is available for over 300 California native grasses (http://www.dot.ca.gov/ hq/LandArch/grass.html), and it includes historical geography, preferred soil type, elevation, and species characteristics. This could help the practitioner select species for use. For instance, only a few native perennial grasses can tolerate several days of flooding. Some species can tolerate clay soils and periodic inundation; others will only thrive in upland, well-drained soils with generally lower soil moisture. That information can be found on the same Web site. Increasingly restorationists are seeking to include as many “functional groups” of plants (Dukes 2001a; Hooper and Dukes 2004) as possible on a given site and include broad categories such as “deep-rooted perennial grasses,” “deep-rooted biennial native forbs.” “native annual forbs,” “nitrogenfixing forbs,” or “early” and “late” phenology forbs (Lulow 2004). Ideally, an experienced plant ecologist should be consulted to help determine a species list that encompasses the full breadth of functional groups. Although some useful sample species lists for certain common grassland types are available (Anderson and Anderson 1996), these should be tailored to the individual site. One of the most important considerations in choosing a source for seeds or other plant materials is to preserve the genetic integrity of what is potentially a community of locally adapted populations. The number of studies of genetic variation and adaptation in California grassland plants is fairly small (see Rice and Espeland, Chapter 11), but what information is available suggests a fairly large amount of variation between populations of most species. For example, in Elymus glaucus (blue wild rye), the population structure suggests reduced gene flow associated with genetic differences between distant populations (Knapp and Rice 1996). In addition to long-distance genetic differences, there is also substantial within-population variation, perhaps because the plants are not entirely self-pollinating. In Nassella pulchra (purple needlegrass) there is highly restricted gene flow resulting from limited seed dispersal distances (Dyer and Rice 1997a), but across the landscape there are only weak differences between nearby populations, probably reflecting the much greater dispersal distance of pollen. In general, a species occurs as distinct genotypes (or populations) associated with a specific, local environment, and these “ecotypes” (Hufford and Mazer 2003) can interbreed

with other ecotypes. Simply observing a group of individuals with some unique traits in a species (color, size, enzymes) does not define an ecotype, because individuals of many species, indeed the same ecotype, can exhibit a wide variety of physical traits (phenotypic plasticity) yet have little genetic variation. The classic approach to detecting ecotypes (Clausen et al. 1940; Linhart and Grant 1996; Hufford and Mazer 2003) requires common gardens and reciprocal transplants, and this has been done for a few grasses used in California restoration (Dyer and Rice 1997a; Knapp and Rice 1998). Newer genetic analyses using molecular markers (microsatellites, sequencing, etc.) hold the promise of a less labor-intensive method to determine levels of local adaptation; however, the evidence for their value in this regard is inconsistent (Hufford and Mazer 2003). Molecular genetics are still best used with common garden studies, and for the few grassland restoration species so far studied in California, the suggestions for restoration continue to support the use of very local seed sources (Knapp and Rice 1996, 1997, 1998; Rice and Knapp 2000). With regard to more practical restoration, if enough common garden studies were done, as have been done in conifer forest restoration (Kitzmiller 1990), “seed zones” could be mapped (Parker 1992). Seeds for restoration in each zone could be harvested in the same zone, and the zones would roughly match ecotypic zones for each species (Hufford and Mazer 2003). Efforts were made to establish the required common gardens to establish seed zones for native grasses (Amme 2003). Market demand for source-identified seed was very low, and native grass seed is no longer included in the California Crop Improvement Association programs to certify seed sources and founder populations. Because of the lack of common garden studies, we do not always know whether observed genetic variation reflects a history of adaptation to local conditions. We therefore recommend that every effort should be made to preserve genetic differences that are obvious between populations. We also do not know, in most cases, the extent to which nearby populations are related to one another and what a reasonable zone for genetic similarity should be. Studies of the genetics of our native plants in relation to restoration (McKay et al. 2005) have suggested that restoration should be done, as far as practical, with plant material (seeds, rhizomes, etc.) collected as near to the restoration site as possible. A precautionary principle is to collect seed from within the same watershed as the site. If this is not possible, then match climatic and physical conditions (e.g., soil properties) as closely as possible within a reasonable geographic distance. Use plant materials with genetic characteristics (outcrossing, inbreeding, ploidy level) that match those found on the remnant or adjacent ecosystems. Another consideration in selecting plant materials is how much genetic variation to introduce for each of the species to be planted. Intraspecific heritable genetic variation plays a critical role in the potential for further adaptive change in response to new selective challenges. In the face of global environmental change, allowing for evolution seems critical

G R A S S L A N D R E S T O R AT I O N

273

to the persistence of many species. Although we do not know how rapid the local adaptive selection process is for most California native grassland plants, significant changes can occur over only a few generations (Rice and Emery 2003). Thus, it is important to include a reasonable amount of intraspecific variation in source materials; this may not be too difficult with some of our native grasses (Dyer and Rice 1997a), because even local populations show characteristics found across their range. Additional research on the genetic structure of our grasslands’ plants and their rates of microevolution is needed and could provide relevant information for decisions in restoration. Another issue that can be a problem in restoration is genetic swamping (Rice and Emery 2003); this occurs when a small population of a locally adapted ecotype is surrounded and swamped with pollen or dispersal propagules of a different species or ecotype (intraspecific swamping). Most of the concern related to genetic swamping is related to the potential loss of fitness in a population by introduction of nonadapted ecotypes (Hufford and Mazer 2003). An emerging concern is contamination from either other similar species or genes. If transgenic crops and specialty plants are introduced near restoration sites, they may contribute genes that could be troublesome to nearby native plants (Ellstrand 2006). In California for example, “pollen contamination from cultivated walnut may hybridize the (endangered) Hind’s walnut out of existence” (Ledig 1992). Currently in California, genetically modified grasses are being developed by the turf industry (e.g., Agrostis stolonifera, Poa pratensis) to be resistant to glyphosate. Field trials of other grasses indicate that effective pollen dispersal occurs as far as 0.6 mile (1 kilometer) (Ellstrand 2006). These grasses are currently regulated in California (Ellstrand 2006), but if these transgenes escape to unwanted grasses, or if these genetically modified grasses themselves escape from cultivation, California would be faced with yet another weed, and this time the weed would be resistant to one of the most important restoration tools available: the herbicide glyphosate. Often a remnant does not have a large enough population of native plants to provide enough seed for a nearby restoration project. In these cases, it is common to collect some seed on the local site and then provide it to contract growers who will perfom a “seed increase.” The U.S. Department of Agriculture (USDA)’s Natural Resources Conservation Service (NRCS) has many plant material centers, including one at Lockeford, California. Along with other federal agencies (U.S. Forest Service, National Park Service), they have detailed seed collection and seed increase protocols (D. Dyer 2003). Commercial growers of native seed can also be engaged. Typically, the seeds are raised in small gardens, harvested and planted in larger gardens, and then planted in agricultural fields for large-scale seed production. Seeds for the restoration should be intentionally selected from all parts of the production field over the entire period of time that the seeds are viable. This avoids unintended selection, and thus genetic shifts that can occur when the fields chosen for seed increase

274

POLICY AND MANAGEMENT

are in very different environments than the restoration site. Seed increases should also be done in a production field as physically close and similar (soils, elevations, etc.) as possible to the native seed sources (McKay et al. 2005). Seeds or plant material from commercial growers (CNGA 2003) should be correctly labeled with regard to seed source sites and collection dates, be free of weed seeds, and have a “percent live seed (PLS)” rating, or at least informal germination tests to ensure viability (see subsequent discussion). Because PLS can vary for the same species among years and among growers, restorationists should always ask for current PLS data from their supplier. Seed companies in California and elsewhere may have special “reclamation” seed mixes that represent statewide collections of particularly vigorous selections, which may not be suitable to individual sites in terms of either species or ecotypes. Care should be taken to obtain site-specific seed sources from seed vendors.

Planting Technique The choice of planting techniques will depend largely on the size and nature of the project, the budget, and local site conditions (Robins 2002a). Small projects can be effectively planted entirely with small transplants or “plugs” (Discovery Park, Table 21.1). Larger areas can be planted using vehicles (tractor-operated seed drills or hydroseeders). In either case, the planting must be properly timed so that plants, whether from seed or from plugs, are able to take advantage of available rainfall to become fully established prior to summer drought. Irrigation, common for individual plantings of woody species, is rarely done for broadly planted grasses. In general, restorations using seed are more successful if planted after the first germinating rain in the late fall or early winter (Colusa, Table 21.1). If seeds are planted before the rains arrive, many are lost to seed predators, or they lose viability as they dry out. Waiting until after the first germinating rains also allows managers to eradicate the first flush of weeds prior to or just after seeding native grasses. For instance, a site can be tilled after the first germinating rain, but just before planting. If the site is planted before the first germinating rain, herbicides (2,4-D, glyphosate, etc.) can be applied if there is a window between the germination of weeds and that of the native plants or if the herbicide can be selectively applied to weeds only, for example, using wick applicators to swipe taller weeds (Llano Seco Project, Table 21.1). Planting from seed in spring is likely to be unsuccessful because many native grasses and forbs require a longer period of warm and wet soils for germination and establishment than is generally available in the spring, and some forbs may require a cold stratification period. Using plugs is cost effective in smaller projects (
2 acres) or in those in which establishment must be rapid. Both grasses and forbs can be established from plugs (Brown and Bugg 2001). One benefit of plug planting is a very high (often
90%) survival rate (Cunliffe and Meyer 2002; Corbin and D’Antonio 2004b; Huddleston and Young 2004), perhaps

offsetting its higher cost. Another advantage is that plug planting can be done later in the season (mid- to late-winter), allowing site managers to eliminate (by spraying or tillage) a larger component of weed cover. They can also be used in conjunction with pre-emergent herbicide that would otherwise inhibit native grass seed. Grass plugs can be contractgrown at greenhouses using locally collected seed or even rhizomes from obligate vegetative reproducing species. Alternatively, plugs may be ordered from a seed company using commercially available native grass seed. Typically, plugs of grass are planted in arrays 12 to 18 inches (30–46 centimeters) apart to yield an appropriate final density desired for the adult plants (Stromberg and Kephart 1996; Huddleston and Young 2004). However, some situations such as erosion control projects may require higher initial density. Planting from seed is generally less expensive and therefore more appropriate for larger sites (Robins 2002a). Seed can be broadcast with hand-held or all-terrain vehicle (ATV)-towed rotary seed spreaders and then lightly covered with soil and compacted using harrows or ring-rollers. An example of this kind of planting is the restoration of Holsclaw Levee in the Colusa area, under the supervision of Jessica Groves, NRSC, USDA. Seed can be sown with 8-foot or 12-foot wide tractortowed seed drills. Seed germination is generally higher if the seeds are drilled shallowly into the soil and covered (Stromberg and Kephart 1996; Kephart 2001; Stromberg et al. 2002). Seeds should not be buried deeply (e.g., with a disc), but good seed/soil contact is required. Cross-drilling the area (seeding twice, but with the second drill lines at right angles to the first) will improve the seed distribution and seed/soil contact, but the drill should then be calibrated appropriately. Several no-till drills have been manufactured (e.g., Truax) that can cut through litter and sod, open a slot in the soil at a set depth, drop the seeds in, and roll the soil back tightly over the slot. Very light or fuzzy native grass seed can be mixed with coarse vermiculite or wheat bran to improve flow through the drill. Seeds with large awns, such as Nassella spp., Hordeum brachyantherum, or Elymus multisetus, often need to have the awns removed before sowing, but de-awning can cause higher seed mortality in long-term storage through desiccation. When planting from seed, it is very important to determine both the appropriate overall seeding rate and the relative proportions of seeds from different species in the seed mix (Russian Ridge, Table 21.1). Overall seeding rate will vary depending on the method and the objectives of a particular site. Drill seeding rates are generally lower than broadcast seed rates because drilling usually results in better seed/soil contact and therefore higher germination rates. Drill seeding rates can be adjusted by calibrating the seed drill (drive the drill over a swept hard surface or tarp, weigh or count the number of seeds being dropped, and adjust the drill accordingly) and are usually set between 12 and 20 lbs/acre (see Table 21.1). Broadcast seeding rates are mostly determined visually by the experience of the applicator and may be as high as 35 pounds per acre. Determining the relative proportion of different species in a mix, however, is complicated

by the fact that species vary widely in the PLS per pound, larger-seeded species tend to have greater success per seed than smaller-seeded species (Lulow et al. 2007), and some species are “fast-starters” and tend to dominate early while others may have delayed germination. A typical seeding density for California native perennial grasses with large seeds (e.g., Nassella, Elymus) is 60 live seeds per square foot (600/m2). When seeding a single species, the calculation of pounds per acre needed to yield roughly 60 live seeds per square foot is relatively straightforward using the PLS number multiplied by 43,560 square feet per acre. However, to achieve a mixed-species stand in which the relative density of the species varies according to the desired species mix in the resulting stand, the calculations are complicated by the fact that species vary widely in their PLS rating (for example, Nassella lepida can have 10 times as many live seeds per pound than N. pulchra), and this number will vary between growers and between years. Experienced grassland restorationists often use spreadsheets in which PLS ratings are used to calculate ratios of pounds of seed in a mix. The most important thing to remember, however, is to base the seed mix on the current PLS rating, not simply on pounds per acre. Hydroseeding native grass is an option, especially on slopes that are too steep for equipment or on sites with very wet soils that preclude access by tractors with drills. Most effective hydroseeding is done in two passes. The first pass sprays seed and water, often with a very light binding agent to ensure good seed/soil contact. The next pass will typically distribute 3,000 to 4,000 pounds per acre of straw, compost, or 3/8” chipped wood (CalTrans 2004). Mulch increases available soil moisture, provides thermal insulation (night radiation cooling), and is often required as a soil erosion control measure. If the seed and mulch are mixed in one pass, much of the seed is suspended above the soil in the mulch. Subsequent wetting- drying cycles will see the mulch expand and shrink, uprooting the grass seedlings. Swathing and baling native grass when it has mature seeds, either in seed production plots or in particularly dense native grass stands, allows one to move native grass seeds in bales and then plant them by spreading the native straw (Dye Creek, Diablo Canyon, Table 21.1). This method is particularly useful for rocky sites where traditional seeding methods are impossible (Dye Creek, Table 21.1). In bales of Nassella pulchra straw, the seeds retain their awns, which naturally “drill” the seed into the ground as the awns curl in response to changes in moisture. Such native straw also provides mulch and erosion protection for the seedlings (Hujik 1999) and may provide seeds of additional native species from diverse source communities. Native hay may spread weedy species, so care should be taken to harvest in areas relatively free of weeds.

Erosion Control Sites on slopes or on historically eroded areas will be vulnerable to soil erosion during winter storms. Many erosion control

G R A S S L A N D R E S T O R AT I O N

275

methods have proven effective in limiting loss of planted seed to erosion (Goldman et al. 1986; ABAG 1995). Generally, erosion is most severe after the soils are saturated in the late winter. If the seeding was done in the fall at the time of the first germinating rains, by late winter a relatively good mat of seedling roots should be present to prevent minor erosion if the soil was not compacted. If the restoration site is on a slope, control of run-on from the slopes above the restoration site (ditches, drains, etc.) may be required. Erosion control measures should be in place by mid-October to prevent largescale soil losses. In California, restoration projects subject to erosion might be deferred if a strong El Niño winter is predicted. Erosion can be mitigated by changing length of slope, steepness of slope, and vegetative mulch to promote water infiltration into the soils. On highly disturbed soils, incorporation of organic matter is critical for plant growth (Curtis and Claassen 2005). Erosion control measures (Discovery Park, Table 21.1) often include installation of filter cloth or fiber mats/blankets pinned to the slope, contoured furrows, fiber rolls, vegetated gabions, application of thatch or tub-ground wood shreds, woven willow check dams, and similar devices (ABAG 1995).

Long-term Site Management Whether a restoration consists of planting new material or is based on managing what is already there, long-term site management is the key to successful grassland restoration. Unfortunately, it is typically the most challenging part of a restoration project, because of the intensely competitive environment created by exotic species. In restoration sites where the primary weeds are annuals, managers can use the differences in life history and growth characteristics between annuals and perennial species to promote the native species. However, if the primary weeds are exotic perennial grasses (such as sites dominated by Holcus lanatus, Festuca arundinaceae, or Ammophila arenaria), other strategies are necessary and in general have not been developed. Most of the annual, exotic grasses in California produce far more seeds than needed for replacement (Young and Evans 1989) and far more seeds than native perennial grasses. In a typical grassland the germinable seed pool is typically dominated by non-native species (see, e.g., Major and Pyott 1966) that germinate earlier than native perennials and form dense stands early in the growing season (Deering and Young 2006). Exotic annual grasses can limit the growth and survivorship of native seedlings (Brown and Rice 2000). Higher soil N availability, which may characterize some restoration sites, often favors annual exotic species over native perennials (Huddleston and Young 2005; Corbin et al. 2006). Because of the intense competitive environment created by the high density of exotic annual species (Dyer and Rice 1997b; Hamilton et al. 1999; Brown and Rice 2000), management treatments have focused on trying to shift this competitive balance. This can be achieved by trying to force nitrogen immobilization through additions of carbon to soil,

276

POLICY AND MANAGEMENT

timed mowing, timed livestock grazing, prescribed fire (Corbin et al. 2006), herbicide application, and supplemental irrigation. Here, we briefly review these various management options. Addition of nitrogen to soils during restoration in California grassland sites should be avoided (Brown et al. 2000). Although productivity of natives may increase with additional N, there are often even greater increases in weed density and biomass (Brown et al. 2000). In grasslands where N availability has been increased by fertilization, N-rich mulch (Huddleston and Young 2005), or the presence of some nitrogen-fixing invasive plants (e.g., Genista, Lupinus), there are large-scale shifts in grassland composition from native perennial to exotic annual species (Maron and Connors 1996). This has led some restorationists to suggest the impoverishment of soils as a restoration technique, through topsoil removal, burning, or carbon supplementation. At sites where native grasses have been seeded on N-poor subsoil resulting from mine reclamation or other excavations, native species have tended to perform well because of lower weed competition (see for example, Maxwell Flat project, Table 21.1). One way to decrease nitrogen availability to plants is to add carbon to provide an energy source for increased microbial growth and associated nitrogen uptake by microbes. Sawdust is a low-cost carbon source that is widely available for potential use in restoration. In coastal dune systems enriched by N from Lupinus, addition of sawdust to change the C:N ratio reduced the annual invasive grasses, but increased the frequency of both native and non-native forbs (Alpert and Maron 2000). In general, increasing the ratio of C:N with the addition of sawdust or other carbon sources on restoration sites has had little long-term effects in efforts to shift the competitive advantage to native perennial grasses in California (Corbin and D’Antonio 2004a; Huddleston and Young 2005; Haubensak and D’Antonio 2006), but more research may be productive. In tallgrass prairie restoration, there have been some successful reductions in competition with exotics by changing the C:N ratio (Averett et al. 2004). Three conditions must be met for carbon supplementation to be an effective tool in prairie restorations (Blumenthal et al. 2003): Weeds must suppress native species in the absence of C addition, weeds must be nitrophilic relative to native species, and C addition must result in a decrease in available N sufficient in magnitude and duration to alter the balance of competition between native species and weeds. There is some evidence that exotic invasive species leave a biochemical legacy of toxins in the soil that prevent germination of native grasses (Robinson 1971; Callaway and Aschehougdagger 2000). If so, the addition of activated carbon (charcoal) may be an effective soil amendment for restoration (Kulmatiski and Beard 2006), particularly in abandoned agricultural fields. However, this has yet to be broadly tested in restoration settings. Mulch on restoration sites should be used carefully. Mulch is detrimental to establishment of native annual wildflowers (Hayes and Holl 2003a) but can improve biomass accumulation

in the establishment phase of a restoration site (Curtis and Claassen 2005). Mulch (or standing biomass from the previous growing season) also provides some winter thermal insulation that promotes establishing seedling grasses and reduces soil erosion (Jackson and Bartolome 2002). Mulch should be as free as possible from weeds and should decompose slowly. Rice straw is often used in restorations for these benefits (Brown et al. 2000), and it is widely available in the Central Valley. Mowing as a tool for restoration in California grasslands may be valuable only in certain conditions (Hayes and Holl 2003b). Mowing has been effective in controlling non-native grasses in some cases (Maron and Jefferies 2001; Wilson and Clark 2001). Mowing might control exotic annuals through several mechanisms. Mowing after exotic annuals have immature seed, but before the native perennials have initiated annual growth, can control some non-native annuals such as yellow starthistle, Centaurea solstitialis (Benefield et al. 1999). In some sites, frequent mowing favors nonnative forbs over exotic grasses but has no effect on native grasses in the short term (Hayes and Holl 2003b). Because the soil seed bank of annual exotics is generally not long-lived (Marañon and Bartolome 1989; Rice 1989b), repeated welltimed mowing might favor native perennials, but more studies are needed. In restoring old fields, swathing (mowing with a sickle bar and leaving windrows) the elongating stems and flowers of weeds (wild oats, yellow starthistle, medusahead, etc.) just before seed maturity allows either baling and removing much of the seed, or fall fires hot enough in the windrows to kill the seeds (Hedgerow Farms, Table 21.1). Mowing is generally not effective in shifting the competitive balance toward native perennial grasses if non-native perennial grasses are the primary source of competition, as in many coastal prairie locations (see D’Antonio et al., Chapter 6). More research is needed to determine whether long-term mowing programs are effective in reducing available soil N in degraded perennial grasslands (Corbin et al. 2006). Grazing as a tool for restoration is controversial (see Huntsinger et al., Chapter 20). Goats, sheep, or cattle are used to selectively “mow” exotic annual species, which are often more palatable than native perennials early in the growing season. Grazing animals can also be chosen depending on their forage preferences. Cattle, for example, generally select grasses, whereas sheep (or goats) may be used to target problem forbs (Russian Ridge, Table 21.1). Grazing may also be used to target thistles later in the growing season. Grazing must be carefully timed and intensive enough (high density of animals) to force selective removal of exotic annuals (Huntsinger et al., Chapter 20). Such grazing can reduce exotic grass cover and thus favor native plants, as has been done in Bay Area serpentine grasslands to favor native forbs that support rare butterflies (Weiss 1999) and in a wide array of rangeland projects (see Table 21.1). Livestock grazing in central valley grasslands improved species richness in native plant species and aquatic invertebrates in vernal pools (Marty 2005). Livestock grazing, or vegetation clipping and removal, has had variable results in regard to increasing native grassland species (Hayes and Holl

2003b; Corbin et al. 2006), and additional, controlled experiments are needed. If used correctly, selective post-emergent herbicides may remove broadleaf forbs or annual grasses without harm to perennial grasses (see DiTomaso et al., Chapter 22). Some preemergent herbicides are also effective on broad-leafed plants but not grasses (Lanini et al. 1996), but all herbicides should be used carefully in a planned restoration that may include native broad-leafed plants. Herbicides are generally the most costeffective method of those discussed here in restoration projects in which stand-dominating exotic invasive weeds are to be selectively removed (DiTomaso et al. 1999b; Kephart 2001). The long-term and unintended environmental effects of herbicides are variable, and manufacturers are increasingly trying to develop more short-term, targeted products. Nonetheless, many land managers still restrict herbicide use. Prescribed fire is often a difficult option because of air pollution and safety considerations (Russian Ridge, Table 21.1), but it has been effective in some instances in restoring native grasslands (DiTomaso et al. 1999a; Kephart 2001; Kyser and DiTomaso 2002), especially when a particular weed such as Centaurea solstitialis is targeted (Keeley 2006). Site managers need to be aware of the life history strategies of all species guilds at the site in order to avoid inflexible prescriptions that may benefit one native guild (e.g., grasses) at the expense at of others (e.g., forbs; see Reiner, Chapter 18). For example, research at the Santa Rosa Plateau suggested that repeat spring burning over three years increased native grass cover and frequency while reducing that of annual forbs, even desired native forbs (Wills 2000). Management plans should vary treatments among years to more closely mimic the stochasticity inherent in natural disturbance regimes. Some weeds that are highly susceptible to fire, such as medusahead, may reinvade burned sites within 3–5 years (Ranchette 1, Table 21.1). In such cases, managers should be prepared to commit to a long-term site fire management plan or other long-term strategies. For more detailed discussion of the role of fire, see Reiner, Chapter 18. Where more certainty and speed in restoration of a grassland are required (after road construction, urban development, or landscaping), irrigation may provide an attractive alternative to depending upon the erratic California rainfall. To best promote native perennial grasses, irrigation should mimic natural patterns in rainfall that favor native perennial grasses (Jackson and Roy 1986). For example, years with early saturating (fall) rains followed by prolonged winter drought favor perennials and are associated with reduced total biomass of exotic, annual grasses. However, these kinds of winters can be devastating on newly seeded restoration sites. By contrast, years with continual rainfall following fall wet-up are associated with relatively high standing crop of exotic, annual grasses (Pitt and Heady 1978) and are the kinds of winters that increase the success of restoration plantings. Timing and amount of rainfall can be more closely correlated with establishment and growth of native, perennial grasses than any grazing or burning treatments (Marty et al. 2005).

G R A S S L A N D R E S T O R AT I O N

277

Sometimes the establishment and management of different functional groups can require conflicting management actions. For instance, the a broadleaf-specific post-emergent herbicide may be applied only over establishing native perennial grasses before native forbs are introduced to the restoration. A more complete restoration may take several years of planting and management, adding a new functional group at each cycle. This is complicated by the fact that many grassland restoration sites in California that have lost most or all of their native grasses still have substantial native forb diversity (Lulow 2004; Keeler-Wolf et al., Chapter 3). In these situations management options include fire, timed herbicide application, and timed mowing or grazing.

Monitoring Monitoring responses to treatments as well as changes going on in reference communities is critical to continued management of a restoration project. There are a multitude of monitoring procedures and protocols that can be tailored to the objectives and resources of a given restoration project (Elzinga et al. 1998). Monitoring protocols are typically laid out along with the specific objectives of the project and should be designed to be able to determine whether specific objectives are being met. SER recommends monitoring of a wide range of ecosystem properties. Yet a review of over 460 restorations described in the journal Restoration Ecology showed that only 68 projects evaluated success after planting or treating sites using SER suggestions to measure diversity, vegetation structure, or ecological processes (Ruiz-Jaen and Mitchell Aide 2005). This proportion is likely even lower among projects that are not described in scientific journals. Because diversity is one of the most commonly measured attributes, it is worth a brief discussion. There are different mathematical measures of diversity including the most simple measure, species richness. Each measure depends on the area sampled (Gotelli and Colwell 2001). A reasonable scale for comparing diversity among California grassland sites is to sample 1 meter  1 meter (Stromberg et al. 2001; Harrison et al. 2003) plots. An appropriate area of the restoration site should be sampled using a stratified, random design and recording species presence and cover in the 1 m2 quadrats. Vegetation structure and change over time can be monitored with a variety of indices including plant cover, plant density, biomass, plant height, or a combination of these. Data should be collected and recorded, not only for the restoration site but for a comparable site that is not being manipulated and may be the reference site used for the particular restoration. Such comparisons to reference sites could be included as a contractual obligation. Photographic monitoring from fixed locations is a relatively simple, cost-effective measure that should be instituted at all restoration sites in addition to quantitative data collection (Merenlender et al. 2001; Hall 2002), although it should not substitute for quantitative methods. Sometimes paired photographs (Figure 21.1) can be dramatic tools in explaining complex changes over

278

POLICY AND MANAGEMENT

time. Additional ecological processes that can be measured using relatively well-developed methods (Bonham 1989; Elzinga et al. 1998; Knapp et al. 1998) include describing changes in soil processes, animal use of sites, herbivory on planted species, seed dispersal into sites, pollination, predation, and parasitism. Hundreds of restoration projects that are currently under way could be transformed, with relatively little modification, into experiments in local adaptation (McKay et al. 2005). However, to do so, information would be needed on (1) where the plant material came from, (2) where it was planted on the site, (3) how it performed, using some index of individual or stand-level plant performance, and (4) site environmental characteristics. If metadata and quantitative measures of ecological processes and composition were available from a large number of sites, it would be possible to infer a great deal more about what influences success of particular restoration techniques (Young 2000; Young et al. 2005). Finding and retrieving metadata (Michener and Brunt 2000) in ecological research requires cooperation and participation of restoration ecologists. A data registry has been developed for discovering and sharing information on restoration projects in California (NRPI 1997), and readers are urged to register their restoration projects. In addition, the U.S. Fish and Wildlife Service (USFWS), The Nature Conservancy, the USDA, and others are compiling records of restoration projects in California (see Table 21.1). If restoration ecology is to develop into a mature science, such monitoring and reporting are critical, especially when combined with thoughtful initial restoration designs.

Managing Remnants When a system has a significant natural component that is still functional, the restoration may include a combination of management techniques without the resource-intensive step of planting new vegetation or intensively managing invasive species. The intensity of these measures will vary depending on the ecological integrity of the site, the weed species present both on site and in the nearby area, and available resources. In any case, management strategies should be long-term and flexible to adapt to changing conditions. Although intensive planting is usually not necessary at such sites, many restoration methods can be used to augment both the grasses and the forbs of a remnant. The success of each of these methods has varied (see Reiner, Chapter 18 and Huntsinger et al., Chapter 20), and additional monitoring of the trials of these management tools is needed. We encourage restoration practitioners to implement long-term monitoring programs, record observations of what works, and write up their observations in the publications of the CNGA or SER. There are two special cases in which restoration or recreation of native California grasslands will occur in places that the citizens of California use or see on a daily basis. These are large roof surfaces (Box 21.1) and roadsides (Box 21.2). Although such small-scale patches may be viewed as artificial, they still fulfill some functions desired in ecological

BOX 21.2 FARM EDGE AND ROADSIDE RESTORATION

Highway rights-of-way are almost all covered with noxious, invasive weeds (Wrysinski 2002). California’s Department of Transportation, at various regional offices, has supported research and demonstration plots for restoring roadside vegetation to a stable state that requires little annual management, and it is committed to trying to use native plants along new or replacement state highways. The California Department of Transportation maintains a native grass database and restoration guidelines (http://www.dot.ca.gov/hq/LandArch/index.htm) and supports many university-level research projects in grassland restoration (Brown and Rice 2000). Replacing weedy roadsides with native grass buffers between roadsides and farm fields (Figure 21.6) is an alternative to typical roadside management regimes of annual grading, and spraying (Anderson and Anderson 1996; Robins 2002a). By restoring most of the road edge area to relatively long-lived native perennial grasses and halting the annual sources of disturbances (grading, etc.), farm edges can be improved dramatically and may resist reinvasion. Although long-term studies of roadsides revegetated with native grass have not been done in California, field research at the University of California at Davis has shown that Elymus glaucus can successfully maintain dominance even when seeds of yellow starthistle are introduced into the plot (unpublished data; Joe DiTomaso, personal communication, 2006). Yolo County probably leads the state of California in establishing native grassland vegetation along irrigation ditches (Figure 21.7), between road surfaces and farmed fields, and in tailwater ponds. The county has published a useful landowner handbook (Robins 2002a), also available online (http://www.yolorcd.org/library/index.shtml), which describes in detail successful methods to convert ditch edges and road edges to native vegetation, often with native grasses. There is an amazing opportunity here; there are over 10,000 miles of roadside, ditches and levees in California’s central valley (Steve Greco, UC Davis, Landscape Architecture, personal communication). Detailed and useful instructions are included for landowners seeking agency cost sharing (Robins 2002b) for restoring hedgerows, tailwater ponds, and road and ditch edges. Conversions of these edges to strips of native grassland provide a wide variety of ecological services, including: •

Reducing annual costs of “clean” agriculture associated with removing all vegetation from road and canal edges.

•

Providing weed-free areas adjacent to agriculture and roads

•

Providing habitat for beneficial insects and a wide variety of wildlife

•

Reducing soil erosion, stabilizing earthen levees

•

Improving water quality through biological filtering

•

Recharging groundwater

Roadsides and levee restoration are widespread in California, and observant readers with time to stop— or the ability to do grass species identification at 60 miles per hour—can appreciate these restoration efforts as they travel through California.

F I G U R E 21.6. Roadside planted with native grasses and, after

F I G U R E 21.7. Irrigation ditch planted with native grasses for

several years of establishment, native forbs drilled to increase diversity. Photograph by John Anderson.

wildlife, soil stabilization, and reduced annual maintenance costs. Photograph by John Anderson.

restoration, such as supporting a diverse array of successfully reproducing native plant species, reducing local weed populations, and supporting insects and other higher trophic levels. Also, they can be used as a resource for public education about grassland biodiversity and its values.

Future Challenges Restoration of California’s grasslands is proceeding along many fronts. This review of the current field has revealed some major knowledge gaps, and research needs. These include: •

280

Better definitions of success and clearer, broader-scale criteria for evaluating characteristics of “restored” sites. This would allow evaluation of the extent to which grassland restorations in California comply with SER goals (SER 2004).

POLICY AND MANAGEMENT

•

Better tracking of restoration projects, both to measure success and to add to the knowledge base regarding causes of successes and failures. To do this, it is necessary to implement long-term collection and compilation of monitoring data and metadata.

•

More information on genetic variation and local adaptation in native species on which to base seed collection criteria.

•

Strategies to engage private landowners in long-term management approaches that favor native grasslands across the landscape.

•

Evaluation of which restoration techniques are both practically effective and cost-effective.

•

More detailed evaluation of soil processes and their effects on restoration.

TW E NTY TWO

Exotic Plant Management in California Annual Grasslands J O S E P H M. D I TO MAS O, STE P H E N F. E N LO E, M I C HAE L J. P ITCAI R N

California grasslands have experienced tremendous changes in vegetative composition over the last few centuries, with significant declines in native perennial vegetation and concomitant increases in aggressive, non-native, weedy species. The changing grassland composition is the result of numerous processes that are reviewed elsewhere in this volume, including many plant introductions, primarily from Europe and Asia (Heady 1977). Burcham (1956) characterized four waves of invasion that were dominated by European annual grasses and shallow-rooted winter annual forbs. In the last half century, a fifth wave of annual invaders, characterized by a later season phenology and deeper rooting pattern, has moved across the grasslands with unrelenting progress. However, individual species from all of these waves of invasion still dominate large grassland areas, with wide population fluxes from year to year, depending upon environmental conditions and local- and landscape-scale processes (Pitt and Heady 1978). As observed in every wave of invasion, probably the most striking shared characteristic of invaders in California’s valley and coast range grasslands has been the success of the annual life history. This chapter will focus primarily on annual grassland systems occupying the coastal ranges, central valley, and Sierra Nevada foothills. To be consistent with terminology elsewhere in the book, annual grassland systems will be referred to as “valley grassland” systems. In contrast, some of the more dominant invasive species in coastal prairies are perennial grasses and shrubs (see D’Antonio et al., Chapter 6). Similarly, perennial species represent the vast majority of invasive non-native species in wetland or riparian areas within grassland settings. Over 73% of the major invasive non-native species in valley grassland are winter annuals, about 11% are biennials, and 16% typically act as perennials, although members in both of these lesser groups can sometimes act as biennials or annuals. Although some woody species, such as Himalaya blackberry (Rubus armeniacus) and the native junipers

(Juniperus spp.), can encroach into grasslands, they are not considered as significant a threat as herbaceous species in valley grasslands. In some coastal range grasslands, including some that are annual-dominated, it is not uncommon for invasive shrubs, such as the brooms (Genista monspessulana, Cytisus scoparius, Spartium junceum) or gorse (Ulex europaeus), to invade these systems (see Chapter 6). Of the 44 most commonly encountered invasive valley grassland species in the state, about 80% belong to either the sunflower (Asteraceae) or grass (Poaceae) family, with an equal distribution between the two families (Table 22.1). Other important families of non-native plants in these grasslands include the Brassicaceae, Geraniaceae, and Fabaceae. Of the interior valley grassland invasive plants recently listed in the “California Invasive Plant Inventory” published by the California Invasive Plant Council (Cal-IPC 2006), eight species are classified in the “High” concern category (Table 22.1). These are considered the species having the greatest impact on valley grassland habitats and include yellow starthistle (Centaurea solstitialis), artichoke thistle (Cynara cardunculus), Scotch thistle (Onopordum acanthium), barb goatgrass (Aegilops triuncialis), red brome (Bromus madritensis ssp. rubens), downy brome (Bromus tectorum), and medusahead (Taeniatherum caput-medusae). For close-ups showing distinctive features of yellow starthistle, barb goatgrass, and medusahead see Figures 22.1 and 22.2. All of these species belong to either the sunflower or the grass family. Among the remaining species occurring in valley grassland on the Cal-IPC list, 48% are listed as “Moderate” in their impacts, and 34% are categorized as “Limited.” Among the species listed as “High” in their impacts, downy brome and yellow starthistle are considered the two most invasive species in the 17 western states, with 56 and 15 million acres infested, respectively (Duncan and Clark 2005). Medusahead is considered the ninth most invasive species in the western United States, with an estimated infestation of 2.4 million acres. Among the most problematic

281

TA B L E 22.1 Most Common Non-native Invasive Species in California Valley and Foothill Grasslands, Including their Growth Form and Cal-IPC Classification

Common name

Scientific name

Family

Growth habit

Cal-IPC category

Fennel Italian thistle Slenderflower thistle Woolly distaff thistle Purple starthistle Malta starthistle (tocalote) Yellow starthistle Squarrose knapweed

Foeniculum vulgare Carduus pycnocephalus Carduus tenuiflorus Carthamus lanatus Centaurea calcitrapa Centaurea melitensis Centaurea solstitialis Centaurea virgata var. squarrosa Chondrilla juncea Cirsium vulgare Cynara cardunculus Hypochaeris glabra Hypochaeris radicata Onopordum acanthium Picris echioides Senecio jacobaea Silybum marianum Isatis tinctoria Medicago polymorpha Trifolium hirtum Erodium botrys Erodium brachycarpum Erodium cicutarium Hypericum perforatum Aegilops triuncialis Aira caryophyllea Avena barbata Avena fatua Briza maxima Briza minor Bromus diandrus Bromus hordeaceus Bromus madritensis ssp. rubens Bromus tectorum Cynosurus echinatus Dactylis glomerata Festuca arundinacea Hordeum marinum Hordeum murinum

Apiaceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae

Perennial Winter annual Winter annual Winter annual Annual to perennial Winter annual Winter annual Perennial

High Moderate Limited Moderate alert Moderate Moderate High Moderate

Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Asteraceae Brassicaceae Fabaceae Fabaceae Geraniaceae Geraniaceae Geraniaceae Hypericaceae Poaceae Poaceae Poaceae Poaceae Poaceae Poaceae Poaceae Poaceae Poaceae

Biennial Biennial Perennial Winter annual Winter annual Biennial Annual or biennial Biennial Winter annual Biennial Winter annual Winter annual Winter annual Winter annual Winter annual Perennial Winter annual Winter annual Winter annual Winter annual Winter annual Winter annual Winter annual Winter annual Winter annual

Moderate Moderate Moderate Limited Moderate High Limited Limited Limited Moderate Limited Moderate Not listed Not listed Limited Moderate High Not listed Moderate Moderate Limited Not listed Moderate Limited High

Poaceae Poaceae Poaceae Poaceae Poaceae Poaceae

Winter annual Winter annual Perennial Perennial Winter annual Winter annual

High Moderate Low Moderate Moderate Moderate

Poaceae Poaceae

Winter annual Winter annual

Moderate High

Poaceae Poaceae Scrophulariaceae

Winter annual Winter annual Winter annual or biennial

Not listed Moderate Limited

Rush skeletonweed Bull thistle Artichoke thistle Smooth catsear Common catsear Scotch thistle Bristly oxtongue Tansy ragwort Blessed milk thistle Dyer’s woad California burclover Rose clover Broadleaf filaree Shortfruited filaree Redstem filaree Common St. Johnswort Barb goatgrass Silver hairgrass Slender oat Wild oat Big quakinggrass Little quakinggrass Ripgut brome Soft brome Red brome Downy brome (cheatgrass) Hedgehog dogtailgrass Orchardgrass Tall fescue Mediterranean barley Hare, smooth and wall barley Italian ryegrass Medusahead Squirreltail fescue Rattail fescue Bellardia

Lolium multiflorum Taeniatherum caput-medusae Vulpia bromoides Vulpia myuros Bellaria trixago

F I G U R E 22.1 Close-ups of flowerheads of two common invasive Centaurea species in California. Left: C. melitensis (tocalote or Malta starthistle). Right: C. solstitialis (yellow starthistle). Note that the spines on C. solstitialis are approximately twice the length of those found on C. melitensis.

invasive plants in California’s valley grasslands, the two most common species are yellow starthistle and medusahead (see Figures 22.1, 22.2). Downy brome, despite its widespread western impact, has caused significant problems only in the northeastern (Modoc Plateau) part of the state, so it will not be discussed further here.

Impacts Invasive non-native plants in California grasslands can have significant ecological and economic impacts. Although a more thorough treatment of ecological impacts can be found in Chapter 6, a number of key impacts will be discussed here. Since the California grasslands have historically played a major role in livestock production, weed impacts have often been strongly associated with that industry. Impacts include interference with grazing practices; reductions in forage productivity or quality; increased costs of managing and producing livestock; reduced animal weight gains; reduced quality of meat, milk, wool, and hides; and livestock poisoning. Medusahead, for example, is of low value because of its high silica content (George 1992). This reduces the forage quality and makes it less palatable to livestock and wildlife compared to other forage grasses. In areas heavily infested with medusahead, livestock carrying capacity can be reduced by as much as 75 to 80% (Major et al. 1960; Hironaka 1961; George 1992). One of the more recent animal issues to come to light is the impact of many invasive plants on the horse industry, especially where small acreage development is occurring. Invasive forbs such as yellow starthistle, Russian knapweed, houndstongue, and tansy ragwort can result in poisoning and death to horses when consumed at high levels (Cordy 1954).

Direct impacts of grassland weeds on humans and quality of life include increased allergens in the air, sickness or death from inadvertent consumption of poisonous plants, damage to recreational equipment, and injury or discomfort from physical contact with spiny thistles such as milk thistle and yellow starthistle or abrasive parts such as the awns of ripgut brome (Bromus diandrus). With increased use of grasslands for recreational activity, including hiking and biking, direct impacts on humans are now much more prevalent (DiTomaso 2000). In addition to impacts on humans and animals, many invasive plants may alter ecosystem structure and functional processes, including hydrologic, fire, and nutrient cycles. Structural changes in invaded plant communities typically cause reduced native species richness and diversity and changes in canopy structure (Belcher and Wilson 1989; Parmenter and MacMahon 1983; Rikard and Cline 1980; Wallace et al. 1992). In one study reported on by DeLoach (1991), the number of plant species present in California grasslands increased by 35% following biological control of common St. Johnswort (Hypericum perforatum). When large scale conversions of vegetational life history strategies and phenology occur, the potential for functional changes to hydrology, nutrient, and fire cycles probably increases. This has been observed with the late-maturing, deeply rooted forb yellow starthistle when it invades the earlymaturing, shallow-rooted annual grass communities. Yellow starthistle has been shown to deplete soil moisture reserves and alter water cycles in annual grasslands (Enloe et al. 2005). This could cause large annual economic losses in water conservation costs in California. In Siskiyou County, for example, it was estimated that the potential water loss due to yellow starthistle would be more than 100,000 m3, or 26,400,000 gallons, per year (Enloe 2002). The depletion of soil moisture

EXOTIC PLANT MANAGEMENT

283

F I G U R E 22.2 Close-ups of inflorescences of two “high-impact” invasive annual grasses in California.

Left: Taeniatherum caput-medusae (medusahead). Right: Aegilops triuncialis (barb goatgrass).

by yellow starthistle on invaded sites compared to annual grasslands is equivalent to a loss of 15–25% of mean annual precipitation (Jetter et al. 2003). Thus, yellow starthistle infestations can actually create drier than normal conditions even in subsequent years with average rainfall (Gerlach et al. 1998). Another lesser studied but demonstrable negative effect of this type of structural change is the impact on soil erosion. Spotted knapweed (Centaurea maculosa), a deeply rooted forb that has invaded millions of acres in Montana, has been shown to cause reduced water infiltration rates and subsequent increased surface water runoff and increased stream sediment yields compared to bunchgrass communities in Montana (Lacey et al. 1989). Although not yet quantified for yellow starthistle in the context of California’s annual grasslands, it is possible that similar effects may be occurring.

Management Techniques There are several tools for invasive plant management in the California grasslands. It is important to recognize the following key issues for weed management in the grasslands. First, weed management is a long-term process, and there are no “silver bullets” for immediate success. The biological attributes of most invaders provide mechanisms that allow for some survival in subsequent years after periods where reproduction is completely inhibited (which is also equated to a successful weed control event). These mechanisms include some survival via soil seed banks, temporal windows of resistance to fire, and herbivory and asexual reproduction via adventitious bud formation on the roots of many perennial species. If managers initiate control methods with little follow-up, failure is inevitable, as has been repeatedly demonstrated. Second, in many cases there may be limitations to the tools available for weed management. Not every tool can be

284

POLICY AND MANAGEMENT

used in every grassland area. For example, steep, rocky landscapes may limit reseeding with rangeland drills; proximity to urban areas may prevent the use of prescribed fire; and local ordinances have banned herbicide use in certain areas. Additionally, special regulations may apply if threatened or endangered species are present in the management area. These limitations are realities in many areas of California and serve to increase the challenges of successful weed management. Third, the principles of adaptive weed management should be applied whenever possible. These include establishing land management goals, identifying and prioritizing those species that threaten the goals, assessing available weed management techniques, developing and implementing a weed management plan, post treatment monitoring and assessment of impacts, and review and modification to get better results (Klinger and Randall 1997). To be successful, adaptive weed management requires flexibility and persistence. The fourth key concept, which has only recently been embraced by many land managers, is early detection and rapid response (EDRR) to new invaders. This concept entails immediate and aggressive action to eradicate incipient populations of exotic plants in a given area. Successful eradication has been clearly shown to be possible when invasions cover very small areas (Rejmánek and Pitcairn 2002). EDRR does not always allow for a thorough risk assessment to occur before investing resources toward eradication, especially for unfamiliar or novel taxa. Although there is a good chance that many incipient populations may fail to become serious problems (Williamson and Fitter 1996), there is still an increasing consensus among land managers that it is better to be safe now than sorry later. Thus land managers are beginning to use EDRR efforts to essentially “draw a line in the sand” to prevent new harmful invaders from becoming

widespread problems. While many weeds, such as yellow starthistle, are well beyond the scope of EDRR in California, there are still many species to which this approach can be applied locally and regionally. In grasslands these include Scotch thistle, woolly distaff thistle, artichoke thistle, and spotted knapweed. EDRR may also be integrated into adaptive management strategies with little conflict.

Mechanical Control A number of mechanical methods are used to control herbaceous grassland weeds, including hand labor, mowing, and cultivation techniques. In many cases these techniques are not practical or cost-effective, but there are situations in which they can be used very effectively. They also can be used effectively and with little training for volunteer-based stewardship programs or “weed pulling days.”

Hand Labor Hand labor methods for weed control in grasslands include hand pulling and tools such as weed whips, sling blades, clippers, shovels, hoes, mattocks, and Weed Wrenches™. There has been little published research comparing hand labor to other weed control methods in the California grasslands, so most available information has been translated from agricultural systems or is anecdotal in nature. Hand labor is widely used for controlling small weed patches but is difficult and expensive to use on large infestations. Hand labor is also more commonly used where volunteer help is available and in follow-up control programs where few plants remain after several years of intensive management (Sheley et al.1998). The relative success of hand labor in grasslands is dependent upon removal of a plant’s growing points. For annuals and biennials, severing the plants below the crown (i.e., cutting or breaking plants off a few inches below the soil surface) is all that is necessary. For creeping perennials, removing the vertical and lateral roots or rhizomes is essential for success. The difficulty of doing this in most soil types is immense, with the exception of moist, sandy soils. Therefore, hand weeding techniques are typically more effective on annual and biennial species and less effective on perennials, which often regenerate from adventitious buds on deep lateral and vertical roots. Plant height is also important, as lowgrowing rosettes are generally more difficult to remove by hand pulling than plants with bolted or elongated stems as long as the soil remains moist. These factors result in optimal control timing, which is when plants reach the late bolting to early bud stage before soils become too dry. This timing also often coincides with the end of the germination period for many winter annual weeds that dominate California grasslands. This reduces the potential for new cohorts to emerge following the disturbance caused by hand labor. A benefit of hand removal is that desirable species, if present, can be left in place. In Marin County, repeated hand pulling with a Weed Wrench not only proved to be an

effective method for French and Scotch broom control in small infestations, but also encouraged native plant recovery (Alexander and D’Antonio 2003a). It was also more effective than mowing.

Mowing and Clipping Mowing is a common vegetation management technique primarily used along roadsides and right-of-ways throughout California. Its main purposes include maintaining the safety recovery zone or “clear zone,” keeping visibility high, and reducing fuel loads to prevent wildfires. Mowing has also been used for weed management in both the interior and coastal California grasslands. Although not generally effective for weed eradication, mowing has primarily been used to reduce seed production of both exotic grasses and forbs, and proper timing is critical for its effectiveness. Mowing too early in the spring increases light penetration without removing a significant proportion of weed biomass. This generally benefits weedy species by stimulating rapid recovery of growth while soil moisture is still abundant. Mowing too late in the summer does not prevent seed production and may serve to work the seeds down into the seed bank better and disseminate weed seeds to new areas. The optimum time for mowing most annual species is in the flowering stage before seed development. This generally results in the greatest reduction of seed production. However, when soil moisture is plentiful following mowing, the effectiveness of control may be greatly reduced. For example, mowing diffuse knapweed (Centaurea diffusa) under adequate soil moisture resulted in compensatory growth and ultimately greater seed production compared to plants in an unmowed area (Sheley et al. 1999). Although mowing is more often used as a tool for control of noxious annual weeds, it can successfully control some biennial and perennial weeds (Benefield et al. 1999; Tyser and Key 1988). Repeated mowing on perennial broadleaf species can prevent seed production, reduce root carbohydrate reserves, and give advantages to desirable perennial grasses. Properly timed mowing has been demonstrated to be a successful tool for the control of yellow starthistle (Benefield et al. 1999). However, the growth form of the plants is critical for success. If plant architecture is characterized by profuse basal branching, then mowing tends to make the growth form prostrate, and the result is limited control. However, if the growth form is primarily elongated stems with little basal branching, such as those found within dense cover, mowing can be very effective (Benefield et al. 1999). Consequently, mowing for yellow starthistle control is best employed where competition for light results in elongated yellow starthistle stems. Mowing or clipping for vegetation management has also been shown to shift species composition in California coastal prairie from exotic annual grasses to exotic forbs (Hayes and Holl 2003b) or mixes of native and exotic forbs (Maron and Jefferies 2001). In pastures, mowing may reduce grass canopy

EXOTIC PLANT MANAGEMENT

285

cover and release more desirable short-statured legumes (DiTomaso 2002). In addition, it can remove the flowering stems of late-season undesirable invasive species, thus preventing or reducing new seed recruitment into the soil seed bank. When desirable perennial grasses are present, mowing can maintain their vigor and remove the unpalatable lowerquality growth or accumulated thatch. In many California grassland settings, mowing is of limited use because of safety concerns associated with steep terrain and physical damage that may occur to equipment. Also, mowing can create a fire hazard because of sparks generated by contact of the equipment with rocks. Even when mowing is employed as a control technique, it is not always successful and can decrease the reproductive efforts of insect biocontrol agents, injure late-growing native forb species, and reduce fall and winter forage for wildlife and livestock.

Tillage One of the most common mechanical weed control techniques used in agricultural systems is cultivation or tillage. Tillage equipment can include plows or discs, which control annual weeds by burying plant parts, including seeds. In contrast, the use of harrows, knives, and sweeps will damage root systems or separate shoots from roots (DiTomaso 2002). Tillage must be conducted when the surface soil is dry; otherwise, fragmented plant segments will regrow and thereby exacerbate the problem. For example, in new seedings of alfalfa in northeastern California, the spread of perennial pepperweed (Lepidium latifolium) from small populations was greatly exacerbated by preplant tillage operations (R. Wilson, personal communication). Despite its effectiveness in the control of annual weeds, tillage can also have the negative effect of increasing atmospheric dust levels and soil erosion. Tillage can, on occasion, effectively control some invasive species in grasslands. For example, early summer tillage can damage yellow starthistle and give adequate control. In most California grasslands, however, cultivation is not practical and does not achieve the intended objective, since it tends to select against desired species as well as undesirable ones. In addition, tillage can enhance a perennial weed problem, such as Canada thistle (Cirsium arvense), by spreading root fragments or stimulating emergence of new shoots from roots just below the tillage line in soil (Young et al. 1998). Tillage in California grasslands is most commonly used to create firebreaks just beyond highway right-of-ways, where wildfires are frequently started, or to create hayfields that are grazed after harvest. Tillage may also serve as an important tool in the early stages of restoring historically farmed lands to native grasslands (see Jantz et al., Chapter 23). In these areas, tillage can effectively be used to eliminate the early fall or spring cohorts of weeds before reseeding with natives (Stromberg et al., Chapter 21).

Thatch Removal The competitive ability of medusahead in annual grasslands is primarily due to the slow breakdown of its silica-rich

286

POLICY AND MANAGEMENT

thatch. It has been shown that the thatch layer is the main component responsible for suppressing other competing species (Kyser et al. 2007). Removing the thatch by either tillage or mowing in the fall can reduce the competitiveness of medusahead and provide better than 50% reduction in medusahead the following year. In addition, thatch removal can dramatically improve the efficacy of the herbicide imazapic, regardless of whether the removal technique is through burning, tillage, or mowing followed by thatch removal (Kyser et al. 2007). Thatch buildup is also associated with dominance of other sometimes undesirable non-native annual grasses in California. For example, ripgut brome, a species that is palatable to livestock when young but not when in fruit (DiTomaso 2000), if not grazed or mowed, can accumulate a great deal of thatch (Biswell 1956). This, in turn, may inhibit the germination of other species in subsequent years—a trait that contributes to the listing of this species by the California Invasive Plant Council as a threat to native grassland species (Table 22.1).

Biological Control Methods Biological control agents (generally insects or pathogens) are mobile and are expected to move from the release area and spread throughout the region. As a result, this control method is not specific to an invaded site or weed infestation. The goal of biological control is to establish self-sustaining populations of beneficial organisms that build up high numbers on the target weed. It is hoped that attack by the biological control agents will reduce the invasiveness of the host weed and result in a substantial reduction in its abundance. A key requirement of the control agents is their high level of specificity to the target weed. Many years of research are necessary to find the appropriate natural enemies and to perform the necessary host specificity tests. Once completed, the results are submitted for review by the U.S. Department of Agriculture (USDA)-Animal and Plant Health Inspection Service (APHIS), the agency that approves permits for the introduction of living organisms into the United States. In California, over 50 noxious and invasive weeds have been the target of biological control efforts. Of these, 16 species are considered grassland weeds (Table 22.2); 11 species are annual or biennial forbs, and five are perennial forbs. No exotic grass has been the recipient of a biological control agent release; however, efforts to explore for biological control agents against medusahead are currently under way by USDA. The results of successful biological control have been mixed: Five weeds are considered to be under successful biological control, four weeds are thought to be under moderate control, and eight weeds have shown little or no control or their level of control is unknown. Control of some weeds in grassland settings, such as common St. Johnswort and tansy ragwort (Senecio jacobaeae), has been spectacular (Huffaker and Kennett 1959; McEvoy et al. 1991). Although preliminary, another success appears to be developing against

TA B L E 22.2 List of Grassland Weeds Targeted for Biological Control, in Chronological Order

Year of first agent

Weed

Scientific name

Growth habit

Level of control

Common St. Johnswort Tansy ragwort Puncturevine Canada thistle Yellow starthistle Blessed milk thistle Slenderflower thistle Rush skeletonweed Spotted knapweed

Hypericum perforatum Senecio jacobaea Tribulus terrestris Cirsium arvense Centaurea solstitialis Silybum marianum Carduus tenuiflorus Chondrilla juncea Centaurea maculosa (C. biebersteinii) Salvia aethiopis Cirsium vulgare Carduus pycnocephalus Onopordum acanthium Centaurea diffusa Centaurea virgata var. squarrosa Centaurea calcitrapa

Perennial forb Biennial Annual Perennial forb Annual Annual Annual Biennial Perennial forb

High High High Low Moderate to low Low Unknown High Moderate

1945 1959 1961 1966 1969 1971 1973 1975 1976

Perennial forb Biennial Annual Biennial Biennial Perennial forb Biennial

Unknown Unknown Moderate None Moderate High Low

1976 1976 1976 1976 1980 1995 1998

Growth habit category

Proportion

Control category

Mediterranean sage Bull thistle Italian thistle Scotch thistle Diffuse knapweed Squarrose knapweed Purple starthistle

Perennial forb Biennial Annual

NOTE:

0.31 0.38 0.31

High Medium Low None Unknown

Proportion 0.31 0.25 0.19 0.06 0.19

Further information on the biological control efforts against these weeds is available in Coombs et al. (2004).

squarrose knapweed (Centaurea virgata var. squarrosa) (Woods and Villegas 2004). Releases of two seed head weevils have resulted in the destruction of nearly all annual seed production at several locations in eastern Shasta County. Even though the plant is a perennial, annual monitoring of field populations shows a steady decline in seedling recruitment and adult plant abundance. The biological control agents approved for use in California are listed in Table 22.3. Most of the weeds have had more than one biological control agent released against them. Usually, only one or two of the agents have been observed to show some effectiveness against their target weed. These have been rated as “good” or “excellent,” and it is recommended that land managers use only these agents if they are not already present in their area. All of the biological control agents in Table 22.3 are available through county agricultural commissioners’ offices in California. The most significant grassland weed in California is yellow starthistle. Biological control research efforts against this weed have been on-going since the 1960s. A total of six biological control insects have been approved for use in the United States: the gall flies Urophora jaculata and U. sirunaseva;

the weevils Bangasternus orientalis, Eustenopus villosus, and Larinus curtus; and the fruit fly Chaetorellia australis (Pitcairn et al. 2004). Of these, five have established in California and three are widespread, occurring almost wherever yellow starthistle grows (Pitcairn et al. 2002). A seventh insect, the false peacock fly, Chaetorellia succinea, was accidentally introduced in the early 1990s (Balciunas and Villegas 1999); hence its absence from Table 22.3. This insect has a strong affinity for yellow starthistle and has also spread throughout California. The weevil E. villosus and the fly C. succinea are the two most common insects found on yellow starthistle in California (Pitcairn et al. 2003). Annual monitoring data at two long-term study sites show a steady increase in the seed head attack rate and a concomitant decrease in seed production, seedling recruitment, and adult plant abundance. Both study sites are located in undisturbed grasslands, so these results may be limited to this kind of habitat. However, these data do suggest that some level of control by the combined attack of E. villosus and C. succinea may occur in the appropriate habitats. Most recently, the autoecious, brachycyclic rust fungus (Puccinia jaceae var. solstitialis) received an experimental use

EXOTIC PLANT MANAGEMENT

287

Scientific name Hypericum perforatum

Centaurea diffusa

Centaurea maculosa (C. biebersteinii)

Centaurea virgata var. squarrosa

Tribulus terrestris Senecio jacobaea

Salvia aethiopis

Weed

Common St. Johnswort

Diffuse knapweed

Spotted knapweed

Squarrose knapweed

Puncturevine

Tansy ragwort

Mediterranean sage

Phrydiuchus tau

Longitarsus jacobaea Pegohylemyia seneciella Tyria jacobaeae

Established limited

Established widely Established widely Established widely

Established widely Established widely

No establishment Established widely Established limited No establishment No establishment Established widely

Cyphocleonus achates Larinus minutus Sphenoptera jugoslavica Terellia virens Urophora affinis Urophora quadrifasciata Microlarinus lareynii Microlarinus lypriformis

Established widely

Established limited Established limited Established limited Established widely Established limited

Cyphocleonus achates Larinus minutus Terellia virens Urophora affinis Urophora quadrifasciata Bangasternus fausti

Established limited

Established widely Established widely Established widely Established widely Established limited

Established widely Unknown Established widely No establishment Established limited

Distribution

Agapeta zoegana

Bangasternus fausti Larinus minutus Sphenoptera jugoslavica Urophora affinis Urophora quadrifasciata

Agrilus hyperici Chrysolina hyperici Chrysolina quadrigemina Chrysolina varians Zeuxidiplosis giardi

Biological control agent

1976

1969 1966 1959

1961 1961

1995 1997 1998 1998 1998 1998

1996

1993 1995 1995 1976 1990

1993

1994 1995 1980 1976 1990

1950 1945 1946 1952 1950

Year of intro

TA B L E 22.3 List of Biological Control Agents Against Grassland Weeds, Approved for Release in California

Light

Heavy Light Light

Heavy Heavy

Absent Heavy Moderate Absent Absent Light

Moderate

Light Light Light Light Light

Light

Moderate Moderate Heavy Light Slight

Moderate Unknown Heavy Absent Light

Infestation

Unknown

Excellent Poor Good

Excellent Excellent

None Excellent Fair None None Poor

Excellent

Poor Good Poor Poor Poor

Poor

Good Good Unknown Poor Poor

Unknown Unknown Excellent Unknown Poor

Control

Centaurea solstitialis

Carduus tenuiflorus Cirsium vulgare Cirsium arvense

Carduus pycnocephalus Silybum marianum Onopordum acanthium

Yellow starthistle

Slenderflower thistle

Bull thistle

Canada thistle

Italian thistle

Blessed milk thistle

Scotch thistle

Rhinocyllus conicus

Rhinocyllus conicus

Rhinocyllus conicus

Altica carduorum Ceutorhynchus litura Urophora cardui

Rhinocyllus conicus Urophora stylata

Rhinocyllus conicus

Bangasternus orientalis Chaetorellia australis Eustenopus villosus Larinus curtus Puccinia jaceae var. solstitialis Urophora jaculata Urophora sirunaseva

Bangasternus fausti Larinus minutus Terellia virens

Cystiphora schmidti Eriophyes chondrillae Puccinia chondrillina

Further information on each of these agents is available in Coombs et al. (2004).

Centaurea calcitrapa

Purple starthistle

NOTE:

Chondrilla juncea

Rush skeletonweed

No establishment

Established widely

Established widely

No establishment No establishment Established limited

No establishment Established limited

1976

1971

1976

1966 1971 1977

1976 1993

1973

1969 1984

No establishment Established widely Established widely

1985 1988 1990 1992 2003

1999 1998 1998

1975 1977 1976

Established widely Established widely Established widely Established limited Initial release

No establishment No establishment No establishment

Established widely Established widely Established widely

Absent

Light

Moderate

Absent Absent Light

Absent Moderate

Moderate

Absent Moderate

Light Light Heavy Light Unknown

Absent Absent Absent

Moderate Moderate Moderate

None

Poor

Fair

None None Poor

None Unknown

Fair

None Poor

Poor Poor Good Poor Unknown

None None None

Poor Fair Good

permit for release on yellow starthistle in California. The rust was originally collected from yellow starthistle in its native range of Turkey in 1978 (Woods and Villegas 2004). Puccinia jaceae var. solstitialis is an obligate parasite of yellow starthistle and is associated with the thistle over a wide area, at least from Spain to Turkey (Savile 1970). It completes its life cycle on a single host plant and has all five spore forms. It causes nonsystemic foliar infections that can reduce fresh and dry weights of inoculated yellow starthistle in controlled studies (Bruckart 1989; Shishkoff and Bruckart 1993). Spores, however, may not persist over the winter. Despite considerable research on host specificity of Puccinia jaceae var. solstitialis and nontarget plant safety at the USDAAgricultural Research Service (ARS) facility in Ft. Detrick, Maryland, little is known about the plant-pathogen interaction under field conditions, including information on the most effective inoculation timing window to maximize infection rates and foliar damage. Furthermore, while the rust has been shown to reduce root biomass, it is unknown whether this is due to a decrease in lateral root production or to the ability of yellow starthistle to produce deep roots. Such information may significantly impact our capability to predict the effect of the pathogen on yellow starthistle’s ability to develop tolerance to water stress. Moisture levels have been shown to correlate directly with seed production. If root depth is limited by pathogen stress, this may subsequently impact plant growth, seedhead production, and ultimately reproductive output. In addition, nothing is known on how the Puccinia rust will interact with the established biological control insects or how it will change the competitive ability of yellow starthistle compared to other grassland vegetation under different environmental conditions. Results of these studies can greatly improve the capacity to predict the potential effectiveness of the Puccinia rust under a number of abiotic and biotic conditions.

Chemical Control Herbicides are an important method of weed control in grassland systems. Herbicides can be applied to grasslands by a number of methods, including fixed-wing aircraft, helicopters, ground applicators, backpack sprayers, and rope wick applicators. Herbicides registered for use in grasslands of the western United States are listed in Table 22.4, along with pertinent information on each compound. Some of these products, including picloram and metsulfuron, are not registered in California. Of these compounds used in grasslands, the auxinic, or growth regulator, herbicides have played the most important role in broadleaf weed control. For large-scale use, herbicides are typically considered to be the most economical option. The most widely used grassland herbicides in California are those that have postemergence activity (DiTomaso 2000). These include 2,4-D, triclopyr, dicamba, clopyralid, chlorsulfuron, and glyphosate. Of these, clopyralid, 2,4-D, triclopyr, and dicamba are growth regulator herbicides that are selective on broadleaf species

290

POLICY AND MANAGEMENT

and have little activity on grasses. Clopyralid also has excellent preemergence activity and is highly effective for control of yellow starthistle and other members of the Asteraceae. However, it may have some negative impacts on some native plants within the Asteraceae, Fabaceae, Polygonaceae, and Apiaceae and short-term impacts on Violaceae (Reever Morghan et al. 2003). Chlorsulfuron is an amino acid inhibitor and is also very effective on most broadleaf species, particularly members of the mustard (Brassicaceae) and figwort (Scrophulariaceae) families. It has both preemergence and postemergence activity on most species, but only preemergence activity on yellow starthistle. Like the growth regulator compounds, chlorsulfuron is fairly safe on most grasses. Glyphosate is a nonselective aromatic amino acid inhibitor and provides excellent control of annual and perennial grasses and broadleaf weeds, but it will also damage desirable plants. Aminopyralid is a new growth regulator herbicide registered in California in 2006. It has about three times the activity of clopyralid on yellow starthistle. In addition to its activity against yellow starthistle, aminopyralid has a broader spectrum of selectivity and has also been shown to be very effective on knapweeds, many other thistles, and fiddlenecks (Amsinckia spp.) (Kyser et al. 2007). Both products will injure native legumes during the growing season, but some can be used safely when treatments are made after senescence or during the dormant phase of perennial legumes. Another newly registered herbicide (not yet in California), imazapic, has proven to be very effective on medusahead, downy brome, ripgut brome, barb goatgrass, and other annual grasses, without significantly injuring seedlings of many native perennial grass or broadleaf species (Kyser et al. 2007). Timing of herbicide applications can determine the effectiveness of the treatment. For yellow starthistle control, the best timing for application of clopyralid and aminopyralid seems to be between December and the end of March, depending on the location (DiTomaso et al. 1999b). Fall or spring applications of imazapic can be effective for the control of annual grasses, depending on the location. In areas with snowpack, spring applications may be more desirable, but in typical Central Valley or foothill conditions, fall applications have proven successful. For perennials, timing of application can depend on the herbicide. Chlorsulfuron can control perennial species with spring, summer, or fall treatments (Drake and Whitson 1989; Whitson et al. 1989; Young et al. 1998), whereas glyphosate is best applied in spring, when invasive plants are at the late bud to early flowering stages (Waterhouse and Mahoney 1983; Young et al. 1998). Herbicides are generally applied as broadcast treatments over the entire field or directed (spot) applications to control early weed invasions or to prevent the spread of small infestations. However, it is also possible to achieve selective control of a particular weed with otherwise nonselective or relatively nonselective postemergence herbicides by employing a rope wick or wick applicator. These can be either hand-held or vehicle-mounted boom wipers. As a benefit, this application method reduces the potential for herbicide drift and

Mode of action Growth regulator

Growth regulator

Amino acid synthesis inhibitor Growth regulator

Growth regulator Amino acid synthesis inhibitor Amino acid synthesis inhibitor

Amino acid synthesis inhibitor Growth regulator

Growth regulator

Trade name

Weedar®, Weedone® and many others

Milestone®

Telar®

TranslineTM

Banvel®, Vanquish®

Roundup®, and others

Plateau®

Escort®

TordonTM

Garlon®, Remedy®

Common name

2,4-D

Aminopyralid

Chlorsulfuron

Clopyralid

Dicamba

Glyphosate

Imazapic

Metsulfuron

Picloram

Triclopyr

Broadleaf species

Broadleaf species, weak on mustards

Broadleaf species

Nonselective, but best on annual grasses

Nonselective

Broadleaf species

Certain broadleaf families (e.g., Asteraceae, Fabaceae, Apiaceae, Solanaceae, Polygonaceae)

Mainly broadleaf species

Certain broadleaf families (between clopyralid and picloram)

Broadleaf species

Weed spectrum

Less than 1 month

Up to 3 years

At least 2 months

Full season

None

Less than 1 month

Most of the season

At least 2 months

Full season

Less than 2 weeks

Soil residual

TA B L E 22.4 Commonly Used Herbicides for Grassland Invasive Weed Control

Yes

No

No

Registration expected in 2007 or 2008

Yes

Yes

Yes

Yes

Yes

Yes

Registered in California

Postemergence only, good on seedlings, fair on mature plants

Effective both pre- and postemergence; applied fall, winter or spring

Fairly effective; preemergence only

Mainly as a preemergence treatment, postemergence control with seedlings or rosettes

Postemergence only, from seedling to early flowering

Postemergence only, from seedling to bolting

Effective both pre- and postemergence; applied fall, winter or spring

Preemergence only

Effective both pre- and postemergence; applied fall, winter, or spring

Postemergence only, from seedling to bolting

Effective timing

injury to adjacent sensitive agricultural crops and can be used to selectively control invasive species around vernal pools, streams, and other bodies of water, or in areas with rare and endangered species or other desirable plants. Residual thatch can influence the effectiveness of herbicides with preemergence activity. Some herbicides, such as imazapic, can adsorb to standing thatch or other dried debris on the soil surface, thus reducing the effectiveness of the application. However, this does not appear to be a characteristic of aminopyralid or clopyralid (DiTomaso et al. 1999b). Although herbicides are effective for the control of invasive grassland weeds, they generally do not provide longterm control of weeds when used alone (Bussan and Dyer 1999). In the absence of a healthy plant community composed of desirable species, one noxious weed may be replaced by another equally undesirable species insensitive to the herbicide treatment (DiTomaso 2000). Thus, herbicides in grasslands are best used as part of an integrated weed management system.

Cultural Control

more balanced competitive relationships among native and invasive species (Olson 1999). High-intensity, short-duration grazing, practiced on a rotational basis, is a management system widely adopted in other countries (DiTomaso 2002) that is becoming increasingly recommended in California grasslands (see Jackson and Bartolome, Chapter 17). This can be logistically difficult and generally requires electric fencing to keep animals confined to a specific area. Once grassland has been intensively grazed, it is allowed to recover for about a month before being grazed again. This system usually leads to more uniform forage use (including many weeds) by the grazer. In many cases, this method has been shown to provide much better control of specific weeds than season-long livestock grazing (see review by DiTomaso 2000). High-intensity grazing of both cattle and sheep has been tested experimentally for the control of medusahead. George et al. (1989a) found that two years of intensive grazing with cattle significantly reduced medusahead from 45% of the total species composition to only 10%. In another study using sheep, intensive mid-spring grazing (April/May) reduced medusahead by greater than 80% the following year (DiTomaso, Kyser and Doran, unpublished data).

Grazing Grazing can be an effective way of managing undesirable species in some grasslands both in California and elsewhere (see also Huntsinger et al., Chapter 20). The effectiveness depends on the plant species present, the type of grazer used, and the timing and intensity of the grazing program. Under some conditions grazing can increase undesirable nonnative species, or other less palatable or poisonous species. Intensive grazing can also disturb soil and enhance weed seed germination, reduce competition from more desirable species, and increase soil compaction (Elmore 1992). So, if grazing is to be used, it must be used judiciously with prescriptions designed for individual sites. Successful invasive weed management can also depend on the type of grazer and timing of grazing. For example, the foraging behaviors of both cattle and goats are conducive to the management of yellow starthistle when plants are grazed at the bolting stage, whereas only goats are effective when plants are in the spiny stages of growth (Thomsen et al. 1993). The ideal time to graze is when the noxious species are most susceptible to defoliation or when the impact on the desirable vegetation is minimal (Kennett et al. 1992). Stocking rates of livestock can also be adjusted to maximize invasive plant management. Lower stocking rates will generally allow livestock to graze preferred plants and avoid less palatable species. If the invasive species is preferred, then lower stocking rates can be effective. In most cases, however, grasslands with low cattle stocking rates have higher weed infestations compared to areas that are more intensively grazed. Higher stocking densities can minimize the grazers’ ability to avoid less palatable invasive weed species. This can lead to a more uniform composition of plant species and

292

POLICY AND MANAGEMENT

Prescribed Burning There has been increasing interest in the use of prescribed fire for vegetation management in recent years, including in California grasslands (see Reiner, Chapter 18). Purposes for prescribed burning have included invasive weed control, thatch removal, nutrient release from dead, dried plant matter, and stimulation of early growth of desirable species in the spring. However, while most of these purposes are best accomplished with late fall burns, burning designed to control invasive species in California grasslands generally needs to be conducted in late spring or summer. Unfortunately, this timing coincides with the period when there is high risk of fire escapes. Moreover, air quality problems and liability issues can also present a problem when burns are conducted near populated areas. In areas where biological control agents are present, burning may cause damage to these insect populations. In some areas, burning can lead to rapid invasion by other undesirable postfire colonizers with wind-dispersed seeds, particularly members of the sunflower family. The main objective of using controlled burns for invasive plant control is to deplete the soil seed bank, destroy seeds that are on the plant, and prevent sexual reproduction, which would in turn replenish soil seed reserves. To successfully control annual species with fire, it is critical to either kill plants before their seeds become viable (DiTomaso et al. 1999a) or destroy the seeds before they disperse (Allen 1995; Menke 1992). For annual species, burns should be conducted when the target plant’s seeds are still undispersed and are exposed to direct flames in the canopy, but desirable species have dispersed their seeds to the ground. Seeds on the soil surface are not generally exposed to lethal temperatures in

grassland fires (Sweet 2005). For perennial or herbaceous plants with protected meristems (e.g., rosettes or rhizomes), the burn must be hot enough to damage the vegetative reproductive tissues and prevent resprouting. For this reason, prescribed burning is rarely effective for the management of perennials and in most cases can stimulate their growth (DiTomaso et al. 1999a). It has, however, been used to reduce the seed bank of French broom within California grassland settings (Alexander and D’Antonio 2003b). Fire stimulates broom seed germination, and the flush of seedlings is then cut, treated with an herbicide, or reburned prior to plants becoming reproductively mature. Annual species that are most susceptible to control with prescribed burning are those that produce seeds after the fire season begins, have flowering structures either embedded within the fuel bed or exposed to direct flames, and have short-lived seed banks. For control of these late-season annual grasses and forbs, the timing of the burns is critical. Burns conducted before the target species have fully cured their seeds are most effective (Brooks 2001). To be successful, however, fine fuels to carry the fire should not be limiting. However, fine fuels are often patchy across landscapes, and the result is a mosaic burn pattern that results in variable weed control. Invasive grasses with long-awned seeds (e.g., medusahead, downy brome, ripgut brome, red brome, and barb goatgrass) rely on animal dispersal. In many of these species, the seeds remain in the inflorescence longer than most desirable grasses, and, as a result, they are more susceptible to destruction by the direct heat of burning (Dahl and Tisdale 1975, Young et al. 1970). Effective control of medusahead with prescribed burning (more than 90%) was demonstrated as far back as 1953 (Furbush 1953) and has been demonstrated on a number of other occasions (George 1992; McKell et al. 1962; Pollak and Kan 1998; Sharp et al. 1957; Betts 2003; DiTomaso et al., unpublished). However, in some cases burning has not proven successful on medusahead. Young et al. (1972) found that repeated annual burning in mid-summer increased medusahead infestations while decreasing the population of more desirable annual grasses. This inconsistency is probably due to differences in the length of flame exposure and to the heat of the burn. Other invasive long-awned annual grasses, including barb goatgrass (DiTomaso et al. 2001; Hopkinson et al. 1999) and ripgut brome (DiTomaso et al. 1999a; Kyser and DiTomaso 2002) have also been controlled with one or multiple years of burning, although exception can also be found (A. Levine and C. D’Antonio, unpublished data). In contrast, downy brome and red brome are difficult to control with burning because their seedheads begin to shatter and the seeds fall to the soil surface before enough fuel is available (Brooks 2002; Young and Evans 1978). Late season forbs, particularly yellow starthistle, can also be controlled by repeated early summer burns (DiTomaso et al. 1999a; Kyser and DiTomaso 2002). Because the seed of starthistle survives for more than two years in the soil and germination is enhanced by a preceding burn (DiTomaso and Kyser, unpublished data), a single year of burning will

not control an infestation. In one case, three consecutive years of burning were required to reduce the yellow starthistle seed bank by 99% (DiTomaso et al. 1999a). In other studies (DiTomaso and Kyser, unpublished data; Miller 2003), integrating a first year burn with a second year herbicide treatment was the most effective strategy. Like other control strategies, prescribed burning requires a follow-up program to prevent escaped or isolated plants from completing their life cycle. Where the seed bank is shortlived, a follow-up program may take only a couple of years; in other cases, it may take longer. Since multiple burns are not usually practical or permitted, and fuel loads may not be sufficient to allow multiple year burns, integrated approaches are often more appropriate than using burning as a sole control option.

Revegetation The goal of grassland management should be to improve degraded communities and make them less susceptible to noxious weed invasion. Revegetation with desirable and competitive plant species is one approach to achieving longterm, sustainable suppression of weed population growth, while providing high forage production or desirable plant diversity (Borman et al. 1991; Lym and Tober 1997). The choice of species used in a revegetation effort is critical to its success. Seeded species need to be adapted to the soil conditions, elevation, climate, and precipitation level of the site (Jacobs et al. 1999). Only a limited number of species have proven to be aggressive enough to resist establishment of problematic invasive species, and the proper species choice varies depending on the location and objective. Perennial bunchgrasses are among the most commonly used species for revegetating western rangelands and grasslands, and they have been shown to reduce the growth and reproduction of weeds such as yellow starthistle (Lym and Tober 1997; Roché et al. 1994; Dukes 2001a; Reever Morghan and Rice 2005). Some native broadleaves, including Hemizonia congesta, have a similar life cycle to and can suppress the growth of yellow starthistle (Duke 2001a). Introduced broadleaf species such as legumes can also be used in revegetation programs to suppress rangeland or grassland weeds. For example, Thomsen et al. (1997) and Thomas (1996, 1997) tested several legume species for their competitive effect on yellow starthistle. Thomsen et al. (1997) found subterranean clover (Trifolium subterraneum) varieties to be the most competitive when combined with grazing and mowing, but they did not provide adequate seasonal control of yellow starthistle in the absence of other control options. Thomas (1996, 1997), however, used a combination of subterranean clover and/or crimson clover (Trifolium incarnatum) as a cover crop in yellow starthistle–infested pasture. In a completely infested field, he reported an 80–90% reduction in yellow starthistle one year after planting with crimson clover. However, use of legumes may increase soil nitrogen, which can cause other potentially undesirable effects (see Dukes and Shaw, Chapter 19).

EXOTIC PLANT MANAGEMENT

293

S I D E B A R 2 2 . 1 C A S E S T U D I E S O F I N T E G R AT E D M A N A G E M E N T S T R AT E G I E S

Combination of Herbicides and Revegetation An integrated approach combining herbicide treatments and perennial grass revegetation was tested in a heavily infested yellow starthistle grassland site near Yreka, California (Siskiyou County). The goal was to provide ranchers and land managers with economical and sustainable management programs that maximized forage production or restored and preserved desired ecosystem functions, including reducing the susceptibility of their lands to reinvasion or invasion by other noxious weeds. In this severely degraded site, a mid-February treatment with glyphosate (one treatment) and clopyralid (one, two, or three annual spring treatments) was used to provide a window of reduced competition for the subsequent establishment of pubescent wheatgrass drill-seeded in early March (Enloe 2002; Enloe et al. 2005). This seeding timing, while not appropriate for much of California, has been shown to work well for far Northern California (Kay and Street 1961), where sufficient rain falls after early March to support wheatgrass establishment. The study area was monitored for six years (Enloe et al. 2005). Clopyralid treatment significantly reduced yellow starthistle, and glyphosate gave control of the annual grasses. This combination allowed pubescent wheatgrass seedlings to establish with a single year of treatment. Once pubescent wheatgrass seedlings survived the first year, additional applications of clopyralid to control the starthistle did not improve their establishment. In the absence of clopyralid and glyphosate, pubescent wheatgrass establishment was very limited. The integrated approach gave long-term suppression of yellow starthistle and other exotic annual grasses and forbs over the six year period. Treatments with clopyralid alone (e.g., without seeding of wheatgrass), gave good control of yellow starthistle, but the plant community initially became dominated by undesirable annual grasses, particularly downy brome. Downy brome, in turn, offered little competitive resistance to starthistle reinvasion, and within a couple of years after the final clopyralid application this site reverted to yellow starthistle (Enloe et al. 2005).

Long-term Management Using Prescribed Burning and Clopyralid As was previously discussed, repeated burning is generally impractical and can negatively impact air quality as well as compromise establishment of biocontrol agents. The continuous use of clopyralid can also have undesired outcomes. As a result, an integrated strategy was developed combining clopyralid and prescribed burning for management of yellow starthistle (DiTomaso et al. 2003). Results of small-scale plot studies indicate that prescribed burning stimulated the germination of yellow starthistle seed in the subsequent rainy season. This helped to deplete the seed bank more rapidly. Thus, a first-year prescribed burn followed by a second-year clopyralid treatment gave nearly complete control of yellow starthistle in the year after the last treatment. This strategy may reduce the number of years necessary to intensively manage yellow starthistle and allow land managers to transition into a follow-up management and reseeding program sooner. The reverse order gave very poor control, suggesting that an integrated approach should not end with a prescribed burn. An additional benefit of integrating prescribed burning into a yellow starthistle management program is the control of noxious annual grasses. When ripgut brome and medusahead coexisted with yellow starthistle, burning contributed to the reduction of all three invasive species (DiTomaso et al. 2003). With the results of these small-scale studies, a large-scale integrated approach was used at Fort Hunter Liggett in Monterey County, California (Miller 2003; Torrence et al. 2003a, b). After two or three years of treatment, which included a first-year burn followed by one or two years of clopyralid,

yellow starthistle control was excellent. At that time, a follow-up maintenance plan was implemented to prevent any potential reinfestation. This project demonstrated that yellow starthistle populations could be controlled with two years of properly-timed, intensive management. The most successful long-term, large-scale yellow starthistle control treatment was to follow a first-year prescription burn with a broadcast clopyralid application treatment the next year. However, a follow-up program should be instituted immediately in order to prevent invasive plant resurgence to previous levels.

Because of the ecological diversity within California, no single species or combination of species will be effective in providing ecological resistance against invasive weeds under all circumstances. While pubescent wheatgrass (Thinopyrum intermedium) has proven successful for yellow starthistle suppression in Siskiyou County (DiTomaso et al. 2000; Enloe et al. 2005), it may not be appropriate in many other areas of the state that lack summer rainfall or where native grasses are the landscape objective (see Stromberg et al., Chapter 21) Though perennial grasses have been shown to be most successful in competing with grassland weeds, a combination of species with various growth forms can also be effective. This diversity allows for maximum niche occupation and more sustained resource capture over the growing season (Sheley et al. 1999; Dukes 2001a, b). For example, seed mixtures of grasses with legumes improved the rate of microbial and soil structure recovery compared to grasses alone (Jacobs et al. 1999). Seeding with a variety of species, however, makes it difficult to choose control techniques (such as herbicides) that will not harm one of the seeded species. This is particularly true if grasses are seeded with broadleaf species and the weeds that are being controlled are broadleaf species. A revegetation program may require initial seeding with perennial grasses during the weed management phase, followed by subsequent seeding with desirable broadleaf species. Revegetation is generally a slow process and may take several years to be successful. It is most successful when combined with other management techniques.

Developing a Management Strategy The major elements of a grassland weed management program are preventing introduction or reinvasion of invasive weed seed, reducing the susceptibility of the ecosystem to

invasive plant establishment, developing effective educational materials and activities, and establishing a program for early detection and monitoring (DiTomaso 2000). An effective invasive weed management strategy should include three major goals: (1) controlling the weed; (2) achieving land use objectives such as forage production, wildlife habitat and ecosystem preservation, protecting diversity or endangered species, or recreational land maintenance; and (3) preventing reinvasion or invasion of other noxious species. All these goals are tied together with improving the degraded grassland community and reestablish a functioning ecosystem.

Integrated Approaches to Weed Management As previously discussed, a single method does not generally give sustainable control of a grassland weed. A successful long-term management program should be designed to include combinations of mechanical, cultural, biological, and chemical control techniques. There are many possible combinations that can achieve the desired objectives, and these choices must be tailored to the site, economics, and management goals. See “Case studies of integrated management strategies” for more elaboration (Sidebar 22.1). Even when a single control method does provide effective control over a number of years, it may not be practical. For example, repeated burning can be effective for the control of some annual grasses and yellow starthistle, but this approach is often prohibited. Thus, it may be necessary to incorporate other control methods, along with burning, into a long-term management strategy (DiTomaso et al. 2006; Kyser and DiTomaso 2002). When an integrated strategy is employed, it may be important to employ a particular sequence of approaches. For

EXOTIC PLANT MANAGEMENT

295

example, in a revegetation effort along a yellow starthistle–infested canal and roadside, the first step was to intensively manage starthistle (Brown et al. 1993; Thomsen et al. 1994). The second step was to reseed with competitive, deeprooted native perennial grasses. In the final stage, native broadleaf forbs such as California poppy and lupines were seeded into the system. Biological control can also play a key part in the success of an integrated control program. For example, Huffaker and Kennett (1959) reported on the success of the biological control program for the management of common St. Johnswort. They noted that maximal improvement of the rangeland was achieved when the biocontrol agent was used in combination with moderate timely grazing. This combination prevented the expansion of ripgut brome in the grassland sites previously occupied by common St. John’s wort. In a review, Lym (2005) described the numerous situations in which the successful use of biological control insects

296

POLICY AND MANAGEMENT

(Aphthona spp.) for leafy spurge (Euphorbia esula) depend on the use of other conventional weed control methods. Integration of Aphthona spp. with herbicides, grazing, or burning gave more rapid and better leafy spurge control that any method used alone. Finally, it is important to emphasize that the ultimate objective of any control strategy is to develop a “healthy” functional ecosystem. Competitive background vegetation in grasslands will not only reduce the potential establishment of invasive plants, but can also enhance the effectiveness of other control strategies, including biological control (McEvoy and Coombs 1999). As an example, the seed feeding insects for yellow starthistle can have attack rates of greater than 90% (DiTomaso et al. 2006), yet reduce seedset by only about 50–60% (Woods et al. 2004). In a system with competitive vegetation, a reduction of this level may suppress starthistle to an acceptable and sustainable level.

TWENTY THREE

Regulatory Protection and Conservation PATR I C K A. JANTZ, B E R N HAR D F. L. P R E U S S E R, J E S S E K. F UJ I K AWA, J O S E P H A. K U H N, C H R I STO P H E R J. B E R S BAC H, J O NATHAN L. G E LBAR D, AN D F RAN K W. DAVI S

In this chapter the current environmental policy setting for conserving and managing grassland ecosystems in California is reviewed. Grassland ecosystems are perhaps the most human-altered terrestrial ecosystem nationally, and in California, but they have been relatively overlooked in terms of conservation policy, especially in the western United States (Connor et al. 2001). In spite of the ecological, economic, and scenic value that grassland ecosystems provide, large expanses of habitat are still being degraded, fragmented, and converted to agriculture and subdivisions because of inadequate protection (Holland and Keil 1995; Stromberg and Griffin 1996; Harper et al. 1998; Gelbard 2003). Only 4% of extant California grasslands are in formally designated reserves (Davis et al. 1998). Finally, because more than 80% of California grasslands are privately owned, land use and land management policy resides primarily at the county and local levels. This review covers the spectrum of current policies that could affect future ecological trends in California grasslands. Based on recent statewide mapping efforts, grasslands cover nearly 11 million acres of California, or almost 11% of the state (Davis et al. 1998; CDF-FRAP 2002) (Figure 23.1). More than 99% of mapped grassland, which excludes vegetation with more than 10% tree or shrub cover, is classified as annual grassland, with the balance consisting of perennial grassland. This data underestimates the amount of perennial grasslands, which often occur in small patches below the resolution of statewide maps. Oak woodland and savanna ecosystems, which comprise another 5 – 6% of the state, also encompass extensive grassland habitat, but the presence of the tree stratum places these ecosystems into a different sociopolitical and ecological context because of the special conservation attention afforded oaks as well as their pronounced influence on wildlife and plant community assemblages (Giusti et al. 2004). For these reasons, oak woodland types are excluded from this analysis, whose focus is on

Mediterranean-climate grasslands, excluding desert and Great Basin grasslands of eastern California that fall outside of the California floristic province (but see Keeler-Wolf et al., Chapter 3). It is not possible with existing statewide data to discriminate among different grassland community types or to distinguish between grasslands dominated by native versus exotic species, so for mapping purposes all grasslands in the study region are treated as “California grasslands.” However, many federal, state, and county policies do differentiate between native and exotic grasses. Therefore, when possible and appropriate, the discussion is broken down along the lines set by the policies in question. First, the hierarchy of federal-to-local land use policies that potentially affect the use and management of California grassland are summarized, focusing in particular on how these policies are currently being applied to grasslands and grassland associated species. Next, existing and emerging conservation incentives for private grassland managers are examined as a complement to regulatory approaches. Policies and incentives based on the protection they afford to the biological integrity of grassland ecosystems are evaluated. In conclusion, recommendations for future policy needs and opportunities are offered.

Federal, State, and Local Regulation of California Grasslands Federal, state, and local governments have regulatory authority impacting the protection of California grasslands. Table 23.1 lists some of the key federal, state, and local laws and policies influencing the protection of California grasslands. The laws and policies are organized roughly in descending order of magnitude of protection, as determined by a combination of the strength of protection afforded to grasslands and the approximate grassland area protected. As discussed subsequently, federal, state and local governments

297

F I G U R E 23.1. Grassland distribution in California. Courtesy of Jesse Fukikawa.

also impact protection of California grasslands through ownership and/or management of land.

many non-DoD federal lands. Many of the areas mapped as perennial grasslands occur on DoD lands, in particular Camp Pendleton.

Federal Authority F E DE RAL E N DANG E R E D S P ECI E S ACT F E DE RAL OW N E R S H I P AN D MANAG E M E NT OF G RAS S LAN DS

Roughly 9% of California grassland is owned and managed by the federal government: 2.1% by the Department of Defense (DoD) and 7.1% by other federal agencies (Table 23.2). Examples of extensive grassland areas under federal management include the Carrizo Plain National Monument (Bureau of Land Management [BLM], San Luis Obispo County), the Panoche and Tumey Hills (BLM, Fresno County), and Beale Air Force Base (Yuba County). Management of livestock grazing and off-road vehicle use remain contentious issues on

298

POLICY AND MANAGEMENT

The U.S. Endangered Species Act (ESA) plays an especially prominent role in California grassland conservation and management. Section 7 of the ESA requires federal agencies that authorize, fund, or carry out activities that affect federally listed threatened or endangered species or designated critical habitat to consult with the Fish and Wildlife Service (USFWS) or the National Marine Fisheries Service (NMFS) to ensure “that actions authorized, funded or carried out by them do not jeopardize the continued existence” of listed species (Goble et al. 2005). Section 9 prohibits any person

Usually for open space or agricultural preservation. Sometimes explicitly for grassland habitat incl. native perennials.

Easements

Requires mitigation for projects that negatively affect the environment.

CEQA

Protection provided only by state statute.

Grassland species protected by CESA

Protection provided only by federal statute.

Grassland species protected by CWA

Protection provided only by federal statute.

Grassland species protected by ESA

There are four grassland associated species listed under CESA and not ESA: San Joaquin antelope squirrel (Ammospermophilus nelsoni), Swainson’s hawk (Buteo swainsoni), bank swallow (Riparia riparia), and Belding’s savannah sparrow (Passerculus sandwichensis beldingi). A building project at the University of California, San Diego, was required to mitigate for impacts to native grassland habitat.

The Nature Conservancy has at least two projects that protect grasslands: the Mount Hamilton Project east of San Jose and the Merced Grassland, conserving 81,000 and 5,000 acres respectively.

13 vertebrates, 45 plants

Unknown

Unknown

Prohibits indirect or direct “take” for listed speices. NCCPs offer exemptions to this rule.

Incompatible land use in easement areas can decrease the conservation value of these areas.

(Continued)

A 1000-acre complex of vernal pools at the University of California, Merced site. These vernal pools fall within federal jurisdiction due to swale connection to the San Joaquin River.

All federal jurisdictional wetland and vernal pool habitat throughout California

Section 404 restricts discharge of fill material into the “waters of the United States” without a permit.

Evaluated on a case-by-case basis; not preventative.

There are five grassland associated species listed under ESA and not CESA: San Clemente loggerhead shrike (Lanius ludovicianus mearnsi), California red-legged frog (Rana aurora draytonii), California tiger salamander (Ambystoma californiense), riparian woodrat (Neotoma fuscipes riparia), Buena Vista Lake shrew (Sorex ornatus relictus)

10 vertebrates, 14 invertebrates, 51 plants. 3.9 million acres in approved HCPs/NCCPs, 7.8 million acres in planned HCPs/NCCPs.* It was not possible to differentiate between HCP and NCCP acreage.

Prohibits indirect or direct “take” for listed species. HCPs offer exemptions to this rule. Does not prohibit “take” of plant species on private ly owned land.

San Joaquin kit fox, Stevens kangaroo rat

8 vertebrates, 36 plants

Prohibits take or destruction of critical habitat.

Grassland species protected by both ESA and CESA

Protection provided by both state and federal statutes.

Example

Extent

Features

Mechanism

TA B L E 23.1 Listing of Federal, State and County Policies/Regulations That Affect Grasslands and Grassland-Associated Species

Provides weak controls on development. A small amount of this land is strongly protected (conservation easements, etc.). This designation allows incompatible land uses such as motorized sports, golf courses, etc.

Incompatible land uses can limit conservation value. Desire to preserve prime farmland can funnel development to nonprime rangeland/grassland.

Affects the second-largest area of grassland with 20% zoned statewide (varies by county).

Affects the largest area of grassland with 50% zoned statewide (varies by county).

Each county must produce a general plan with mandated elements such as land use, conservation, and open space. Recognition of grassland communities varies greatly between counties.

Fresno County OS-D.5 states “The County shall strive to identify and conserve upland habitat areas adjacent to wetland and riparian areas that are critical to the feeding, hibernation or nesting of wildlife species associated with these wetland and riparian areas.”

Humboldt County general plan Policy 2523 in the general plan states that agricultural lands “shall be preserved” and lists several actions to fulfill this policy. However, policy 2523 1.D. directs development to “uneconomical or marginally viable agricultural lands” when development must occur.

Sacramento County recognizes native grasses as a significant biological resource in the conservation element of their general plan.

a General plan policies are often implemented through zoning ordinances such as those described in the text of the chapter. Although there are many zoning types, the majority of grassland is zoned as either agriculture or open space and are the categories focused on.

Lower density zoning that can constrain the extent of development.

Open space zoninga

Lower density zoning that can constrain the extent of development.

Agricultural zoninga

Counties can emphasize grasslands as communities of concern in their planning documents.

General Plana Recognition of the biological importance of grasslands in general plans can set the stage for increased grassland conservation. General plans policies are often weakly worded.

Humboldt County adopted Williamson Act guidelines in 1969 and issued a guideline update in 2002. Most of the grazing lands in the county are enrolled in the program.

Covers roughly 9.9 million acres of nonprime farmland as of 2003. Within counties with available saptial data, an average of 60% of all grasslands were in nonprime contracts from 2000–2004 .

Conservation value compromised by intensive land uses.

Williamson Act

Subset of agriculturally zoned areas. Land required to remain in agricultural for ten-year contracts, preventing development for a limited time.

Example

Extent

Features

Mechanism

TA B L E 23.1 ( C O N T I N U E D ) Listing of Federal, State and County Policies/Regulations That Affect Grasslands and Grassland-Associated Species

TA B L E 23.2 Area of Federal and State Owned Lands by Grassland Type

Ownership type

Grassland area (acres)

Percent of total grasslands in california

Private

9,462,200

88.08%

Federal

759,427

7.07%

Military

227,305

2.12%

State

131,517

1.22%

Other

162,819

1.52%

NOTE: GIS analysis done using layers from CDF-FRAP (2002) and GAP analysis.

from “taking” (e.g., killing or harming) endangered animal (but not plant) species, thereby extending the Act’s protection to projects with no federal government involvement, including projects on privately owned land. By contrast, for plant species, Section 9 prohibits removal or malicious destruction only on federal lands. Generally, the ESA affords less direct protection of plant, as compared to animal, species. However, protection of endangered animal species often carries protection of endangered plant species along with it, as discussed in the following paragraphs. Section 9 also prohibits any person from engaging in commerce of endangered species. Section 10 provides exemptions, permits, and exceptions to Section 9’s regulations, including allowance for “incidental” takes as long as such takes do not jeopardize the continued

existence of the species. Under Section 10, the Secretary of Interior can issue an incidental take permit to an applicant in conjunction with the development of a habitat conservation plan (HCP), so long as the applicant’s proposed project (resulting in a take) does not “appreciably reduce the likelihood of the survival and recovery of the species in the wild” (ESA sec. 10(a)(2)(B)). The HCP process has been used extensively in California to mitigate incidental take of endangered species, including grassland-dependent animal species, and has thereby directly affected grassland conservation in many areas of the state. For development to proceed within the established HCP area, HCPs require minimization of impact and mitigation of remaining impact to the maximum extent practicable. Such mitigation can be handled through privately owned mitigation or conservation banks, which protect and/or restore similar habitat in another location and sell credits that allow for the development of otherwise protected habitat. Currently 75 grassland-associated1 species, including 10 vertebrates, 14 invertebrates, and 51 plants, are listed as threatened or endangered under the ESA (Appendix 23.1). On a county basis, the number of listed species associated with grassland habitat ranges from 0–18 (Figure 23.2a). Unfortunately, the total habitat area affected by the listed species has not been determined on a consistent basis.

1. Grassland-associated species were determined by referencing the California Department of Fish and Game wildlife habitat relationship (WHR) database for vertebrates, the Calflora database for plants, and life history descriptions for invertebrates.

F I G U R E 23.2. Number of federal (A) and state (B) listed endangered or threatened grassland associated

species by county.

R E G U L A T O R Y P R O T E C T I O N A N D C O N S E R VA T I O N

301

Approved and proposed HCPs for grassland-dependent species range in size from small plans covering a single species, such as the Wildcat Line LP HCP in Monterey County, which only covers Smith’s blue butterfly in a planning area of 11.5 acres, to large subcounty plans such as the East Contra Costa County HCP/NCCP,2 protecting multiple species and covering at least 23,500 acres, about 19,000 acres of which is grasslands. HCP/NCCPs may also cover both listed and non-listed species. An example of the latter is Solano County’s HCP/NCCP, which covers 25 listed species, eight of which are grassland associated, as well as 51 nonlisted species (Solano County Water Agency, http://www. scwa2.com/hcp.html). Although numerous grassland-associated species have been included in HCPs, there are a few species that appear more often than others, including San Joaquin kit fox (Vulpes macrotis mutica), blunt-nosed leopard lizard (Gambelia silus), and California red-legged frog (Rana aurora draytonii). Some species such as the San Joaquin kit fox and the California condor (Gymnogyps califonianus) serve as “umbrella species” because of their large area requirements. For example, the kit fox figures prominently in the design of HCPs, multispecies HCPs and NCCP reserves and habitat corridors in East Contra Costa, Santa Clara, western Stanislaus, San Joaquin, and Kern Counties. Increasingly, plans are designed to meet the requirements of both the HCP and NCCP planning processes. Of the 97 approved HCPs and NCCPs in California listed on the USFWS and California Department of Fish and Game (CDFG) websites, 42 of them affect grassland species (Table 23.3). These plans cover nearly four million acres, and plans covering more than seven million additional acres are currently being developed. It should be noted that these are the total planning areas of the HCP/NCCPs and include areas that are not part of habitat reserves. Inclusion in an HCP/NCCP does not necessarily ensure strong protections for grasslands or grassland species. In the case of the San Diego Multiple Species Conservation Program (MSCP), for example, grasslands are just one of 26 natural vegetation communities in the plan area. In this case, grasslands were clearly not the main focus of the plan, which was especially concerned with the coastal sage scrub ecosystem. According to San Diego’s land development codes within the San Diego MSCP area, native grasslands are classified as Tier I habitat and non-native grasslands as a Tier III habitat type (more mitigation is required for Tier I habitat than for Tier III habitat). This is expected to lead to higher levels of preservation for native grasslands than for non-native grasslands. F E DE RAL CLEAN WATE R ACT

Section 404 of the federal Clean Water Act, which regulates fill of wetlands, also provides protection for California grasslands. 2. Natural Community Conservation Plans (NCCPs), provided for under the California Endangered Species Act, allow for voluntary cooperation of landowners and agencies in developing efforts to protect listed and nonlisted species.

302

POLICY AND MANAGEMENT

Section 404 is administered by the Army Corps of Engineers. Wetlands and vernal pool wetlands in particular can occur in or near grassland habitat, potentially increasing protection of grasslands. Federal jurisdiction over isolated wetlands and ponds was diminished following the Supreme Court’s 2001 SWANCC v. Army Corps of Engineers decision. However, many of California’s vernal pools, for example those at the University of California’s new Merced campus, are still considered “waters of the United States” and thus come under federal authority and section 404 processes, because of surface connections through swales. Isolated wetlands not connected to waters of the United States are under jurisdiction of the state, which may delegate responsibility to local regulatory bodies. Although it specifically addresses wetlands, Section 404 of the CWA can affect grassland conservation to the extent that grassland adjacent to a federal jurisdictional wetland is less likely to be developed, as a result of the 404 permitting process restricting dredging or fill of wetlands. Most of California’s vernal pools occur in grasslands, and CWA as well as ESA (and to a lesser extent the U.S. Migratory Bird Treaty Act) protections have led to some conservation of vernal pool/grassland complexes in areas such as western Merced County. Perhaps more important has been the application of Federal Clean Water State Revolving Funds to protect grassland/vernal pool complexes. For example, agencies, land trusts, and conservancies have collaborated to protect more than 40,000 acres of vernal pool landscapes in the Central Valley (See reports by the U.S. Environmental Protection Agency, http://www.epa.gov/owow/wetlands/pdf/ state_rev_fund.pdf). If a vernal pool or wetland is not deemed “waters of the United States,” the permitting process is conducted variously by the counties. The extent of wetland buffer zones also varies depending on wetland type, the specific jurisdiction, or county (National Academy of Sciences 2001). Often the wording of the policy specifies a distance within which removing, filling, dredging, or altering certain resource areas, such as vernal pools, is prohibited. By virtue of its presence in the landscape, vernal pool habitat has the potential to conserve large areas of grasslands, either through federal 404 or county level jurisdictional permitting processes.

F E DE RAL C ON S E RVATION R E S E RVE P RO G RAM

The Conservation Reserve Program (CRP) is a voluntary program to remove sensitive and erodible cropland and pasture from agricultural production. Farmers enroll their land for 10 or 15 years in exchange for rental payments to compensate for lost revenue. The enrolled acreage is planted with natural, although not necessarily native, cover to reduce erosion and provide wildlife habitat. According to the USDA Farm Service Agency, as of 2005, 144,438 acres of land in California were in CRP contracts (http://www.fsa.usda.gov/ Internet/FSA_File/fy2005.pdf). Of these, 126,760 acres were in either grass or legume cover. Because its influence is minor, no further discussion of the CRP is included in this review.

TA B L E 23.3 HCPs and NCCPs Affecting Grassland Species

Number of HCPs/NCCPs

Area covered (acres)

Number of planned HCPs/NCCPs

Area planned (acres)

Multiple Locationsa

1

2,937

0

0

Alameda

1

32.3

0

0

Contra Costa

0

0

1

175,435

Fresno

2

104.12

1

10

Kern

8

286,790.5

1

1,900,000

Kern/Tulare/ Kings

1

19,900

0

0

Los Angeles

1

14

1

8,661

Merced

0

0

2

250,630

Monterey

1

11.5

0

0

Orange

1

208,000

1

91,000

Placer

0

0

1

273,983

Riverside

10

1,853,644.5

1

11,785

Sacramento

0

0

1

340,000

Sacramento/ Sutter

1

53,342

0

0

San Benito

0

0

1

888,960 (County-wide Plan)

San Diego

5

608,938.9

3

1,585,938

San Joaquin

2

896,300

0

0

San Mateo

2

3,525.4

0

0

Santa Clara

2

5.02

1

440,318

Santa Cruz

3

168.7

0

0.0

Shasta

0

0

1

160,000

Solano

0

0

1

530,560 (County-wide Plan)

Tulare

1

9.7 Linear Miles

0

0

Yolo

0

0

1

400,000

Yuba/Sutter

0

0

1

200,100

Totalb

42

3,933,713.94

19

7,257,380.0

County

NOTE: These HCPs and NCCPs exclude constructed wetlands because we assumed that they would not contain grassland habitat. Data for this table from http://ecos.fws.gov/conserv_plans/servlet/gov.doi.hcp.servlets.PlanReportSelect?region
8&type=HCP and http://www.dfg.ca.gov/nccp/status.htm. For plans listed by both the FWS and DFG, FWS area numbers were used. For plans containing multiple subarea plans, all subareas plans were considered to be on HCP/NCCP aThis plan covers 25 existing and 4 planned sites in California. bTotal Area Covered does not include the Tulare County plan. Some planning areas overlap, so these numbers may be overestimates. Planning areas include both reserve and non-reserve areas covered by these plans.

R E G U L A T O R Y P R O T E C T I O N A N D C O N S E R VA T I O N

303

State Authority STATE OW N E R S H I P AN D MANAG E M E NT OF G RAS S LAN DS

The state of California owns and manages roughly 1.2% of California’s annual grasslands (Table 23.2) (Davis et al. 1998). Noteworthy examples include the annual grasslands of the Department of Fish and Game’s Carrizo Plain Ecological Reserve and coastal grasslands at Montana de Oro State Park (San Luis Obispo County), and Wilder Ranch State Park (Santa Cruz County). The University of California’s Natural Reserve System manages several significant grassland areas for research, education, and conservation, notably the Jepson Prairie Reserve (Solano County), McLaughlin Natural Reserve (Napa, Lake, and Yolo Counties), and Sedgwick Reserve (Santa Barbara County). The Jepson Prairie Reserve includes extensive native grassland and vernal pool complexes, while McLaughlin and Sedgwick include extensive areas of serpentine grassland. STATE LEG I S L ATION

The State of California has enacted several pieces of legislation impacting California grasslands. Many of California’s environmental laws mirror their federal counterparts, and some are more protective than federal law. State law also frequently shapes county-level legislation and mandates county-level management responsibilities. Some of these are discussed in more detail in the following paragraphs. CALI FOR N I A E N DANG E R E D S P E C I E S ACT

The California Endangered Species Act (CESA) mirrors the federal ESA in many ways. Notably, however, CESA extends protection of listed plant species onto privately owned lands (ESA sec. 9(a)(2) and CESA sec. 2062, sec. 2068). Natural Community Conservation Plans (NCCPs) serve the same function under CESA that HCPs serve under the ESA, the main differences being that permitting is through the CDFG, rather than the USFWS, and that the NCCP process requires independent scientific review. Currently 58 grassland-associated species are listed under CESA. Of these species, 13 are vertebrates and 45 are plants (Appendix 23.1). Grassland-associated insects are not eligible for listing under CESA (Appendix 23.1) (http://ag.ca.gov/ opinions/pdfs/98-105.htm). As with the federal ESA, the amount of protection afforded to grassland habitat by CESA varies by county based on the number of listed species present; more species or species requiring large home ranges can lead to larger amounts of protected grassland habitat. The number of state listed species present in any given county ranges from 0–11 (Figure 23.2b). The extent of habitat for these species within each county has not been consistently estimated. The effects of NCCPs on grasslands are similar to the effects of HCPs. As was discussed in the federal ESA section, there is

304

POLICY AND MANAGEMENT

considerable overlap between HCPs and NCCPs, and there are many similarities between the two types of plans. Both require minimization of impact and limit “take” to levels that will not endanger the survival of the species. Both also require that adequate funding for implementation of the plan be guaranteed before the plan is approved. Although HCPs and NCCPs are generally similar, there are a few differences. In particular, HCPs under ESA require mitigation of impacts to the maximum extent practicable, while NCCPs require mitigation to be roughly proportional to the take. OTH E R STATE LAWS I M PACTI NG G RAS S LAN DS

Additional state laws impacting the protection of California grasslands include the California Environmental Quality Act (CEQA), the Williamson Act, the Coastal Act, and the Government Code. These statutes are discussed in the following paragraphs in the section concerning county government authority since county governments frequently implement these laws.

County Authority C OU NT Y OW N E R S H I P AN D MANAG E M E NT OF G RAS S LAN DS

County parks, water districts, and open space districts account for less than 1% of the state’s lands, although some parks and utility districts protect significant native grasslands. Examples include the East Bay Regional Park District’s Point Pinole Park (Contra Costa County), Redwood Regional Park (Alameda County), the East Bay Municipal Utility District’s Siesta Valley watershed management area, and the Ramona Grassland Preserve (San Diego County). C OU NT Y G E N E RAL P LAN S AN D P OLICY D O CU M E NTS

Counties have authority to create policies and ordinances that determine how land can be used and what conditions can be placed on use. County land use goals and policies are specified in general plan documents, which the state requires counties to create and maintain. General plans include both mandatory and optional elements. The mandatory elements are land use, circulation, housing, open space, conservation, safety, and noise. The land use, open space, and conservation elements have special relevance to conservation of grasslands. Examples of optional elements with potential to impact grasslands include the agriculture, scenic highways, and biological resources elements. General plans are binding documents but may be amended to accommodate changing county needs. The goals and policies of the general plan are implemented by county zoning ordinances. These ordinances specify the actions in which land owners may engage and the use to which land may be put. For example, they specify allowable densities for different land use types in the county. In order to assess the level of protection afforded grasslands at the county level, readily available3 county policy

documents in 21 counties were surveyed. These 21 counties account for 76% of the total grassland area in California (Appendix 23-2). Privately owned grassland in many of these counties comprises more than 90% of total grassland area within each county (Appendix 23.3). General plans and zoning ordinances and conducted phone and email interviews with county planning departments were reviewed. Sixteen of the counties had readily accessible general plans. As discussed in the following paragraphs, the number of counties that included conservation, open space or land use elements, as well as other optional elements that recognize grassland resources were quantified.

Conservation Element California Government Code Section 65302(d) requires that a conservation element (some legal guidelines regarding conservation) be present in all county general plans. This element provides guidance for the conservation, development, and utilization of natural resources within the county. Counties frequently combine this element with the open space element of their general plan. The county conservation element often identifies specific habitats of concern such as wetlands, rare plant community types, and habitat for sensitive species. Fourteen of the counties examined had readily available conservation elements, of which six were combined conservation/open space elements. Of these, five of the conservation elements and five of the combined elements recognized grasslands and/or native grasslands. The Kern County general plan combines land use, conservation, and open space elements and contains no recognition of grassland ecosystems (Appendix 23-4). When grasslands are mentioned in the conservation element of the general plan, it is often in context of preserving significant natural areas. For example, in Madera County’s combined natural resources and agricultural element Policy 5.F.3, grassland could be included in “natural areas of outstanding vegetation.” In practice Madera County’s policy is implemented by maintaining a list of significant species, including species listed under CESA, species included in the California Native Plant Society’s Inventory of Rare and Endangered Plants, and Department of Fish and Game species of special concern. The list is kept current to fulfill the requirements of federal and state endangered species laws and to guide the county’s efforts in conserving important natural areas. Instead of providing additional protections, conservation elements generally defer to federal and state regulations. However, some counties appear to be moving toward more proactive grassland conservation. The 2004 Monterey County general plan update, for example, contains language that would give special protection to native perennial grass habitat within the county. If the draft plan is approved, coastal terrace prairie/valley needlegrass grassland will be defined as an 3. “Readily available” documents consisted of those available on county Web sites and those that could be acquired by mail within the time frame of this analysis.

environmentally sensitive area, which “shall be protected against any significant disruption of habitat values” according to policy CZ-5.2 (Monterey County General Plan, Draft, Environmental Resources Management, Coastal Zone Goal 5).

Open Space Element California Government Code Section 65302(e) requires each county to provide a plan for comprehensive and long-range preservation and conservation of open space land within its jurisdiction. This element is frequently combined with the conservation element in the general plan. Grasslands can in principle be protected by open space land use restrictions. However, open space land can be managed for varied uses ranging from natural resource production to outdoor recreation or the promotion of public health and safety. TA B L E 23.4 Area of Grassland in General Plan Zones

Grassland area (acres)

Percent

Agriculture

7,831,093

53.25%

Open Space

2,991,640

20.34%

Very Low Density Residential

1,888,981

12.84%

Low Density Residential

1,316,524

8.95%

14,467,465

4.62%

Zoning

Other

All of the plans examined contained either open space elements or combined elements that included open space information and policy goals. Grasslands and/or native grasslands specifically were recognized in four open space elements, as well as the five combined elements discussed already (Appendix 23.4). For example, Fresno County’s policy, Fresno County, General Plan, Open Space and Conservation Element (http://www.co.fresno.ca.us/4510/4360/ General_Plan/Gp_Final_policy_doc/Open_Space_Element_ rj.pdf) stipulates that “the County shall require that development on hillsides be limited to maintain valuable natural vegetation, especially forests and open grasslands, and to control erosion.” Sonoma County’s Open Space Element 3.1 designates critical habitat areas that “require special protection because they are highly sensitive to change and could be adversely affected by development,” including vernal pools, native bunch grasses, and oak savannahs. Currently approximately 3 million acres of grassland—3% of the state land area and 20% of all California grasslands— are in open space designation (Table 23.4). The ecological significance of this designation is difficult to ascribe given the wide range of management goals for these areas. In the longer term, as in the conservation element, open space affords little protection in excess of what federal and state regulations supply.

R E G U L A T O R Y P R O T E C T I O N A N D C O N S E R VA T I O N

305

Land Use Element California Code section 65302(a) requires a general plan land use element that “designates the general location and intensity of housing, business, industry, open space, education, public buildings and grounds, waste disposal facilities, and other land uses.” Of the 16 general plans examined, 15 had land use elements and one had a combined land use, open space, and conservation element. Grasslands were recognized in four of the available land use elements, usually associated with agricultural and historical land uses within county boundaries. No land use elements included recognition of native grasslands (Appendix 23.4). In the Paradise Urban Reserve area of Butte County, for example, native grasslands are recognized as part of the heritage and aesthetic environment of local communities. More broadly, the land use element provides guidance for zoning densities for particular land uses, as will be discussed at greater length subsequently.

Optional Elements: Agriculture Element Twenty-six counties have adopted agriculture elements in their general plans, including 11 of the 22 counties investigated. Agriculture elements are primarily designed to protect and promote agriculture and use a variety of approaches to accomplish these goals. These approaches include technical assistance to farmers, establishing agricultural support zones to ensure the economic viability of agricultural operations, promoting agricultural products, and striving to minimize the conflict between residential and agricultural areas. Because a large percentage of California’s grasslands are found in agriculturally zoned grazing lands (Table 23.4), agriculture elements can have significant effects on grassland ecosystems. Land use restrictions for preservation of agricultural lands can keep residential development from encroaching on grasslands. On the other hand, policies that promote retention of prime farmland over less valuable rangelands may direct development onto grasslands. Few agriculture elements directly address grassland or native ecosystems, and none mentions grasslands or native ecosystems in agriculture element policies. Merced County mentions “native pasture” when describing trends in land conversion. Stanislaus County mentions “native ecological systems” and “grasslands” (in Appendix C and Appendix A of its general plan, respectively).

Coastal Element Under the California Coastal Act Section 30108.55, local governments are permitted to prepare coastal plans in order to protect, maintain, and enhance coastal zone resources, plan for public recreation in the coastal zone, and coordinate between local and state initiatives that affect the coastal zone. CDFG-listed sensitive grassland types, including coastal terrace prairie, are recognized as Environmentally Sensitive Habitat Areas (ESHA), and as such are provided with some protection.

306

POLICY AND MANAGEMENT

Grasslands are occasionally mentioned in county coastal plans. The Mendocino County coastal element, for example, acknowledges the presence of grassland habitat within the coastal zone but does not recognize native grassland as a habitat of special consequence or afford any special protections to coastal grassland systems. Alternatively, the Santa Barbara County coastal element lists specific coastal management areas in which native grassland species are known to exist.

Other Optional Elements Optional elements that may affect grassland ecosystems include resource conservation, scenic highways, growth management, parks and recreation, and biology elements. Counties that have scenic highway elements include Butte, Madera, Mariposa, Mendocino, Sacramento, San Benito, Shasta, and Tehama. The contents of this element are often paralleled in the mandatory elements of other counties, especially the open space element. Scenic highway policies often overlap with grassland systems but generally only limit zoning density in included regions rather than forestalling development entirely. None of the counties examined had a growth management or parks and recreation element that afforded any special recognition to grassland or native ecosystems. None of the counties that we examined had biological elements in their general plans. ZON I NG

Zoning ordinances are a reflection of the goals and policies found in general plan elements. Thus, the lack of specific attention to grasslands in the general plans is mirrored in zoning ordinances. As with general plans, zoning ordinances can be changed, and variances are often granted, which can reduce the efficacy of zoning as a tool for grassland conservation. The zoning types most likely to indirectly preserve grassland habitats are those that promote large lot size or low housing densities such as open space, agriculture, and very low-density residential. In current general plans, more than 80% of grasslands fall in one of these three categories, with 54% of grasslands presently zoned for agricultural use, and 20% zoned for open space (Table 23.4). C OU NT Y I M P LE M E NTATION OF TH E CALI FOR N IA E NVI RON M E NTAL QUALIT Y ACT (CEQA)

Counties must comply with CEQA, which requires them to prepare reports examining the significant environmental impacts of proposed development projects before approving such projects. CEQA allows counties to set thresholds when determining what a significant environmental impact is. Furthermore, counties can create resource protection ordinances with specific thresholds that can be used as the thresholds of significance in CEQA. Generally, significant environmental impacts, as defined under CEQA or the county’s implementing rules, must either be avoided or mitigated.

The two areas of CEQA review of particular relevance to the protection of California grasslands are the agricultural and biological reviews. Agricultural impacts are generally considered significant if they conflict with Williamson Act contracts (see following paragraphs) or could result in the conversion of farmland to nonagricultural use. Because large areas of grassland are either under Williamson Act contracts, included in agricultural zones, or adjacent to agricultural areas, CEQA review has the potential to influence large areas of grassland. Biological impacts in grasslands are generally considered significant if they threaten sensitive species or habitat types, including:(1) candidate, sensitive, or special status species identified in regional or local plans or by the CDFG or USFWS; (2) sensitive communities identified in regional or local plans or by the CDFG or USFWS; (3) areas covered by CWA Section 404; (4) wildlife corridors; or (5) areas covered by habitat conservation plans. In addition, counties may use other criteria, such as listing by the California Native Plant Society (CNPS), to determine whether the biological impacts of a project would be significant. Annual grasslands are generally not considered sensitive communities; thus, impacts to them do not generally meet the significance threshold for which CEQA mandates mitigation or avoidance. However, if individual species of concern, such as endangered animals or plants, are associated with grasslands, then CEQA’s mandates may apply. Certain native grassland communities, such as valley needlegrass, have been identified as rare by the CDFG and thus impacts to these communities may be significant and require mitigation or avoidance. For such communities, thresholds of significance, which are applied on a case by case basis, will determine whether project impacts are significant enough to require mitigation or avoidance. In the absence of explicit guidelines, counties rely on the judgment of county agencies or hired consultants to determine whether a project’s impacts to biological resources would be significant. Inventorying and listing of significant grassland communities by county agencies, in combination with scientifically based thresholds of significance, would increase the likelihood that impacts to such communities would be avoided or mitigated. For example, the County of Santa Barbara Environmental Thresholds and Guidelines Manual states that impacts to “small acreages” of non-native (annual) grassland are not considered significant if “wildlife values are low.” Santa Barbara defines native grasslands as containing at least 10% native grass cover. This specific guideline is used with the general CEQA guidelines to determine whether impacts to native grassland communities in the county are significant. Counties could also define a significance threshold concerning cumulative impacts to grassland communities, which could help mitigate the piecemeal development of annual grasslands. For example, a county could define some proportion of its total grassland area that it desired to retain. The impact of an individual project could then be evaluated

relative to this threshold. This would help to set an upper limit on the amount of grassland developed in the county. TH E W I LLIAM SON ACT

The Williamson Act empowers California county governments to offer contracts to private landowners to preserve land for agricultural and open space use in exchange for a reduced property tax assessment. Williamson Act contracts limit contracted parcels of land to agricultural uses for ten years, after which time the contract is automatically renewed unless the landowner files for non-renewal at least nine years prior to the contract’s expiration. The Williamson Act also provides for the establishment of Farmland Security Zones (FSZs), which put land under contract for twenty years in return for greater property tax reduction.

TA B L E 23.5 Grassland Area Currently under Williamson Act Contracts

County

Annual grasslands in Williamson Act (acres)

Percent annual grassland in Williamson Act

Glenn

215,671

99.9%

San Benito

345,930

77.9%

Tehama

371,992

74.5%

Madera

192,720

72.6%

Santa Barbara

181,845

64.4%

Fresno

287,035

54.3%

Humboldt

90,302

40.9%

Sacramento

64,954

32.3%

Currently all California counties except Del Norte, San Francisco, Inyo, and Yuba offer Williamson Act contracts. Of the nearly 27 million acres of agricultural land in California in 2002, slightly more than 5 million acres were classified as prime farmland. Close to another 7 million acres were classified as important farmland of other types. The balance, roughly 14 million acres, was classified as nonprime farmland, which is primarily grazing land, which does not require high-quality soil types. Nearly 11 million is held under Williamson Act contracts as nonprime farmland. In the eight counties for which GIS data are available online (Fresno, Glenn, Humboldt, Madera, Sacramento, San Benito, Santa Barbara, Tehama), an average of 64.6% of the total grassland area is held under Williamson Act contracts (Table 23.5). In recent years the acreage held under Williamson Act contracts has remained relatively stable. In the period from 1990–2003, the California Division of Land Resource Protection reports that acreage increased only slightly, from approximately 15.97 million acres in 1990 to 16.56 acres in 2003, but

R E G U L A T O R Y P R O T E C T I O N A N D C O N S E R VA T I O N

307

these numbers do not reflect changes to the spatial layout of holdings (http://conservation.ca.us.gov/DLRP/lca/stats_ reports/index.htm). Some landowners may choose to file for nonrenewal when the profit of developing land exceeds the tax savings under the Williamson Act, whereas speculators may purchase land far from existing urban centers and place it under contract in order to hold it at a reduced tax rate until development becomes profitable.

Interactions between Federal, State, and County Policy There are significant interactions among federal, state, and county policies. For example under CEQA, it is up to the counties to set thresholds for determining significant biological impacts. However, often the biological elements that are the foci of CEQA actions are designated as sensitive by the state and/or federal government. Similarly, the Williamson Act is often closely related to the general plan. Some counties specifically encourage entry into Williamson Act contracts in their general plans. Humboldt County Agriculture Element Policy 2523, for example, states that “Agricultural lands shall be conserved and conflicts minimized between agricultural and nonagricultural uses . . . by broadening the utility of agricultural preserves and the Williamson Act Program to accommodate and encourage intensely managed farms” (Humboldt County Community Development Services Department Planning Division 1984). In addition, counties can use their authority under the Williamson Act to encourage preservation of significant open space lands and wildlife habitat by including such areas in Agriculture Preserve delineations, which circumscribe lands eligible for enrollment in the program. ESA, CESA, and CEQA play somewhat complementary roles in conserving grassland habitats. The ESA has strong prohibitions against indirect take but does not apply those protections to plants, whereas CESA has weaker indirect take provisions but stronger plant protections. ESA and CESA provide protection to listed species, while CEQA applies to species or habitats of concern which have not yet been listed and therefore do not receive protection from the other regulations. More generally, ESA, CESA, and CEQA strictures are forcing counties to reexamine their land planning approaches to take fuller account of wildlife and natural resource issues that have traditionally been the domain of state and federal resource agencies (Scott et al., 2006).

Summary of Public Policies and Regulations Several important conclusions can be drawn from the preceding analysis of public policies and regulations. First, most grasslands are privately owned, and at least two of every three grassland acres are zoned for agricultural use or open space (Table 23.4) and/or managed under Williamson Contracts. To a large extent the future of these grasslands depends on private land management priorities and approaches. Second, in areas where grasslands are under the greatest pressure of land development— notably in the Bay area, southern San Joaquin Valley, Riverside Basin, and the

308

POLICY AND MANAGEMENT

South Coast—hundreds of thousands if not millions of acres of grassland are being protected as habitat for endangered species. The future of grasslands in these areas depends largely on the design and ongoing management of habitat reserves, which at best is a contentious and expensive negotiated settlement between the agencies and stakeholders (Davis et al. 2005). Third, in the absence of regulated species, grassland ecosystems, particularly annual grasslands dominated by non-native species, are protected mainly for open space or rangeland values. Such protection will certainly contribute to conserving habitat for many grassland-dependent species, although in the absence of explicit biological management goals and ecological monitoring the extent of conservation remains conjectural. Furthermore, given increasing restrictions on development in prime farmland, oak woodlands, and coastal shrublands, grasslands could receive an increasing share of development in the future.

Additional Tools for Grassland Conservation on Private Lands Tools outside of government regulation and management that promote sustainable use and restoration of grassland ecosystems are being increasingly applied in California. A compilation of web links to these and other public and private incentive programs, as well as further information on federal, state and local policies previously discussed, is provided in Appendix 23.5, which is posted online at http://www2.bren.ucsb.edu/~ca_grasslands/appendix5.htm.

Socioeconomic Benefits of Grassland Conservation Many private landowners currently perceive conservation measures as unfairly expensive and a threat their livelihood, freedom, and property rights (Ling 1998; Esseks and Drozd 2002). To help alleviate these concerns, there are several organizations that promote conservation from an agricultural land user’s viewpoint, including the California Rangeland Trust and American Farmland Trust. Even those who are interested in conservation actions typically lack the time, money, and technical expertise to implement them (Ingram and Lewandrowski 1999). In addition, they are often unaware of the significant benefits that such actions can bring them, including economic benefits and the incentive programs that are available to help make conservation actions both affordable and money saving (Ingram and Lewandrowski 1999; Phillips 2001; Balmford et al. 2002). Conservation actions can help landowners avoid economic costs of poor land management, boost profits, and generate new income opportunities (Richards and George 1996; Balmford et al, 2002). For example, overgrazing that causes weed invasions (Hobbs and Humphries 1995) may increase control costs and reduce land value and livestock forage capacity (Sheley and Petroff 1999; Naylor 2000). Weeds cost the State of Montana $100 million per year, and an invasion reduced the real estate value of one ranch by nearly 60% (Sheley et al. 1998). Costs may also include water

loss. Gerlach (2004) estimated that in the Sacramento River watershed alone, yellow starthistle may cause losses of soil moisture of 15 – 25% of mean annual precipitation, with the cost of lost water estimated to range from $16 million to $75 million (U.S.) per year. Devising a management plan that minimizes invasions can simultaneously reduce these costs, boost profits, and benefit native species and ecosystem processes (Hobbs and Humphries 1995; Mack et al. 2000; Naylor 2000; Phillips 2001). Conservation actions can also help grassland owners diversify their income. Sustainable management can qualify them to tap into the growing market for sustainably produced products (Kennard 2005). Maintaining healthy fish, wildlife, and wildflower populations improves prospects for landowners seeking to generate income through hunting, fishing, and ecotourism (Esseks and Drozd 2002; Tate 2003), which can greatly exceed income generated through agriculture and ranching (Balmford et al. 2002). It also makes them eligible for incentives that reward good stewardship (McQueen and McMahon 2003). Concerning development pressures, there is a prevailing thought that as California grows, it is inevitable that more homes must be built. While this is true, county agencies and developers that use subdivision for conservation purposes allow for a higher degree of efficiency in development, minimizing potential for destruction of open tracts of grassland. This holds particularly when alternative scenarios involve large parcel creation. Conservation easement language can also reflect this, specifying that the land be subdivided within a boundary, easing development rights for the remainder of the tract.

DI R ECT F U N DI NG

Financial incentives for grassland conservation

R EG U LATORY STR EAM LI N I NG

Numerous incentives are offered by federal, state, local, and private organizations to provide grassland owners, local governments, and NGOs with financial and technical assistance for implementing conservation measures (Appendix 23.5), as briefly summarized in the following paragraphs.

Incentives include permitting landowners to conduct otherwise restricted activities, such as hunting and fishing, in exchange for implementing a management plan accepted by a federal or state conservation agency (Appendix 23.5). Examples of such programs include Safe Harbor Agreements (Bean et al. 2001), aimed at granting landowners more flexibility in promoting and protecting endangered species.

TAX I NCE NTIVE S

Tax incentive programs help landowners reduce state and federal income and other taxes with a credit for part or all of the costs of conservation practices (Appendix 23-5). Perhaps the most common type of program is the conservation easement, which is designed on a case-by-case basis and can provide income, property, and estate tax credits (McQueen and McMahon 2003). Easements reward landowners for conserving grasslands for agricultural and open space purposes by assessing lands at a reduced tax rate that is based on the parcel’s agricultural or open space value instead of its higher development value. McLaughlin (2004) estimated federal income tax savings for donating a conservation easement worth $500,000 to be $157,500 for a high-income ($250,000/year) landowner, $36,450 for a middle-income ($75,000/year) landowner, and $9,450 for a low-income ($35,000/year) landowner.

Opportunities include federal, state, local, and private grants, purchases of conservation easements, cost sharing, and reimbursement of conservation-related expenses (Appendix 23-5). Federal programs offered by the Natural Resource Conservation Service (NRCS) through the Farm Bill, for example, provide funding and cost-share opportunities to improve water management and quality, erosion control, wildlife habitat protection, and overall quality of land management by incorporating conservation into ranching and farming operations. P U B LIC F I NANCI NG

Support among the public for conservation of natural lands and open space has been growing in recent years. In 2005 alone, ballot measures across the country created a total of $1.7 billion earmarked for conservation (Trust for Public Land 2005). As demand for open space in California grows, public financing for grassland conservation will likely increase in importance. TECH N ICAL AS S I STANCE

Landowners often need assistance with identifying and understanding relevant programs, understanding regulations, applying for permits or programs, or developing conservation, restoration, monitoring, and sustainable management plans. This type of advice is available through personal consultations, extension short courses, and other government and nongovernment sources (Appendix 23.5).

EC OSYSTE M S E RVICE S I NCE NTIVE S

An emerging category of incentives revolves around the stewardship of ecosystem services whose economic value is often comparable to or even greater than the value of goods and services that have a market value, such as production of meat and crops (Daily 1997; Balmford et al. 2002). Ecosystem services provided by grasslands include pollination of crops by insects, soil fertility and stabilization, water filtration and storage, livestock forage, species diversity, carbon storage, and genetic material for improving our food crops (Daily et al. 1997; White et al. 2000). Sala and Paruelo (1997) noted that not recognizing these services has resulted in management systems aimed at maximizing production of marketable goods and services such as meat. As Balmford et al. (2002)

R E G U L A T O R Y P R O T E C T I O N A N D C O N S E R VA T I O N

309

noted, the development of market instruments that capture the social values of relatively undisturbed ecosystems—for instance, through carbon or biodiversity credits or through premium pricing for sustainably produced products — is a crucial step toward sustainability. E DUCATI NG G RAS S LAN D MANAG E R S ABOUT TH E B E N E F ITS OF I NCE NTIVE S

In a recent survey of agricultural landowners in California, 45% of respondents were not making an effort to minimize overgrazing, 55% were not making an effort to minimize soil erosion, and two-thirds or more were failing to make an effort to protect or improve wildlife habitats (66%), minimize flow of chemical fertilizers or pesticides into surface waters or groundwaters (69%), prevent or minimize flow of livestock wastes into surface waters or groundwater (75%), or protect or improve wetlands (83%) (Esseks and Drozd 2002). In many cases, this reflects that landowners remain unaware of the conservation options available to them (Bean et al. 2001). Unfortunately, awareness of incentives is sufficiently low that programs are sometimes defunded on grounds of lack of use (Dawn Afman, U.S. NRCS, personnal. communication). This suggests that government agencies and NGOs need to develop better strategies to increase public awareness of and support for them ( Jacobsen 2003; Farrior 2005). Such strategies will have to appeal to personal concerns and values such as financial health, independence, and ingenuity by framing conservation actions as opportunities (as opposed to threats)

to boost profits and solve time-consuming management problems (e.g., Farrior 2005). Schultz and Zelezney (2003) noted that when targeting the voting public in California as part of a campaign to protect open space, “our children’s future” and quality of life (for our family and future generations) were foremost among reasons people cared about protecting land. Educational messages and materials can be made available through such channels as direct mailings, the Internet, information booths at farmers’ markets, extension short courses, and presentations at meetings of ranching, farming, and outdoor recreation organizations (Richards and George 1996; Jacobsen 2003; Tate 2003).

Concluding Remarks A large number of policies currently operate to influence the future of privately owned grasslands in California, ranging from relatively strong regulations such as ESA and CESA to weaker but potentially influential guiding policies in county general plans to nonregulatory financial incentives such as tax breaks and financial assistance programs. The association of grasslands with wide-ranging endangered species such as the California condor and San Joaquin kit fox (Vulpes macrotis mutica) and locally important species such as the Stephens kangaroo rat (Dipodomys stephensi) and Bay checkerspot butterfly (Euphydryas editha bayensis) has led to the establishment of large grassland reserves in many areas of the state undergoing large-scale development, with the trend

F I G U R E 23.3. Comparison of threatened and endangered grassland species distributions with projected

development in California. The threatened and endangered species map was derived from California Natural Diversity Database point data and was generalized to 10-kilometer grid cells. Projected development was derived from FRAP data (CDF-FRAP 2002). The original FRAP data combined historical growth patterns with population projections to forecast growth for 5-kilometer grid cells. For each 10kilometer grid cell in our map, we calculated the area of underlying 5-kilometer cells projected to reach a density of 1 house per 20 acres by the year 2020.

310

POLICY AND MANAGEMENT

being toward larger, subcounty- or countywide biological conservation programs. Clearly, ESA and CESA are the major drivers of grassland protection in urbanized and urbanizing areas of California. In more rural areas, the majority of grasslands are presently under agricultural and open space zoning, and a large fraction is managed under Williamson Act contracts. Nevertheless, the future of these grasslands is tenuous, as continued rapid population growth and demand for rural residential development raise the market value of these lands. Counties have authority to protect habitat types that they consider important, but in the absence of legally protected species or community types, few counties currently attribute special significance or conservation value to grassland ecosystems. Comparing the projected location of future development with the current distribution of grassland-associated threatened and endangered species reveals areas, particularly central and southern California, where grasslands are likely to be heavily impacted (Figure 23.3). Counties where rapid growth and sensitive species distributions overlap will be the focus of intense planning efforts but also present an opportunity for implementation of more proactive grassland conservation policies. The future of California grasslands also depends in part on public policies that tend to operate against grassland conservation. Such policies have not been analyzed here, but it should at least be mentioned that subsidies for agriculture, livestock grazing, water use, and road building and maintenance can promote cropping in marginal areas better suited for rangeland, grazing, and subdivision of rural lands (Myers and Kent 1998; Dale et al. 2000; GAO 2005). Public policies pertaining to road network development and maintenance have a pervasive influence, because roads serve as corridors

for the spread of invasive species, fragment grassland habitats, and degrade roadside environments (Forman et al. 2003; Gelbard and Belnap 2003; Gelbard and Harrison 2003). Financial incentives are playing an increasingly important role in the protection and management of privately owned grasslands. Public agencies and conservancies such as The Nature Conservancy, the Trust for Public Land, the California Rangeland Trust, and dozens of county and local land trusts have already invested hundreds of millions of public and private dollars to protect grasslands through outright acquisition and conservation easements. Other kinds of incentives can also help improve the effectiveness of conservation partnerships. For example, The Malpais Borderlands Group, a coalition of ranchers, agencies, The Nature Conservancy, and foundations in southern New Mexico and Arizona developed a grass banking approach to protecting and restoring grasslands (Page 1997). By joining the group, ranchers whose lands have become degraded gain access to a grass bank on which to graze their livestock (the 150,000 ha Gray Ranch), providing rest from grazing that allows their grasses to recover. They can also receive native reseeding, technical assistance, and monitoring from management experts. In exchange, they take steps such as (1) donating a conservation easement to protect their land as open space (which reduces their taxes) and (2) allowing natural fires to burn, which helps beat back invasive plants and stimulates the recovery of native grasses. Perhaps such an approach to landscapescale grassland conservation could be tailored for use in California. For example, grass banking could give ranchers a place to graze during years that their land is treated with fire and reseeding to eradicate weeds and restore native species (Corbin et al. 2004).

R E G U L A T O R Y P R O T E C T I O N A N D C O N S E R VA T I O N

311

San Mateo thorn-mint San Diego thorn-mint Munz’s onion Large-flowered fiddleneck Hearst’s manzanita Presidio manzanita Braunton’s milk-vetch Clara Hunt’s milk-vetch Baker’s blennosperma Indian Valley brodiaea Thread-leaved brodiaea Kaweah brodiaea Chinesescamp brodiaea Tiburon Mariposa lily Tiburon Indian paintbrush California jewelflower Coyote ceanothus Purple amole

Acanthomintha ilicifolia

Allium munzii

Amsinckia grandiflora

Arctostaphylos hookeri ssp. hearstiorum

Arctostaphylos hookeri ssp. ravenii

Astragalus brauntonii

Astragalus clarianus

Blennosperma bakeri

Brodiaea coronaria ssp. rosea

Brodiaea filifolia

Brodiaea insignis

Brodiaea pallida

Calochortus tiburonensis

Castilleja affinis ssp. neglecta

Caulanthus californicus

Ceanothus ferrisae

Chlorogalum purpureum var. purpureum

Plants

Common name

Acanthomintha duttonii

Scientific name

SE

ST

ST

SE

SE

SE

SE

SE

ST

SE

SE

SE

ST

SE

SE

State

FT

FE

FE

FE

FE

FT

FT

FE

FE

FE

FE

FE

FE

FT

FE

Federal

Status

San Joaquin kit fox

Riparian brush rabbit

Bank swallow

Salt-marsh harvest mouse

California red-legged frog

California condor

Sandhill crane

Blunt-nosed leopard lizard

Peregrine falcon

Stephens kangaroo rat

Fresno kangaroo rat

Giant kangaroo rat

Swainson’s hawk

San Joaquin antelope squirrel

California tiger salamander

Longhorn fairy shrimp

Conservancy fairy shrimp

Invertebrates

Branchinecta longiantenna

Branchinecta conservatio

Vulpes macrotis mutica

Sylvilagus bachmani riparius

Riparia riparia

Reithrodontomys raviventris

Rana aurora draytonii

Gymnogyps californianus

Grus canadensis tabida

Gambelia sila

Falco peregrinus anatum

Dipodomys stephensi

Dipodomys nitratoides exilis

Dipodomys ingens

Buteo swainsoni

Common name Vertebrates

Ammospermophilus nelsoni

Ambystoma californiense

Scientific name

A P P E N D I X 23.1 Listed Threatened and Endangered Grassland-Associated Species

ST

SE

ST

SE

SE

ST

SE

SE

ST

SE

SE

ST

ST

State

FE

FE

FE

FE

FE

FT

FE

FE

FE

FE

FE

FE

Federal

Status

Howell’s spineflower Robust spineflower Sonoma spineflower Ashland thistle Fountain thistle Presidio clarkia Vine Hill clarkia Pismo clarkia Springville clarkia Palmate-bracted bird’s-beak Yellow larkspur Conejo dudleya Santa Clara Valley dudleya Laguna Beach dudleya Hoover’s woollystar San Diego button-celery Roderick’s fritillary Striped abobe lily Otay tarplant Gaviota tarplant Marin dwarf flax Lake County dwarf-flax Santa Cruz tarplant Contra Costa goldfields San Joaquin woollythread Western lilly Douglas’ meadowfoam

Chorizanthe howellii

Chorizanthe robusta

Chorizanthe valida

Cirsium ciliolatum

Cirsium fontinale var. fontinale

Clarkia franciscana

Clarkia imbricata

Clarkia speciosa ssp. immaculata

Clarkia springvillensis

Cordylanthus palmatus

Delphinium luteum

Dudleya abramsii ssp. parva

Dudleya setchellii

Dudleya stolonifera

Eriastrum hooveri

Eryngium aristulatum var. parishii

Fritillaria roderickii

Fritillaria striata

Hemizonia conjugens

Hemizonia increscens ssp. villosa

Hesperolinon congestum

Hesperolinon didymocarpum

Holocarpha macradenia

Lasthenia conjugens

Lembertia congdonii

Lilium occidentale

Limnanthes douglasii ssp. sulphurea

SE

SE

SE

SE

ST

SE

SE

ST

SE

SE

ST

SE

SE

SE

SE

SE

SE

SE

ST

FE

FE

FE

FT

FT

FE

FT

FE

FT

FT

FE

FT

FE

FE

FT

FE

FE

FE

FE

FE

FE

FE

Streptocephalus woottoni

Speyeria zerene behrensii

Speyeria callippe callippe

Lepidurus packardi

Icaricia icarioides missionensis

Euphydryas editha quino

Euphydryas editha bayensis

Euphilotes enoptes smithi

Elphrus viridis

Cicindela ohlone

Branchinecta lynchi

Branchinecta sandiegoensis

Riverside fairy shrimp

Behren’s silverspot butterfly

Callippe silverspot butterfly

Vernal pool tadpole shrimp

Mission blue butterfly

Quino checkerspot

Bay checkerspot butterfly

Smith’s blue butterfly

Delta green ground beetle

Ohlone tiger beetle

Vernal pool fairy shrimp

San Diego fairy shrimp

(Continued)

FE

FE

FE

FE

FE

FE

FT

FE

FT

FE

FT

FT

Milo Baker’s lupine Bakersfield cactus Lake County stonecrop White-rayed pentachaeta Lyon’s pentachaeta San Francisco popcorn-flower Calistoga popcorn-flower Hartweg’s golden sunburst San Joaquin adobe sunburst Keck’s checkerbloom Metcalf Canyon jewel-flower Tiburon jewel-flower Kneeland prairie pennycress Showy Indian clover California vervain

Lupinus milo-bakeri

Opuntia basilaris var. treleasei

Parvisedum leiocarpum

Pentachaeta bellidiflora

Pentachaeta lyonii

Plagiobothrys diffusus

Plagiobothrys strictus

Pseudobahia bahiifolia

Pseudobahia peirsonii

Sidalcea keckii

Streptanthus albidus ssp. albidus

Streptanthus niger

Thlaspi californicum

Trifolium amoenum

Verbena californica

ST

SE

SE

SE

ST

SE

SE

SE

SE

SE

ST

SE

State

FT

FE

FE

FE

FE

FE

FT

FE

FE

FE

FE

FE

FE

FE

Federal

Status

Scientific name Vertebrates

Common name State

NOTE: FE: Federally listed as Endangered FT: Federally listed as Threatened SE: State listed as Endangered ST: State listed as Threatened Plant Selection: State and Federal threatened and endangered grassland associated species were selected from the Calflora database. Plants from communities m24 or m25 (Coastal Prairie or Valley Grassland) were selected. From this list, T  E sp. were selected. The regioncode was used to populate a list of T  E sp. by county. State “rare” species were excluded. Animal Selection: State and Federal threatened and endangered grassland associated species were selected from the Wildlife Habitat Relationship database from the CDFG. Invertebrate Selection: Federal list of California invertebrates was compared with life history records to determine association with grassland ecosystems.

Butte County meadowfoam

Plants

Common name

Limnanthes floccosa ssp. californica

Scientific name

A P P E N D I X 23.1 ( C O N T I N U E D ) Listed Threatened and Endangered Grassland-Associated Species

Federal

Status

A P P E N D I X 23.2 Counties Surveyed and Grassland Areas

Acres

Proportion of total grassland area

Cumulative proportion of total grassland area

1,343,693

12

12

San Luis Obispo

991,126

9

21

Monterey

638,047

6

27

Fresno

529,088

5

32

Tehama

499,305

5

37

Merced

497,017

5

41

San Benito

443,567

4

45

Tulare

340,365

3

49

Stanislaus

319,211

3

52

Santa Barbara

282,500

3

54

Mendocino

277,068

3

57

Madera

263,414

2

59

Sonoma

227,310

2

61

Humboldt

221,038

2

63

Glenn

215,657

2

65

Kings

209,388

2

67

Sacramento

201,053

2

69

Siskiyou

187,571

2

71

San Joaquin

169,555

2

73

Mariposa

165,737

2

74

San Diego

163,106

2

76

County Kern

R E G U L A T O R Y P R O T E C T I O N A N D C O N S E R VA T I O N

315

24,354

San Luis Obispo

38,044

Merced

8,616

3,393

Santa Clara

8,238

4,094

Alameda

Butte

6,889

Contra Costa

Riverside

3,630

Solano

59

3,348

San Diego

Calaveras

1,122 2,199

Mariposa

6,580

San Joaquin

2,703

Siskiyou

0

Kings

Sacramento

7,443

6,657

Sonoma

4,055

2,619

Madera

Glenn

3,595

Mendocino

Humboldt

2,735 3,773

Santa Barbara

17,381

Stanislaus

Tulare

6,299

24,824

Tehama

San Benito

25,234

Fresno

5,458

36,158

Kern

Monterey

Acres

County

1

0

0

0

1

1

1

0

0

0

0

0

0

0

0

1

0

0

0

0

1

1

3

1

1

0

1

1

% of County

Public reserve

1,376

16,042

5,159

6,200

2,422

11,018

3,484

54,861

7,929

237

16,141

4,596

7,289

5,935

6,227

8,179

7,082

11,977

30,144

205

9,422

34,520

5,347

19,118

76,569

72,477

163,014

137,511

Acres

0

0

1

1

0

2

1

2

1

0

0

1

1

1

0

1

1

1

2

0

0

4

0

1

2

3

8

3

% of County

Public non reserve

1,053

3,244

4,379

600

15,674

15,520

1,567

7

0

3,902

620

4,977

0

539

30

1,384

509

64

1,685

1,688

2,488

346

7,116

20,887

3,138

726

1,127

5,056

Acres

0

0

1

0

3

3

0

0

0

0

0

1

0

0

0

0

0

0

0

0

0

0

1

1

0

0

0

0

% of County

Private reserve

127,709

112,038

135,295

140,812

132,512

121,544

148,673

104,889

155,609

164,294

164,230

188,777

202,098

205,128

207,339

211,090

253,203

261,432

246,897

314,583

311,074

402,402

446,510

434,476

424,147

559,385

802,630

1,164,968

Acres

12

2

20

17

25

24

26

4

17

18

4

30

23

24

9

21

18

12

15

32

10

45

35

23

11

26

38

22

% of County

Private non reserve

A P P E N D I X 23.3 Ownership and Management Profile of California Grasslands

138,754

139,562

144,892

151,005

154,702

154,971

157,353

163,106

165,737

169,555

187,571

201,053

209,388

215,657

221,038

227,310

263,414

277,068

282,500

319,211

340,365

443,567

497,017

499,305

529,088

638,047

991,126

1,343,693

Acres

Total

93

83

96

94

96

88

95

64

94

99

88

96

97

95

94

93

96

94

88

99

92

91

91

91

81

88

81

87

% Private

7

17

4

6

4

12

5

36

6

1

12

4

3

5

6

7

4

6

12

1

8

9

9

9

19

12

19

13

% Public

4,092

Colusa

5,112

San Bernardino

420

Del Norte

Percent of Total Grassland Area 3.25

349,046

0

Imperial

Total

0

Inyo

10

430

Sierra

San Francisco

2,162

1,050

Modoc

1,665

875

Nevada

Alpine

2,424

Orange

Santa Cruz

1,856

Lassen

655

657

Sutter

San Mateo

12,113

Mono

598

2,676

Trinity

Plumas

2,301

Ventura

Lake

6,128

Los Angeles

588

4,225

El Dorado

Napa

153 3,791

Amador

489 2,928

Yuba

4,174

Placer

Tuolumne

521

361

Shasta

Yolo

27,090

Marin

0

0

0

0

0

0

1

0

0

0

0

0

0

1

0

0

0

0

0

1

0

0

0

1

0

0

0

1

0

7

8.14

875,577

0

12

42

919

4,114

6,682

57

991

11,500

1,712

1,836

14,164

242

22,306

19,323

2,147

4,416

4,967

2,006

3,141

3,272

1,070

1,250

20,588

1,228

12,760

1,539

1,703

1,700

5,407

0

0

0

0

1

1

0

0

0

0

0

0

0

1

1

0

0

1

0

0

0

0

0

5

0

1

0

0

0

1

0.97

103,962

0

0

0

72

5

0

22

2,538

2

2

432

35

72

0

35

0

15

62

35

67

32

0

0

37

0

0

0

1,181

5

988

0

0

0

0

0

0

0

1

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

0

87.64

9,422,536

0

0

185

2,656

3,810

2,019

12,249

18,967

14,060

28,370

28,819

19,694

38,303

4,957

20,939

34,302

36,529

43,905

59,942

55,528

59,586

71,004

76,986

58,224

99,606

86,159

105,482

102,870

107,787

85,852

0

0

0

0

1

0

4

5

1

5

6

1

10

0

1

0

2

5

12

5

2

6

20

14

10

6

16

14

4

23

100

10,751,121

0

12

237

4,067

8,359

10,863

13,993

23,151

26,613

30,959

33,512

35,748

39,274

39,375

40,895

41,562

43,635

51,234

62,571

64,864

67,115

75,865

78,390

81,778

101,323

103,093

107,543

109,846

109,853

119,337

0

0

78

67

46

19

88

93

53

92

87

55

98

13

51

83

84

86

96

86

89

94

98

71

98

84

98

95

98

73

0

100

22

33

54

81

12

7

47

8

13

45

2

87

49

17

16

14

4

14

11

6

2

29

2

16

2

5

2

27

1980 –1998 1972–1981

Fresno

Merced

San Benito

Tularec

Stanislaus

Santa Barbara

Mendocino

Madera

Sonoma

Humboldt

Kings

Sacramento

Mariposa

San Diego

4

6

7

8

9

10

11

12

13

14

16

17

20

21

2

1

0

0

0

0

0

1

0

0

0

0

1

0

1,2

Land use

3

1a

3

2

0

1

1,2

2a

0

1

0

0

3

2

1

DNE

1

0

DNE

0

0

DNE

DNE

0

1

1,3

DNE

3a

DNE

1a

2

1,2

NA

DNE

1,3

DNE

Agriculture

1a

1,3

Open space

3a

NA

0a

Conservation

Mandatory elements

NA

DNE

DNE

DNE

0

DNE

DNE

1

3

DNE

DNE

DNE

DNE

DNE

NA

DNE

Coastal

1

0

NA

DNE

DNE

DNE

NA

1

NA

DNE

0

NA

DNE

NA

DNE

DNE

Scenic highways

Optional elements

DNE

0

DNE

DNE

1

DNE

NA

DNE

0

DNE

DNE

DNE

DNE

NA

DNE

DNE

Resource conservation

NOTE: Information on whether counties recognize grassland ecosystems, recognize native ecosystems, or recognize native grassland ecosystems was drawn from current and available county general plans (16 total, excluding Monterey, Tehama, Glenn, Siskiyou, and San Joaquin). 0
No recognition of grasslands or native ecosystems, 1
specific recognition of grassland ecosystems, 2
specific recognition of native ecosystems, 3
specific recognition of native grasslands, NA
element not readily available for this county, DNE
element does not exist a Several counties combine mandatory elements. For these counties the columns have been merged. b Only Part 4 of the land use element for San Luis Obispo County was examined c The land use element for Tulare County was unavailable. This information is from a 1991 general plan policy summary of policies from 1964 to 1998.

2000

2003

1974 –1993

1996 –2004

1982 –1984

1989

1995

1981

1975 – 2004

1992 –1994

1990 – 2002

2000

1974 –1998

San Luis

2004

2

Obispob

Year

Kern

County

1

Rank by grassland area

A P P E N D I X 23.4 General Plan Element Profile

TWENTY FOUR

Epilogue Future Directions MAR K R. ST R O M B E R G, CA R LA M. D’ANTO N I O, AN D J E F F R EY D. C O R B I N

The preceding chapters provide an understanding of our current knowledge of the ecological, evolutionary, and cultural forces influencing grasslands of the California floristic province and their conservation and management. Many of the chapters provide discussions that support or elaborate on some of the syntheses found in the first, and only other edited volume on California grasslands, published by Huenneke and Mooney in 1989. Other chapters provide wholly new approaches or syntheses of both basic and applied research. In this epilogue we discuss some of the ways in which our understanding of California grasslands has grown in the 18 years since the publication of this earlier synthesis. We conclude by speculating on some future areas of study. With the advent of molecular techniques and the proliferation of approaches to delineating genetic relationships, there has been a dramatic increase in our information on systematic relationships among species and lineages. Hypothesized relationships among the grasses found in California have likewise changed, and new proposed phylogenies are presented by Peterson and Soreng (Chapter 2). Preliminary releases of the taxonomic treatment of these species in the Flora of North America reveal dramatic changes at both many generic and species levels (FNA 2007). We also now have a greater understanding of how the non-native and native grass floras’ compare and are related to one another. Other technological advances in the past 18 years have changed our ability to study broad aspects of grassland, such as their distribution and ecology. For example, recent advances in remote sensing, as well as more detailed mapping of vegetation units in selected sites (primarily grasslands purchased for conservation) have supported the original estimates of the area of grassland plant communities in California (Huenneke 1989). Huenneke’s observations of the mosaic nature of grasslands as patches in the landscape have been supported by recent mapping at more detailed levels (Keeler-Wolf, in press).

Huenneke and Mooney’s book, as well as Heady’s (1988) description of Valley Grassland, focused on Eurasian annual species as dominants of vast areas of grassland. While this is clearly true, over the past 16 years there has been a dramatic rise the study of California native grasslands and an effort to find and restore sites where native species are still abundant. Yet we still have relatively little information on the overall distribution and abundance of native-species-dominated grasslands across the state. Although it was once stated that native California grasslands were restricted to uncommon, chemically unique soils (Murphy and Ehrlich 1989), we now understand that native grass stands occur as well on a wide variety of soils and climate zones (Reever Morghan et al., Chapter 7; Eviner and Firestone, Chapter 8; Jackson et al., Chapter 9; Harpole et al., Chapter 10; Harrison and Viers, Chapter 12) and that many native forbs remain within sites otherwise dominated by non-native annual species (Keeler-Wolf et al., Chapter 3). Understanding factors limiting the distribution of native grasses has been a focus of research in the past 16 years, and we expect that interest to continue in the future. For example, competitive interactions between native perennial and nonnative annual grasses have been extensively studied (Corbin et al., Chapter 13). The importance of land use history as a limitation on the distribution of native grasslands has become clearer (Stromberg and Griffin 1996; Steenwerth et al. 2002; Carmel and Flather 2004), although the mechanism behind the lack of recovery of native perennial grasses on formerly tilled soils is still poorly understood. Recent studies have supported the idea that native grasses are seed-limited in some circumstances (Hamilton et al. 1999; Seabloom et al. 2003a; Divittorio 2007). Recent developments in soil microbiology have led to new questions about soil conditions and the microbial and biogeochemical legacies of land use. These are described by Jackson et al. (Chapter 9), where the authors show that land use history leaves a clear imprint on the soil

319

microbial community. The importance of this microbial “legacy” to future restoration or recovery is, however, not understood. In addition to the rise in interest in native perennial grasses in California, there has been a rise in interest in studying nonnative plant invaders into California grasslands. The documentation of waves of species invasion has gone on for decades (Burcham 1957; Heady 1958; Heady et al. 1991), but a recent emerging focus has been on understanding community susceptibility to particular invaders (Dukes 2001a; Reever Morghan and Rice 2005) and the impacts of certain plant invaders on grassland communities and ecosystem functioning. Intriguing new research suggests a potential role that plant viruses might play in mediating interactions between native and non-native species (Malmstrom et al. 2006). Indeed, it is notable how many of the studies cited by D’Antonio et al. (Chapter 6) have been published in the last 6–10 years. Whether these basic ecological studies can be applied to better manage and restore grasslands remains to be explored. Several chapters note the potential for invasive species to further dominate California grassland under future climate and chemical environmental change scenarios. Experimental studies suggest that certain ecologically and potentially economically significant invaders such as yellow starthistle (Centaurea solstitialis) will increase with rising CO2 or that anthropogenic nitrogen deposition is already enhancing the abundance and persistence of invasive grasses (see Dukes and Shaw, Chapter 19). The latter situation is particularly threatening for the persistence of rare species on serpentine soils, which have otherwise served as refuges for native grassland species in the face of non-native invaders (see Harrison and Viers, Chapter 12). The recent expansion of unpalatable, noxious invaders such as barbed goatgrass (Aegilops triuncialis) onto serpentine raises the issue of whether we can predict what invaders will be here in another 16 years. Genetic changes within existing populations of nonnative species that are not currently invasive or damaging to grassland habitat may cloud our ability to predict which species will be invasive in what settings. The word genetics did not occur in the index of the previous review of California grasslands (Huenneke and Mooney 1989a). Yet studies of the genetics of selected species of California grasses and selected non-native, invasive grassland plants have contributed important new understanding of compositional change, local adaptation, and mechanisms of evolution in grassland species (Rice and Espeland, Chapter 11). As climate change continues and grasslands potentially become more fragmented, genetics will play a role in determining which species will persist, which invaders will spread, and which genotypes will be used in restoration. Another new index term since 1989 is restoration (Stromberg et al., Chapter 21), which denotes a field where both ecology and genetics have been critical. Although restoration arose initially as a practice and not as a science, ecological and genetic studies are increasingly being used as the basis for restoration decision making. California grassland restoration arose largely during the 1990s with the

320

POLICY AND MANAGEMENT

emergence of the California Native Grass Association and native grass industry. Today, it is a diverse field with projects being conducted from vernal pool grassland restoration to upland rangeland restoration. Nonetheless, although the linkages between scientific studies and the practice of restoration will help to achieve goals, particularly as they relate to species and landscape preservation, ecologists still tend to work on small-plot scales while restoration practitioners are increasingly tackling large-scale projects. It will be interesting to see how ecological experimentation and theory and ecological restoration interface in the coming decade. Since 1989, new views of the interactions between people and California grasslands (Wigand, Edwards, and Schiffman, Chapter 4; Anderson, Chapter 5) have been published. Anderson’s landmark book, Tending the Wild (Anderson 2005), as well as her chapter in this volume, explore the dominant effect of the original human occupants of California on grassland environments. From their contributions to the changing postglacial megafauna (see Edwards, Chapter 4), to their impact using fire to manage large landscapes, evidence suggests that Native Californians created an anthropogenic grassland landscape throughout many regions of Sierran foothill, central valley, coastal range, and prairie environments. Europeans then tremendously altered these landscapes through their widespread use of agriculture and introduction of livestock and Eurasian plant species (Chapters 4–6). We can expect that the effects of humans will be equally important in the future. If the models of climate change are as dramatic as predicted for the California flora, the next 20 years will see significant changes in California grasslands (Dukes and Shaw, Chapter 19; S. R. Loarie et al., in review). Further, the human population of California is expected to increase to 45 million by 2020, and much of that growth will be in parts of California where grasslands occur (Public Policy Institute of California 2006). Thus, land conversion will diminish grassland area, and impacts from human populations such as nitrogen deposition, fragmentation, and altered movement of species will contribute to ongoing compositional and functional changes.

Future Directions Research on California grassland species or ecosystems has been an exceptionally dynamic field in the last 16 years. Research areas that might be particularly dynamic in the coming decade are those with applications to managing grasslands for the multiple values we gain from them. These likely include impacts of global environmental change on grassland ecosystems, the importance of genetics and further species invasions to structural and functional change, and the development of science based, context-specific tools to restore native biodiversity or ecosystem functioning. Rising temperatures, altered precipitation patterns, and changes to the chemical systems of the earth will influence our ability to predict the condition of California grasslands over the coming decades. Some of these ongoing changes, such as atmospheric N deposition, appear to have fairly clear-cut

effects, while others, such as climate change, are more difficult to forecast because of the uncertainty of future conditions and the interactions between various components (e.g., temperature, precipitation, and CO2 concentrations) on grassland vegetation. Although Dukes and Shaw (Chapter 19) provide an up-to-date assessment of our knowledge of global change impacts and research in California grasslands, this is an area where our understanding could increase greatly in the coming decade. Existing climate models suggest either increased winter rainfall or increased spring and early summer rainfall in California over the coming decades, with potentially important consequences for grasslands (Dukes and Shaw, Chapter 19). Increased winter rainfall may enhance the growth of woody species within grassland settings, leading to a loss of grasslands. However, feedbacks and community interactions can reverse increases in individual species in response to a changing climate with additional summer rainfall (Suttle et al. 2007). There are considerable differences in predicted rainfall between climate models, and information available for new modeling efforts is growing rapidly (California Climate Change Center 2006). The design of successful restoration strategies will require more effective tools to reduce the influence of non-native species and improve the establishment and persistence of native species. Grazing (Jackson and Bartolome, Chapter 17;

Huntsinger et al., Chapter 20) and fire (Reiner, Chapter 18) are the only tools we currently have to manage grassland composition over large scales. Yet these are also two of the most complex and challenging issues and are ones that will require intensive, and long-term, observations because of the species-, year-, and site-specific nature of their effects. Effects of various grazing and/or fire treatments also will differ between the annual grasslands (largely non-native, introduced species) and existing or restored native grasslands, and there is much to learn about how to use these tools to manage composition within particular bounds or drive it toward specific goals. Use of either of these tools is controversial in many settings, for social and cultural as well as ecological reasons. A more clear understanding of how to interface grazing or fire with other management techniques and cultural constraints will become even more critical as human population growth continues its upward trajectory in California. While the future of California grasslands is largely unknown, it is surely linked to an unprecedented human occupancy of the state. Management, conservation, and restoration will result both from increased scientific knowledge, such as has been reviewed in this book, and from how the people of California regard and invest in grasslands as a natural resource.

EPILOGUE: FUTURE DIRECTIONS

321

REFERENCES

ABAG. 1995. Manual of standards for erosion and sediment control measures. Association of Bay Area Governments, Oakland, CA. Aber, J., R. P. Neilson, S. McNulty, J. M. Lenihan, D. Bachelet, and R. J. Drapek. 2001. Forest processes and global environmental change: Predicting the effects of individual and multiple stressors. Bioscience 51:735–751. Adam, D. P. 1988. Palynology of two upper Quaternary cores from Clear Lake, Lake County, California. USGS Professional Paper 1363. U.S. Geological Survey, Denver, CO. Adam, D. P., A. M. Sarna-Wojcicki, H. J. Rieck, J. P. Bradbury, W. E. Dean, and R. M. Forester. 1989. Tulelake, California: The last 3 million years. Paleoclimatology, Paleogeography, Paleoecology 72:89–103. Adam, D. P., J. D. Sims, and C. K. Throckmorton. 1981. 130,000-yr. continuous pollen record from Clear Lake, Lake County, California. Geology 9(8):373 – 377. Adam, D. P., and G. L. Vagenas. 1990. Pollen data for a 3-m.y. core record from Tulelake, Siskiyou County, California. USGS Open-File Report 90-65. U.S. Geological Survey, Denver, CO. Adams, J. S., and T. O. McShane. 1996. The myth of wild Africa. University of California Press, Berkeley. Adams, T. E., P. B. Sands, W. H. Weitkamp, N. K. McDougald, and J. Bartolome. 1987. Enemies of white oak regeneration in California. Pages 459–462 in T. R. Plumb and N. H. Pillsbury, editors. Proceedings of the symposium on multiple-use management of California’s hardwood resources, November 12–14, 1986, San Luis Obispo, California. General Technical Report PSW-GTR-100. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Adams, T. E., P. B. Sands, W. H. Weitkamp, and M. E. Stanley. 1997. Oak seedling establishment by artificial regeneration on California rangelands. Pages 213–222 in N. H. Pillsbury, J. Verner, and W. D. Tietje, editors. Proceedings of a symposium on oak woodlands: Ecology, management, and urban interface issues, March 19–22, 1996, San Luis Obispo, California. General Technical Report PSW-GTR-160. USDA Forest Service, Pacific Southwest Research Station, Albany, CA.

Adams, T. E., and W. H. Weitkamp. 1992. Gophers love oaks—to death. California Agriculture 46:27–29. Adams, W. T., and R. W. Allard. 1982. Mating system variation in Festuca microstachys. Evolution 36:591–595. Adovasio, J. M. 1974. Prehistoric North American basketry. Pages 98–148 in D. R. Tuohy and D. L. Rendall, editors. Collected Papers on Aboriginal Basketry. Anthropological Papers No. 16. Nevada State Museum, Carson City. Ahmed, E. O. 1983. Fire ecology of Stipa pulchra on California annual grassland. Ph.D. dissertation, University of California, Davis. Aiken, S. G., M. J. Dallwitz, C. L. McJannet, and L. L. Consaul. 2005. Festuca of North America: descriptions, illustrations, identification, and information retrieval. http://delta-intkey.com/ festuca/www/arun.htm (accessed March 26, 2007). Air Resources Board. 1995. Atmospheric deposition to agricultural soil. Research Note No. 95-17. Air Resources Board, Sacramento, CA. Akersten, W., T. M. Foppe, and G. T. Jefferson. 1988. New sources of dietary data for extinct herbivores. Quaternary Research 30:92–97. Alexander, J. M., and C. M. D’Antonio. 2003a. Control methods for the removal of French and Scotch broom tested in coastal California. Ecological Restoration 21:191–198. ———. 2003b. Seed bank dynamics of French broom in coastal California grasslands: Effects of stand age and prescribed burning on control and restoration. Restoration Ecology 11:185–197. Allan, C. J., and P. J. Holst. 1996. The ecological role of the goat in maintaining pasture and range. Pages 427– 435 in Proceedings of the VI international conference on goats, May 6–11, China, Vol. 1. International Academic Publishers, Beijing. Allard, R. W. 1996. Genetic basis of the evolution of adaptedness in plants. Euphytica 92:1–11. Allard, R. W., and J. Adams. 1969. Population studies in predominantly self-pollinating species. XII. Intergenotypic competition and population structure in barley and wheat. American Naturalist 103:621–645.

323

Allard, R. W., G. R. Babbel, M. T. Clegg, and A. L. Kahler. 1972. Evidence for coadaptation in Avena barbata. Proceedings of the National Academy of Sciences USA 69:3043–3048. Allard, R. W., and S. K. Jain. 1962. Population studies in predominantly self-pollinated species. II. Analysis of quantitative genetic changes in a bulk-hybrid population of barley. Evolution 16:90–101. Allard, R. W., S. K. Jain, and P. L. Workman. 1968. The genetics of inbreeding populations. Advances in Genetics 14:55–131. Allard, R. W., and L. W. Kannenberg. 1968. Population studies in predominantly self-pollinated species. XI. Genetic divergence among members of the Festuca microstachys complex. Evolution 22:517–528. Allen, A. G., S. C. Jarvis, and D. M. Headon. 1996. Nitrous oxide emissions from soils due to inputs of nitrogen from excreta return by livestock on grazed grassland in the UK. Soil Biology & Biochemistry 28:597–607. Allen, E. B. 1995. Restoration ecology: limits and possibilities in arid and semiarid lands. Pages 7–15 in B. A. Roundy, editor. Proceedings, wildland shrub and arid land restoration symposium, Las Vegas, NV, October 19–21, 1993. General Technical Report INT-GTR-315. USDA Forest Service, Intermountain Research Station, Ogden, UT. Allen, E. B., M. F. Allen, D. J. Helm, J. M. Trappe, R. Molina, and E. Rincon. 1995. Patterns and regulation of mycorrhizal plant and fungal diversity. Pages 47–62 in H. P. Collins, G. P. Robertson, and M. J. Klug, editors. The significance and regulation of soil biodiversity. Kluwer Academic Press, Dordrecht, The Netherlands. Allen, E. B., and J. A. MacMahon. 1985. Impact of disturbance of cold desert fungi: Comparative microscale dispersion patterns. Pedobiologica 28:215–224. Allen, E. B., P. E. Padgett, A. Bytnerowicz, and R. A. Minnich. 1998. Nitrogen deposition effects on coastal sage vegetation of southern California. Pages 131–140 in Proceedings of the international symposium on air pollution and climate change effects on forest ecosytems, February 5–9, 1996, Riverside, CA. General Technical Report PSW-GTR-166, USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Allen, M. F., D. A. Jasper, and J. C. Zak. 2002. Micro-organisms. Pages 257–278 in M. R. Perrow and A. J. Davy, editors. Handbook of ecological restoration. Cambridge University Press, Cambridge, UK, and New York. Allen-Diaz, B., and J. W. Bartolome. 1998. Sagebrush-grass vegetation dynamics: Comparing classical and state-transition models. Ecological Applications 8:795–804. Allen-Diaz, B. H., J. W. Bartolome, and M. P. McClaran. 1997. California oak savanna. Pages 322–339 in J. M. Baskin, editor. Savanna, barren and rock outcrop plant communities of North America. Cambridge University Press, Cambridge, UK. Allen-Diaz, B., and R. D. Jackson. 2000. Grazing effects on spring ecosystem vegetation of California’s hardwood rangelands. Journal of Range Management 53:215–220. Allen-Diaz, B., R. D. Jackson, J. W. Bartolome, K. W. Tate, and L. G. Oates. 2004. Long-term grazing study in spring-fed wetlands reveals management tradeoffs. California Agriculture 58(3):144–148. Allen-Diaz, B., R. D. Jackson, and J. S. Fehmi. 1998. Detecting channel morphology change in California’s hardwood rangeland spring ecosystems. Journal of Range Management 51: 514–518.

324

REFERENCES

Alpert, P., and J. L. Maron. 2000. Carbon addition as a countermeasure against biological invasion by plants. Biological Invasions 2:33–40. Alroy, J. 2001. A multispecies overkill simulation of the end-Pleistocene megafaunal mass extinction. Science 292:1893–1896. AMEC Earth and Environmental Inc. 2005. Los Peñasquitos watershed management plan. City of San Diego, San Diego, CA. Amme, D. 2003. The native grass Sierra seed zone program: A progress report. Grasslands 13:9–12. Amor, R. L. 1974. Ecology and control of blackberry (Rubus fruticosus L. Agg). 2. Reproduction. Weed Research 14:231–238. Amundson, R. 2001. The carbon budget in soils. Annual Review of Earth and Planetary Sciences 29:535–562. Anderson, B. 1989a. Beginnings of agriculture in the Salinas Valley 1770–1849. Coastal Grower (Winter 1989):32. ———. 1989b. Salinas Valley agriculture 1850–1885. Coastal Grower (Spring 1989):38. ———. 1989c. Salinas Valley agriculture 1885–1919. Coastal Grower (Summer 1989):29. Anderson, E. 1984. Who’s who in the Pleistocene: A mammalian bestiary. Pages 40–89 in P. S. Martin and R.G. Klein, editors. Quaternary extinctions. University of Arizona Press, Tucson. Anderson, H. D. 1982. Aids to determining fuel models for estimating fire behavior. General Technical Report INT-GTR-122. USDA Forest Service, Intermountain Research Station, Ogden, UT. Anderson, J. H., and J. L. Anderson. 1996. Establishing permanent grassland habitat with California native perennial grasses. Valley Habitats 14. Ducks Unlimited, Western Regional Office, Rancho Cordova, CA. Anderson, M. K. 1993a. Indian fire-based management in the sequoia–mixed conifer forests of the central and southern Sierra Nevada. Cooperative Agreement Order No. 8027-002. Yosemite Research Center, U.S. Department of Interior, National Park Service. Western Region, Yosemite National Park. ———. 1993b. The experimental approach to assessment of the potential ecological effects of horticultural practices by indigenous peoples on California wildlands. Ph.D. dissertation, University of California, Berkeley. ———. 1996. The ethnobotany of deergrass, Muhlenbergia rigens (Poaceae): Its uses and fire management by California Indian tribes. Economic Botany 50:409–422. ———. 2005. Tending the wild: Native American knowledge and the management of California’s natural resources. University of California Press, Berkeley. ———. 2006. The use of fire by Native Americans in California. Pages 417–430 in N. G. Sugihara, J. W. van Wagtendonk, J. Fites-Kaufman, K. E. Shaffer, and A. E. Thode, editors. Fire in California’s ecosystems. University of California Press, Berkeley. Anderson, S. J., and C. P. Stone. 1993. Snaring to control feral pigs Sus scrofa in a remote Hawaii rain forest. Biological Conservation 63:195–201. Andrews, P. L. 1986. BEHAVE: fire behavior prediction and fuel modeling system—BURN subsystem, part 1. General Technical Report INT-GTR-194. USDA Forest Service, Intermountain Research Station, Ogden, UT. Anger, M., C. Hoffmann, and K. Uhbauch W. 2003. Nitrous oxide emissions from artificial urine patches applied to different N-fertilized swards and estimated annual N2O emissions for differently fertilized pastures in an upland location in Germany. Soil Use & Management 19:104–111.

Aplet, G. H., S. J. Anderson, and C. P. Stone. 1991. Association between feral pig disturbance and the composition of some alien plant assemblages in Hawaii Volcanoes National Park. Vegetatio 95:55–62. Arguello, L. A. 1994. Effects of prescribed burning on two perennial bunchgrasses in the Bald Hill of Redwood National Park. Master’s thesis, Humboldt State University, Arcata, CA. Arizona-Sonora Desert Museum. n.d. Onionweed (Asphodelus fistulosus). http://www.desertmuseum.org/invaders/invaders_ onionweed.htm (accessed March 26, 2007). Armstrong, J., and L. F. Huenneke. 1992. Spatial and temporal variation in species composition in California grasslands: The interaction of drought and substratum. Pages 207–212 in A. J. M. Baker, J. Proctor, and R. D. Reeves, editors. The vegetation of ultramafic (serpentine) soils. Intercept, Ltd., Andover, UK. Arno, S. F., and G. E. Gruell. 1986. Douglas-fir encroachment into mountain grasslands in southwestern Montana. Journal of Range Management 39:272–276. Aschmann, H. 1977. Aboriginal use of fire. Pages 132–141 in H. A. Mooney and C. E. Conrad, editors. Symposium on environmental consequences of fire and fuel management in Mediterranean ecosystems, August 1–5, 1977, Palo Alto, CA. General Technical Report WO-3. USDA Forest Service, Washington, DC. Aslam, T., M. A. Choudhary, and S. Saggar. 1999. Tillage impacts on soil microbial biomass C, N, and P, earthworms and agronomy after two years of cropping following permanent pasture in New Zealand. Soil and Tillage Research 51:103–111. Augustine, D. J., and S. J. McNaughton. 1998. Ungulate effects on the functional species composition of plant communities: Herbivore selectivity and plant tolerance. Journal of Wildlife Management 62:1165–1183. Austin, A. T., and L. Vivanco. 2006. Plant litter decomposition in a semi-arid ecosystem controlled by photodegradation. Nature 442:555 –558. Averett, J. M., R. A. Klips, L. E. Nave, S. D. Frey, and P. S. Curtis. 2004. Effects of soil carbon amendment on nitrogen availability and plant growth in an experimental tallgrass prairie restoration. Restoration Ecology 12:568–574. Axelrod, D. I. 1944. The Oakdale flora. Carnegie Institution of Washington Publication 553:147–166. ———. 1985. Rise of the grassland biome, central North America. Botanical Review 51:163–201. Azevedo, J., and D. L. Morgan. 1974. Fog precipitation in coastal California forests. Ecology 55:1135–1141. Bachelet, D., R. P. Neilson, J. M. Lenihan, and R. J. Drapek. 2001. Climate change effects on vegetation distribution and carbon budget in the United States. Ecosystems 4:164–185. Bailey, R. G. 1995. Description of the ecoregions of the United States. 2nd edition. Miscellaneous Publication 1391. USDA Forest Service, Washington, DC. Bakeman, M. E., and T. Nimlos. 1985. The genesis of Mollisols under Douglas-fir. Soil Science 140:449–452. Baker, A. J. M., J. Proctor, and R. D. Reeves. 1992. The vegetation of ultramafic (serpentine) soils. Intercept Ltd., Andover, UK. Baker, H. G. 1989. Sources of the naturalized grasses and herbs in California grasslands. Pages 29–38 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Dordrecht, The Netherlands. Baker, H. G., and G. L. Stebbins. 1965. The genetics of colonizing species. Academic Press, New York.

Baker, M. A. 1981. The ethnobotany of the Yurok, Tolowa and Karok Indians of northwest California. M.A. thesis, Humboldt State University, Arcata, CA. Baker, W. L. 1994. Restoration of landscape structure altered by fire suppression. Conservation Biology 8:763–769. Balciunas, J. D., and B. Villegas. 1999. Two new seed head flies attack yellow starthistle. California Agriculture 53:8–11. Balmford, A., A. Bruner, P. Cooper, R. Costanza, S. Farber, R. E. Green, M. Jenkins, P. Jeffriss, V. Jessamy, J. Madden, K. Munro, N. Myers, S. Naeem, J. Paavola, M. Rayment, S. Rosendo, J. Roughgarden, K. Trumper, and R. K. Turner. 2002. Economic reasons for conserving wild nature. Science 297:950–953. Balser, T. C., and M. K. Firestone. 2005. Linking microbial community composition and soil processes in a California annual grassland and mixed-conifer forest. Biogeochemistry 73: 395–415. Barbour, M., A. Solomeshch, C. Witham, R. Holland, R. Macdonald, S. Cilliers, J. A. Molina, J. Buck, and J. Hillman. 2003. Vernal pool vegetation of California: variation within pools. Madroño 50: 129–146. Barbour, M. G., J. H. Burk, W. D. Pitts, F. S. Gilliam, and M. W. Schwartz. 1999. Terrestrial plant ecology, 3rd edition. Addison Wesley Longman, New York. Barbour, M. G., and N. L. Christensen. 1993. Vegetation. Pages 97–131 in Flora of North America north of Mexico, Vol. 1. Oxford University Press, New York. Barbour, M. G., and J. Major, editors. 1977. Terrestrial vegetation of California. John Wiley and Sons, New York. Barkworth, M. E., K. M. Capels, S. Long, and M. B. Piep. 2006. Grass manual on the Web. http://herbarium.usu.edu/webmanual/ default.htm. Barr, C. M. 2004. Hybridization and regional sex ratios in Nemophila menziesii. Journal of Evolutionary Biology 17: 786–794. Barrett, R. H. 1978. The feral hog in Dye Creek Ranch, California. Hilgardia 46:281–355. ———. 1993. Feral swine: the California experience. Pages 107–116 in C. W. Hanselka and J. F. Cadenhead, editors. Feral swine: a compendium for resource managers. Proceedings of a conference. Texas Agricultural Extension Service, San Angelo. Barrett, R. H., B. L. Goatcher, P. J. Gogan, and E. L. Fitzhugh. 1988. Removing feral pigs from Annadel State Park. Transactions of the Western Section of the Wildlife Society 24:47–52. Barrett, S. A. 1910. The material culture of the Klamath Lake and Modoc Indians of northeastern California and southern Oregon. University of California Publications in American Archeology and Ethnology 5(4):239– 292. ———. 1952. Material aspects of Pomo culture. Bulletin of the Public Museum of the City of Milwaukee 20:1–260. Barrett, S. A., and E. W. Gifford. 1933. Miwok material culture. Bulletin of the Public Museum of the City of Milwaukee 2:117–376. Barrows, D. P. 1967. Ethnobotany of the Coahuilla Indians. Malki Museum Press, Morongo Indian Reservation, Banning, CA. Barry, S. J. 1998. Managing the Sacramento Valley vernal pool landscape to sustain native flora. Pages 236-240 in E. T. Witham, C. W. Bauder, D. Belk, and W. R. Ferren Jr., editors. Ecology, conservation, and management of vernal pool ecosystems: Proceedings of a 1996 conference. California Native Plant Society, Sacramento, CA.

REFERENCES

325

———. 2005. Public agency land use in the San Francisco Bay Area. Report of Bay Area Natural Resources and Livestock Advisor, University of California Cooperative Extension, Davis, CA. Barry, S. J., and L. Huntsinger. 2002. Will California’s landscapes keep working? Rangelands 24:6–10. Barry, S. J., R. E. Larsen, G. A. Nader, M. Doran, K. Guenther, and G. Hayes. 2006. Understanding livestock grazing impacts: strategies for the California annual grassland and oak woodland vegetation series. Publication 21626, University of California Division of Agriculture and Natural Resources, Oakland. Barry, W. J. 1972. The Central Valley Prairie. Vol. 1. California Prairie Ecosystem. California State Parks and Recreation Department, Sacramento. Bartholomew, B. 1970. Bare zone between California shrub and grassland communities: Role of animals. Science 170:1210–1212. Bartlett, E. T., R. G. Taylor, J. R. McKean, and J. G. Hof. 1989. Motivation of Colorado ranchers with federal grazing allotments. Journal of Range Management 42:454–457. Bartolome, J. W. 1976. Early rains alter range forage. California Agriculture 30:14–15. ———. 1979. Germination and seedling establishment in California annual grassland. Journal of Ecology 67:273–281. ———. 1987. California annual grassland and oak savannah. Rangelands 9:122–125. ———. 1989. Local temporal and spatial structure. Pages 73–80 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academics, Dordrecht, The Netherlands. ———. 1993. Application of herbivore optimization theory to rangelands of the western United States. Ecological Applications 3:27–29. Bartolome, J. W., B. H. Allen-Diaz, and W. D. Tietje. 1994. The effect of Quercus douglasii removal on understory yield and composition. Journal of Range Management 47:151–154. Bartolome, J. W., W. J. Barry, T. Griggs, and P. Hopkinson. In press. Valley grassland. In M. G. Barbour, T. Keeler-Wolf, and A. Schoenherr, editors. Terrestrial vegetation of California. 3rd edition. University of California Press, Berkeley. Bartolome, J. W., J. S. Fehmi, R. D. Jackson, and B. Allen-Diaz. 2004. Response of a native perennial grass stand to disturbance in California’s Coast Range grassland. Restoration Ecology 12:279–289. Bartolome, J. W., W. E. Frost, N. K. McDougald, and J. M. Connor. 2002. California guidelines for Residual Dry Matter (RDM) management on coastal and foothill annual rangelands. Rangeland Management Series 8092. University of California Division of Agriculture and Natural Resources, Oakland. Bartolome, J. W., and B. Gemmill. 1981. The ecological status of Stipa pulchra (Poaceae) in California. Madroño 28: 172–184. Bartolome, J. W., R. D. Jackson, A. D. K. Betts, J. M. Connor, G. A. Nader, and K. W. Tate. In press. Residual dry matter effects on net primary production and plant functional groups in California annual grassland. Grass and Forage Science. Bartolome, J. W., S. E. Klukkert, and W. J. Barry. 1986. Opal phytoliths as evidence for displacement of native California grassland. Madroño 33:217–222. Bartolome, J. W., and M. P. McClaran. 1992. Composition and production of California oak savanna seasonally grazed by sheep. Journal of Range Management 45:103–107.

326

REFERENCES

Bartolome, J. W., M. C. Stroud, and H. F. Heady. 1980. Influence of natural mulch on forage production on differing California annual range sites. Journal of Range Management 33:4–8. Baskin, C. C., and J. M. Baskin. 1998. Seeds: Ecology, biogeography, and evolution of dormancy and germination. Academic Press, San Diego, CA. Bates, C. D. 1982. Coiled basketry of the Sierra Miwok: A study of regional variation. San Diego Museum Papers San Diego Museum of Man, San Diego, CA. Batten, K. M. 2004. Plant invasion and the soil microbial community: Interactions and implications for native plant performance. Ph.D. dissertation, University of California, Davis. Batten, K. M., K. M. Scow, K. F. Davies, and S. P. Harrison. 2006a. Two invasive plants alter soil microbial community composition in serpentine grasslands. Biological Invasions 8:217–230. Batten, K. M., J. Six, K. M. Scow, and M. C. Rillig. 2006b. Plant invasion of native grassland on serpentine soils has no major effects upon selected physical and biological properties. Soil Biology and Biogeochemistry 37:2277–2282. Batzli, G. O., and F. A. Pitelka. 1970. Influence of meadow mouse populations on California grassland. Ecology 51: 1027–1039. Bauer, A., and A. L. Black. 1994. Quantification of the effect of soil organic matter content on soil productivity. Soil Science Society America Journal 58:185–193. Bazilchuk, N. 2006. Skyscraper habitats. Conservation in Practice 7:38–39. Bazzaz, F. A. 1990. The response of natural ecosystems to the rising global CO2 levels. Annual Review of Ecology and Systematics 21:167–196. Bean, L. J., and K. S. Saubel. 1972. Temalpakh: Cahuilla Indian knowledge and usage of plants. Malki Museum Press, Morongo Indian Reservation, Banning, CA. Bean, M. J., J. P. Jenny, and B. van Eerden. 2001. Safe harbor agreements: Carving out a new role for NGOs. Conservation 2:8–16. Beard, Y. S. 1979. The Wappo: A report. Malki Museum Press, Morongo Indian Reservation, Banning, CA. Beattie, A. J. 1989. The effects of ants on grasslands. Pages 105–116 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Dordrecht, The Netherlands. Beetle, A. A. 1947. Distribution of the native grasses of California. Hilgardia 17:309–357. Beever, E. A., and P. F. Brussard. 2000. Examining the ecological consequences of feral horses using exclosures. Western North American Naturalist 60:236–254. Beever, E. A., R. J. Tausch, and P. F. Brussard. 2003. Characterizing grazing disturbance in semiarid ecosystems across broad scales, using diverse indices. Ecological Applications 13: 119–136. Begg, E. L. 1968. Soil survey of Glenn County, California. U.S. Government Printing Office, Washington, DC. Begon, M., J. L. Harper, and C. R. Townsend. 1996. Ecology: Individuals, populations, and communities. 3rd edition. Blackwell Science, Oxford, UK. Belcher, J. W., P. A. Keddy, and L. Twolan-Strutt. 1995. Root and shoot competition intensity along a soil depth gradient. Journal of Ecology 83:673–682.

Belcher, J. W., and S. D. Wilson. 1989. Leafy spurge and species composition of a mixed-grass prairie. Journal of Range Management 42:172–175. Belkovsky, G. E., and A. Joern. 1995. The dominance of different regulating factors for rangeland grasshoppers. Pages 359–386 in N. Cappuccino and P. W. Price, editors. Population dynamics: New approaches and synthesis. Academic Press, New York. Belkovsky, G. E., and J. B. Slade. 1993. The role of vertebrate and invertebrate predators in a grasshopper community. Oikos 68: 193–201. ———. 1995. Dynamics of two Montana grasshopper populations: Relationships among weather, food abundance and intraspecific competition. Oecologia 101:383–396. ———. 2000. Insect herbivory accelerates nutrient cycling and increases plant production. Proceedings of the National Academy of Sciences USA 97:14412–14417. Bell, S. S., E. D. McCoy, and H. R. Mushinsky. 1991. Habitat structure: The physical arrangement of objects in space. Chapman and Hall, London. Belsky, A. J. 1986. Does herbivory benefit plants? A review of the evidence. American Naturalist 127:870–892. Belsky, A. J., and J. L. Gelbard. 2000. Livestock grazing and weed invasions in the arid West. Oregon Natural Desert Association, Bend, OR. Belsky, A. J., A. Matzke, and S. Uselman. 1999. Survey of livestock influences on stream and riparian ecosystems in the western United States. Journal of Soil and Water Conservation 54:419–431. Bendell, J. F. 1974. Effects of fire on birds and mammals. Pages 73–137 in T. T. Kozlowski and C. E. Ahlgren, editors. Fire and ecosystems. Academic Press, New York. Benefield, C. B., J. M. DiTomaso, G. B. Kyser, S. B. Orloff, K. R. Churches, D. B. Marcum, and G. A. Nader. 1999. Success of mowing to control yellow starthistle depends on timing and plant’s branching form. California Agriculture 53:17–21. Benson, L. V., D. R. Currey, R. I. Dorn, K. R. Lajoie, C. G. Oviatt, S. W. Robinson, G. I. Smith, and S. Stine. 1990. Chronology of expansion and contraction of four Great Basin lake systems during the past 35,000 years. Paleoclimatology, Paleogeography, Paleoecology 78:241–286. Bentham, G. 1881. Notes on Gramineae. Botanical Journal of the Linnean Society 19:14–134. Bentley, J. R., and R. L. Fenner. 1958. Soil temperatures during burning related postfire seedbeds on woodland range. Journal of Forestry 56:737–740. Bentley, J. R., and L. R. Green. 1954. Stimulation of native clovers through application of sulfur on California foothill range. Journal of Range Management 7:25–30. Bentley, J. R., L. R. Green, et al. 1958. Herbage production and grazing capacity on annual-plant range pastures fertilized with sulfur. Journal of Range Management 11:133–140. Bentley, J. R., and M. W. Talbot. 1948. Annual-plant vegetation of the California foothills as related to range management. Ecology 29:72–79. Berg, A. W. 1990a. Formation of mima mounds: a seismic hypothesis. Geology 18:281–284. ———. 1990b. Reply on “Formation of mima mounds: a seismic hypothesis.” Geology 18:1260–1261. Berg, N. 1986. Snow chemistry in the central Sierra-Nevada, California. Water, Air, & Soil Pollution 30:1015–1021.

Bernhardt, E. A., and T. J. Swiecki. 1997. Effects of cultural inputs on survival and growth of direct seeded and naturally occurring valley oak seedlings on hardwood rangeland. Pages 310–311 in N. H. Pillsbury, J. Verner, and W. D. Tietje, editors. Proceedings of a symposium on oak woodlands: Ecology, management, and urban interface issues 1997. General Technical Report PSW-GTR-160. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Betancourt, J. L., T. R. Van Devender, and P. S. Martin. 1990. Fossil packrat middens: The last 40,000 years of biotic change. University of Arizona Press, Tucson. Betterini, I., F. P. Vaccari, and F. Miglietta. 1998. Elevated CO2 concentrations and stomatal density: Observations from 17 plant species growing in a CO2 spring in central Italy. Global Change Biology 4:17–22. Betts, A. D. K. 2003. Ecology and control of goatgrass (Aegilops triuncialis) and medusahead (Taeniatherum caput-medusae) in California annual grasslands. Ph.D. dissertation, University of California, Berkeley. Bicknell, S. H., A. T. Austin, D. J. Bigg, and R. P. Godar. 1992. Late prehistoric vegetation patterns at six sites in coastal California. Bulletin of the Ecological Society of America 73 Supplement:112 (abstract). Biswell, H. H. 1953. The brush control problem in California. Journal of Range Management 7(2):57–62. ———. 1956. Ecology of California grasslands. Journal of Range Management 9:19–24. ———. 1967. The use of fire in wildland management in California. Pages 71–86 in S. V. Ciriacy-Wantrup and J. J. Parsons, eds. Natural Resources, Quality and Quantity. University of California Press, Berkeley. ———. 1989. Prescribed burning in California wildlands vegetation management. University of California Press, Berkeley. Biswell, H. H., and C. A. Graham. 1956. Plant counts and seed production on California annual-type ranges. Journal of Range Management 9:116– 118. Biswell, H. H., R. D. Taber, D. W. Hendrick, and A. M. Schultz. 1952. Management of chamise brushlands in the North Coast Region of California. California Fish and Game 38: 453–484. Black, T. A., and D. R. Montgomery. 1991. Sediment transport by burrowing mammals, Marin County, California. Earth Surface Processes and Landforms 16:163–172. Blackburn, T. C., and K. Anderson, editors. 1993. Before the wilderness: Environmental management by native Californians. Ballena Press, Menlo Park, CA. Blanchard, C. L., and K. A. Tonneson. 1993. Precipitation chemistry measurements from the California Acid Deposition Monitoring Program, 1985–1990. Atmospheric Environment 27A:1755–1763. Blattner, F. R. 2006. Multiple intercontinental dispersals shaped the distribution area of Hordeum (Poaceae). New Phytologist 169:603–614. Bliss, S. A., and P. H. Zedler. 1998. The germination process in vernal pools: Sensitivity to environmental conditions and effects on community structure. Oecologia 113:67–73. BLM. 2005. Working together for healthy lands and thriving communities. 2004 annual report. Bureau of Land Management, Washington, DC. http://www.blm.gov/nhp/info/stratplan/ BLMarFY04.pdf.

REFERENCES

327

———. 2006. Public Lands Statistics 2005. Bureau of Land Management, Washington, D.C. http://www.blm.gov/natacq/ pls05/. Bloom, A. J., F. S. Chapin, III, and H. A. Mooney. 1985. Resource limitation in plants: An economic analogy. Annual Review of Ecology and Systematics 16:363–392. Bloom, M. G., editor. 1958. The Hazelton letters: a contribution to western Americana. California History Foundation, Stockton. Blumenthal, D. M., N. R. Jordan, and M. P. Russelle. 2003. Soil carbon addition controls weeds and facilitates prairie restoration. Ecological Applications 13:605–615. Blumler, M. A. 1992. Some myths about California grasslands and grazers. Fremontia 20:22–27. ———. 1995. Invasion and transformation of California’s valley grassland, a Mediterranean analogue ecosystem. Pages 308–332 in R. A. Butlin and N. Roberts, editors. Ecological relations in historical times: Human impact and adaptation. Blackwell, Cambridge, MA. Bocek, B. R. 1984. Ethnobotany of Costanoan Indians, California, based on collections by John P. Harrington. Economic Botany 38:240–255. Bock, C. E., and J. H. Bock. 1992. Response of birds to wildfire in native versus exotic Arizona grassland. The Southwestern Naturalist 37:73–81. Bock, C. E., J. H. Bock, and H. M. Smith. 1993. Proposal for a system of federal livestock exclosures on public rangelands in the western United-States. Conservation Biology 7:731–733. Bolander, H. N. 1866. The grasses of the state. Transactions California State Agricultural Society 1864–65:131–145. Bolsinger, C. L. 1988. The hardwoods of California’s timberlands, woodlands, and savannas. Resource Bulletin PNWRB-148. USDA Forest Service, Pacific Northwest Research Station, Portland, OR. Bolton, H. E. 1927. Fray Juan Crespi, missionary explorer on the Pacific coast, 1769–1774. University of California Press, Berkeley. ———. 1930. Anza’s California Expeditions Vol. III. The San Francisco Colony diaries of Anza, Font, and Eixarch and narratives by Palou and Moraga. University of California Press, Berkeley. ———. 1931. In the south San Joaquin ahead of Garces. Quarterly of the California Historical Society 10:211–219. ———. 1966. Anza’s California expeditions. Vol. IV. Font’s complete diary of the second Anza expedition. Russell and Russell, New York. Bond, W. J. 2005. Large parts of the world are brown or black: A different view on the “green world” hypothesis. Journal of Vegetation Science 16:261–266. Bond, W. J., F. I. Woodward, and G. F. Midgley. 2005. The global distribution of ecosystems in a world without fire. New Phytologist 165:525–537. Bonham, C. D. 1989. Measurements for terrestrial vegetation. John Wiley and Sons, New York. Boose, D. L., B. R. Harmeling, and R. T. Turcotte. 2005. Genetic variation in eastern Washington populations of Navarretia leucocephala (Polemoniaceae), a vernal pool endemic. Madroño 52:99–106. Borchert, M. I., F. W. Davis, J. Michaelsen, and L. D. Oyler. 1989. Interactions of factors affecting seedling recruitment of blue oak (Quercus douglasii) in California. Ecology 70: 389–404.

328

REFERENCES

Borchert, M. I., and S. K. Jain. 1978. The effect of rodent seed predation on four species of California annual grasses. Oecologia 33:101–113. Borman, M. M., D. E. Johnson, and W. C. Krueger. 1992. Soil moisture extraction by vegetation in a Mediterranean/maritime climatic regime. Agronomy Journal 84:897–904. Borman, M. M., W. C. Krueger, and D. E. Johnson. 1991. Effects of established perennial grasses on yields of associated annual weeds. Journal of Range Management 44:318–322. Bossard, C. C., J. M. Randall, and M. C. Hoshovsky. 2000. Invasive plants of California’s wildlands. University of California Press, Berkeley. Bossema, I. 1979. Jays and oaks: Eco-ethological study of a symbiosis. Behaviour 70:1–117. Boulton, A. M., and K. D. Amberman. 2006. How ant nests increase soil biota richness and abundance: a field experiment. Biodiversity and Conservation 15:69–82. Boulton, A. M., K. F. Davies, and P. S. Ward. 2005. Species richness, abundance, and composition of ground-dwelling ants in northern California grasslands: Role of plants, soil, and grazing. Environmental Entomology 34:96–104. Boulton, A. M., B. A. Jaffee, and K. M. Scow. 2003. Effects of a common harvester ant (Messor andrei) on richness and abundance of soil biota. Applied Soil Ecology 23:257–265. Bovey, R., D. L. Tourneau, and L. C. Erickson. 1961. The chemical composition of medusahead and downy brome. Weeds 9:307–311. Bradley, B. A., and J. F. Mustard. 2005. Identifying land cover variability distinct from land cover change: Cheatgrass in the Great Basin. Remote Sensing of the Environment 94:204–213. Bradshaw, A. D. 1965. Evolutionary significance of phenotypic plasticity in plants. Advances in Genetics 13:115–155. Bradshaw, H. D., Jr. 2005. Mutations in CAX1 produce phenotypes characteristic of plants tolerant to serpentine soils. New Phytologist 167:81–88. Brady, K. U., A. R. Kruckeberg, and H. D. Bradshaw. 2005. Evolutionary ecology of plant adaptation to serpentine soils. Annual Review of Ecology, Evolution, and Systematics 36:243–266. Bratton, S. P. 1975. The effect of the European wild boar, Sus scrofa, on gray beech forest in the Great Smoky Mountains. Ecology 56:1356–1366. Brenchley, W. E., and K. Warington. 1958. The Park Grass plots at Rothamsted 1856–1949. Rothamsted Experimental Station, Harpenden. Brenzel, K. N. 2001. Sunset western garden book. Sunset Publishing, Menlo Park, CA. Breshears, D. D., J. W. Nyhan, C. E. Heil, and B. P. Wilcox. 1998. Effects of woody plants on microclimate in a semi-arid woodland: soil temperature and evaporation in canopy and intercanopy patches. International Journal of Plant Sciences 159:1010–1017. Brewer, W. H. 1966. Up and down California in 1860–1864. The journal of William H. Brewer, professor of agriculture in the Sheffield Scientific School from 1864 to 1903. Edited by F. P. Farquhar. 3rd edition. University of California Press, Berkeley. Breysse, P. A. 1984. Health hazards of smoke. Journal of Forestry 82:89. Briden, L. E., A. Archon, and D. L. Chesemore. 1992. Ecology of the San Joaquin kit fox (Vulpes macrotis mutica) in western Merced County, California. Pages 81–87 in D. F. Williams, S. Byrne, and T. A. Rado, editors. Endangered and sensitive

species of the San Joaquin Valley: Their biology, management and conservation. California Energy Commission, Sacramento. Briske, D. D. 1991. Developmental morphology and physiology of grasses. Pages 85–108 in R. K. Heitschmidt and J. W. Stuth, editors. Grazing management: An ecological perspective. Timber Press, Portland, OR. Brooks, C. N., and A. M. Merenlender. 2001. Determining the pattern of oak woodland regeneration for a cleared watershed in northwest California: A necessary first step for restoration. Restoration Ecology 9:1–12. Brooks, M. L. 1999. Alien annual grasses and fire in the Mojave Desert. Madroño 46:13–19. ———. 2001. Fire and invasive plants in the wildlands of California. Noxious Times 3:4–5. ———. 2002. Peak fire temperatures and effects on annual plants in the Mojave Desert. Ecological Applications 12:1088–1102. ———. 2003. Effects of increased soil nitrogen on the dominance of alien annual plants in the Mojave Desert. Journal of Applied Ecology 40:344–353. Brooks, M. L., C. M. D’Antonio, D. M. Richardson, J. B. Grace, J. E. Keeley, J. M. DiTomaso, R. J. Hobbs, M. Pellant, and D. Pyke. 2004. Effects of invasive alien plants on fire regimes. Bioscience 54:677–688. Brooks, M. L., and T. Esque. 2000. Alien grasses in the Mojave and Sonoran deserts. Pages 39–44 in M. Kelly, M. Howe, and B. Neill, editors. Proceedings, California Exotic Pest Plant Council Symposium, Vol. 5. California Invasive Plant Council, Berkeley. Brooks, M. L., and J. R. Matchett. 2003. Plant community patterns in unburned and burned blackbrush shrublands in the Mojave Desert. Western North American Naturalist 63(3):283–298. Brooks, M. L., and R. A. Minnich. 2007. Fire in the southeastern desert bioregion. Chapter 16 in N. G. Sugihara, J. W. van Wagtendonk, J. Fites-Kaufman, K. Shaffer, and A. E. Thode, editors. Fire in California’s ecosystems. University of California Press, Berkeley. Brooks, M. L., and D. A. Pyke. 2001. Invasive plants and fire in the deserts of North America. Pages 1–14 in K. E. M. Galley and T. P. Wilson, editors. Proceedings of the Invasive Species Workshop: The role of fire in the control and spread of invasive species, held at Fire Conference 2000: The first national congress on fire ecology, prevention, and management. Miscellaneous Publication 11. Tall Timbers Research Station, Tallahassee, FL. Brooks, R. R. 1987. Serpentine and its vegetation: a multidisciplinary approach. Dioscorides Press, Portland, OR. Broughton, J. M. 1994. Late Holocene resource intensification in the Sacramento Valley, California: The vertebrate evidence. Journal of Archeological Science 21:501–514. ———. 2002. Pre-Columbian human impact on California vertebrates: Evidence from old bones and implications for wilderness policy. Pages 44–71 in C. E. Kay and R. T. Simmons, editors. Wilderness and political ecology: Aboriginal influences and the original state of nature. University of Utah Press, Salt Lake City. Brown, C. S. 1998. Restoration of California central valley grasslands: Applied and theoretical approaches to understanding interactions among prairie species. Ph.D. dissertation, University of California, Davis.

Brown, C. S., and R. L. Bugg. 2001. Effects of established perennial grasses on introduction of native forbs in California. Restoration Ecology 9:38–48. Brown, C. S., R. L. Bugg, and J. H. Anderson. 1993. Roadside planting of perennial grasses. Grasslands 3:1–3. Brown, C. S., and K. J. Rice. 2000. The mark of Zorro: Effects of the exotic annual grass Vulpia myuros on California native perennial grasses. Restoration Ecology 8:10–17. Brown, C. S., K. J. Rice, and V. Claassen. 1998. Competitive growth characteristics of native and exotic grasses. New Technology and Research Program FHWA/CA/ESC-98/07. California Department of Transportation, Sacramento, CA. ———. 2000. The effects of soil amendments and mulches on establishment of California native perennial grasses: A summary of selected results. Grasslands 10:1–17. Brown, D. E. 1982. Biotic communities of the American southwest—United States and Mexico. Desert Plants 4:1–315
appendices (also available from the Boyce Thompson Southwestern Arboretum, Superior, AZ). Brown, M. J. F., and K. G. Human. 1997. Effects of harvester ants on plant species distribution and abundance in a serpentine grassland. Oecologia 112:237–243. Brown, P. M., and T. W. Swetnam. 1994. A cross-dated fire history from coast redwood near Redwood National park, California. Canadian Journal of Forest Research 24:21–31. Brown, R. 1814. Gramineae. Pp. 580–583 in Mathew Flinder. A voyage to Terra Australis. London. Browning, P. 1991. Bright gem of the western seas: California 1846–1852. Great West Books, Lafayette, CA. Bruckart, W. L. 1989. Host range determination of Puccinia jaceae from yellow starthistle. Plant Disease 73:155–160. Brunner, R. D., and T. W. Clark. 1997. A practice-based approach to ecosystem management. Conservation Biology 11:48–58. Bryant, E. 1848. What I saw in California. D. Appleton and Co. (reissued 1985 by University of Nebraska Press, Lincoln, NE). Bryant, H. C. 1929. Outdoor heritage. Powell Publishing, Los Angeles. Bugg, R. L., C. S. Brown, and J. H. Anderson. 1997. Restoring native perennial grasses to rural roadsides in the Sacramento Valley of California: Establishment and evaluation. Restoration Ecology 5:214–228. Buol, S. W., R. J. McCracken, and R. J. Southard. 1997. Soil genesis and classification. Iowa State University Press, Ames. Burcham, L. T. 1955. Recent trends in range improvement on California foothill ranges. Journal of Range Management 8:121–125. ———. 1956. Historical backgrounds of range land use in California. Journal of Range Management 9:19–24. ———. 1957. California range land: An historico-ecological study of the range resource of California. Division of Forestry, Department of Natural Resources, State of California, Sacramento. ———. 1961. Cattle and range forage in California 1770–1880. Agricultural History 35:140–149. ———. 1975. Climate, structure, and history of California’s annual grassland ecosystem. Pages 16–34 in R. M. Love, editor. The California annual grassland ecosystem. Institute of Ecology, University of California, Davis. ———. 1981. California range land. Center for Archaeological Research at the University of California, Davis. Burke, I. C., C. M. Yonker, W. J. Parton, C. V. Cole, K. Flach, and D. S. Schimel. 1989. Texture, climate, and cultivation effects

REFERENCES

329

on soil organic matter content in USA grassland soils. Soil Science Society of America Journal 53:800–805. Bussan, A. J., and W. E. Dyer. 1999. Herbicides and rangeland. Pages 116–132 in R. L. Sheley and J. K. Petroff, editors. Biology and Management of Noxious Rangeland Weeds 1999. Oregon State University Press, Corvallis. Butler, D. R. 1995. Zoomorphology: Animals as geomorphic agents. Cambridge University Press, Cambridge, UK. Bytnerowicz, A., and M. E. Fenn. 1996. Nitrogen deposition in California forests: A review. Environmental Pollution 92: 127–146. Calderón, F. J., L. E. Jackson, K. M. Scow, and D. E. Rolston. 2000. Microbial responses to simulated tillage in cultivated and uncultivated soils. Soil Biology & Biochemistry 32:1547–1559. Caldwell, R. M., J. W. Menke, and D. A. Duncan. 1985. Effects of sulfur fertilization on productivity and botanical composition of California annual grassland. Journal of Range Management 38:108–113. Caley, P., and B. Ottley. 1995. The effectiveness of hunting dogs for removing feral pigs (Sus scrofa). Wildlife Research 22: 147–154. California Department of Conservation. 2006. Farmland Mapping and Monitoring Program of the California Department of Conservation. http://www.consrv.ca.gov/DLRP/fmmp/ California Land Cover Mapping and Monitoring Program. 2003. Vegetation data. California Department of Forestry and Fire Protection and USDA Forest Service, Sacramento, CA. Cal-IPC. 2006. California Invasive Plant Inventory Publ. No. 2006–02. California Invasive Plant Council. http://www.calipc.org/ip/. Callaway, R. M. 1992. Effect of shrubs on recruitment of Quercus douglasii and Quercus lobata in California. Ecology 73: 2118–2128. Callaway, R. M., and E. T. Aschehougdagger. 2000. Invasive plants versus their new and old neighbors: A mechanism for exotic invasion. Science 20:521–523. Callaway, R. M., and C. M. D’Antonio. 1991. Shrub facilitation of coast live oak establishment in central California. Madroño 38:158–169. Callaway, R. M., and F. W. Davis. 1993. Vegetation dynamics, fire, and the physical environment in coastal central California. Ecology 74:1567–1578. ———. 1998. Recruitment of Quercus agrifolia in central California: The importance of shrub-dominated patches. Journal of Vegetation Science 9:647–656. Callaway, R. M., B. E. Mahall, C. Wicks, J. Pankey, and C. Zabinski. 2003. Soil fungi and the effects of an invasive forb on grasses: Neighbor identity matters. Ecology 84:129–135. Callaway, R. M., N. M. Nadkarni, and B. E. Mahall. 1991. Facilitation and interference of Quercus douglasii on understory productivity in central California. Ecology 72:1484–1499. Callaway, R. M., B. Newingham, C. A. Zabinski, and B. E. Mahall. 2001. Compensatory growth and competitive ability of an invasive weed are enhanced by soil fungi and native neighbors. Ecology Letters 4:429–433. CalTrans. 2004. Effective Planting Techniques to Minimize Erosion. CTSW-RT-04-004.69.01. Department of Transportation, State of California, Sacramento. Canals, R. M., V. T. Eviner, D. J. Herman, and F. S. Chapin III. 2005. Plant colonizers shape early N dynamics in gopher mounds. Plant and Soil 276:327–334.

330

REFERENCES

Canals, R. M., D. J. Herman, and M. K. Firestone. 2003. How disturbance by fossorial mammals alters N cycling in a California annual grassland. Ecology 84:875–881. Carcaillet, C., Y. Bergeron, P. Richard, B. Frechette, S. Gauthier, and Y. Prairie. 2001. Change of fire frequency in the eastern Canadian boreal forests during the Holocene: Does vegetation composition or climate trigger the fire regime? Journal of Ecology 89:930–946. Carlisle, E. A., K. L. Steenwerth, and D. R. Smart. 2006. Effects of land use on soil respiration: conversion of oak woodlands to vineyards. Journal of Environmental Quality 35:1396–1404. Carlsen, T. M., J. W. Menke, and B. M. Pavlik. 2000. Reducing competitive suppression of a rare annual forb by restoring native California perennial grasslands. Restoration Ecology 8:18–29. Carmel, Y., and C. H. Flather. 2004. Comparing landscape scale vegetation dynamics following recent disturbance in climatically similar sites in California and the Mediterranean basin. Landscape Ecology 19:573–590. ———. 2006. Constrained range expansion and climate change assessments. Frontiers in Ecology 4:178–179. Carmen, W. J. 1988. Behavioral ecology of the California scrub jay (Aphelocoma coerulescens Californica): a non-cooperative breeder with close cooperative relatives. Ph.D. dissertation, University of California, Berkeley. Carney, K. M., P. A. Matson, and B. J. M. Bohannan. 2004. Diversity and composition of tropical soil nitrifiers across a plant diversity gradient and among land-use types. Ecology Letters 7:684–694. Carpenter, C. C. 1982. The bullsnake as an excavator. Journal of Herpetology 16:394–401. Carpenter, E. J., and S. W. Cosby. 1925. Soil survey of the Salinas area, California. Bureau of Chemistry and Soils, U.S. Department of Agriculture, Washington, DC. Carr, J. C. 1892. Among the basket makers. The Californian 2:597–610. Carran, R. A., and P. W. Theobald. 2000. Effects of excreta return on properties of a grazed pasture soil. Nutrient Cycling in Agroecosystems 56:79–85. Castelli, J. P., and B. B. Casper. 2003. Intraspecific AM fungal variation contributes to plant-fungal feedback in a serpentine grassland. Ecology 84:323–336. Caswell, H. 1989. Matrix population models. Sinauer Associates, Sunderland, MA. Cayan, D., A. L. Luers, M. Hanemann, G. Franco, and B. Croes. 2006. Scenarios of climate change in California: An overview. CEC-500-2005-186-SF. California Climate Change Center, Sacramento. CCWD. 2005. Using livestock grazing as a resource management tool in California. Contra Costa Water District, Oakland, CA. http://www.ccwater.com/files/LivestockGrazingFinal72005.pdf (accessed July 2, 2007). CDF-FRAP. 2002. California Department of Forestry and Fire Protection. Multi-source landcover data v. 02_2. http://frap.cdf.ca. gov/data/frapgisdata/select.asp. ———. 2003. Changing California: Forest and range 2003 assessment. State of California Resources Agency, California Department of Forestry and Fire Protection—Fire and Resource Assessment Program, Sacramento. CDFA. 2007. Encycloweedia. Notes on identification, biology, and management of plants defined as noxious weeds by

California law. http://www.cdfa.ca.gov/phpps/ipc/encycloweedia/ encycloweedia_hp.htm (accessed March 26, 2007). CDFG. 2005. The status of rare, threatened and endangered plants and animals in California, 2000–2004. California Department of Fish and Game, Sacramento. Center, D. M., and M. B. Jones. 1983. Economics of gypsum and elemental sulfur as fertilizers on subclover-grass pastures in northern California Trifolium subterraneum. Journal of Range Management 36:378–381. Center, D. M., C. E. Vaughn, and M. B. Jones. 1989. Effects of management on plant production and nutrient cycling on two annual grassland sites. Hilgardia 57:1–40. Chadden, A., E. Dowksza, and L. Turner. 2004. Adaptive management for southern California rangelands. University of California, Santa Barbara. Challies, C. N. 1975. Feral pigs (Sus scrofa) on Auckland Island: status, and effects on vegetation and nesting sea birds. New Zealand Journal of Zoology 2:479–490. Chapin, F. S., A. J. Bloom, C. B. Field, and R. H. Waring. 1987. Plant response to multiple environmental factors. BioScience 37:49–57. Chapin, F. S., P. A. Matson, and H. A. Mooney. 2002. Principles of terrestrial ecosystem ecology. Springer-Verlag, New York. Chase, J. M., and M. A. Leibold. 2003. Ecological niches: Linking classical and contemporary approaches. University of Chicago Press, Chicago. Chaves, M., and J. Pereira. 1992. Water-stress, CO2 and climate change. Journal of Experimental Botany 43:1131–1139. Cheng, S. 2004. Forest Service research natural areas in California. General Technical Report PSW-GTR-188, USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Cheng, X., and C. S. Bledsoe. 2002. Contrasting seasonal patterns of fine root production for blue oaks (Quercus douglasii) and annual grasses in California oak woodland. Plant and Soil 240:263–274. Chiariello, N. R. 1989. Phenology of California grasslands. Pages 47–58 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Dordrecht, The Netherlands. Chiariello, N. R., and C. B. Field. 1996. Annual grassland responses to elevated CO2 in long-term community microcosms. Pages 139–175 in C. Körner and F. A. Bazzaz, editors. Community, population and evolutionary responses to elevated carbon dioxide concentration. Academic Press, San Diego, CA. Chocquenot, D., J. Hone, and G. Saunders. 1999. Using aspects of predator-prey theory to evaluate helicopter shooting for feral pig control. Wildlife Research 26:251–261. Chocquenot, D., and W. A. Ruscoe. 2003. Landscape complementation and food limitation of large herbivores: Habitat-related constraints on the foraging efficiency of wild pigs. Journal of Animal Ecology 72:14–26. Christensen, N. L., and C. H. Muller. 1975. Effects of fire on factors controlling plant growth in Adenostoma chaparral. Ecological Monographs 45:29–55. Christenson, L. N. 2004. An overview of the history of Los Peñasquitos Canyon. History Archives of San Diego County 8. Christian, C. E. 2003. A plan for monitoring and evaluating the effects of fire on plant communities at the Lassen Foothills project. The Nature Conservancy, Chico, CA. Cione, N. K., P. E. Padgett, and E. B. Allen. 2002. Restoration of a native shrubland impacted by exotic grasses, frequent fire,

and nitrogen deposition in southern California. Restoration Ecology 10:376–384. City of San Diego. 2003. Vernal pool inventory, 2002–2003. http://www.sandiego.gov/planning/mscp/vpi/index.shtml (accessed March 27, 2007). Clark, T. W., and J. R. Cragun. 1994. Organizational and managerial guidelines for endangered species restoration programs and recovery teams. Pages 9–33 in M. L. Bowles and C. J. Whelan, editors. Restoration of endangered species: Conceptual issues, planning, and implementation. Cambridge University Press, New York. Clary, W. P. 1995. Vegetation and soil responses to grazing simulation on riparian meadows. Journal of Range Management 48:18–25. ———. 1999. Stream channel and vegetation responses to late spring cattle grazing. Journal of Range Management 52: 218–227. Clausen, J., D. D. Keck, and W. M. Hiesey. 1940. Experimental studies on the nature of species. I. Effects of varied environments on western North America plants. Carnegie Institution of Washington Publication 520:1–452. Clausnitzer, D., and J. H. Huddleston. 2002. Wetland determination of a southeast Oregon vernal pool and management implications. Wetlands 22:677–685. Clegg, M. T., and R. W. Allard. 1972. Patterns of genetic differentiation in the slender wild oat species Avena barbata. Proceedings of the National Academy of Sciences USA 69:1820–1824. Clegg, M. T., R. W. Allard, and A. L. Kahler. 1972. Is the gene the unit of selection? Evidence from two experimental plant populations. Proceedings of the National Academy of Sciences USA 69:2474–2478. Cleland, E. E. 2005. The influence of multiple interacting global changes on the structure and function of a California annual grassland ecosystem. Ph.D. dissertation, Stanford University, Stanford, CA. Cleland, E. E., N. R. Chiariello, S. R. Loarie, H. A. Mooney, and C. B. Field. 2006. Diverse responses of phenology to global changes in a grassland ecosystem. Proceedings of the National Academy of Sciences 103:13740– 13744. Clements, F. E. 1916. Plant succession: analysis of the development of vegetation. Carnegie Institution of Washington Publication 242. ———. 1934. The relict method in dynamic ecology. Journal of Ecology 22:39–68. ———. 1936. Nature and structure of the climax. Journal of Ecology 24:252–284. Clements, F. E., and V. E. Shelford. 1939. Bio-Ecology. John Wiley and Sons, New York. CNGA. 2003. California Native Grasslands Association: Products and Services Directory. Davis, CA. http://www.cnga.org/psd.php. CNPS. 2001. Plot data from the central and northern Sierra Nevada foothills. California Native Plant Society, Sacramento, CA. Unpublished file. Coblentz, B. E. 1978. The effects of feral goats (Capra hircus) on island ecosystems. Biological Conservation 13:297–286. Cockburn, A., and W. Z. Lidicker. 1983. Microhabitat heterogeneity and population ecology of an herbivorous rodent, Microtus californicus. Oecologia 59:167–177. Colautti, R. I., A. Ricciardi, I. A. Grigorovich, and H. J. MacIsaac. 2004. Is invasion success explained by the enemy release hypothesis? Ecology Letters 7:721–733.

REFERENCES

331

Collier, E. T., and S. B. Thalman, editors. 1991. Interviews with Tom Smith and Maria Copa: Isabel Kelly’s ethnographic notes on the Coast Miwok Indians of Marin and Southern Sonoma Counties, California. Miwok Archeological Preserve of Marin, Marin, CA. Collinge, S. K., C. A. Wise, and B. Weaver. 2003. Germination, early growth, and flowering of a vernal pool annual in response to soil moisture and salinity. Madroño 50:83–93. Collins, S. L. 1987. Interactions of disturbances in tallgrass prairie: a field experiment. Ecology 68:1243–1250. Collins, S. L., A. K. Knapp, J. M. Briggs, J. M. Blair, and E. M. Steinauer. 1998. Modulation of diversity by grazing and mowing in native tallgrass prairie. Science 280:745–747. Conant, R. T., and K. Paustian. 2002. Potential soil carbon sequestration in overgrazed grassland ecosystems. Global Biogeochemical Cycles 16(4), 1143, doi:10.1029/2001GB001661. Conant, R. T., K. Paustian, and E. T. Elliott. 2001. Grassland management and conversion into grassland: effects on soil carbon. Ecological Applications 11:343–355. Connell, J. H. 1954. Home range and mobility of brush rabbits in California chaparral. Journal of Mammalogy 35:392–405. ———. 1978. Diversity in tropical rainforests and coral reefs. Science 199:1302–1310. Conner, J. K., and D. L. Hartl. 2004. A primer of ecological genetics. Sinauer Associates, Sunderland, MA. Conner, J. M., and B. L. Willoughby. 1997. Effects of blue oak canopy on annual forage production. Pages 321–326 in N. H. Pillsbury, J. Verner, and W. D. Tietje, editors. Proceedings of a symposium on oak woodlands: Ecology, management, and urban interface issues 1997. Gen. Tech. Rep. PSW-GTR-160. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Connor, R., A. Seidl, L. VanTassell, and N. Wilkins. 2001. United States grasslands and related resources: An economic and biological trends assessment. Texas A&M University, College Station. http://landinfo.tamu.edu/presentations/grasslands. cfm. Conrad, C. E., E. J. Woolfolk, et al. 1966. Fertilization and mangement implications on California annual-plant range. Journal of Range Management 19:20–26. Conrad, J. P., H. L. Hall, and B. A. Chaugule. 1948. Sulfur fertilization of legumes in the upper Ojai Valley, California, and the resulting effects on the following nonlegumes. Soil Science Society of America Proceedings 12:275–277. Conroy, R. M., and D. L. Chesemore. 1992. Food habits of burrowing owl (Athene cunicularia) in Fresno County, California. Pages 105–110 in D. F. Williams, S. Byrne, and T. A. Rado, editor. Endangered and sensitive species of the San Joaquin Valley: Their biology, management and conservation. California Energy Commission, Sacramento. Cook, S. A. 1965. Population regulation of Eschscholzia californica by competition and edaphic conditions. Journal of Ecology 53:759–769. Coombs, E. M., J. K. Clark, G. L. Piper, and A. F. Cofrancesco, editors. 2004. Biological control of invasive plants in the United States. Oregon State University Press, Corvallis. Cooper, D. W. 1960. Fort Baker ranges return to champagne grasses. Journal of Range Management 13:203–205. Cooper, W. S. 1922. The broad-sclerophyll vegetation of California: an ecological study of the chaparral and its related communities. Carnegie Institution of Washington Publication 319.

332

REFERENCES

Corbin, J. D., and C. M. D’Antonio. 2004a. Can carbon addition increase competitiveness of native grasses? A case study from California. Restoration Ecology 12:36–43. ———. 2004b. Competition between native perennial and exotic annual grasses: Implications for an historical invasion. Ecology 85:1273–1283. Corbin, J. D., C. M. D’Antonio, and S. Bainbridge. 2004. Tipping the balance in the restoration of native plants: Experimental approaches to changing the exotic:native ratio in California grassland. Pages 154–179 in M. S. Gordon and S. M. Bartol, editors. Experimental approaches to conservation biology. University of California Press, Berkeley. Corbin, J. D., M. A. Thomsen, T. E. Dawson, and C. M. D’Antonio. 2005. Summer water use by California coastal prairie grasses: fog, drought, and community composition. Oecologia 145: 511–521. Cordy, D. R. 1954. Nigropallidal encephalomalacia in horses associated with ingestion of yellow star thistle. Journal of Neuropathology and Experimental Neurology 13:330–342. Cook, S. A. 1961. Genetic system, variation, and adaptation in Eschscholzia californica. Evolution 16:278– 299. Coues, E., editor. 1900. On the trail of a Spanish pioneer: The diary and itinerary of Francisco Garcés in his travels through Sonora, Arizona, and California 1775–1776. Francis P. Harper, New York. Cox, G. W. 1984. The distribution and origin of mima mound grassland in San Diego County, California. Ecology 65(5): 1397–1405. ———. 1990. Comment on “Formation of mima mounds: a seismic hypothesis.” Geology 18:1259–1260. ———. 1999. Alien species in North America and Hawaii: impacts on natural ecosystems. Island Press, Washington, DC. ———. 2004. Alien Species and Evolution. Island Press, Washington, DC. Cox, G. W., and D. W. Allen. 1987. Soil translocation of pocket gophers in a mima moundfield. Oecologia 72:207–210. Cox, G. W., L. C. Contreras, and A. V. Milewski. 1995. Role of fossorial animals in community structure and energetics of Pacific Mediterranean ecosystems. Pages 383–398 in M. T. Arroyo, P. H. Zedler, and M. D. Fox, editors. Ecology and biogeography of Mediterranean ecosystems in Chile, California, and Australia. Springer-Verlag, New York. Cox, G. W., and V. B. Sheffer. 1991. Pocket gophers and mima terrain in North America. Natural Areas Journal 11:193–198. CPIF. 2000. The grassland bird conservation plan: A strategy for protecting and managing grassland habitats and associated birds in California. California Partners in Flight and Point Reyes Bird Observatory, Stinson Beach, CA. Cramer, W., A. Bondeau, F. I. Woodward, I. C. Prentice, R. A. Betts, V. Brovkin, P. M. Cox, V. Fisher, J. A. Foley, A. D. Friend, C. Kucharik, M. R. Lomas, N. Ramankutty, S. Sitch, B. Smith, A. White, and C. Young-Molling. 2001. Global response of terrestrial ecosystem structure and function to CO2 and climate change: Results from six dynamic global vegetation models. Global Change Biology 7:357–373. Crampton, B. 1974. Grasses in California. University of California Press, Berkeley. Crawley, M. J. 1989. The relative importance of vertebrate and invertebrate herbivores in plant population dynamics. Pages 45–71 in E. A. Bernays, editor. Insect-Plant Interactions. CRC Press, Boca Raton, FL.

Crespí, J. 2001. A description of distant roads: Original journals of the first expedition into california, 1769–1770. Edited and translated by A. K. Brown. San Diego State University Press, San Diego, CA. Cronin, T. M. 1999. Principles of paleoclimatology. Perspectives in paleobiology and earth history. Columbia University Press, New York. Cronise, T. F. 1868. The natural wealth of California. H. H. Bancroft and Co., San Francisco, CA. Cronquist, A., A. H. Holmgren, J. H. Holmgren, J. L. Reveal, and P. K. Holmgren. 1977. Intermountain flora. Vascular plants of the intermountain West, USA. Columbia University Press, New York. Cross, A. F., and W. H. Schlesinger. 1999. Plant regulation of soil nutrient distribution in the northern Chihuahuan Desert. Plant Ecology 145:11–25. Crouchley, G. 1980. Regrowth control by goats — plus useful meat returns. New Zealand Journal of Agriculture (1980, December):9–14. CRT. 2005. California Rangeland Trust, Home Page California Rangeland Trust http://www.rangelandtrust.org/crt/. Cruden, R. W. 1974. The adaptive nature of seed germination in Nemophila menziesii Aggr. Ecology 55:1295–1305. ———. 1977. Pollen-ovule ratios: a conservative indicator of breeding systems in flowering plants. Evolution 31:32–46. Cruz, F., C. J. Donlan, K. Campbell, and V. Carrion. 2005. Conservation action in the Galapagos: feral pig (Sus scrofa) eradication from Santiago Island. Biological Conservation 121:473–478. Cui, M., and M. M. Caldwell. 1997. A large ephemeral release of nitrogen upon wetting of dry soil and corresponding root responses in the field. Plant and Soil 191:291–299. Culin, S. 1975. Games of the North American Indians. Dover Publications, Inc., New York. Cunliffe, B. A., and M. H. Meyer. 2002. Propagation time affects winter survival and finishing date for ornamental grasses. Journal of Environmental Horticulture 20:201–203. Curtis, M. J., and V. P. Claassen. 2005. Compost incorporation increases plant available water in a drastically disturbed serpentine soil. Soil Science 170:939–953. Cushman, J. H., T. A. Tierney, and J. M. Hinds. 2004. Variable effects of feral pig disturbances on native and exotic plants in a California grassland. Ecological Applications 14:1746–1756. CVIS. 2006. California vegetation information system, database query. Vegetation Classification and Mapping Program, Department of Fish and Game, Sacramento, CA. Cypher, B. L., K. A. Spencer, and J. H. Scrivner. 1996. Use of food items by sex and age classes of coyotes. California Fish and Game 82:42–47. D’Antonio, C. M. 1990. Invasion and dominance of coastal plant communities by the introduced succulent, Carpobrotus edulis. Ph.D. dissertation. University of California, Santa Barbara. ———. 1993. Mechanisms controlling invasion of coastal plant communities by the alien succulent Carpobrodus edulis. Ecology 74:83–95. D’Antonio, C. M., S. J. Bainbridge, C. Kennedy, J. W. Bartolome, and S. Reynolds. 2006. Ecology and restoration of California grasslands with special emphasis on the influence of fire and grazing on native grassland species. A report to the David and Lucille Packard Foundation. University of California, Santa Barbara. D'Antonio, C. M., and J. D. Corbin. 2003. Effects of plant invaders on nutrient cycling: using models to explore the link

between invasion and development of species effects. Pages 363– 384 in C. D. Canham, J. J. Cole, and W. K. Lauenrothy, editors. Models in ecosystem science. Princeton University Press, Princeton, NJ. D’Antonio, C. M., T. L. Dudley, and M. Mack. 1999. Disturbance and biological invasions: Direct effects and feedbacks. In L. R. Walker, editor. Ecosystems of disturbed ground. Elsevier, New York. D’Antonio, C. M., and S. E. Hobbie. 2005. Plant system effects on ecosystem processes: Insights from invasive species. Pages 65–84 in D.F. Sax, J. J. Stachowitz, and S.D. Gaines, editors. Species invasions, insights into ecology, evolution and biogeography. Sinauer Associates, Sunderland, MA. D’Antonio, C. M., and P. M. Vitousek. 1992. Biological invasions by exotic grasses, the grass/fire cycle, and global change. Annual Review of Ecology and Systematics 23: 63–87. D’Arcy, C. J. 1995. Symptomology and host range of barley yellow dwarf. Pages 9–28 in C. J. D’Arcy and P. A. Burnett, editors. Barley yellow dwarf: 40 years of progress. The American Phytopathological Society, St. Paul, MN. Da Silva, P. G., and J. W. Bartolome. 1984. Interaction between a shrub, Baccharis pilularis subsp. consanguinea (Asteraceae), and an annual grass, Bromus mollis (Poaceae) in coastal California. Madroño 31:93–101. Daar, S. 1983. Using goats for brush control. The IPM Practitioner 5:4–6. Daehler, C. C. 1998. The taxonomic distribution of invasive angiosperm plants: ecological insights and comparison to agricultural weeds. Biological Conservation 84:167–180. Dahl, B. E., and E. W. Tisdale. 1975. Environmental factors related to medusahead distribution. Journal of Range Management 28: 463–478. Dahlgren, R. A., W. R. Horwath, K. W. Tate, and T. J. Camping. 2003. Blue oak enhance soil quality in California oak woodlands. California Agriculture 57:2:42–47. Dahlgren, R. A., M. J. Singer, and X. Huang. 1997. Oak tree and grazing impacts on soil properties and nutrients in a California oak woodland. Biogeochemistry 39:45–64. Dahlgren, R. A., K. W. Tate, D. Lewis, E. R. Atwill, J. M. Harper, and B. Allen-Diaz. 2001. Watershed research examines rangeland management effects on water quality. California Agriculture 55:64–71. Daily, G. C., editor. 1997. Nature’s services: societal dependence on natural ecosystems. Island Press, Washington DC. Daily, G. C., S. Alexander, P. R. Ehrlich, L. Goulder, J. Lubchenko, P. A. Matson, H. A. Mooney, S. Postel, S. H. Schneider, D. Tilman, and G. M. Woodwell. 1997. Ecosystem services: Benefits supplied to human societies by natural ecosystems. Issues in Ecology 2:1–18. Dakin, S. B., editor. 1939. A Scotch paisano: Hugo Reid’s life in California 1832–1852 derived from his correspondence. University of California Press, Berkeley. Dale, V. H., S. Brown, R. A. Haeuber, N. T. Hobbs, N. Huntly, R. J. Naiman, W. E. Riebsame, M. G. Turner, and T. J. Valone. 2000. Ecological principles and guidelines for managing the use of land. Ecological Applications 10:639–670. Dalrymple, L. 2000. Indian basket makers of California and the Great Basin. Museum of New Mexico Press, Santa Fe. Damhoureyeh, S. A., and D. C. Hartnett. 2002. Variation in grazing tolerance among three tallgrass prairie plant species. American Journal of Botany 89:1634–1643.

REFERENCES

333

Daniel, J. A., K. Potter, W. Altom, H. Aljoe, and R. Stevens. 2002. Long-term grazing density impacts on soil compaction. Transactions of the ASAE 45:1911–1915. Danielsen, K. C. 1990. Seedling growth rates of Quercus lobata Nee. (Valley oak) and the competitive effects of selected grass species. M.A. thesis, California State University, Los Angeles. Dao, T. H. 1998. Tillage and crop residue effects on carbon dioxide evolution and carbon storage in a Paleustoll. Soil Science Society of America Journal 62:250–256. Dasmann, R. 1973. The destruction of California. The Macmillan Company, New York. Daubenmire, R. F. 1970. Steppe vegetation of Washington. Agricultural Experiment Station Technical Bulletin 62, Washington State University, Pullman. Daubenmire, R. F. 1975. An analysis of structural and functional characters along a steppe-forest catena. Northwest Science 49:120–140. Davidson, E. A., S. C. Hart, C. A. Shanks, and M. K. Firestone. 1991. Measuring gross nitrogen mineralization, immobilization, and nitrification by N-15 isotopic pool dilution in intact soil cores. Journal of Soil Science 42:335–349. Davidson, E. A., D. J. Herman, A. Schuster, and M. K. Firestone. 1993. Cattle grazing and oak trees as factors affecting soil emissions of nitric oxide from an annual grassland. Pages 100–119 in R. J. Luxmoore, editor. Agricultural ecosystem effects on trace gases and global climate change: Proceedings of a symposium. ASA Special Publication No. 55. American Society of Agronomy, Crop Science Society of America, Soil Science Society of America, Madison, WI. Davidson, E. A., J. M. Stark, and M. K. Firestone. 1990. Microbial production and consumption of nitrate in an annual grassland. Ecology 71:1968–1975. Davis, F. W., M. I. Borchert, L. E. Harvey, and J. Michaelsen. 1991. Factors affecting seedling survivorship of blue oak (Quercus douglasii H.&A.) in central California. Pages 81–86 in R. B. Standiford, editor. Proceedings of the symposium on oak woodlands and hardwood rangeland management 1991. General Technical Report PSW-GTR-126. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Davis, F. W., J. M. Scott, and D. D. Goble. 2005. Renewing the conservation commitment. Pages 299–306 in D. D. Goble, J. M. Scott, and F. W. Davis, editors. The Endangered Species Act at thirty. Vol. 1: Renewing the conservation promise. Island Press, Washington DC. Davis, F. W., D. M. Stoms, A. D. Hollander, K. A. Thomas, P. A. Stine, D. Odion, M. I. Borchert, J. H. Thorne, M. V. Gray, R. E. Walker, K. Warner, and J. Graae. 1998. The California Gap Analysis Project: Final Report, June 30, 1998. University of California, Santa Barbara. http://www.biogeog.ucsb.edu/ projects/gap/gap_rep.html. Davis, M. A., and K. Thompson. 2001. Invasion terminology: Should ecologists define their terms differently than others? No, not if we want to be of any help! Ecological Society of America Bulletin 82:206. Davis, M. A., J. P. Grime, and K. Thompson. 2000. Fluctuating resources in plant communities: A general theory of invasibility. Journal of Ecology 88:528–534. Davis, O. K. 1999. Pollen analysis of Tulare Lake, California: Great Basin-like vegetation in central California. Review of Palaeobotany and Palynology 107:249–257.

334

REFERENCES

———. 2002. Southern California pollen diagrams. http:// www.geo.arizona.edu/palynology/socalif/map_soca.html (accessed March 20, 2007). Davis, O. K., and M. J. Moratto. 1988. Evidence for a warm dry early Holocene in the western Sierra Nevada of California: Pollen and plant macrofossil analysis of Dinkey and Exchequer meadows. Madroño 35:132–149. Dawson, T. E. 1998. Fog in the California redwood forest: Ecosystem inputs and use by plants. Oecologia 117:476–485. Dawson, T. E., S. Mambelli, A. H. Plamboeck, P. H. Templer, and K. P. Tu. 2002. Stable isotopes in plant ecology. Annual Review of Ecology and Systematics 33:507–559. De Bach, P., and D. Rosen. 1991. Biological control by natural enemies. Cambridge University Press, Cambridge, UK. de la Vega, M. P. 1996. Plant genetic adaptedness to climatic and edaphic environment. Euphytica 92:27–38. de la Vega, M. P., P. Garcia, and R. W. Allard. 1991. Multilocus genetic structure of ancestral Spanish and colonial Californian populations of Avena barbata. Proceedings of the National Academy of Sciences USA 88:1202–1206. Deering, R. H., and T. P. Young. 2006. Germination speeds of exotic annual and native perennial grasses in California and the potential benefits of seed priming for grassland restoration. Grasslands 16:1,14–17. Del-Val, E. K., and M. J. Crawley. 2005. Are grazing increaser species better tolerators than decreasers? An experimental assessment of defoliation tolerance in eight British grassland species. Journal of Ecology 93:1005–1016. Delmas, A. 1999. The effect of fire on California’s native grassland in the absence of grazing at the Nature Conservancy’s Vina Plains Preserve, in southern Tehama County. Master’s thesis, California State University, Chico. DeLoach, C. J. 1991. Past successes and current prospects in biological control of weeds in the United States and Canada. Natural Areas Journal 11:129–142. Denef, K., J. Six, H. Bossuyt, S. D. Frey, E. T. Elliott, R. Merckx, and K. Paustian. 2001. Influence of dry-wet cycles on the interrelationship between aggregate, particulate organic matter, and microbial community dynamics. Soil Biology & Biochemistry 33:1599–1611. Dennis, A. 1989. Effects of defoliation on three native perennial grasses in the California annual grassland. Ph.D. dissertation, University of California, Berkeley. Derlet, R. W., and J. R. Carlson. 2006. Coliform bacteria in Sierra Nevada wilderness lakes and streams: What is the impact of backpackers, pack animals, and cattle? Wilderness & Environmental Medicine 17:15–20. DeSimone, S. A., and P. H. Zedler. 1999. Shrub seedling recruitment in unburned Californian coastal sage scrub and adjacent grassland. Ecology 80:2018–2032. ———. 2001. Do shrub colonizers of southern Californian grassland fit generalities for other woody colonizers? Ecological Applications 11:1101–1111. Devlin, B., and N. C. Ellstrand. 1990. The development and application of a refined method for estimating gene flow from angiosperm paternity analysis. Evolution 44:248–259. DeWitt, T. J., A. Sih, and D. S. Wilson. 1998. Costs and limits of phenotypic plasticity. Trends in Ecology and Evolution 13:77–81. Di, H. J., and K. C. Cameron. 2002. Nitrate leaching in temperate agroecosystems: sources, factors and mitigating strategies. Nutrient Cycling in Agroecosystems 64:237–256.

Dickey, P. B., O. K. Hoglund, and B. A. Madson. 1948. Effect of fertilizers on the production and season of use of annual grass range in California. Agronomy Journal 40:186–188. Dickinson, W., J. Ferreyra, K. L. Imbesi, S. Joshi, C. Kingsolver, E. Klein, N. Lessios, A. Ng, T. Stamp, K. White, D. Xu, and R. L. Vidra. 2006. The ethical challenges faced by ecological restorationists. Ecological Restoration 24:102–104. Diffenbaugh, N. S., and L. C. Sloan. 2004. Mid-Holocene orbital forcing of regional-scale climate: A case study of western North America using a high-resolution RCM. Journal of Climate 17:2927–2937. DiTomaso, J. M. 1997. Risk analysis of various weed control methods. Pages 34–39 in M. Kelly, E. Wagner, and P. Warner, editors. Proceedings, California Exotic Pest Plant Council Symposium. Vol. 3. California Exotic Pest Plant Council, Trabuco Canyon, CA. ———. 2000. Invasive weeds in rangelands: Species, impacts, and management. Weed Science 48:255–265. ———. 2002. Irrigated pastures. Pages 329–336 in Principles of Weed Control. California Weed Science Society, Thomson Publications, Fresno, CA. ———. 2006. Yellow starthistle information. Weed Research and Information Center, University of California, Davis. http://wric. ucdavis.edu/yst/ DiTomaso, J. M., K. L. Heise, G. B. Kyser, A. M. Merenlender, and R. J. Keiffer. 2001. Carefully timed burning can control barb goatgrass. California Agriculture 55:47–53. DiTomaso, J. M., and D. W. Johnson, editors. 2006. The use of fire as a tool for controlling invasive plants. Cal-IPC Publication 2006-01. California Invasive Plant Council, Berkeley, CA. DiTomaso, J. M., G. B. Kyser, and M. S. Hastings. 1999a. Prescribed burning for control of yellow starthistle (Centaurea solstitialis) and enhanced native plant diversity. Weed Science 47:233–242. DiTomaso, J. M., G. B. Kyser, J. R. Miller, S. Garcia, R. F. Smith, G. A. Nader, J. M. Connor, and S. B. Orloff. 2006. Integrating prescribed burning and clopyralid for the management of yellow starthistle (Centaurea solstitialis). Weed Science 54:757–767. DiTomaso, J. M., G. B. Kyser, and S. B. Orloff. 1999b. New growth regulator herbicide provides excellent control of yellow starthistle. California Agriculture 53:12–16. DiTomaso, J. M., G. B. Kyser, S. B. Orloff, and S. F. Enloe. 2000. Integrated strategies offer site-specific control of yellow starthistle. California Agriculture 54:30–36. DiTomaso, J. M., G. B. Kyser, S. Orloff, S. Garcia, G. Nader, and R. Smith. 2003. Integrated management of yellow starthistle and medusahead using prescribed burning and clopyralid. Pages 62–63 in M. Kelly, editor. Proceedings, California Exotic Pest Plant Council Symposium. Vol. 6. California Exotic Pest Plant Council, Berkeley. DiTomaso, J. M., and W.T. Lanini. 2001a. UC IPM online statewide integrated pest management program: yellow starthistle. University of California, Division of Agriculture and Natural Resources. Davis. http://www.ipm.ucdavis.edu/ PMG/PESTNOTES/pn7402.html. ———. 2001b. UC IPM online statewide integrated pest management program: poison oak. http://www.ipm.ucdavis.edu/ PMG/PESTNOTES/pn7431.html. DiVittorio, C., J. D. Corbin, and C. M. D’Antonio. 2007. Spatial and temporal patterns of seed dispersal: An important determinant of grassland invasion and restoration potential. Ecological Applications 17:311–316.

Dobzhanksy, T. 1970. Genetics of evolutionary process. Columbia University Press, New York. Dompierre, H., and C. S. Churcher. 1996. Premaxillary shape as an indicator of the diet of seven extinct late Cenozoic New World camels. Journal of Vertebrate Paleontology 16:141–148. Donovan, L. A., and J. R. Ehleringer. 1992. Contrasting water-use patterns among size and lilfe-history classes of a semi-arid shrub. Functional Ecology 6:482–488. Doran, J. W. 1980. Soil microbial and biochemical changes associated with reduced tillage. Soil Science Society of America Journal 5:765–771. Downing, J. A., C. W. Osenberg, and O. Sarnelle. 1999. Metaanalysis of marine nutrient-enrichment experiments: Variation in the magnitude of nutrient limitation. Ecology 80:1157–1167. Drake, J. A., H. A. Mooney, F. di Castri, R. H. Groves, F. J. Kruger, M. Rejmanek, and M. Williamson, editors. 1989. Biological invasions: A global perspective. John Wiley and Sons, Chichester, UK. Drake, K. R., and T. D. Whitson. 1989. Control of hoary cress (Cardaria draba [L.] Desv.) on pastures. Proceedings of Western Society of Weed Science 42:40. Driver, H. E. 1939. Culture element distributions X: Northwest California. University of California Anthropological Records 1:297–433. Drucker, P. 1937. The Tolowa and their southwest Oregon kin. University of California Publications in American Archaeology and Ethnology 36:221–300. Du Bois, C. A. 1935. Wintu ethnography. University of California Publications in American Archaeology and Ethnology 36: 1–148. Dukes, J. S. 2000. Will the increasing atmospheric CO2 concentration affect the success of invasive species? Pages 95–113 in H. A. Mooney and R. J. Hobbs, editors. Invasive Species in a Changing World. Island Press, Washington. ———. 2001a. Biodiversity and invasibility in grassland microcosms. Oecologia 126:563–568. ———. 2001b. Productivity and complementarity in grassland microcosms of varying diversity. Oikos 94:468–480. ———. 2002a. Comparison of the effect of elevated CO2 on an invasive species (Centaurea solstitialis) in monoculture and community settings. Plant Ecology 160:225–234. ———. 2002b. Species composition and diversity affect grassland susceptibility and response to invasion. Ecological Applications 12:602–617. Dukes, J. S., N. R. Chiariello, E. E. Cleland, L. A. Moore, M. R. Shaw, S. Thayer, T. Tobeck, H. A. Mooney, and C. B. Field. 2005. Responses of grassland production to single and multiple global environmental changes. Public Library of Science Biology 3:1829–1837. Dukes, J. S., and C. B. Field. 2000. Diverse mechanisms for CO2 effects on grassland litter decomposition. Global Change Biology 6:145–154. Dukes, J. S., and B. A. Hungate. 2002. Elevated carbon dioxide and litter decomposition in California annual grasslands: which mechanisms matter? Ecosystems 5:171–183. Dukes, J. S., and H. A. Mooney. 1999. Does global change increase the success of biological invaders? Trends in Ecology & Evolution 14:135–139. Duncan, C. L., and J. K. Clark, editors. 2005. Invasive plants of range and wildlands and their environmental, economic and societal impacts. Weed Science Society of America, Lawrence, KS.

REFERENCES

335

Duncan, D. A., and R. G. Woodmansee. 1975. Forecasting forage yield from precipitation in California’s annual rangeland. Journal of Range Management 28:327–329. Dwire, K. A. 1984. What happens to native grasses when grazing stops? Fremontia 12:23–25. Dyer, A. R. 1993. Response of native bunchgrass to burning and grazing. Restoration and Management Notes 11:150–151. ———. 2002. Burning and grazing management in a California grassland: Effect on bunchgrass seed viability. Restoration Ecology 10:107–111. ———. 2003. Burning and grazing management in a California grassland: Growth, mortality, and recruitment of Nassella pulchra. Restoration Ecology 11:291–296. Dyer, A. R., A. Fenech, and K. J. Rice. 2000. Accelerated seedling emergence in interspecific competitive neighborhoods. Ecology Letters 3:523–529. Dyer, A. R., H. C. Fossum, and J. W. Menke. 1996. Emergence and survival of Nassella pulchra in a California grassland. Madrono 43:316–333. Dyer, A. R., and K. J. Rice. 1997a. Evidence of spatial genetic structure in a California bunchgrass population. Oecologia 112:333–339. ———. 1997b. Intraspecific and diffuse competition: the response of Nassella pulchra in a California grassland. Ecological Applications 7:484–492. ———. 1999. Effects of competition on resource availability and growth of a California bunchgrass. Ecology 80:2697–2710. Dyer, D. 2003. A technical summary of plant materials studies at the Lockeford Plant Materials Center, Lockeford, California. USDA NRCS Lockeford Plant Materials Center Annual Technical Report, Lockeford, CA. Dyer, M. I., C. L. Turner, and T. R. Seastedt. 1993. Herbivory and its consequences. Ecological Applications 3:10–16. Dyksterhuis, E. J. 1949. Condition and management of rangeland based quantitative ecology. Journal of Range Management 2:104–115. Eastman, Q. 2005. Cattle, birds, endangered species coexist in Ramona grasslands. San Diego North County Times, San Diego, CA, December 24, 2005. EBRPD. 2005. Grazing animals in the East Bay Regional Park District. East Bay Regional Parks District. http://www.ebparks. org/resources/stewards/grazing.htm. EDF. 2007. Alameda County’s rancher and wildlife-friendly program restores ponds and helps rare animals. Environmental Defense Fund, Center for Conservation Incentives. http:// www.environmentaldefense.org/article.cfm?contentID6295. Edwards, S. W. 1992. Observations on the prehistory and ecology of grazing in California. Fremontia 20:3–11. ———. 1995. Notes on grazing and native plants in central California. Four Seasons 10:61–65. ———. 1996. A Rancholabrean-age, latest Pleistocene bestiary for California botanists. The Four Seasons 10:5–34. ———. Further notes on Rancholabrean ecology. The Four Seasons 10:25–28. ———. A botanical exploration of the Tilden Park areas of San Pablo Ridge, Contra Costa County, CA. The Four Seasons 11: 3–52. ———. 2004. Paleobotany of California. The Four Seasons 12(2): 3–75. Egan, D., and E. Howell, editors. 2001. The historical ecology handbook: A restorationist’s guide to reference ecosystems. Society for Ecological Restoration, Washington, D.C.

336

REFERENCES

Ehleringer, J. R., and R. K. Monson. 1993. Evolutionary and ecological aspects of photosynthetic pathway variation. Annual Review of Ecology and Systematics 24:411–439. Ehrenfeld, J. G. 2003. Effects of exotic plant invasions on soil nutrient cycling processes. Ecosystems 6:503–523. Ehrenfeld, J. G., and N. Scott. 2001. Invasive species and the soil: effects on organisms and ecosystem processes. Ecological Applications 11:1259–1260. Ehrlich, P. R., D. S. Dobkin, and D. Wheye. 1988. The birder’s handbook: A field guide to the natural history of North American birds. Simon and Schuster, Inc., New York. Ehrlich, P. R., and I. Hanski. 2004. On the wings of the checkerspots: A model system for population biology. Oxford University Press, New York. Eigenheer, R. A. 1976. Early perceptions of agricultural resources in the Central Valley of California. Ph.D. dissertation, Univeristy of California, Davis. Eliason, S. A., and E. B. Allen. 1997. Exotic grass competition in suppressing native shrubland re-establishment. Restoration Ecology 5:245–255. Eller, K. G. 1994. The potential value of fire for managing Stephen’s kangaroo rat habitat at Lake Perris. University of California, Riverside. Elliott, H. W., and R. H. Barrett. 1985. Dietary overlap among axis, fallow, and blacktailed deer and cattle. Journal of Range Management 38:546–550. Elliott, H. W., and J. D. Wehausen. 1974. Vegetational succession on coastal rangeland of Point Reyes Peninsula. Madroño 22:231–238. Ellstrand, N. C. 2006. When crop transgenes wander in California, why should we worry? California Agriculture 60:116–125. Elmore, W. 1992. Riparian responses to grazing practices. Pages 442–457 in R. J. Naiman, editor. Watershed management: Balancing sustainability and environmental change. SpringerVerlag, New York. Elton, C. S. 1958. The ecology of invasions. Methuen, London, UK. Ely, L. L., Y. Enzel, V. R. Baker, and D. R. Cayan. 1993. A 5000-year record of extreme floods and climate change in the southwestern United States. Science 262:410–412. Elzinga, C. D., D. Salzer, J. W. Willoughby, and J. Gibbs. 2001. Monitoring plant and animal populations: A handbook for field biologists. Blackwell Publishing. Elzinga, C. L., D. W. Salzer, and J. W. Wiloughby. 1998. Measuring and monitoring plant populations. BLM/RS/ST-98/005
1730. USDI Bureau of Land Management and The Nature Conservancy, Denver, CO. Endler, J. A. 1977. Geographic variation, speciation, and clines. Princeton University Press, Princeton, NJ. Enloe, S. F. 2002. The ecology and management of yellow starthistle in annual and perennial grassland communities. Ph.D. dissertation, University of California, Davis. Enloe, S. F., J. M. DiTomaso, S. B. Orloff, and D. J. Drake. 2004. Soil water dynamics differ among rangeland plant communities dominated by yellow starthistle (Centaurea solstitialis) annual grasses, or perennial grasses. Weed Science 52: 929–935. ———. 2005. Perennial grass establishment integrated with clopyralid treatment for yellow starthistle management on annual range. Weed Technology 19:94–101. EPA. 2001. Heat island reduction initiative: Inside the green house. EPA-430-N-01-002, Environmental Protection Agency, Washington, DC.

———. 2003. Endangered and threatened wildlife and plants; listing of the Central California distinct population segment of the California tiger salamander; reclassification of the Sonoma County and Santa Barbara County distinct populations from endangered to threatened; special rule. Federal Register 68(100):28647–28670. ———. 2005. Endangered and threatened wildlife and plants; designation of critical habitat for the California tiger salamander in Sonoma County. Federal Register 70(147): 44301–44322. Epstein, H. E., W. K. Lauenroth, and I. C. Burke. 1997. Effects of temperature and soil texture on ANPP in the U.S. Great Plains. Ecology 78:2628–2631. Erlandson, J. M. 1984. A case study in faunal turbation: Delineating the effects of the burrowing pocket gopher on the distribution of archeological materials. American Antiquity 49:785–790. ———. 1994. Early hunter-gatherers of the California coast. Plenum Press, New York. Erwin, W. J., and R. H. Stasiak. 1979. Vertebrate mortality during the burning of a reestablished prairie in Nebraska. The American Midland Naturalist 101:247–249. Espeland, E. K., T. M. Carlsen, and D. MacQueen. 2005. Fire dynamics of granivory on a California grassland forb. Biodiversity and Conservation 14:267–280. Espeland, E. K., and R. G. Myatt. 2001. Evidence for a sand hills ecotype of Eschscholzia californica (Papaveraceae). Madroño 48:25–32. Esseks, J. D., and D. J. Drozd. 2002. Policy insights from two sets of surveys: A public opinion poll about agricultural stewardship issues conducted nation-wide and in seven regions June and July 2001 and surveys of owners of urban-edge agricultural land in five important farm states in July 2001 to February 2002. American Farmland Trust, Washington, DC. Evans, R. A., B. L. Kay, and J. A. Young. 1975. Microenvironment of a dynamic annual community in relation to range improvement. Hilgardia 43:78–102. Evans, R. A., and J. A. Young. 1972. Germination and establishment of Salsola in relation to seedbed environment. II. Seed distribution, germination, and seedling growth of Salsola microenvironmental monitoring of seedbed. Agronomy Journal 64: 219–224. ———. 1989. Characterization and analysis of abiotic factors and their influence on vegetation. Pages 13–28 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Dordrecht, The Netherlands. Evens, J. M., E. Kentner, and J. Klein. 2006. Classification of vegetation associations from the Mount Tamalpais Watershed, Nicasio Reservoir, and Soulajule Reservoir in Marin County, California. Report to Marin Municipal Water District. California Native Plant Society, Sacramento. Evens, J. M., and S. San. 2004. Vegetation associations of a serpentine area: Coyote Ridge, Santa Clara County, California. Report to the Marin Municipal Water District. California Native Plant Society, Sacramento. ———. 2005. Vegetation alliances of the San Dieguito River Park region, San Diego County, California. California Native Plant Society, Sacramento. Eviner, V. T. 2001. Linking plant community composition and ecosystem dynamics: Interactions of plant traits determine

the ecosystem effects of plant species and plant species mixtures. Ph.D. dissertation, University of California, Berkeley. ———. 2004. Plant traits that influence ecosystem processes vary independently among species. Ecology 85:2215–2229. Eviner, V. T., and F. S. Chapin. 2003a. Biogeochemical interactions and biodiversity. Pages 151–173 in J. Melillo, C. B. Field, and B. Moldan, editors. Interactions of the major biogeochemical cycles: Global change and human impacts. Island Press. ———. 2003b. Functional matrix: A conceptual framework for predicting multiple plant effects on ecosystem processes. Annual Review of Ecology and Systematics 34:455–485. ———. 2003c. Gopher-plant-fungal interactions affect establishment of an invasive grass. Ecology 84:120–128. ———. 2005. Selective gopher disturbance influences plant species effects on nitrogen cycling. Oikos 109:154–166. Eviner, V. T., F. S. Chapin, and C. E. Vaughn. 2000. Nutrient manipulations in terrestrial ecosystems. Pages 289–307 in O. E. Sala, R. B. Jackson, H. A. Mooney, and R. W. Howarth, editors. Methods in ecosystem science. Springer, New York. ———. 2006. Seasonal variations in species effects on N and P cycling. Ecology 87:974–986. Ewing, A. L., and J. W. Menke. 1983. Reproductive potential of Bromus mollis and Avena barbata under drought conditions. Madroño 30:159–167. Ewing, R. A., N. Tosta, R. Tuazon, L. Huntsinger, R. Morose, K. Nielson, R. Motroni, and S. Turan. 1988. California’s forests and rangelands: Growing conflict over changing uses. California Dept. of Forestry and Fire Protection, Forest and Rangeland Resources Assessment Program, Sacramento, CA. Facelli, J. M., and T. A. Pickett. 1991a. Plant litter: Its dynamics and effects of plant community and structure. Botanical Review 57:1–32. ———. 1991b. Plant litter: Light interception and effects on an old-field plant community. Ecology 71:1024–1031. Fan, A. M., C. C. Willhite, and S. A. Book. 1987. Evaluation of the nitrate drinking water standard with reference to infant methemoglobinemia and potential reproductive toxicity. Regulatory Toxicology and Pharmacology 7:135–148. Farrior, M. 2005. Breakthrough strategies for engaging the public: Emerging trends in communications and social science. Biodiversity Project. http://www.biodiversityproject.org/ bpemergingtrendspaper.doc (accessed June 8, 2007). Fehmi, J. S., and J. W. Bartolome. 2002. Species richness and California voles in an annual and a perennial grassland. Western North American Naturalist 62:73–81. ———. 2003. Impact of livestock grazing and burning on the spatial patterns of the grass Nassella pulchra (Poaceae). Madroño 50:8–14. Fehmi, J. S., K. J. Rice, and E. A. Laca. 2004. Radial dispersion of neighbors and the small-scale competitive impact of two annual grasses on a native perennial grass. Restoration Ecology 12:63–69. Felson, A. J., and S. T. A. Pickett. 2005. Designed experiments: New approaches to studying urban ecosystems. Frontiers in Ecology and the Environment 3:549–556. Fenn, M. E., J. S. Baron, E. B. Allen, H. M. Rueth, K. R. Nydick, L. Geiser, W. D. Bowman, J. O. Sickman, T. Meixner, D. W. Johnson, and P. Neitlich. 2003a. Ecological effects of nitrogen deposition in the western United States. Bioscience 53: 404–420.

REFERENCES

337

Fenn, M. E., R. Haeuber, G. S. Tonnesen, J. S. Baron, S. GrossmanClarke, D. Hope, D. A. Jaffe, S. Copeland, L. Geiser, H. M. Rueth, and J. O. Sickman. 2003b. Nitrogen emissions, deposition, and monitoring in the western United States. Bioscience 53:391–403. Feranec, R. S. 2003. Stable isotopes, hypsodonty, and the paleodiet of Hemiauchenia (Mammalia: Camelidae): A morphological specialization creating ecological generalization. Paleobiology 29:230–242. ———. 2004. Geographic variation in the diet of hypsodont herbivores from the Rancholabrean of Florida. Paleoclimatology, Paleogeography, Paleoecology 207:359–369. Feranec, R. S., and B. J. MacFadden. 2000. Evolution of the grazing niche in Pleistocene mammals from Florida: evidence from stable isotopes. Paleoclimatology, Paleogeography, Paleoecology 162:155–169. Ferraro, D. O., and M. Oesterheld. 2002. Effect of defoliation on grass growth. A quantitative review. Oikos 98:125–133. Fiedler, P. L., and R. A. Leidy. 1987. Plant communities of Ring Mountain Preserve, Marin County, California. Madroño 34:173–192. Field, C. B., F. S. Chapin, N. R. Chiariello, E. A. Holland, and H. A. Mooney. 1996. The Jasper Ridge CO2 experiment: Design and motivation. Pages 121–145 in G. W. Koch and H. A. Mooney, editors. Carbon dioxide and terrestrial ecosystems. Academic Press, San Diego, CA. Field, C. B., G. C. Daily, F. B. Davis, S. Gaines, P. A. Matson, J. Melack, and N. L. Miller. 1999. Confronting climate change in California: Ecological impacts on the Golden State. The Union of Concerned Scientists and The Ecological Society of America, Cambridge, MA. Field, C. B., C. P. Lund, N. R. Chiariello, and B. E. Mortimer. 1997. CO2 effects on the water budget of grassland microcosm communities. Global Change Biology 3:197–206. Field, C. B., and M. R. Raupach, editors. 2004. The global carbon cycle: Integrating humans, climate, and the natural world. Island Press, Washington. Fierer, N., A. S. Allen, J. P. Schimel, and P. A. Holden. 2003a. Controls on microbial CO2 production: A comparison of surface and subsurface soil horizons. Global Change Biology 9:1322–1332. Fierer, N., and J. P. Schimel. 2002. Effects of drying-rewetting frequency on soil carbon and nitrogen transformations. Soil Biology & Biochemistry 34:777–787. Fierer, N., J. P. Schimel, and P. A. Holden. 2003b. Influence of drying-rewetting frequency on soil bacterial community structure. Microbial Ecology 45:63–71. Firestone, M. K., L. J. Halverson, and D. J. Herman. 1995. Nutrient cycling in managed and non-managed oak woodland-grass ecosystems: Final report. Integrated Hardwood Range Management Program. University of California, Berkeley. Fitch, H. S., and J. R. Bentley. 1949. Use of California annual-plant forage by range rodents. Ecology 30:306–321. Flahault, C., and C. Schroeter. 1910. Nomenclature phytogéographique. Rapports et propositions. IIIe. Congrés International de Botanique, Brussels. Fleischner, T. L. 1994. Ecological costs of livestock grazing in western North America. Conservation Biology 8:629–644. Fleshman, E. 2000. Monitoring the response of butterfly communities to prescribed fire. Environmental Management 26: 685–695.

338

REFERENCES

Flessa, H., M. Potthoff, and N. Loftfield. 2002. Laboratory estimates of CO2 and N2O emissions following surface application of grass mulch: Importance of indigenous microflora of mulch. Soil Biology & Biochemistry 34:875–879. FNA. 2007. Flora of North America. Flora of North America Association. http://www.fna.org/FNA. Foin, T. C., and M. M. Hektner. 1986. Secondary succession and the fate of native species in a California coastal prairie community. Madroño 33:189–206. Follett, R. F., J. M. Kimble, and R. Lal. 2001. The potential of U.S. grazing lands to sequester carbon and mitigate the greenhouse effect. Lewis Publishers, New York. Ford, L. D., and G. Hayes. In press. Northern coastal scrub and coastal prairie. In M. G. Barbour, T. Keeler-Wolf, and A. Schoenherr, editors. Terrestrial vegetation of California. 3rd edition. University of California Press, Berkeley. Ford, L. D., and L. Huntsinger. 2006. Interview notes. Central Coast Rangeland Coalition Sustainability Indicators Project. Ford, R. I. 1985. Patterns of food production in North America. Pages 341–364 in R. I. Ford, editor. Prehistoric food production in North America. Anthropological Papers No. 74. Museum of Anthropology, University of Michigan, Ann Arbor. Forman, R. T. T., D. Sperling, J. A. Bissonette, A. A. Clevenger, C. D. Cutshall, V. H. Dale, L. Fahrig, R. France, C. R. Goldman, K. Heaune, J. A. Jones, F. J. Swanson, T. Turrentine, and T. C. Winter. 2003. Road ecology: Science and solutions. Island Press, Washington, D.C. Fossum, H. C. 1990. Effects of prescribed burning and grazing on Stipa pulchra (Hitch.) seeding emergence and survival. Master’s thesis, University of California, Davis. Foster, B. L. 1999. Establishment, competition, and the distibution of native grasses among Michigan old-fields. Journal of Ecology 87:476–489. Foster, B. L., and K. L. Gross. 1998. Species richness in a successional grassland: effects of nitrogen enrichment and plant litter. Ecology 79:2593–2603. Foster, G. M. 1944. A summary of Yuki culture. Anthropological Records 5(3): 155–244. Franck, V. M., B. A. Hungate, F. S. Chapin, and C. B. Field. 1997. Decomposition of litter produced under elevated CO2: Dependence on plant species and nutrient supply. Biogeochemistry 36:223–237. Frank, D. A., R. S. Inouye, N. Huntly, G. W. Minshall, and J. E. Anderson. 1994. The biogeochemistry of a north-temperate grassland with native ungulates: Nitrogen dynamics in Yellowstone National Park. Biogeochemistry 26:163–188. Frank, D. A., and S. J. McNaughton. 1993. Evidence for the promotion of aboveground grassland production by native large herbivores in Yellowstone National Park. Oecologia 96:157–161. Franklin, I., and R. C. Lewontin. 1970. Is the gene the unit of selection? Genetics 65:707–734. Franklin, J., A. D. Syphard, D. J. Mladenoff, H. S. He, D. K. Simons, R. P. Martin, D. Deutschman, and J. F. O’Leary. 2001. Simulating the effects of different fire regimes on plant functional groups in Southern California. Ecological Modelling 142:261–283. FRCC. 2005. Fire Regime Condition Class. http://www.frcc.gov. Fredeen, A. L., J. T. Randerson, N. M. Holbrook, and C. B. Field. 1997. Elevated atmospheric CO2 increases water availability in a water-limited grassland ecosystem. Journal of the American Water Resources Association 33:1033–1039.

Freitas, H., and H. A. Mooney. 1996. Effects of water stress and soil texture on the performance of two Bromus hordeaceus ecotypes from sandstone and serpentine soils. Acta Oecologia 17:307–317. Freilich, J. E., J. M. Emlen, J. J. Duda, D. C. Freeman, and P. J. Cafaro. 2003. Ecological effects of ranching: a six-point critique. BioScience 53:759 – 765. Fried, J., and L. Huntsinger. 1998. Managing for naturalness at Mt. Diablo State Park. Society and Natural Resources 11:505–516. Fried, J. S., M. S. Torn, and E. Mills. 2004. The impact of climate change on wildfire severity: A regional forecast for northern California. Climatic Change 64:169–191. Friends of Los Peñasquitos Canyon Preserve. 1997. Grasses of Los Peñasquitos Canyon. San Diego: Friends of Los Peñasquitos Canyon Preserve. Frost, W. E., and N. K. McDougald. 1989. Tree canopy effects on herbaceous production of annual rangeland during drought. Journal of Range Management 42:281–283. Frost, W. E., N. K. McDougald, and M. W. Demment. 1991. Blue oak canopy effect on seasonal forage production and quality. Pages 307–311 in R. Standiford, editor. Proceedings of the symposium on oak woodlands and hardwood rangeland management 1991. PSW-GTR-126. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Fuhlendorf, S. D., and D. M. Engle. 2001. Restoring heterogeneity on rangelands: Ecosystem management based on evolutionary grazing patterns. Bioscience 51:625–632. Furbush, P. 1953. Control of medusahead on California ranges. Journal of Forestry 51:118–121. Gabel, J. R. 1972. Further observations of theraphosid tarantula burrows. Pan-Pacific Entomologist 48:72–73. Gabet, E. J. 2000. Gopher bioturbation: Field evidence for nonlinear hillslope diffusion. Earth Surface Processes and Landforms 25:1419–1428. GAO. 2005. Livestock grazing: Federal expenditures and receipts vary, depending on the agency and the purpose of the fee charged. GAO-05-869. U.S. General Accounting Office, Washington, DC. Garcelon, D. K., S. J. Escover, and S. F. Timm. 1993. Feral pig control methods on Santa Catalina Island. Pages 44–45 in W. Tietje and R. H. Barrett, editors. The wild pig in California oak woodland: Ecology and economics, Workshop Proceedings, May 25 and 26, 1993. Integrated Hardwood and Range Management Program, University of California, Berkeley. Garcia, P., F. J. Vences, M. P. de la Vega, and R. W. Allard. 1989. Allelic and genotypic composition of ancestral Spanish and colonial Californian gene pools of Avena barbata: Evolutionary implications. Genetics 122:687–694. Garcia-Crespo, D. 1983. Multiple treatments to renovate depleted bunchgrass (Muhlenbergia rigens and Stipa pulchra) range sites in southern California. Ph.D. dissertation, Loma Linda University, Loma Linda, CA. Gardner, K. M., and R. G. Latta. 2006. Identifying loci under selection across contrasting environments in Avena barbata using quantitative trait locus mapping. Molecular Ecology 15:1321–1333. Gasith, A., and V. H. Resh. 1999. Streams in Mediterranean climate regions: Abiotic influences and biotic responses to predictable seasonal events. Annual Review of Ecology and Systematics 30:51–81.

Gayton, A. H. 1936. Estudillo among the Yokuts: 1819. Pages 67–85 in R. H. Lowie, editor. Essays in anthropology presented to A. L. Kroeber in celebration of his sixtieth birthday. University of California Press, Berkeley. ———. 1948. Yokuts and Western Mono ethnography I: Tulare Lake, Southern Valley, and Central Foothill Yokuts. University of Califorina Press, Berkeley. ———. 1976. Culture-environment integration: External references in Yokuts life. Pages 790–798 in L. J. Bean and T. C. Blackburn, editors. Native Californians: A theoretical retrospective. Ballena Press, Menlo Park, California. Gelbard, J. L. 2003. Grasslands at a crossroads: Protecting and enhancing resilience to climate change. Pages 13–40 in L. J. Hanson, J. L. Biringer, and J. R. Hoffman, editors. Buying time: A user’s manual for building resistance and resilience to climate change in natural systems. World Wildlife Fund, Washington, DC. Gelbard, J. L., and J. Belnap. 2003. Roads as conduits for exotic plant invasions in a semiarid landscape. Conservation Biology 17:420–432. Gelbard, J. L., and S. Harrison. 2003. Roadless habitats as refuges for native grasslands: interactions with soil, aspect, and grazing. Ecological Applications 13:404–415. ———. 2005. Invasibility of roadless grasslands: An experimental study of yellow starthistle. Ecological Applications 15: 1570–1580. Gentner, B. J., and J. A. Tanaka. 2002. Classifying federal public land grazing permittees. Journal of Range Management 55:2–11. George, M. R. 1992. Ecology and management of medusahead. Agricultural Experiment Station Series 32. Department of Agronomy and Range Science, University of California, Davis, CA. George, M. R., J. R. Brown, and W. J. Clawson. 1992. Application of non-equilibrium ecology to management of Mediterranean grasslands. Journal of Range Management 45: 436–440. George, M. R., R. S. Knight, P. B. Sands, and M. W. Demment. 1989a. Intensive grazing increases beef production. California Agriculture 43:16–19. George, M. R., R. E. Larsen, N. K. McDougald, K. W. Tate, J. D. Gerlach, and K. O. Fulgham. 2002. Influence of grazing on channel morphology of intermittent streams. Journal of Range Management 55:551–557. George, M. R., C. A. Raguse, W. J. Clawson, C. B. Wilson, R. T. Willoughby, N. K. McDougald, D. A. Duncan, and A. H. Murphy. 1988. Correlation of degree-days with annual herbage yields and livestock gains. Journal of Range Management 41: 193–196. George, M. R., W. A. Williams, N. K. McDougald, W. J. Clawson, and A. H. Murphy. 1989b. Predicting peak standing crop on annual range using weather variables. Journal of Range Management 42:508–513. Gerlach, J. D. 2004. The impacts of serial land-use changes and biological invasions on soil water resources in California, USA. Journal of Arid Enviroments 57:365–379. Gerlach, J. D., A. Dyer, and K. J. Rice. 1998. Grassland and foothill woodland ecosystems of the Central Valley. Fremontia 26: 39–43. Gerlach, J. D., and K. J. Rice. 2003. Testing life history correlates of invasiveness using congeneric plant species. Ecological Applications 13:167–179.

REFERENCES

339

Gifford, E. W. 1939. Karok field notes. Ethnological Document No. 174 in Bancroft Library, Department and Museum of Anthropology, University of California, Berkeley. Gill, A. M. 1977. Plant traits adaptive to fires in Mediterranean land ecosystems. Pages 17–36 in H. A. Mooney and C. E. Conrad, editors. Proceedings of the symposium on environmental consequences of fire and fuel management in Mediterranean ecosystems, August 1–5, 1977, Palo Alto, CA. General Technical Report WO-3. USDA Forest Service, Washington, DC. Gillespie, I. G., and E. B. Allen. 2004. Fire and competition in a southern California grassland:impacts on the rare forb Erodium macrophyllum. Journal of Applied Ecology 41: 643–652. Gillespie, L. J., and R. J. Soreng. 2005. A phylogenetic analysis of the bluegrass genus Poa based on cpDNA restriction site data. Systematic Botany 30:84–105. GISI. 2005. Global Invasive Species Initiative. Invasive plant documents & photographs. http://tncweeds.ucdavis.edu/esadocs. html (accessed March 26, 2007). Giusti, G. A., R. B. Standiford, D. D. McCreary, A. Merenlender, and T. Scott. 2004. Oak woodland conservation in California’s changing landscape. University of California Integrated Hardwood Range Management Program, Berkeley. Gleeson, S. K., and R. E. Good. 2003. Root allocation and multiple nutrient limitation in the New Jersey Pinelands. Ecology Letters 6:220–227. Gleeson, S., and D. Tilman. 1992. Plant allocation and the multiple limitation hypothesis. American Naturalist 139: 1322–1343. Goble, D. D., M. Scott, and F. W. Davis. 2005. The Endangered Species Act at thirty: Vol. 1. Renewing the conservation promise. Island Press, Washington, D.C. Gobster, P. H., and R. B. Hull, editors. 2000. Restoring nature: Perspectives from the social sciences and humanities. Island Press, Washington DC. Goerrissen, J. H. 2005. Grassland birds in California : An investigation into the influence of season, floristic composition, and artificial structures on avian community structure. Ph.D. dissertation, University of California, Davis. Gogan, P. J. P., and R. H. Barrett. 1995. Elk and deer diets in a coastal prairie-scrub mosaic, California. Journal of Range Management 48:327–335. Goldberg, D., and A. Novoplansky. 1997. On the relative importance of competition in unproductive environments. Journal of Ecology 85:409–418. Goldman, S., K. Jackson, and T. A. Bursztynsky. 1986. Erosion and sediment control handbook. McGraw Hill, New York. Golet, H. G., M. D. Robert, R. A. Luster, G. Werner, E. W. Larsen, R. Unger, and G. G. White. 2006. Assessing societal impacts when planning restoration of large alluvial rivers: A case study of the Sacramento River Project, California. Environmental Management 37(6):862–879. Golubiewski, N. E. 2006. Urbanization increases grassland carbon pools: Effects of landscaping in Colorado’s front range. Ecological Applications 16:555–571. Gong, P. 2004. Hardwood rangeland monitoring with aerial photography. Oaks ’n’ Folks 20. Goodell, K., D. R. Elam, J. D. Nason, and N. C. Ellstrand. 1997. Gene flow among small populations of a self-incompatible plant: An interaction between demography and genetics. American Journal of Botany 84:1362–1371.

340

REFERENCES

Goodrich, J., C. Lawson, and V. P. Lawson. 1980. Kashaya Pomo plants. Page 171 in American Indian Monograph Series, No. 2. American Indian Studies Center, Los Angeles. Gordon, A., and A. W. Sampson. 1939. Composition of common California foothill plants as a factor in range management. Agricultural Experiment Station Bulletin 627. University of California, Berkeley. Gordon, B. L. 1974. Monterey Bay area: Natural history and cultural imprints. Boxwood Press, Pacific Grove, California. Gordon, D. R., and K. J. Rice. 1992. Partitioning of space and water between two California annual grassland species. American Journal of Botany 79:967–976. Gordon, D. R., and K. J. Rice. 1993. Competitive effects of grassland annuals on soil water and blue oak (Quercus douglasii) seedlings. Ecology 74:68–82. ———. 2000. Competitive suppression of Quercus douglasii (Fagaceae) seedling emergence and growth. American Journal of Botany 87:986–994. Gordon, D. R., J. M. Welker, J. W. Menke, and K. J. Rice. 1989. Competition for soil water between annual plants and blue oak (Quercus douglasii) seedlings. Oecologia 79:533–541. Gordon, W. S., and R. B. Jackson. 2000. Nutrient concentrations in fine roots. Ecology 81:275–280. Gotelli, N. J., and R. K. Colwell. 2001. Quantifying biodiversity: Procedures and pitfalls in the measurement and comparison of species richness. Ecology Letters 4:379–391. Gould, F. W., and R. Moran. 1981. The grasses of Baja California, Mexico. Memoirs San Diego Society of Natural History 12:1–140. Gowans, K. D. 1967. Soil survey of Tehama County, California. USDA Soil Conservation Service and Forest Service and University of California Experiment Station, U.S. Government Printing Office, Washington, DC. GPWG. 2001. (Grass Phylogeny Working Group) Phylogeny and subfamilial classification of the grasses (Poaceae). Annals of the Missouri Botanic Garden 88:373–457. Graham, C. A. 1956. Some reactions of annual vegetation to fire on Sierra Nevada foothill rangeland. M. S.thesis, University of California, Berkeley. Graham, R. W. 2003. Pleistocene tapir from Hill Top Cave, Trigg County, Kentucky, and a review of Plio-Pleistocene tapirs of North America and their paleoecology. Pages 87–118 in B. W. Schubert, J. I. Mead, and R. W. Graham, editors. The ice cave faunas of North America. Indiana University Press, Bloomington. Gram, W. K., E. T. Borer, K. L. Cottingham, E. W. Seabloom, V. L. Boucher, L. Goldwasser, F. Micheli, B. E. Kendall, and R. S. Burton. 2004. Distribution of plants in a California serpentine grassland: Are rocky hummocks spatial refuges for native species? Plant Ecology 172:159–171. Grant, G., L. Engleback, and B. Nicholson. 2003. Green roofs: Their existing status and potential for conserving biodiversity in urban areas. English Nature Research Report 498. http://naturalengland.twoten.com/naturalenglandshop/docs/R498.pdf. Grayson, A. J. 1920. Game in the San Joaquin Valley in 1853. California Fish and Game 6:104–107. Grayson, D. 1991. Late Pleistocene mammalian extinctions in North America: Taxonomy, chronology, and explanations. Journal of World Prehistory 5:193–231. Green, J. L., J. Harte, and A. Ostling. 2003. Species richness, endemism and abundance patterns: Tests of two fractal models in a serpentine grassland. Ecology Letters 6:919–928.

Green, L. R., and C. A. Graham. 1957. Observations on growth and control of tarweed. Forest Research Notes 130. California Forest and Range Experiment Station, University of California, Berkeley. Green, L. R., K. A. Wagnon, et al. 1958. Diet and grazing habits of steers on foothill range fertilized with sulfur. Journal of Range Management 11:221–227. Greenlee, J. M., and J. H. Langenheim. 1980. The history of wildfires in the region of Monterey Bay. California State Department of Parks and Recreation, Sacramento. ———. 1990. Historic fire regimes and their relation to vegetation patterns in the Monterey Bay area of California. American Midland Naturalist 124:239–253. Gregory, S. D., E. K. Espeland, T. M. Carlsen, and E. K. Bissell. 2001. Demography and population biology of a rare tarplant, Blepharizonia plumosa (Asteraceae), a California summer annual forb. Madroño 48:272–285. Griesbach, J. A., R. Creamer, G. Lorens, B. W. Falk, C. O. Qualset, and L. E. Jackson. 1989. Barley yellow dwarf of California cereals. California Agriculture 43:23–24. Griesbach, J. A., B. W. Falk, and R. A. Valverde. 1990a. Incidence of barley yellow dwarf viruses in California cereals. Plant Disease 74:111–114. Griesbach, J. A., B. J. Steffenson, M. P. Brown, B. W. Falk, and R. K. Webster. 1990b. Infection of grasses by barley yellow dwarf viruses in California. Crop Science 30:1173–1177. Griffin, J. R. 1971. Oak regeneration in the upper Carmel Valley, California. Ecology 52:862–868. ———. 1976. Regeneration in Quercus lobata savannas, Santa Lucia Mountains, California. American Midland Naturalist 95:422–435. ———. 1977. Oak woodland. Pages 383–415 in M. G. Barbour and J. Major, editors. Terrestrial vegetation of California. John Wiley and Sons, New York. Griffin, J. R. 1980. Sprouting in fire-damaged valley oaks, Chews Ridge, California. Pages 216–219 in T. R. Plumb, editor. Proceedings of the symposium on the ecology, management, and utilization of California oaks 1980. General Technical Report PSW-GTR-44. USDA Forest Service, Pacific Southwest Forest and Range Experiment Station, Albany, CA. Griggs, F. T. 2000. Vina Plains Preserve: Eighteen years of adaptive management. Fremontia 27:48–51. Grime, J. P. 1973. Competitive exclusion in herbaceous vegetation. Nature 242:213–225. ———. 1979a. Evidence of the existence of three primary strategies in plants and its relevance to ecological and evolutionary theory. American Naturalist 111:1169–1194. ———. 1979b. Plant strategies and vegetation processes. John Wiley and Sons, New York. Grinnell, J. 1923. The burrowing rodents of California as agents of soil formation. Journal of Mammalogy 4(3):137–149. ———. 1932. Habitat relations of the giant kangaroo rat. Journal of Mammalogy 13:305–320. ———. 1933. Review of the recent mammal fauna of California. University of California Press Publications in Zoology 40: 71–234. ———. Up-hill planters. Condor 38:80–82. ———. 1938. California’s grizzly bears. Sierra Club Bulletin 23: 70–81. Grissino-Mayer, H., and T. Swetnam. 2000. Century-scale climate forcing of fire regimes in the American Southwest. Holocene 10:213–220.

Grogan, P., and F. S. Chapin. 2000. Nitrogen limitation of production in a Californian annual grassland: The contribution of arbuscular mycorrhizae. Biogeochemistry 49:37–51. Grossman, D. H., D. Faber-Langendoen, A. S. Weakley, M. Anderson, P. Bourgeron, R. Crawford, K. Gooding, S. Landaal, K. Metzler, K. D. Patterson, M. Pyne, M. Reid, and L. Sneddon. 1988. International classification of ecological communities: Terrestrial vegetation of the United States. The Nature Conservancy, Arlington, VA. Grover, J. P. 1997. Resource competition. Chapman & Hall, London. Grubb, P. J. 1977. The maintenence of species-richness in plant communities: the importance of the regeneration niche. Biological Reviews 52:107–145. Grundy, A. C., and W. Bond. 1998. Managing the weed seedbank. Nutrition and Food Science 98:80–83. Guggenberger, G., E. T. Elliott, S. D. Frey, J. Six, and K. Paustian. 1999. Microbial contributions to the aggregation of a cultivated grassland soil amended with starch. Soil Biology & Biochemistry 31:407–419. Gulmon, S. L. 1979. Competition and coexistence: Three annual grass species. American Midland Naturalist 101:403–416. ———. 1992. Patterns of seed germination in Californian serpentine grassland species. Oecologia 89:27–31. Gulmon, S. L., N. R. Chiariello, H. A. Mooney, and C. C. Chu. 1983. Phenology and resource use in three co-occurring grassland annuals. Oecologia 58:33–42. Gurevitch, J., P. S. Curtis, and M. H. Jones. 2001. Meta-analysis in ecology. Advances in Ecological Research 32:199–247. Gurevitch, J., and L. V. Hedges. 1999. Statistical issues in ecological meta-analysis. Ecology 80:1142–1149. Gusewell, S., W. Koerselman, and J. T. A. Verhoeven. 2003. Biomass N:P ratios as indicators of nutrient limitation for plant populations in wetlands. Ecological Applications 13:372– 384. Guthrie, R. D. 1990. Frozen fauna of the mammoth steppe. University of Chicago Press, Chicago, IL. Gutowski, W. J., Z. Pan, C. J. Anderson, R. W. Arritt, F. Otieno, E. S. Takle, J. H. Christensen, and O. B. Christensen. 2000. What RCM data area available for California impacts modeling? (How good is it?) Presented at California Energy Commission Workshop on Climate Change Scenarios for California. California Energy Commission, Sacramento, California. http://www.pircs. iastate.edu/people/Gutowski/Presentations/CalEnergy/. Hack-Ten-Broeke, M. J. D., and A. H. J. Van Der Putten. 1997. Nitrate leaching affected by management options with respect to urine-affected areas and groundwater levels for grazed grassland. Agriculture Ecosystems & Environment 66: 197–210. Haidinger, T. L., and J. E. Keeley. 1993. Role of high fire frequency in destruction of mixed chaparral. Madrono 40:141–147. Hall, E. R. 1927. An outbreak of house mice in Kern County, California. University of California Press Publications in Zoology 30:189–203. Hall, F. C. 2002. Photo point monitoring handbook: Part B— concepts and analysis. General Technical Report PNW-GTR-526. USDA Forest Service, Pacific Northwest Research Station, Portland, OR. Hall, L. M., M.R. George, D.D. McCreary, and T.E. Adams. 1992. Effects of cattle grazing on blue oak seedling damage and survival. Journal of Range Management 45:503–506.

REFERENCES

341

Halligan, J. P. 1973. Bare areas associated with shrub stands in grassland—case of Artemisia californica. Bioscience 23: 429–432. ———. 1976. Toxicity of Artemisia californica to 4 associated herb species. American Midland Naturalist 95:406–421. Halsey, R. W. 2004. In search of allelopathy: An eco-historical view of the investigation of chemical inhibition in California coastal sage scrub and chamise chaparral. Journal of the Torrey Botanical Society 131:343–367. Hamilton, J. G. 1997a. Changing perceptions of pre-European grasslands in California. Madroño 44:311–333. ———. 1997b. Environmental and biotic factors affecting the occurrence of the native bunchgrass Nassella pulchra in California. Ph.D. dissertation, University of California, Santa Barbara. Hamilton, J. G., J. R. Griffin, and M. R. Stromberg. 2002. Long-term population dynamics of native Nassella (Poaceae) bunchgrasses in unmanaged stands in central California. Madroño 49:274–284. Hamilton, J. G., C. Holzapfel, and B. E. Mahall. 1999. Coexistence and interference between a native perennial grass and nonnative annual grasses in California. Oecologia 121:518–526. Hamrick, J. L., and R. W. Allard. 1972. Microgeographical variation in allozyme frequencies in Avena barbata. Evolution 29:438–442. Hamrick, J. L., and M. J. W. Godt. 1989. Allozyme diversity in plant species. Pages 43–63 in A. H. D. Brown, M. T. Clegg, A. L. Kahler, and B. S. Weir, editors. Plant population genetics, breeding, and genetic resources. Sinauer Associates, Sunderland, MA. Hamrick, J. L., and L. R. Holden. 1979. Influence of microhabitat heterogeneity on gene frequency distribution and gametic phase disequilibrium in Avena barbata. Evolution 33: 521–533. Hamrick, J. L., and J. M. Lee. 1987. Effect of soil surface topography and litter cover on the germination, survival, and growth of musk thistle (Carduus nutans). American Journal of Botany 74:451–457. Hansell, M. H. 1993. The ecological impact of animal nests and burrows. Functional Ecology 7:5–12. Hansen, R. 1978. Shasta ground sloth food habits, Rampart Cave, Arizona. Paleobiology 4:302–319. ———. 1980. Late Pleistocene plant fragments in the dung of herbivores at Cowboy Cave. Pages 179–189 in J. Jennings, editor. Cowboy Cave. University of Utah Anthropological Papers No.104. University of Utah Press, Salt Lake City. Hansen, R. B. 1986. The effect of fire and fire frequency on grassland species composition in California’s Tulare Basin. Master’s thesis, California State University, Fresno. Harding, J., C. B. Mankinen, and M. H. Elliott. 1974. Genetics of Lupinus VII. Outcrossing, autofertility, and variability in natural populations of the group. Taxon 23:729–738. Harding, J., and K. Barnes. 1977. Genetics of Lupinus. X. Genetic variability, heterozygosity and outcrossing in colonial populations of Lupinus succulentus. Evolution 31:247–255. Harper, J., K. Tate, and M. George. 1998. Grazing effects on riparian areas. Rangeland Watershed Program Fact Sheet 14. University of California, Davis. http://danr.ucop.edu/uccelr/h14. htm (accessed June 8, 2007). Harpole, W. S. 2005. Limiting resources and patterns of species abundance and diversity. Ph.D. dissertation, University of Minnesota, St. Paul.

342

REFERENCES

———. 2006. Resource-ratio theory and the control of invasive plants. Plant and Soil 280:23–27. Harpole, W. S., and D. Tilman. 2007. Grassland species loss due to reduced niche dimension. Nature 446:791–793. Harrington, J. P. 1942. Culture element distributions: XIX. Central California coast. Anthropological Records 7(1):1–46. Harris, J., and G. T. Jefferson. 1985. Rancho La Brea: treasures of the tarpits. Los Angeles County Museum, Los Angeles. Harris, N. R., W. E. Frost, N. K. McDougald, M. R. George, and D. L. Nielsen. 2002. Long-term residual dry matter mapping for monitoring California hardwood rangelands. Pages 87–96 in R. B. Standiford, D. McCreary, and K. L. Purcell, editors. Proceedings of the fifth symposium on oak woodlands: Oaks in California’s changing landscape. October 22–25, San Diego, CA. General Technical Report PSW-GTR-184. USDA Forest Service, Albany, CA. Harrison, D. 1982. The visual quality of the natural prairie environment in prairie fiction. Pages 143-157 in B. Sadler and A. Carlson, editor. Environmental aesthetics: Essays in interpretation. Western Geographical Series 20. University of Victoria, Victoria, British Columbia. Harrison, S. 1999a. Local and regional diversity in a patchy landscape: Native, alien, and endemic herbs on serpentine. Ecology 80:70–80. ———. 1999b. Native and alien species diversity at the local and regional scales in a grazed California grassland. Oecologia 121:99–106. Harrison, S., K. F. Davies, H. D. Safford, and J. H. Viers. 2006a. Beta diversity and the scale-dependence of the productivitydiversity relationship: A test in the Californian serpentine flora. Journal of Ecology 94:110–117. Harrison, S., J. B. Grace, K. F. Davies, H. D. Safford, and J. H. Viers. 2006b. Invasion in a diversity hotspot: Exotic cover and native richness in the Californian serpentine flora. Ecology 87: 695–703. Harrison, S., C. Hohn, and S. Ratay. 2002. Distribution of exotic plants along roads in a peninsular nature reserve. Biological Invasions 4: 425–430. Harrison, S., and B. D. Inouye. 2002. High beta diversity in the flora of Californian serpentine “islands.” Biodiversity and Conservation 11:1869–1876. Harrison, S., B. D. Inouye, and H. D. Safford. 2003. Ecological heterogeneity in the effects of grazing and fire on grassland diversity. Conservation Biology 17:837–845. Harrison, S., K. Rice, and J. Maron. 2001. Habitat patchiness promotes invasion by alien grasses on serpentine soil. Biological Conservation 100:45–53. Harrison, S., H. D. Safford, J. B. Grace, J. H. Viers, and K. F. Davies. 2006c. Regional and local species richness in an insular environment: Serpentine plants in California. Ecological Monographs 76:41–56. Harrison, S., J. H. Viers, and J. F. Quinn. 2000. Climatic and spatial patterns of diversity in the serpentine plants of California. Diversity and Distributions 6:153–161. Hart, J. 1991. Farming on the edge: saving family farms in Marin County, California. University of California Press, Berkeley, CA. Hart, S. C., M. K. Firestone, E. A. Paul, and J. L. Smith. 1993. Flow and fate of soil nitrogen in an annual grassland and young mixed-conifer forest. Soil Biology & Biochemistry 25: 431–442. Harte, J., A. Kinzig, and J. L. Green. 1999. Self-similarity in the distribution and abundance of species. Science 284:334–336.

Hartley, W., and C. Slater. 1960. Studies on the origin, evolution, and distribution of the Gramineae. III. The tribes of the subfamily Eragrostideae. Australian Journal of Botany 8: 256–276. Hartmans, M. A., H. Zhang, and E. L. Michalson. 1997. Costs of yellow starthistle management in Idaho. Agricultural Experiment Station Bulletin 793. University of Idaho, Moscow. Harvey, L. E. 1989. Spatial and temporal dynamics of a blue oak woodland. Ph.D. dissertation, University of California, Santa Barbara. Hastings, M. S., and J. M. DiTomaso. 1996. Fire controls yellow star thistle in California grasslands. Restoration and Management Notes 14:124–128. Hatch, D. A., J. W. Bartolome, J. S. Fehmi, and D. S. Hillyard. 1999. Effects of burning and grazing on a coastal California grassland. Restoration Ecology 7:376–381. Hatch, D. A., J. W. Bartolome, and D. S. Hillyard. 1991. Testing a management strategy for restoration of California’s native grasslands. Pages 343–349 in J. Edelbrock and S. Carpenter, editors. Yosemite Centennial Symposium proceedings: Natural areas and Yosemite, prospects for the future, a global issues symposium joining the 17th Annual Natural Areas Conference with the Yosemite Centennial Celebration, October 13–20, 1990. National Park Service, Davis, CA. Haubensak, K. A. 2001. Invasion and impacts of nitrogen-fixing shrubs Genista monspessulana and Cytisus scoparius in grasslands of Washington and coastal California. Ph.D. dissertation, University of California, Berkeley. Haubensak, K. A., and C. M. D’Antonio. 2006. Restoration of a coastal California grassland after Invasion by nitrogen-fixing shrubs, french broom and scotch broom. Ecological Restoration 24:93–99. Haubensak, K. A., C. M. D’Antonio, and J. Alexander. 2004. Effects of nitrogen-fixing shrubs in Washington and coastal California. Weed Technology 18:1475–1479. Haubensak, K. A., and I. M. Parker. 2004. Soil changes accompanying invasion of the exotic shrub Cytisus scoparius in glacial outwash prairies of western Washington. Plant Ecology 175: 71–79. Havlin, J. L., J. D. Beaton, S. L. Tisdale, and W. L. Nelson. 1999. Soil fertility and fertilizers. Pearson Education, Singapore. Hawbecker, A. C. 1951. Small mammal relationships in an Ephedra community. Journal of Mammalogy 32:50–60. Hawkes, C. V., J. Belnap, C. M. D’Antonio, and M. K. Firestone. 2006. Arbuscular mycorrhizal assemblages in native plant roots change in the presence of exotic grasses. Plant and Soil 281:369–380. Hawkes, C. V., I. F. Wren, D. J. Herman, and M. K. Firestone. 2005. Plant invasion alters nitrogen cycling by modifying the soil nitrifying community. Ecology Letters 8:976–985. Hayes, G. F., and K. D. Holl. 2003a. Cattle grazing impacts on annual forbs and vegetation composition of mesic grasslands in California. Conservation Biology 17:1694–1702. Hayes, G. F., and K. D. Holl. 2003b. Site-specific responses of native and exotic species to disturbances in a mesic grassland community. Applied Vegetation Science 6:235–244. Hayhoe, K., D. Cayan, C. B. Field, P. C. Frumhoff, E. P. Maurer, N. L. Miller, S. C. Moser, S. H. Schneider, K. N. Cahill, E. E. Cleland, L. Dale, R. Drapek, R. M. Hanemann, L. S. Kalkstein, J. Lenihan, C. K. Lunch, R. P. Neilson, S. C. Sheridan, and J. H. Verville. 2004. Emissions pathways, climate change, and

impacts on California. Proceedings of the National Academy of Sciences USA 101:12422–12427. Haynes, G. 1991. Mammoths, mastodonts, and elephants: Biology, behavior, and the fossil record. Cambridge University Press, Cambridge, UK. HCPB. 2006. The status of rare, threatened, and endangered plants and animals of California, 2000–2004. Habitat Conservation Planning Branch, California Department of Fish and Game, State of California, Resources Agency, Sacramento. Heady, H. F. 1956. Changes in a California annual plant community induced by manipulation of natural mulch. Ecology 37:798–812. ———. 1958. Vegetational changes in the California annual type. Ecology 39:402–416. ———. 1965. The influence of mulch on herbage production in an annual grassland. Proceedings of the 9th International Grassland Congress 1:391–394. ———. 1972. Burning and the grasslands in California. Pages 97–107 in E. V. Komarek, editor. 12th annual Tall Timbers fire ecology conference 1972. Tall Timbers Research Station, Tallahassee, FL. ———. 1977. Valley grassland. Pages 491–514 in M. G. Barbour and J. Major, editors. Terrestrial vegetation of California. Special Publication Number 9. California Native Plant Society, Sacramento. ———. 1984. Concepts and principles underlying grazing systems. Developing strategies for rangeland management: A report of the Committee on developing strategies for rangeland management of the National Research Council National Academy of Sciences, Westview Press, Boulder, CO. ———. 1988. Valley grassland. Pages 491–514 in M. G. Barbour and J. Major, editors. Terrestrial Vegetation of California, 2nd ed. California Native Plant Society, Sacramento. Heady, H. F., J. W. Bartolome, M. D. Pitt, G. D. Savelle, and M. C. Stroud. 1992. California prairie. Pages 313–335 in R. T. Coupland, editor. Natural grasslands. Ecosystems of the World 8A. Elsevier, New York. Heady, H. F., and R. D. Child. 1994. Rangeland ecology and management. Westview Press, Boulder, CO. Heady, H. F., T. C. Foin, M. M. Hektner, D. W. Taylor, M. G. Barbour, and W. J. Barry. 1988. Coastal prairie and northern coastal scrub. Pages 733–762 in M. G. Barbour and J. Major, editors. Terrestrial vegetation of California, 2nd ed. California Native Plant Society, Sacramento. Heaton, E., and A. M. Merenlander. 2000. Modeling vineyard expansion, potential habitat fragmentation. California Agriculture 54:12–20. Hedges, L. V., J. Gurevitch, and P. S. Curtis. 1999. The metaanalysis of response ratios in experimental ecology. Ecology 80:1150–1156. Hedrick, D. W. 1948. The mulch layer of California annual ranges. Journal of Range Management 1:22–25. ———. 1951. Studies on the succession and manipulation of chamise brushlands in California. Ph.D. dissertation, Texas A&M College, College Station. Heffner, K. 1984. Following the smoke: Contemporary plant procurement by the Indians of northwest California. Unpublished document. Six Rivers National Forest, Eureka, CA. Heise, K. L., and A. M. Merenlender. 2002. Monitoring a halfcentury of change in a hardwood rangeland. Journal of Range Management 55:412–419.

REFERENCES

343

Heitschmidt, R. K., and J. W. Stuth. 1991. Grazing management: An ecological perspective. Timber Press, Portland, OR http//cnrit.tamu.edu/rlem/textbook/textbook-fr.html (accessed July 7, 2007). Heizer, C. B. 1990. Seed to civilization: The story of food. Harvard University Press, Cambridge, MA. Heizer, R. F. 1959. The California Indians. California Monthly 69(7):19–31. Hektner, M. M., and T. C. Foin. 1977. Vegetation analysis of a northern California coastal prairie: Sea Ranch, Sonoma County, California. Madroño 24:83–103. Hektner, M. M., R. W. Martin, and D. R. Davenport. 1983. The bald hills prairies of Redwood National Park. Pages 70–78 in C. Van Riper, L. D. Whittig, and M. L. Murphy, editors. Proceedings of the first biennial conference of research in California’s National Parks, 1983. Cooperative Parks Studies Unit, University of California. Hellmann, J. J. 2002. The effect of an environmental change on mobile butterfly larvae and the nutritional quality of their hosts. Journal of Animal Ecology 71:925–936. Hendry, G. W. 1931. The adobe brick as a historical source. Agricultural History 5:110–127. Henry, H. A. L., N. R. Chiariello, P. M. Vitousek, H. A. Mooney, and C. Field. 2006. Interactive effects of fire, elevated CO2, and N deposition on a California annual grassland. Ecosystems 9:1066–1075. Henry, H. A. L., E. E. Cleland, C. B. Field, and P. M. Vitousek. 2005. Interactive effects of elevated CO2, N deposition, and climate change on plant litter quality in a California annual grassland. Oecologia 142:465–473. Herman, D. H., K. K. Johnson, D. H. Jaeger III, E. Schwartz, and M. K. Firestone. 2006. Root influence on nitrogen mineralization and nitrification in Avena barbata rhizosphere soil. Soil Science Society of America Journal 70:1504 – 1511. Herman, D. J., L. J. Halverson, and M. K. Firestone. 2003. Nitrogen dynamics in an annual grassland: Oak canopy, climate, and microbial population effects. Ecological Applications 13: 593–604. Hervey, D. F. 1949. Reaction of a California annual-plant community to fire. Journal of Range Management 2:116–121. Hessl, A., and S. Spackman. 1995. Effects of fire on threatened and endangered plants: an annotated bibliography. Information and Technology Report 2. U.S. Dept. of the Interior, National Biological Service, Washington, DC. Hester, T. R. 1978. Salinan. Pages 500–504 in R. F. Heizer, editor. Handbook of North American Indians. Vol. 8: California. Smithsonian Institution, Washington, DC. Hibbard, K. A., D. S. Schimel, S. Archer, D. S. Ojima, and W. Parton. 2003. Grassland to woodland transitions: Integrating changes in landscape structure and biogeochemistry. Ecological Applications 13:911–926. Hibberd, M. 1985. A comparison of water stress and site conditions for Quercus agrifolia in relation to controls on distribution. Master’s thesis, San Diego State University, San Diego, CA. Hickman, J. C., editor. 1993. The Jepson manual: Higher plants of California. University of California Press, Berkeley. Higgins, P. A. T., R. B. Jackson, J. M. Des Rosiers, and C. B. Field. 2002. Root production and demography in a California annual grassland under elevated carbon dioxide. Global Change Biology 8:841–850.

344

REFERENCES

Hik, D. S., and R. L. Jefferies. 1990. Increases in the net aboveground primary production of a salt-marsh forage grass: A test of the predictions of the herbivore-optimization model. Journal of Ecology 78:180–195. Hilgard, E. W. 1895. Origin, value, and reclamation of alkali lands. In Yearbook of the U. S. Department of Agriculture. U.S. Department of Agriculture, Washington, DC. Hill, L. W., and R. M. Rice. 1963. Converting from brush to grass increases water yield in southern California. Journal of Range Management 16:300–305. Hill, M. A. 1979. TWINSPAN: A FORTRAN program for arranging multivariate data in an ordered two-way table by classification of the individuals and attributes. Section of Ecology and Systematics, Cornell University, Ithaca, NY. Hironaka, M. 1961. The relative rate of root development of cheatgrass and medusahead. Journal of Range Management 14:463–467. Hitchcock, A. S. 1935. Manual of grasses of the United States. Miscellaneous Publication 200. U. S. Department of Agriculture, Washington, DC. Hitchcock, A. S., and A. Chase. 1951. Manual of grasses of the United States. Miscellaneous Publication 200 (revised). U. S. Department of Agriculture, Washington, DC. Hobbs, N. T. 1996. Modification of ecosystems by ungulates. Journal of Wildlife Management 60:695–713. Hobbs, R. J. 1985. Harvester ant foraging and plant species distribution in annual grassland. Oecologia 67:519–523. Hobbs, R. J., S. L. Gulmon, V. J. Hobbs, and H. A. Mooney. 1988. Effects of fertilizer addition and subsequent gopher disturbance on a serpentine annual grassland community. Oecologia 75:291–295. Hobbs, R. J., and V. J. Hobbs. 1987. Gophers and grassland: A model of vegetation response to patchy soil disturbance. Vegetatio 69: 141–146. Hobbs, R. J., and L. F. Huenneke. 1992. Disturbance, diversity, and invasion: Implications for conservation. Conservation Biology 6:324–337. Hobbs, R. J., and H. E. Humphries. 1995. An integrated approach to the ecology and management of plant invasions. Conservation Biology 9:761–770. Hobbs, R. J., and H. A. Mooney. 1985. Community and population dynamics of serpentine grassland annuals in relation to gopher disturbance. Oecologia 67:342–351. ———. 1986. Community changes following shrub invasion of grassland. Oecologia 70:508–513. ———. 1991. Effects of rainfall variability and gopher disturbance on serpentine annual grassland dynamics. Ecology 72: 59–68. ———. 1995. Spatial and temporal variability in California annual grassland: Results from a long-term study. Journal of Vegetation Science 6:43–56. Hoglund, O. K., H. W. Miller, and A. L. Hafenrichter. 1952. Application of fertilizers to aid conservation on annual forage range. Journal of Range Management 5:55–61. Hole, F. D. 1981. Effects of animals on soil. Geoderma 25:75–112. Holland, E. A., W. J. Parton, J. K. Detling, and D. L. Coppock. 1992. Physiological responses of plant populations to herbivory and their consequences for ecosystem nutrient flow. The American Naturalist 140:685–706. Holland, R. F. 1986. Preliminary description of the terrestrial natural communities of California. California Department of Fish and Game, Natural Heritage Division, Sacramento.

———. 2000. The Flying M Ranch: Soils, plants, and vernal pools in eastern Merced County. Fremontia 27/28:28–32. Holland, R. F., and S. K. Jain. 1988. Vernal pools. Pages 515– 533 in M. G. Barbour, and J. Major, editors. Terrestrial vegetation of California. Wiley Interscience, John Wiley & Sons, New York. Holland, V. L. 1973. A study of soil and vegetation under Quercus douglasii compared to open grassland. Ph.D. dissertation, University of California, Berkeley. ———. 1980. Effect of blue oak on rangeland forage production in central California. Pages 314–318 in T. R. Plumb, editor. Proceedings of the symposium on the ecology, management, and utilization of California oaks 1980. General Technical Report PSW-GTR-44. USDA Forest Service Pacific Southwest Forest and Range Experiment Station, Albany, CA. Holland, V. L., and D. J. Keil. 1995. California vegetation. Kendall/Hunt Publishing Company, Dubuque, IA. Holland, V. L., and J. Morton. 1980. Effect of blue oak on nutritional quality of rangeland forage in central California. Pages 319–322 in T. R. Plumb, editor. Proceedings of the symposium on the ecology, management, and utilization of California oaks 1980. General Technical Report PSW-GTR-44. USDA Forest Service Pacific Southwest Forest and Range Experiment Station, Albany, CA. Holloway, J. M., R. A. Dahlgren, B. Hansen, and W. H. Casey. 1998. Contribution of bedrock nitrogen to high nitrate concentrations in stream water. Nature 395:785–788. Holmes, T. H., and K. J. Rice. 1996. Patterns of growth and soilwater utilization in some exotic annuals and native perennial bunchgrasses of California. Annals of Botany 78: 233–243. Holmgren, M., M. Scheffer, and M. A. Huston. 1997. The interplay of facilitation and competition in plant communities. Ecology 78:1966–1975. Holst, P. J., C. J. Allan, M. H. Campbell, and A. R. Gilmour. 2004. Grazing of pasture weeds by goats and sheep. 2. Scotch broom (Cytisus scoparius). Australian Journal of Experimental Agriculture 44:553–557. Holstein, G. 2000. Plant communities ecotonal to the baylands. Pages 49–67 in P. Olofson, editor. Bayland ecosystem, species and community profiles. San Francisco Bay Area Wetlands Ecosystem Goals Project, San Francisco, CA. ———. 2001. Pre-agricultural grassland in Central California. Madroño 48:253–264. Hone, J. 2002. Feral pigs in Namadgi Natioanal Park, Australia: Dynamics, impacts and management. Biological Conservation 105:231–242. Hook, P. B., I. C. Burke, and W. K. Lauenroth. 1991. Heterogeneity of soil and plant nitrogen and carbon associated with individual plants and openings in North American shortgrass steppe. Plant & Soil 138:247–256. Hooper, D. U., D. E. Bignell, V. K. Brown, L. Brussaard, J. M. Dangerfield, D. H. Wall, D. A. Wardle, D. C. Coleman, K. E. Giller, P. Lavelle, W. H. Van der Putten, P. C. De Ruiter, J. Rusek, W. L. Silver, J. M. Tiedje, and V. Wolters. 2000. Interactions between aboveground and belowground biodiversity in terrestrial ecosystems: Patterns, mechanisms, and feedbacks. Bioscience 50:1049–1061. Hooper, D. U., and J. S. Dukes. 2004. Overyielding among plant functional groups in a long-term experiment. Ecology Letters 7:95–105.

Hooper, D. U., and L. Johnson. 1999. Nitrogen limitation in dryland ecosystems: Responses to geographical and temporal variation in precipitation. Biogeochemistry 46:247–293. Hooper, D. U., and P. M. Vitousek. 1998. Effects of plant competition and diversity on nutrient cycling. Ecological Monographs 68:121–148. Hoopes, M. F., and L. M. Hall. 2002. Edaphic factors and competition affect pattern and formation and invasion in a California grassland. Ecological Applications 12:24–39. Hopkins, N. A. 1987. Mycorrhizae in a California serpentine grassland community. Canadian Journal of Botany/Revue canadienne de botanique 65:484–487. Hopkinson, P., J. S. Fehmi, and J. W. Bartolome. 1999. Summer burns reduce cover, but not spread, of barbed goatgrass in California grassland. Ecological Restoration 17:168–169. Horovitz, A., and J. Harding. 1972. Genetics of Lupinus. V. Intraspecific variability for reproductive traits in Lupinus nanus. Botanical Gazette 133:155–165. Horz, H. P., A. Barbrook, C. B. Field, and B. J. M. Bohannan. 2004. Ammonia-oxidizing bacteria respond to multifactorial global change. Proceedings of the National Academy of Sciences USA 101:15136–15141. Houghton, J. T., Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden, and D. Xiaosu, editors. 2001. IPCC 2001: The scientific basis. Contribution of Working Group I to the third assessment report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, UK. House, P. K., A. R. Ramelli, and C. T. Wrucke. 2001. Geologic map of Battle Mountain Quadrangle, Lander County, Nevada. Map 130. 1:24,000 scale. Nevada Bureau of Mines and Geology, Carson City. Howe, H. F. and J. S. Brown. 2000. Early effects of rodent granivory on experimental forb communities. Ecological Applications 10:917–924. ———. 2001. The ghost of granivory past. Ecology Letters 4: 371–378. Howe, H. F., J. S. Brown, and B. Zorn-Arnold. 2002. A rodent plague on prairie diversity. Ecology Letters 5:30–36. Howell, E., and F. Stearns. 1993. The preservation, management and restoration of Wisconsin plant communities: The influence of John Curtis and his students. Chapter 4 in J. S. Fralish, R. P. McIntosh, and O. L. Loucks, editors. John T. Curtis: Fifty years of Wisconsin plant ecology. Wisconsin Academy of Sciences, Arts and Letters, Madison. Huang, X. 1997. Watershed hydrology, soil, and biogeochemistry in an oak woodland annual grassland ecosystem in the Sierra Foothills, California. Ph.D. dissertation. University of California, Davis. Huddleston, R. T., and T. P. Young. 2004. Spacing and competition between planted grass plugs and preexisting perennial grasses in a restoration site in Oregon. Restoration Ecology 12: 546–551. ———. 2005. Weed control and soil amendment effects on restoration plantings in an Oregon grassland. Western North American Naturalist 65:507–515. Hudson, J. W. n. d. 20.012. Field notes Yosemite/Central Miwok. Grace Hudson Museum Archive Collection, Ukiah, California. ———. n.d. Field notes 20.017. Grace Hudson Museum Archive Collection, Ukiah, Unpublished. ———. n.d. Field notes 20.021. Grace Hudson Museum Archive Collection, Ukiah, Unpublished.

REFERENCES

345

———. n. d. 20.202. Field notes Yosemite/ Central Miwok. Grace Hudson Museum Archive Collection, Ukiah, California. Huenneke, L. F. 1989. Distribution and regional patterns of California grasslands. Pages 1–12 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Press, Dordrecht, The Netherlands. Huenneke, L. F., J. P. Anderson, M. Remmenga, and W. H. Schlesinger. 2002. Desertification alters patterns of aboveground net primary production in Chihuahuan ecosystems. Global Change Biology 8:247–264. Huenneke, L. F., S. P. Hamburg, R. Koide, H. A. Mooney, and P. M. Vitousek. 1990. Effects of soil resources on plant invasion and community structure in Californian serpentine grassland. Ecology 71:478–491. Huenneke, L. F., and H. A. Mooney. 1989a. Grassland structure and function: California annual grassland. Tasks for Vegetation Science 20. Kluwer Academic Press, Dordrecht, The Netherlands. ———. 1989b. The California annual grassland: An overview. Pages 213–218 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Press, Dordrecht, The Netherlands. Huffaker, C. B., and C. E. Kennett. 1959. A ten-year study of vegetation change associated with biological control of Klamath weed. Journal of Range Management 12:69–82. Hufford, K. M., and S. J. Mazer. 2003. Plant ecotypes: Genetic differentiation in the age of ecological restoration. Trends in Ecology & Evolution 18:147–155. Hujik, P. M. 1999. Low-tech seeding technique for perennial bunchgrasses yields promising results in the Lassen foothills (California). Ecological Restoration 17:75–76. Hull, J. C., and H. A. Mooney. 1990. Effects of nitrogen on photosynthesis and growth rates of four California grasses. Acta Oecologica 11:453–468. Hull, J. C., and C. H. Muller. 1977. The potential for dominance by Stipa pulchra in a California grassland. American Midland Naturalist 97:147–175. Humboldt County Community Development Services Department Planning Division. 1984. Humboldt County General Plan. Humboldt County Community Development Services Department. Eureka, CA. Hungate, B. A., J. Canadell, and F. S. Chapin. 1996. Plant species mediate changes in soil microbial N in response to elevated CO2. Ecology 77:2505–2515. Hungate, B. A., F. S. Chapin, H. Zhong, E. A. Holland, and C. B. Field. 1997a. Stimulation of grassland nitrogen cycling under carbon-dioxide enrichment. Oecologia 109:149–153. Hungate, B. A., E. A. Holland, R. B. Jackson, F. S. Chapin, H. A. Mooney, and C. B. Field. 1997b. The fate of carbon in grasslands under carbon dioxide enrichment. Nature 388:576–579. Hungate, B. A., C. H. Jaeger, G. Gamara, F. S. Chapin, and C. B. Field. 2000. Soil microbiota in two annual grasslands: Responses to elevated atmospheric CO2. Oecologia 124:589–598. Hungate, B. A., C. P. Lund, H. L. Pearson, and F. S. Chapin. 1997c. Elevated CO2 and nutrient addition alter soil N cycling and N trace gas fluxes with early season wet-up in a California annual grassland. Biogeochemistry 37:89–109. Hunt, J. 1992. Feeding ecology of valley pocket gophers (Thomomys bottae sanctidiegi) on a California coastal grassland. American Midland Naturalist 127:41–51.

346

REFERENCES

Hunter, J. E. 1986. Some responses of Sidalcea calycosa (Malvaceae) to fire. Madroño 33:305–307. Huntly, N., and R. Inouye. 1988. Pocket gophers in ecosystems: Patterns and mechanisms. Bioscience 38:786–793. Huntly, N., and O. J. Reichman. 1994. Effects of subterranean mammalian herbivores on vegetation. Journal of Mammalogy 75:852–859. Huntsinger, L. 1996. Grazing in a California silvopastoral system: Effects of defoliation season, intensity, and frequency on deerbrush, Ceanothus integerrimus Hook & Arn. Agroforestry Systems 34:67–82. Huntsinger, L., L. Buttolph, and P. Hopkinson. 1997. Ownership and management changes on California hardwood rangelands: 1985 to 1992. Journal of Range Management 50: 423–430. Huntsinger, L., J. S. Fried, and L. Buttolph. 1995. The elk, the cow, the ranger and the Rolex: Urban/wildland resource management conflict at Mt. Diablo State Park, California. Pages 67–74 in J. Endter-Wada and R. J. Lilieholm, editors. Conflicts in natural resource management: Integrating social and ecological concerns. Natural Resources and Environmental Issues 3. Utah State University, Logan. Huntsinger, L., and P. Hopkinson. 1996. Viewpoint: Sustaining rangeland landscapes: A social and ecological process. Journal of Range Management 49:167–173. Huntsinger, L., M. Johnson, M. Stafford, and J. Fried. 2007. A resurvey of oak woodland landowners: 1985, 1992, and 2004. Proceedings of the sixth symposium on oak woodlands: today's challenges, tomorrow's opportunities. 2006 October 9 – 12:Rohnert Park, CA. General Technical Report PSW-GTR. Pacific Southwest Research Station, Forest Service, U.S. Department of Agriculture, Albany, CA. Huntsinger, L., M. P. McClaran, A. Dennis, and J. W. Bartolome. 1996. Defoliation response and growth of Nassella pulchra (A. Hitchc.) Barkworth from serpentine and non-serpentine populations. Madroño 43:46–57. Huston, M. 1979. A general hypothesis of species diversity. American Naturalist 113:81–101. Huston, M. A. 1994. Biological diversity: The coexistence of species on changing landscapes. Cambridge University Press, Cambridge, UK. Huxman, T. E., and S. D. Smith. 2001. Photosynthesis in an invasive grass and native forb at elevated CO2 during an El Niño year in the Mojave Desert. Oecologia 128:193–201. IHRMP. 1998. 7th Annual Progress Report. Integrated Hardwood Range Management Program, Division of Agriculture and Natural Resources, Berkeley, CA. Imam, A. G., and R. W. Allard. 1965. Population studies in predominantly self-pollinated species. VI. Genetic variability between and within natural populations of wild oats from differing habitats in California. Genetics 51:49–62. Ingraham, N. L., and R. A. Matthews. 1995. The importance of fog-drip water to vegetation: Point Reyes Peninsula, California. Journal of Hydrology 164:269–285. Ingram, K., and J. Lewandrowski. 1999. Wildlife conservation and economic development in the west. Rural Development Perspectives 14(2):44–51. Irwin, M. E., and J. M. Thresh. 1990. Epidemiology of barley yellow dwarf: A study in ecological complexity. Annual Review of Phytopathology 28:393–424.

Ishikawa, C. M., and C. S. Bledsoe. 2000. Seasonal and diurnal patterns of soil water potential in the rhizosphere of blue oaks: Evidence for hydraulic lift. Oecologia 125:459–465. Ivans, C. Y., A. J. Leffler, U. Spaulding, J. M. Stark, R. M. Ryel, and M. M. Caldwell. 2003. Root responses and nitrogen acquisition by Artemisia tridentata and Agropyron desertorum following small summer rainfall events. Oecologia 134:317–324. Jackson, L. E. 1985. Ecological origins of California’s Mediterranean grasses. Journal of Biogeography 12:349–361. Jackson, L. E., F. J. Calderon, K. L. Steenwerth, K. M. Scow, and D. E. Rolston. 2003. Responses of soil microbial processes and community structure to tillage events and implications for soil quality. Geoderma 114:305–317. Jackson, L. E., and J. Roy. 1986. Growth patterns of mediterranean annual and perennial grasses under simulated rainfall regimes of southern France and California. Acta Oecologia/Oecologia Plantarum 7:191–212. Jackson, L. E., J. P. Schimel, and M. K. Firestone. 1989. Short-term partitioning of ammonium and nitrate between plants and microbes in an annual grassland. Soil Biology & Biochemistry 21:409–415. Jackson, L. E., R. B. Strauss, M. K. Firestone, and J. W. Bartolome. 1988. Plant and soil nitrogen dynamics in California annual grassland. Plant and Soil 110:9–18. Jackson, L. E., R. B. Strauss, M. K. Firestone, and J. W. Bartolome. 1990. Influence of tree canopies on grassland productivity and nitrogen dynamics in deciduous oak savanna. Agriculture Ecosystems & Environment 32:89–105. Jackson, R. B., J. L. Banner, E. G. Jobbagy, W. T. Pockman, and D. H. Wall. 2002. Ecosystem carbon loss with woody plant invasion of grasslands. Nature (London) 418:623–626. Jackson, R. B., Y. Luo, Z. G. Cardon, O. E. Sala, C. B. Field, and H. A. Mooney. 1995. Photosynthesis, growth, and density for the dominant species in a CO2-enriched grassland. Journal of Biogeography 22:1225–1229. Jackson, R. B., O. E. Sala, C. B. Field, and H. A. Mooney. 1994. CO2 alters water use, carbon gain, and yield in a natural grassland. Oecologia 98:257–262. Jackson, R. D., and B. Allen-Diaz. 2006. Spring-fed wetland and riparian plant communities respond differently to altered grazing intensity. Journal of Applied Ecology 43:485–498. Jackson, R. D., B. Allen-Diaz, L. G. Oates, and K. W. Tate. 2006a. Grazing removal increases spring-fed wetland nitrate concentrations in a Californian oak savanna. Ecosystems 9:1–15. ———. 2006b. Spring-water nitrate increased with removal of livestock grazing in a california oak savanna. Ecosystems 9: 254–267. Jackson, R. D., and J. W. Bartolome. 2002. A state-transition approach to understanding nonequilibrium plant community dynamics in Californian grasslands. Plant Ecology 162:49–65. Jacobs, B. F., J. D. Kingston, and L. L. Jacobs. 1999. The origin of grass-dominated ecosystems. Annals of the Missouri Botanical Garden 86:590–643. Jacobs, J. S., M. F. Carpinelli, and R. L. Sheley. 1999. Revegetating noxious weed-infested rangeland. Pages 133–141 in R. L. Sheley and J. K. Petroff, editors. Biology and management of noxious rangeland weeds. Oregon State University Press, Corvallis. Jacobsen, W. C. 1918. A history of ground squirrel control in California. Monthly Bulletin of the California State Commission of Horticulture 7:129–169.

Jacobson, S. K. 2003. Communication skills for conservation professionals. Island Press, Washington, DC. Jain, S. K. 1975. Genetic reserves. Pages 15–36 in O. H. Frankel and J. G. Hawkes, editors. Crop genetic resources for today and tomorrow. Cambridge University Press, Cambridge, UK. ———. 1978. Inheritance of phenotypic plasticity in soft chess, Bromus mollis L. (Gramineae). Experientia 34:835–836. ———. 1979. Adaptive strategies: polymorphism, plasticity, and homeostasis. Pages 160–187 in O.T. Solbrig, S. K. Jain, G. B. Johnson, and P. H. Raven, editors. Topics in plant population biology. Columbia University Press, New York. ———. 1981. Biosystmatic and evolutionary studies in genus Limnanthes. Pages 232–242 in S. K. Jain and P. B. Moyle, editors. Vernal pools and intermittent streams. 28. Institute of Ecology, University of California, Davis. ———. 1982. Variation and adaptive role of seed dormancy in some annual grassland species. Botanical Gazette 143:101–106. Jain, S. K., and R. W. Allard. 1960. Population studies in predominantly self-pollinated species. I. Evidence for heterozygote advantage in a closed population of barley. Proceedings of the National Academy of Sciences USA 46:1373–1377. ———. 1966. The effects of linkage, epistasis and inbreeding on population changes under selection. Genetics 53:633–659. Jain, S. K., D. R. Marshall, and K. Wu. 1970. Genetic variability in natural populations of softchess (Bromus mollis L.). Oecologia 130:555–562. Jain, S. K., and K. N. Rai. 1974. Population biology of Avena. IV. Polymorphism in small populations of Avena fatua. Theoretical and Applied Genetics 44:7–11. ———. 1980. Population biology of Avena. VIII. Colonization experiment as a test of the role of natural selection in population divergence. American Journal of Botany 67:1342–1346. Jameson, P., and H. J. Peeters. 1988. California mammals. University of California Press, Berkeley and Los Angeles. Janis, C. M., J. Damuth, and J. M. Theodora. 2002. The origins and evolution of the North American grassland biome: The story from the hoofed mammals. Paleoclimatology, Palaeogeography, Palaeoecology 177:183–198. Jasper, D. A., L. K. Abbott, and A. D. Robson. 1989. Soil disturbance reduces the infectivity of external hyphae of vesiculararbuscular mycorrhizal fungi. New Phytologist 112:93–99. Jefferies, R. L., and J. L. Maron. 1997. The embarrassment of riches: atmospheric deposition of nitrogen and community and ecosystem processes. Trends in Ecology and Evolution 12:74– 78. Jefferson, G. T. 1987. The Camp Cady local fauna: Paleoenvironment of the Lake Manix basin. Pages 3–35 in J. Reynolds, editor. Abstracts and proceedings of the 1987 Mojave Desert Quaternary research symposium. San Bernardino County Museum. ———. 1989. Late Cenozoic tapirs (Mammalia: Perissodactyla) of western North America. Contributions in Science 406. Los Angeles County Museum. Jefferson, G. T., and J. L. Goldin. 1989. Seasonal migration of Bison antiquus from Rancho La Brea, California. Quaternary Research 31:107–112. Jennings, C. W. 1977. Geologic map of California. U.S. Geological Survey, Menlo Park, CA. Jennings, M., O. Loucks, D. Glenn-Lewin, R. Peet, D. FaberLangendoen, D. Grossman, A. Damman, M. Barbour, R. Pfister, M. Walker, S. Talbot, J. Walker, G. Hartshorn, G. Waggoner, M. Abrams, D. Brown, A. Hill, D. Roberts, D. Tart, and M. Rej-

REFERENCES

347

manek. In press. Description, documentation, and evaluation of associations and alliances within the U.S. National Vegetation Classification. Ecological Monographs. Jepson, W. L. 1910. The Silva of California. University of California Memoirs Vol. 2. University of California Press, Berkeley, California. Jetter, K. M. 2005. Economic framework for decision making in biological control. Biological Control 35:348 –357. Jetter, K. M., J. M. DiTomaso, D. J. Drake, K. M. Klonsky, M. Pitcairn, and D. A. Sumner. 2003. Biological control of yellow starthistle: A cost-benefit analysis. Pages 225–241 in D.A. Sumner (ed). Exotic Pests and Diseases: Biology, Economics and Public Policy for Biosecurity. Iowa State University Press, Ames, IA. Jimerson, T. M., J. W. Menke, S. K. Carothers, M. P. Murray, V. VanSickle, and K. H. McClellan. 2000. Field guide to the rangeland vegetation types of the northern province, Klamath, Mendocino, Shasta-Trinity and Six Rivers National Forests. General Technical Report R5-ECOL-TP-014. USDA Forest Service, Pacific Southwest Region, Albany, CA. Joel, G., F. S. Chapin, N. R. Chiariello, S. S. Thayer, and C. B. Field. 2001. Species-specific responses of plant communities to altered carbon and nutrient availability. Global Change Biology 7:435–450. Joern, A. 1989. Insect herbivory in the transition to California annual grasslands: Did grasshoppers deliver the coup de grass? Pages 117–134 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic, Dordrecht, The Netherlands. Joffre, R. 1990. Plant and soil nitrogen dynamics in Mediterranean grasslands: A comparison of annual and perennial grasses. Oecologia 85:142–149. Johnson, B. E., and J. H. Cushman. 2007. Influence of a large herbivore reintroduction on plant invasions and community composition in a California grassland. Conservation Biology 21:515–526. Johnson, D. E. 1977. The late Quaternary climate of coastal California: Evidence of an ice-age refugium. Quaternary Research 8:154–179. Johnson, D. H. 1997. Effects of fire on bird populations in mixedgrass prairie. Pages 181–206 in F. L. Knoff and F. B. Samson, editors. Ecology and conservation of Great Plains vertebrates. Springer-Verlag, New York. Johnson, L. C., and J. R. Matchett. 2001. Fire and grazing regulate belowground processes in tallgrass prairie. Ecology 82: 3377–3389. Johnson, P. S., S. R. Shifley, and R. Rogers. 2002. The ecology and silviculture of oaks. CABI Publishing, New York. Johnson, W., C. M. McKell, R. A. Evans, and L. J. Burgy. 1959. Yield and quality of annual range forage following 2, 4-D application on blue oak trees. Journal of Range Management 12:18–20. Johnson, W. R., and J. Proctor. 1981. Growth of serpentine and non-serpentine races of Festuca rubra in solutions simulating the chemical conditions in a toxic serpentine soil. Journal of Ecology 69:855–869. ———. 1984. The effects of magnesium, nickel, calcium, and micronutrients on the root surface phosphatase activity of a serpentine and non-serpentine clone of Festuca rubra L. New Phytologist 96:95–101. JOI. 2006. Jepson online interchange for California floristics. http://ucjeps.berkeley.edu/interchange.html.

348

REFERENCES

Jones, A. 2000. Effects of cattle grazing on North American arid ecosystems: A quantitative review. Western North American Naturalist 60:155–164. Jones, M. B. 1963. Yield, percent nitrogen, and total nitrogen uptake of various California annual grassland species fertilized with increasing rates of nitrogen. Agronomy Journal:254–257. ———. 1964. Effect of applied sulfur on yield and sulfure uptake of various California dryland pasture species. Agronomy Journal 56:235–237. ———. 1974. Fertilisation of annual grasslands of California and Oregon. Pages 255–275 in D. A. Mays, editor. Forage Fertilization. American Society of Agronomy and Soil Science Society of America, Madison, WI. Jones, M. B., C. C. Delwiche, and W. A. Williams. 1977. Uptake and losses of 15N applied to annual grass and clover in lysimeters. Agronomy Journal 69:1019–1023. Jones, M. B., M. W. Demment, C. E. Vaughn, G. P. Deo, M. R. Dally, and D. M. Center. 1990. Effects of phosphorus and sulfur fertilization on subclover-grass pasture production as measured by lamb gain. Journal of Production Agriculture 3:534–539. Jones, M. B., P. W. Lawler, and J. E. Ruckman. 1970. Differences in annual clover response to phosphorus and sulfur. Agronomy Journal 62:439–442. Jones, M. B., and W. E. Martin. 1964. Sulfate-sulfur concentration as an indicator of sulfur status in various California dryland pasture species. Soil Science Society of America Proceedings 28:539–541. Jones, M. B., C. E. Vaughn, and W. A. Williams. 1995. Soilphosphorus requirements for maximum growth of Northern California subclover-annual grass pastures. Communications in Soil Science and Plant Analysis 26:197–207. Jones, M. B., and R. G. Woodmansee. 1979. Biogeochemical cycling in annual grassland ecosystems. Botanical Review 45:111–144. Jurjavcic, N. L., S. Harrison, and A. T. Wolf. 2002. Abiotic stress, competition, and the distribution of the native annual grass Vulpia microstachys in a mosaic environment. Oecologia 130:555–562. Kannenberg, L. W., and R. W. Allard. 1967. Population studies in predominantly self-pollinated species. VIII. Genetic variability in the Festuca microstachys complex. Evolution 21:227–240. Kartesz, J. T., and C. A. Meacham. 2006. Synthesis of the North American flora. Biota of North America Program (BONAP), North Carolina Botanic Garden, Chapel Hill. Kauffman, J. B., and W. C. Krueger. 1984. Livestock impacts on riparian ecosystems and streamside management implications: a review. Journal of Range Management 37:430–438. Kauffman, J. B., W. C. Krueger, and M. Vavra. 1983a. Impacts of cattle on streambanks in northeastern Oregon. Journal of Range Management 36:683–685. ———. 1983b. Effects of late season cattle grazing on riparian plant communities. Journal of Range Management 36: 685–691. Kay, B. L. 1987. Long-term effects of blue oak removal on forage production, forage quality, soil and oak regeneration. Pages 351–357 in T. R. Plumb and N. H. Pillsbury, editors. Proceedings of the symposium on multiple-use management of California’s hardwood resources, 1987. General Technical Report PSW-GTR-100. USDA Forest Service, Pacific Southwest Forest and Range Experiment Station, Albany, CA. Kay, B. L., and J. E. Street. 1961. Drilling wheatgrass into sprayed sagebrush in Northeastern California. Journal of Range Management 14:271–273.

Keane, R. M., and M. J. Crawley. 2002. Exotic plant invasions and the enemy release hypothesis. Trends in Ecology and Evolution 17:164–170. Keator, G. 1998. The life of an oak. Heyday Books, Berkeley, CA. Keeler, K. H. 1978. Intra-population differentiation in annual plants. II. Electrophoretic variation in Veronica peregrina. Evolution 32:638–645. Keeler-Wolf, T. 1987. An ecological survey of the proposed Crater Creek Research Natural Area, Klamath National Forest, Siskiyou County, California. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. ———. In press. History of vegetation classification and mapping in California. in M. G. Barbour, T. Keeler-Wolf, and A. Schoenherr, editors. Terrestrial vegetation of California. University of California Press, Berkeley. Keeler-Wolf, T., and D. Hickson. 2006. Vegetation classification of the Sacramento-San Joaquin River Delta. California Department of Fish and Game, Bay Delta Division, Sacramento. Keeley, J. E. 1977. Seed production, seed populations in soil, and seedling production after fire for two congeneric pairs of sprouting and non-sprouting chaparral shrubs. Ecology 58: 820–829. ———. 1990. The California valley grassland. Pages 2–23 in A. A. Schoenherr, editor. Endangered plant communities of southern California. Special Publication No. 3. Southern California Botanists, Fullerton, CA. ———. 1991. Seed germination and life history syndromes in the California chaparral. Botanical Review 57:81–116. ———. 1999. Chaparral. Pages 201–251 in M. G. Barbour and W. D. Billings, editors. North American terrestrial vegetation. Cambridge University Press, Cambridge, UK. ———. 2001. Fire and invasive species in Mediterranean-climate ecosystems of California. Pages 81–94 in K. E. M. Galley and T. P. Wilson, editors. Proceedings of the Invasive Plant Workshop: The role of fire in the control and spread of invasive species 2001. Miscellaneous Publication No. 11. Tall Timbers Research Station, Tallahassee, FL. ———. 2002. Native American impacts on fire regimes of the California coastal range. Journal of Biogeography 29:303 –320. ———. 2005. Fire history of the San Francisco East Bay region and implications for landscape patterns. International Journal of Wildland Fire 14:285–296. ———. 2006. Fire management impacts on invasive plants in the western United States. Conservation Biology 20:375–384. Keeley, J. E., C. J. Fotheringham, and M. Morais. 1999. Reexamining fire suppression impacts on brushland fire regimes. Science 284:1829–1832. Keeley, J. E., D. Lubin, and C. J. Fotheringham. 2003. Fire and grazing impacts on plant diversity and alien plant invasions in the southern Sierra Nevada. Ecological Applications 13:1355–1374. Keeley, J. E., and P. W. Rundel. 2005. Fire and the Miocene expansion of C4 grasslands. Ecology Letters 8:683–690. Keeley, J. E., and P. H. Zedler. 1998. Characterization and global distribution of vernal pools. Pages 1–14 in C. W. Witham, editor. Vernal pool ecosystems. California Native Plant Society, Sacramento. Kellogg, E. A. 2001. Evolutionary history of the grasses. Plant Physiology 125:1198–1205. Kelt, D. A., E. S. Konno, and J. A. Wilson. 2005. Habitat management for the endangered Stephens kangaroo rat: The effect

of mowing and grazing. Journal of Wildlife Management 69: 424–429. Kennard, B. 2005. Trend watching: Reframing the environment vs. economics debate. Greenbiz.com. April 2005. Kennedy, P. B. 1928. Goat grass or wild wheat (Aegilops triuncialis). Journal of the American Society of Agronomy 20:1292–1296. Kennedy, P. G., and J. Diaz. 2005. The role of seed dispersal and predation in forest encroachment into a coastal California grassland. Madroño 52:24–31. Kennedy, P. G., and W. P. Sousa. 2006. Forest encroachment into a Californian grassland: Examining the simultaneous effects of facilitation and competition on tree seedling recruitment. Oecologia 148:464–474. Kennett, G. A., J. R. Lacey, C. A. Butt, K. M. Olson-Rutz, and M. R. Haferkamp. 1992. Effects of defoliation, shading and competition on spotted knapweed and bluebunch wheatgrass. Journal of Range Management 45:363–369. Kephart, P. 2001. Resource management demonstration at Russian Ridge Preserve. Grasslands 9:1–8. Kerr, R. A. 2004. Climate change: Three degrees of consensus. Science 305:932–934. ———. 2005. How hot will the greenhouse world be? Science 309:100. Kerstyn, A., and P. Stiling. 1999. The effect of burn frequency on the density of some grasshoppers and leaf miners in a Florida Sandhill community. Florida Entomologist 82: 499–505. Ketellapper, H. 1963. Stomatal physiology. Annual Review of Plant Physiology and Plant Molecular Biology 14:249–270. Kie, J. G., and B. B. Boroski. 1996. Cattle distribution, habitats, and diets in the Sierra Nevada of California. Journal of Range Management 49:482– 488. Kieft, T. L., E. Soroker, and M. K. Firestone. 1987. Microbial biomass response to a rapid increase in water potential when dry soil is wetted. Soil Biology & Biochemistry 19:119–126. Kijkstra, F. A., S. E. Hobbie, and P. B. Reich. 2006. Soil processes affected by sixteen grassland species grown under different environmental conditions. Soil Science Society of America Journal 70:770– 777. Kimball, S., and P. M. Schiffman. 2003. Differing effects of cattle grazing on native and alien plants. Conservation Biology 17: 1681–1693. King, J. L., M. A. Simovich, and R. C. Brusca. 1996. Species richness, endemism and ecology of crustacean assemblages in northern California vernal pools. Hydrobiologia 328:85–116. Kinney, W. C. 1996. Conditions of rangelands before 1905. Chapter 3, pages 1–31 in Sierra Nevada Ecosystem Project: Final report to Congress, Vol. II. 2. Centers for Water and Wildland Resources, University of California, Davis. Kirkpatrick, M., and N. H. Barton. 1997. Evolution of a species’ range. American Naturalist 150:1–23. Kittelson, P. M., and J. L. Maron. 2001. Fine-scale geneticallybased differentiation of life-history traits in the perennial shrub Lupinus arboreus. Evolution 55:2429–2438. Kitzmiller, J. H. 1990. Managing genetic diversity in a tree improvement program. Forest Ecology and Management 35:131–149. Klein, A., and J. M. Evens. 2005. Vegetation alliances of western Riverside County, California. California Native Plant Society, Sacramento. Klinger, R., and I. Messer. 2001. The interaction of prescribed burning and site characteristics on the diversity and composition of

REFERENCES

349

a grassland community on Santa Cruz Island, California. Pages 66–80 in K. E. M. Galley and T. P. Wilson, editors. Proceedings of the Invasive Species Workshop: The role of fire in the control and spread of invasive species, held at Fire Conference 2000: The first national congress on fire ecology, prevention, and management. Miscellaneous Publication 11. Tall Timbers Research Station, Tallahassee, FL. Klinger, R., and J. Randall. 1997. The design and implementation of a non-native species control program on Santa Cruz Island. Proceedings, California Wood Science Society 49:29– 37. Knapp, A. K., J. M. Blair, J. M. Briggs, S. L. Collins, D. C. Hartnett, L. C. Johnson, and E. G. Towne. 1999. The keystone role of bison in north American tallgrass prairie: Bison increase habitat heterogeneity and alter a broad array of plant, community, and ecosystem processes. Bioscience 49:39–50. Knapp, A. K., J. M. Briggs, D. C. Hartnett, and S. L. Collins, editors. 1998. Grassland dynamics. Oxford University Press, New York, NY. Knapp, A. K., P. A. Fay, J. M. Blair, S. L. Collins, M. D. Smith, J. D. Carlisle, C. W. Harper, B. T. Danner, M. S. Lett, and J. K. McCarron. 2002. Rainfall variability, carbon cycling, and plant species diversity in a mesic grassland. Science 298:2202–2205. Knapp, E. E., and P. G. Connors. 1999. Genetic consequences of a single-founder population bottleneck in Trifolium amoenum. American Journal of Botany 86:124–130. Knapp, E. E., and K. J. Rice. 1996. Genetic structure and gene flow in Elymus glaucus (blue wildrye): Implications for native grassland restoration. Restoration Ecology 4:1–10. ———. 1997. Ecotypes of native species: How local is local in restoration plantings? Pages 51–55 in K. M. Wagner and P. Warner, editors. Proceedings, California Exotic Pest Plant Council Symposium. Vol. 3. California Exotic Pest Plant Council, Trabuco Canyon, CA. ———. 1998. Comparison of isozymes and quantitative traits for evaluating patterns of genetic variation in purple needlegrass. Conservation Biology 12:1031–1041. Knapp, P. A. 1996. Cheatgrass (Bromus tectorum) dominance in the Great Basin: history, persistence, and influences to human activities. Global Environmental Change 6:37–52(16). Kneitel, J. M. 1997. The effects of fire and pocket gopher (Thomomys bottae) disturbances on a California valley grassland. Master’s thesis, California State University, Northridge. Knight, T. M., and R. D. Holt. 2004. Fire generates spatial gradients in herbivory: An example from a Florida sandhill ecosystem. Ecology 86:587–593. Knopf, F. L. 1996. Prairie legacies: Birds. Pages 135–148 in F. B. Samson and F. L. Knopf, editors. Prairie Conservation: Preserving North America’s most endangered ecosystem. Island Press, Washington, DC. ———. 1998. Foods of mountain plovers wintering in California. Condor 100:382–384. Knops, J. M. H., J. R. Griffin, and A. C. Royalty. 1995. Introduced and native plants of the Hastings Reservation, Central Coastal California: A comparison. Biological Conservation 71:115–123. Knops, J. M. H., T. H. Nash, and W. H. Schlesinger. 1996. The influence of epiphytic lichens on the nutrient cycling of an oak woodland. Ecological Monographs 66:159–179. Knowles, C. J., C. J. Stoner, and S. P. Gieb. 1982. Selective use of black-tailed prairie dog towns by mountain plovers. Condor 84:71–74.

350

REFERENCES

Knowles, P. F. 1943. Improving an annual brome grass, Bromus mollis L. for range purposes. Journal of the American Society of Agronomy 35:584–594. Knutson, M. G., W. B. Richardson, D. M. Reineke, B. R. Gray, J. R. Parmelee, and S. E. Weick. 2004. Agricultural ponds support amphibian populations. Ecological Applications 14: 669–684. Koenig, W. D., J. M. H. Knops, and W. J. Carmen. 2002. Arboreal seed removal and insect damage in three California oaks. Pages 193–204 in R. B. Standiford, D. McCreary, and K. L. Purcell, editors. Proceedings of the fifth symposium on oak woodlands: Oaks in California’s changing landscape 2002. General Technical Report PSW-GTR-184. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Koenig, W. D., R. L. Mumme, W. J. Carmen, and M. T. Stanback. 1994. Acorn production by oaks in central coastal California: Variation within and among years. Ecology 75:99–109. Koerselman, W., and A. F. M. Meuleman. 1996. The vegetation N:P ratio: A new tool to detect the nature of nutrient limitation. Journal of Applied Ecology 33:1441–1450. Koide, R. T., L. F. Huenneke, S. P. Hamburg, and H. A. Mooney. 1988. Effects of applications of fungicide, phosphorus and nitrogen on the structure and productivity of an annual serpentine plant community. Functional Ecology 2: 335–344. Koide, R. T., L. F. Huenneke, and H. A. Mooney. 1987. Gopher mound soil reduces growth and affects ion uptake of two annual grassland species. Oecologia 72:284–290. Komarek, E. V., Sr. 1965. Fire ecology: grasslands and man. Pages 169–220 in E. V. Komarek, Sr., editor. Proceedings of the Fourth Annual Tall Timbers Fire Ecology Conference 1965. Tall Timbers Research Station, Tallahassee, Florida. Koontz, J. A., and P. S. Soltis. 2001. Polyploidy and segregation analysis in Delphinium gypsophilum (Ranunculaceae). Madroño 48:90–97. Kotanen, P. 1994. Effects of feral pigs on grasslands. Fremontia 22:14–17. ———. 1995. Responses of vegetation to a changing regime of disturbance: Effects of feral pigs in a Californian coastal prairie. Ecography 18:190–199. ———. 1996. Revegetation following soil disturbance in a California meadow: The role of propagule supply. Oecologia 108:652–662. ———. 1997a. Effects of experimental soil disturbances on revegetation by natives and exotics in Coastal California meadows. Journal of Applied Ecology 34:631–644. ———. 1997b. Effects of gap area and shape on recolonization by grassland plants with differing reproductive strategies. Canadian Journal of Botany 75:352–361. ———. 2004. Revegetation following soil disturbance and invasion in a Californian meadow: A 10-year history of recovery. Biological Invasions 6:245–254. Kowarik, I. 1995. Time lags in biological invasions with regard to the success and failure of alien species. Pages 15–38 in P. Pysˇek, K. Prach, M. Rajmánek, and M. Wade, editors. Plant invasions. SPB Academic Publishing, Amsterdam, The Netherlands. Krebs, C. J. 2001. Ecology. Benjamin Cummings, San Francisco. Kroeber, A. L. 1925. Handbook of the Indians of California. Government Printing Office, Washington, DC. 1976 reprint, Dover Publications, New York. ———. 1941. Culture element distributions XV: Salt, dogs, tobacco. Anthropological Records 6:1–20.

———. 1951. A Mohave historical epic. Archaeological Records 11:71–176. Kruckeberg, A. R. 1951. Intrapecific variability in the response of certain native plant species to serpentine soil. American Journal of Botany 38:408–419. ———. 1954. The ecology of serpentine soils: A symposium. III. Plant species in relation to serpentine soils. Ecology 35:267–274. ———. 1957. Variation in in fertility of hybrids between isolated populations of the serpentine species Streptanthus glandulosus Hook. Evolution 11:185–211. ———. 1958. The taxonomy of the species complex Streptanthus glandulosus Hook. Madroño 14:217–227. ———. 1967. Ecotypic responses to ultramafic soils by some plant species of northwestern United States. Brittonia 19:133–151. ———. 1984. California serpentines: flora, vegetation, geology, soils and management problems. University of California Press, Berkeley. Kuchler, A. W. 1964. Potential natural vegetation of the conterminous United States. Special Publication 36. American Geographic Society New York, NY. ———. 1977. The map of the natural vegetation of California. Pages 909–938 in M. G. Barbour and J. Major, editors. Terrestrial vegetation of California. Wiley Interscience, New York, NY. Kueppers, L. M., M. A. Snyder, L. C. Sloan, E. S. Zavaleta, and B. Fulfrost. 2005. Modeled regional climate change and California endemic oak ranges. Proceedings of the National Academy of Sciences USA 102:16281–16286. Kulmatiski, A., and K. H. Beard. 2006. Activated carbon as a restoration tool: Potential for control of invasive plants in abandoned agricultural fields. Restoration Ecology 14: 251–257. Kuminoff, N. V., and D. A. Sumner. 2001. Modeling farmland conversion with new GIS data. Agricultural Issues Center, University of California, Davis. http://aic.ucdavis.edu/research1/ FarmGIS1.pdf. Kurten, B. 1967. Pleistocene bears of North America. 2. Genus Arctodus, short-faced bears. Acta Zoologica Fennica 117. Kyser, G. B., and J. M. DiTomaso. 2002. Instability in a grassland community after the control of yellow starthistle (Centaurea solstitialis) with prescribed burning. Weed Science 50:648–657. Kyser, G. B., J. M. DiTomaso, M. P. Doran, S. B. Orloff, R. G. Wilson, D. L. Lancaster, D. F. Lile, and M. L. Porath. 2007. Control of medusahead (Taeniatherum caput-medusae) and other annual grasses with imazapic. Weed Technology 21:66– 75. Lacey, J. R., C. B. Marlow, and J. R. Lane. 1989. Influence of spotted knapweed (Centaurea maculosa) on surface runoff and sediment yield. Weed Technology 3:627–631. Lal, R. 2002. Soil carbon dynamics in cropland and rangeland. Environmental Pollution 116:353–362. Lamarque, J. F., J. T. Kiehl, G. P. Brasseur, T. Butler, P. CameronSmith, W. D. Collins, W. J. Collins, C. Granier, D. Hauglustaine, P. G. Hess, E. A. Holland, L. Horowitz, M. G. Lawrence, D. McKenna, P. Merilees, M. J. Prather, P. J. Rasch, D. Rotman, D. Shindell, and P. Thornton. 2005. Assessing future nitrogen deposition and carbon cycle feedback using a multimodel approach: Analysis of nitrogen deposition. Journal of Geophysical Research—Atmospheres 110:D19303. Lambert, W. D., and C. S. Holling. 1998. Causes of ecosystem transformation at the end of the Pleistocene: Evidence from mammal body-mass distributions. Ecosystems 1:157–175.

Landis, J., and M. Reilly. 2003. How we will grow: Baseline predictions of California’s urban footprint through the year 2100. IURD Working Paper Series. Paper WP-2003-04. Institute of Urban and Regional Development. University of California, Berkeley. http://repositories.cdlib.org/iurd/wps/WP-2003-04/. Landolt, E. 2001. Orchideen-Wiesen in Wollishofen (Zurich): Ein erstaunliches Relict aus dem Anfang des 20. Jarhrhunderts. [Orchid meadows in Wollishofen (Zurich): an astonishing relict from the beginning of the twentieth century]. Vierteljahrsschrift der Naturforschenden Gesellschaft in Zurich 146: 41–51. Langer, R. H. M. 1990. Pasture plants. In R. H. M. Langer, editor. Pastures: their ecology and management. Oxford University Press, Oxford. Langstroth, R. P. 1991. Fire and grazing ecology of Stipa pulchra grassland: A field study at Jepson Prairie. University of California, Davis. Lanini, T. W., R. F. Long, and J. Anderson. 1996. Pre-emergent herbicides have little effect on vigor of perennial grasses. California Agriculture 50:38–41. Lapham, M. H., and W. H. Heileman. 1901. Soil survey of the lower Salinas Valley, California. USDA Bureau of Soils, Baltimore. Larink, O. 1993. Bait lamina as a tool for testing feeding activity of animals in contaminated soils. Pages 339–345 in M. H. Donker, H. Eijsackers, and F. Heimbach, editors. Ecotoxicology of soil organisms. Lewis Publishers, Baton Rouge, LA. Larsen, R. E., W. C. Krueger, M. R. George, M. R. Barrington, J. C. Buckhouse, and D. E. Johnson. 1998. Viewpoint: Livestock influences on riparian zones and fish habitat: Literature classification. Journal of Range Management 51:661–664. Larson, S., K. Smith, D. Lewis, J. Harper, and M. George. 2005. Evaluation of California’s Rangeland Quality Education Program. Rangeland Ecology and Management 58(5):514–522. Latham, D. J., and J. A. Schlieter. 1989. Ignition probabilities of wildland fuels based on simulated lightning discharges. Research Paper INT-411. USDA Intermountain Research Station, Ogden, UT. Latimer, H. 1984. Eastern Fresno County. Fremontia 11:29–30. Latta, F. F. 1949. Hancock field notes. Frank Latta Papers. Yosemite Archives, National Park Service. ———. 1977. Handbook of the Yokuts Indians. Bear State Books, Santa Cruz, CA. Latta, R. G., J. L. MacKenzie, A. Vats, and D. J. Schoen. 2004. Divergence and variation of quantitative traits between allozyme genotypes of Avena barbata from contrasting habitats. Journal of Ecology 92:57–71. Laude, H. M. 1953. The nature of summer dormancy in perennial grasses. Botanical Gazette 114:284–292. Laude, H. M., and J. C. Meldeen. 1958. Characteristics related to summer growth of prairie three-awn grass. Journal of Range Management 11:298–300. Lawrence, G. E. 1966. Ecology of vertebrate animals in relation to chaparral fire in the Sierra Nevada foothills. Ecology 47: 278–291. Lawrence, W. E. 1945. Some ecotypic relations of Deschampsia caespitosa. American Journal of Botany 32:298–314. Lawson, A. C. and C. Palache. 1902. The Berkeley Hills: A detail of Coast Range ecology. University of California Bulletin of the Department of Geology 2:349–450. Lawson, F. R., and R. L. Piemeisel. 1943. The ecology of the principal summer weed hosts of the beet leafhopper in the San

REFERENCES

351

Joaquin Valley, California. USDA Technical Bulletin No. 848. Government Printing Office, Washington, DC. Ledig, F. T. 1992. Human impacts on genetic diversity in forest ecosystems. Oikos 63:87–108. Leger, E. A. 2004. Local adaptation, increased colonization ability, and resistance to natural enemies: The invasion of the California poppy (Eschscholzia californica) in Chile. Ph.D. dissertation, University of California, Davis. Lemaire, G., and D. Chapman. 1996. Tissue flows in grazed plant communities. Pages 3–36 in J. Hodgson and A. W. Illius, editors. The ecology and management of grazing systems. CAB International, New York. Lemons, J. 1996. Scientific uncertainty. Blackwell Science, Cambridge, MA. Lenihan, J. M., D. Bachelet, R. Drapek, and R. Neilson. 2006. The response of vegetation distribution, ecosystem productivity, and fire in California to future climate scenarios simulated by the MC1 Dynamic Vegetation Model. CEC-500-2005-191-SF. California Climate Change Center, Sacramento. Lenihan, J. M., R. Drapek, D. Bachelet, and R. P. Neilson. 2003. Climate change effects on vegetation distribution, carbon, and fire in California. Ecological Applications 13: 1667–1681. Leonard, O. A. 1956. Effect on blue oak (Quercus douglasii) of 2,4-D and 2,4,5-T concentrates applied to cut trunks. Journal of Range Management 9:15–19. Leong, J. M., and R. W. Thorp. 1999. Colour-coded sampling: The pan trap colour preferences of oligolectic and nonoligolectic bees associated with a vernal pool plant. Ecological Entomology 24:329–335. Leriche, H., X. Le Roux, F. Desnoyers, D. Benest, G. Simioni, and L. Abbadie. 2003. Grass response to clipping in an African savanna: Testing the grazing optimization hypothesis. Ecological Applications 13:1346–1354. Leung, L. R., L. O. Mearns, F. Giorgi, and R. L. Wilby. 2003. Regional climate research: Needs and opportunities. Bulletin of the American Meteorological Society 84:89–95. Levin, D. A., and H. W. Kerster. 1971. Neighborhood structure in plants under diverse reproductive methods. American Naturalist 105:345–354. Levine, J. M., P. B. Adler, and S. G. Yelenik. 2004. A meta-analysis of biotic resistance to exotic plant invasions. Ecology Letters 7:975–989. Levins, R. 1968. Evolution in changing environments. Princeton University Press, Princeton, NJ. Lewis, D., M. J. Singer, R. A. Dahlgren, and K. W. Tate. 2000. Hydrology in a California oak woodland watershed: A 17-year study. Journal of Hydrology 240:106–117. ———. 2006. Nitrate sediment fluxes from a California rangeland watershed. Journal of Environmental Quality 35(6):2202–2211. Lewis, H. T. 1973. Patterns of Indian burning in California: Ecology and ethnohistory. Ballena Press, Ramona, CA. 1993 reprint as pages 55–116 in T. C. Blackburn and K. Anderson, editors. Before the wilderness: Environmental management by Native Californians. Ballena Press, Menlo Park, CA. Lewis, H. T., and M. K. Anderson. 2002. Introduction. Pages 3–16 in O. C. Stewart, H. T. Lewis, and M. K. Anderson, editors. Forgotten fires: Native Americans and the transient wilderness. University of Oklahoma Press, Norman. Lewis, H., and C. Epling. 1959. Delphinium gypsophilum, a diploid species of hybrid origin. Evolution 13:511 – 525.

352

REFERENCES

Lewontin, R. C. 1965. Selection for colonizing ability. Pages 77–94 in H. G. Baker and G. L. Stebbins, editors. The genetics of colonizing species. Academic Press, New York. Lewontin, R. C., and J. L. Hubby. 1966. A molecular approach to the study of genic heterozygosity in natural populations. II. Amount of variation and degree of heterozygosity in natural populations of Drosophila pseudoobscura. Genetics 54: 595–609. Lidicker, W. Z. 1989. Impacts of non-domesticated vertebrates on California grasslands. Pages 135–150 in L. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Boston. Liffmann, R. H., L. Huntsinger, and L. C. Forero. 2000. To ranch or not to ranch: Home on the urban range? Journal of Range Management 53:362–370. Ling, S. 1998. A model of environmental compromise between regulators and landowners under the Endangered Species Act. Ph.D. dissertation. University of California, Davis, CA. Linhart, Y. B. 1988. Intrapopulation diferentiation in annual plants III. The contrasting effects of intra- and interspecific competition. Evolution 42(5):1047–1064. Linhart, Y. B., and M. C. Grant. 1996. Evolutionary significance of local genetic differentiation in plants. Annual Review of Ecology & Systematics 27:237–277. Linsbauer, K. 1917. Beitrage zur Kenntnis der Spaltoffnungsbewegung. Flora 9:100–143. Litwin, R. J., N. O. Frederiksen, D. P. Adam, V. A. S. Andrle, and T. P. Sheehan. 1993. Continental-marine correlation of Late Pleistocene climate change: Census of palynomorphs from core OL-92, Owens Lake, California. USGS Open-File Report 93-683. U.S. Geological Survey, Denver, CO. Liu, Z., and T. D. Herbert. 2004. High-latitude influence on the eastern equatorial Pacific climate in the early Pleistocene epoch. Nature 427:720–723. Lobo, A., K. Moloney, O. Chic, and N. R. Chiariello. 1998. Analysis of fine-scale spatial pattern of a grassland from remotely sensed imagery and field-collected data. Landscape Ecology 13: 111–131. Lombardo, K., J. S. Fehmi, K. J. Rice, and E. A. Laca. 2007. Nassella pulchra survival and water relations depend more on site productivity than on small scale disturbance. Restoration Ecology 15:177–178. Long, W. E. 1993. The Santa Rosa wild pigs eradication project (abstract). Page 40 in W. Tietje and R. H. Barrett, editors. The wild pig in California oak woodland: Ecology and economics, workshop proceedings, May 25 and 26, 1993. Integrated Hardwood and Range Management Program, Univerisity of California, Berkeley. Longcore, T., and C. Rich. 1998. 419 acres: UCLA’s natural history. The Urban Wildlands Group. http://www.urbanwildlands. org/Resources/419acres.pdf. Loucks, O. L., M. L. Plumb-Mentjes, and D. Rogers. 1985. Gap processes and large-scale disturbances in sand prairies. Pages 7183 in S. T. A. Pickett and P. S. White, editors. The ecology of natural disturbance and patch dynamics. Academic Press, New York. Love, R. M. 1944. Preliminary trials on the effect of management on the establishment of perennial grasses and legumes at Davis, California. Journal of the American Society of Agronomy 36:699–703. Love, R. M., and D. C. Sumner. 1952. Rose clover, a new winter legume. California Agricultural Experiment Station Circular 407.

LTA. 2004. Land Trust Alliance. Land Trust Census. http://www. lta.org/census/census_tables.htm. Lulow, M. E. 2004. Restoration in California’s inland grasslands: The role of priority effects and management strategies in establishing native communities and the ability of native grasses to resist invasion by non-native grasses. Ph.D. dissertation, University of California, Davis. Lulow, M. E., T. P. Young, J. L. Wirka, and J. H. Anderson. 2007. Variation in the initial success of seeded native bunchgrasses in the rangeland foothills of Yolo County, California. Ecological Restoration 25:20 – 28. Lund, C. P. 2001. Ecosystem carbon and water budgets under elevated atmospheric carbon dioxide concentration in two California grasslands. Ph.D. dissertation, Stanford University, Stanford. Luo, J., R. W. Tillman, and P. R. Ball. 2000. Nitrogen loss through denitrification in a soil under pasture in New Zealand. Soil Biology & Biochemistry 32:497–509. Lusk, W. C., M. B. Jones, P. J. Torell, and C. M. McKell. 1961. Medusahead palatability. Journal of Range Management 14: 248–251. Lym, R. G. 2005. Integration of biological control agents with other weed management techniques: successes from the leafy spurge (Euphorbia esula) IPM program. Biological Control 35:366 –375. Lym, R. G., and D. A. Tober. 1997. Competitive grasses for leafy spurge (Euphorbia esula) reduction. Weed Technology 11: 787–792. MacArthur, R. H., H. F. Recher, and M. Cody. 1966. On the relation between habitat selection and species diversity. American Naturalist 100:319–332. MacArthur, R. H., and E. O. Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, NJ. MacDonald, I. A. W., D. M. Graber, S. DeBenedetti, R. H. Groves, and E. R. Fuentes. 1988. Introduced species in nature reserves in Mediterranean-type climatic regions of the world. Biological Conservation 44:7–66. MacFadden, B. J., and T. E. Cerling. 1996. Mammalian herbivore communities, ancient feeding ecology, and carbon isotopes: A 10 million-year sequence from the Neogene of Florida. Journal of Vertebrate Paleontology 16:103–115. Mack, M. C., and C. M. D’Antonio. 1998. Impacts of biological invasions on disturbance regimes. Trends in Ecology & Evolution 13:195–198. Mack, R. N. 1989. Temperate grasslands vulnerable to plant invasions: characteristics and consequences. Pages 155–179 in J. A. Drake, H. A. Mooney, F. di Castri, R. H. Groves, F. J. Kruger, M. Rejmanek, and M. Williamson, editors. Biological invasions: A global perspective. John Wiley and Sons, New York. Mack, R. N., D. Simberloff, W. M. Lonsdale, H. Evans, M. Clout, and F. A. Bazzaz. 2000. Biotic invasions: Causes, epidemiology, global consequences, and control. Ecological Applications 10:689–710. MacNair, M. R., and M. Gardner. 1988. The evolution of edaphic endemics. Pages 157–171 in D. J. Howard and S. H. Berlocher, editors. Endless forms: Species and speciation. Oxford University Press, New York. Maddox, D. M., and A. Mayfield. 1985. Yellow starthistle infestations are on the increase. California Agriculture 39:10–12. Madhok, O. P., and R. B. Walker. 1969. Magnesium nutrition of two species of sunflower. Plant Physiology 44:1016–1022.

Maestas, J. D., R. L. Knight, and W. C. Gilgert. 2003. Biodiversity across a rural land-use gradient. Conservation Biology 17: 1425–1434. Magurran, A. E. 1988. Ecological diversity and its measurement. Princeton University Press, Princeton, NJ. Major, J. 1988. California climate in relation to vegetation. Pages 11–74 in M. G. Barbour, and J. Major, editors. Terrestrial vegetation of California, 2nd edition. John Wiley and Sons, New York. Major, J., C. M. McKell, and L. J. Berry. 1960. Improvement of medusahead infested rangeland. Leaflet 123, California Agricultural Experiment Station, Extension Service, University of California, Berkeley. Major, J., and W. T. Pyott. 1966. Buried, viable seeds in two California bunchgrass sites and their bearing on the definition of a flora. Vegetatio 13:253–282. Malmstrom, C. M. 1998. Barley yellow dwarf virus in native California grasses. Grasslands 8:6–10. Malmstrom, C. M., C. C. Hughes, L. A. Newton, and C. J. Stoner. 2005a. Virus infection in remnant native bunchgrasses from invaded California grasslands. New Phytologist 168:217–230. Malmstrom, C. M., A. J. McCullough, H. A. Johnson, L. A. Newton, and E. T. Borer. 2005b. Invasive annual grasses indirectly increase virus incidence in California native perennial bunchgrasses. Oecologia 145:153–164. Malmstrom, C. M., C. J. Stoner, S. Brandenburg, and L. A. Newton. 2006. Virus infection and grazing exert counteracting influences on survivorship of native bunchgrass seedlings competing with invasive exotics. Journal of Ecology 94: 264–275. Malo, J. E., and F. Suárez. 1995. Herbivorous mammals as seed dispersers in a Mediterranean dehesa. Oecologia 104:246– 255. Manzano, P., and J. E. Malo. 2006. Extreme long-distance dispersal via sheep. Frontiers in Ecology and the Environment 4:244–248. Marañon, T., and J. W. Bartolome. 1989. Seed and seedling populations in two contrasted communities: open grassland and oak (Quercus agrifolia) understory in California. Acta Oecologica, Oecol. Plant. 10:147–158. Marañon, T., and J. W. Bartolome. 1993. Reciprocal transplants of herbaceous communities between Quercus agrifolia woodland and adjacent grassland. Journal of Ecology 81:673–682. ———. 1994. Coast live oak (Quercus agrifolia) effects on grassland biomass and diversity. Madroño 41:39–52. Marañon, T., and L. V. Garcia. 1997. The relationship between diversity and productivity in plant communities: facts and artefacts. Journal of Ecology 85:95–96. Margolin, M. 1989. Life in a California mission: Monterey in 1786 by Lean Fra La Perouse. Heyday Books,. Maron, J. L., and P. G. Connors. 1996. A native nitrogen-fixing shrub facilitates weed invasion. Oecologia 105:302–312. Maron, J. L., and R. L. Jefferies. 1999. Bush lupine mortality, altered resource availability, and alternative vegetation states. Ecology 80:443–454. ———. 2001. Restoring enriched grasslands: Effects of mowing on species richness, productivity, and nitrogen retention. Ecological Applications 11:1088–1100. Marschner, H. 1996. Mineral nutrient acquisition in nonmycorrhizal and mycorrhizal plants. Phyton (Horn) 36:61–68.

REFERENCES

353

Marshall, D. R., and S. K. Jain. 1968. Phenotypic plasticity of Avena fatua and Avena barbata. American Midland Naturalist 102:457–467. Martens, D. A., W. Emmerich, J. E. T. McLain, and T. N. Johnsen. 2005. Atmospheric carbon mitigation potential of agricultural management in the southwestern USA. Soil & Tillage Research 83:95 –119. Martin, L. M., K. A. Moloney, and B. J. Wilsey. 2005. An assessment of grassland restoration success using species diversity components. Journal of Applied Ecology 42:327–336. Martin, M. M., and J. Harding. 1981. Evidence for the evolution of competition between two species of annual plants. Evolution 35:975–987. Martin, P. S. 1974. The discovery of America. Science 179: 969–974. Martin, W. E. 1958. Sulfur deficiency widespread in California soils. California Agriculture 12(November):10–12. Martin, W. E., and G. L. Jeffries. 1966. Relating ranch prices and grazing permit values to ranch productivity. Journal of Farm Economics 48:233 – 342. Martins, P. S., and S. K. Jain. 1979. The role of genetic variation in the colonizing ability of rose clover (Trifolium hirtum All.). American Naturalist 114:591–595. Martinsen, G. D., J. H. Cushman, and T. G. Whitham. 1990. Impact of pocket gopher disturbances on the diversity of a prairie community. Oecologia 83:132–138. Marty, J. T. 2005. Effects of cattle grazing on diversity in ephemeral wetlands. Conservation Biology 19:1626–1632. Marty, J. T., S. K. Collinge, and K. J. Rice. 2005. Responses of a remnant California native bunchgrass population to grazing, burning and climatic variation. Plant Ecology 181:101–112. Mason, J. A. 1912. The ethnology of the Salinan Indians. University of California Publications in American Archaeology and Ethnology 10:97–240. Mathewson, M. 1985. Threads of life: Cordage and other fibers of the California Tribes. Senior thesis, University of California, Santa Cruz. Mattson, D. J., and T. Merrill. 2002. Extirpations of grizzly bears in the contiguous United States, 1850–2000. Conservation Biology 16:1123–1136. Mayer, J. M., and I. L. Brisbin. 1991. Wild pigs in the United States: Their history, comparative morphology, and current status. University of Georgia Press, Athens. Mayer, M. S., P. S. Soltis, and D. E. Soltis. 1994. The evolution of the Streptanthus glandulosus complex (Cruciferae): Genetic divergence and gene flow in serpentine endemics. American Journal of Botany 81:1288–1299. Mayfield, T. J. 1993. Indian summer. Heyday Books/California Historical Society, Bekeley, CA. MBG. 2006 Grassbanking. The Malpais Borderlands Group: http://www.malpaiborderlandsgroup.org/gb.asp. McBride, J. R. 1974. Plant succession in the Berkeley Hills, California. Madroño 22:317–329. McBride, J. R., and H. L. Heady. 1968. Invasion of grassland by Baccharis pilularis D.C. Journal of Range Management 21: 106–108. McCarten, N. 1992. Community structure and habitat relations in a serpentine grassland in California. In A. J. M. Baker, J. Proctor, and R. D. Reeves, editors. The vegetation of ultramafic (serpentine) soils. Intercept Ltd., Andover, UK.

354

REFERENCES

McCarthy, J. J., O.F. Canziani, N. A. Leary, D. J. Dokken and K. S. White, editors. 2001. Intergovernmental Panel on Climate Change third assessment report: Climate change 2001: Impacts, adaptation and vulnerability. Cambridge University Press, Cambridge, UK. McClaran, M., and J. Bartolome. 1985. The importance of oak to ranchers in the California foothill woodland. Rangelands 7:158–161. ———. 1989a. Effect of Quercus douglasii (Fagaceae) on herbaceous understory along a rainfall gradient. Madroño 36:141–153. ———. 1989b. Fire-related recruitment in stagnant Quercus douglasii populations. Canadian Journal of Forest Research 19:580–585. McClintock, E., P. Reeberg, and W. Knight. 1990. A flora of the San Bruno Mountains. California Native Plant Society, Sacramento. McCreary, D. D. 2001. Regenerating rangeland oaks in California. University of California Division of Agriculture and Natural Resources Communication Services Publication 21601. University of California, Oakland. McCreary, D. D. 2004. Fire in California’s oak woodlands. Integrated Hardwood Range Management Program, University of California, Berkeley. McCreary, D. D., and M. R. George. 2005. Managed grazing and seedling shelters enhance oak regeneration on rangelands. California Agriculture 59:4 217–222. McCreary, D. D., and J. Tecklin. 1994. Grasshoppers continue to hamper oak restoration efforts. Oaks ’n’ Folks 9(2): 4–5. ———. 1997. Effects of seedling protectors and weed control on blue oak growth and survival. Pages 243–250 in N. H. Pillsbury, J. Verner, and W. D. Tietje, editors. Proceedings of a symposium on oak woodlands: Ecology, management, and urban interface issues. General Technical Report PSW-GTR-160. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. McCullough, D. R. 1969. The tule elk: Its history, behavior, and ecology. University of California Publications in Zoology 88. McDonald, J. N. 1981. North American bison. University of California Press, Los Angeles and Berkeley. McEvoy, P., C. Cox, and E. Coombs. 1991. Successful biological control of ragwort (Senecio jacobaea) by introduced insects in Oregon. Ecological Applications 1:430–442. McEvoy, P. B., and E. M. Coombs. 1999. Biological control of plant invaders: Regional patterns, field experiments, and structured population models. Ecological Applications 9:387– 401. McKell, C. M., J. Major, and E. R. Perrier. 1959. Annual-range fertilization in relation to soil moisture depletion. Journal of Range Management 12:189– 193. McInnes, P. F., R. J. Naiman, J. Pastor, and Y. Cohen. 1992. Effects of moose browsing on vegetation and litter of the boreal forest, Isle Royale, Michigan. Ecology 73:2059–2075. McIntyre, S., and S. Lavorel. 1994. Predicting richness of native, rare, and exotic plants in response to habitat and disturbance variables across a variegated landscape. Conservation Biology 8:521–531. McKane, R. B., L. C. Johnson, G. R. Shaver, K. J. Nadelhoffer, E. B. Rastetter, B. Fry, A. E. Giblin, K. Kielland, B. L. Kwiatkowski, J. A. Laundre, and G. Murray. 2002. Resource-based niches provide a basis for plant species diversity and dominance in arctic tundra. Nature 415:68–71.

McKay, J. K., C. E. Christian, S. Harrison, and K. J. Rice. 2005. “How local is local?” A review of practical and conceptual issues in the genetics of restoration. Restoration Ecology 13:432–440. McKay, J. K., and R. G. Latta. 2002. Adaptive population divergence: Markers, QTL and traits. Trends in Ecology and Evolution 17:285–291. McKell, C. C., A. M. Wilson, and B. L. Kay. 1962. Effective burning of rangelands infested with medusahead. Weeds 10: 125–131. McLauchlan, K. K., S. E. Hobbie, and W. M. Post. 2006. Conversion from agriculture to grassland builds soil organic matter on decadal timescales. Ecological Applications 16:143–153. McLaughlin, J. F., J. J. Hellmann, C. L. Boggs, and P. R. Ehrlich. 2002. Climate change hastens population extinctions. Proceedings of the National Academy of Sciences 99:6070–6074. McLaughlin, N. 2004. Increasing the tax incentives for conservation easement donations: A responsible approach. Ecological Law Quarterly 31:1–116. McMillin, J. H. 1956. The aboriginal human ecology of the mountain meadows area in southwestern Lassen County, California. MS thesis, Sacramento State University, Sacramento, CA. McNaughton, S. J. 1968. Structure and function in California grasslands. Ecology 49:962–972. ———. 1979. Grazing as an optimization process: Grass-ungulate relationships in the Serengeti. American Naturalist 113:691–703. McNaughton, S. J., M. Oesterheld, D. A. Frank, and K. J. Williams. 1989. Ecosystem level patterns of primary productivity and herbivory in terrestrial habitats. Nature 341:142–144. McPhee, J. 1994. Assembling California. Farrar, Straus, and Giroux, New York. McQueen, M., and E. McMahon. 2003. Land conservation financing. Island Press, Washington, DC. Mead, J., L. L. Coats, and B. W. Schubert. 2003. Late Pleistocene faunas from caves in the eastern Grand Canyon, Arizona. Pages 64–86 in B. W. Schubert, J. I. Mead, and R. W. Graham, editors. The Ice Age cave faunas of North America. Indiana University Press, Bloomington. Meimberg, H., J. I. Hammond, C. M. Jorgensen, T. W. Park, J. D. Gerlach, K. J. Rice, and J. K. McKay. 2006. Molecular evidence for an extreme genetic bottleneck during the introduction of barbed goatgrass, Aegilops triuncialis, to California. Biological Invasions 8:1355–1366. Menke, J. W. 1989. Management controls on productivity. Pages 173–200 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Dordrecht, The Netherlands. Menke, J. W. 1992. Grazing and fire management for native perennial grass restoration in California grasslands. Fremontia 20:22–25. Menke, J. W., and K. J. Rice. 1981. Effects of fire on seeds and nutrients in California annual grassland. Society of Range Management 34th Annual Meeting (abstracts) 1981. Mensing, S. A. 1993. The impact of European settlement on oak woodlands and fire: Pollen and charcoal evidence from the Transverse Ranges, California. Ph.D. dissertation, University of California, Berkeley. ———. 1998. 560 years of vegetation change in the region of Santa Barbara, California. Madroño 45:1–11. Mensing, S., and R. Byrne. 1998. Pre-mission invasion of Erodium cicutarium in California. Journal of Biogeography 25:757–762.

———. 1999. Invasion of Mediterranean weeds into California before 1769. Fremontia 27:6–9. Merenlender, A. M., K. L. Heise, J. W. Bartolome, and B. Allen-Diaz. 2001. Monitoring shows vegetation change at multiple scales. California Agriculture 55:42–46. Merenlender, A. M., L. Huntsinger, G. Guthey, and S. K. Fairfax. 2004. Land trusts and conservation easements: Who is conserving what for whom? Conservation Biology 18:65–75. Merriam, C. H. 1955. Studies of California Indians. University of California Press, Berkeley. Merriam, K. E., J. E. Keeley, and J. L. Beyers. 2006. Fuel breaks affect nonnative species abundance in Californian plant communities. Ecological Applications 16:515–527. Merrill, R. E. 1923. Plants used in basketry by the California Indians. University of California Publications in American Archaeology and Ethnology 20:215–242. Merzenich, J., W. A. Kurz, S. J. Beukema, M. Arbaugh, and S. Schilling. 1999. Long range modeling of stochastic disturbances and management treatments using VDDT and TELSA. Pages 349–355 in (ed.). Society of American Foresters Annual Meeting 1999. Society of American Foresters. Meyer, M. D. 1997. The effects of fire season and mulch reduction in a California grassland: a comparison of restoration strategies. M.S. thesis, California State University, Northridge. Meyer, M. D., and P. M. Schiffman. 1999. Fire season and mulch reduction in a California grassland: a comparison of restoration strategies. Madroño 46:25–37. Micallef, S. B. 1998. Grazing effects on the grassland vegetation of Mount Diablo State Park, California. M.S. thesis, San Francisco State University, San Francisco. Michener, W. K., and J. W. Brunt. 2000. Ecological data: Design, management and processing. Blackwell Science, Malden, MA. Milankovitch, M. 1930. Mathematishe Klimalehre und astronomische Theorie der Klimaschwankungen. Pages 1–176 in W. Koppen and R. Geiger, editor. Handbuch der Klimatologie. 1. A. Gebrüder Borntraeger, Berlin. Milchunas, D. G., and W. K. Lauenroth. 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecological Monographs 63:327–366. Miller, J. R. 2003. Integrated management of yellow starthistle (Centaurea solstitialis) at Fort Hunter Liggett, Monterey County, California. M.S. thesis, University of California, Davis. Miller, M. 2000. Fire autecology. Page 257 in J. Brown and J. K. Smith, eds. Wildland fire in ecosystems: Effects of fire on flora. General Technical Report RMRS-GTR-42, Vol. 2. Rocky Mountain Research Station, USDA, Ogden, UT. Miller, T. E. 1987. Effects of emergence time on survival and grown in an early old-field plant community. Oecologia 72: 272–278. Miller, T. E., J. H. Burns, P. Munguia, E. L. Walters, J. M. Kneitel, P. M. Richards, N. Mouquet, and H. L. Buckley. 2005. A critical review of twenty years’ use of the resource-ratio theory. American Naturalist 165:439–448. Miller, T. W., S. L. Shinn, R. H. Callihan, L. W. Lass, and D. C. Thill. 1998. Livestock trampling of grass seedlings for enhanced establishment on yellow starthistle-infested rangelands. Proceedings of the Western Society of Weed Science 51: 114–116. Miller, W. A., and L. Rasochova. 1997. Barley yellow dwarf viruses. Annual Review of Phytopathology 35:167–190.

REFERENCES

355

Minnich, R. A. 1983. Fire mosaics in southern California USA and northern Baja-California Mexico. Science 219:1287–1294. Minnich, R. A., and Y. H. Chou. 1997. Wildland fire patch dynamics in the chaparral of southern California and northern Baja California. International Journal of Wildland Fire 7:221–248. Minnich, R. A., and R. J. Dezzani. 1998. Historical decline of coastal sage scrub in the Riverside-Perris plain, California. Western Birds 29:366–391. Mitchell, C. E., and A. G. Power. 2003. Release of invasive plants from fungal and viral pathogens. Nature 421:625–627. Mitchelson, S. R. 1993. Factors affecting fruit set and survival of Zigadenus fremontii (Liliaceae) growing in a northern California annual grassland. California State University, Chico. Molina-Freaner, F., and S. K. Jain. 1992. Isozyme variation in Californian and Turkish populations of the colonizing species Trifolium hirtum. Journal of Heredity 83:423–430. ———. 1993. Inbreeding effects in a gynodioecious population of the colonizing species Trifolium hirtum All. Evolution 47:1472–1479. Moloney, K. A., and S. A. Levin. 1996. The effects of disturbance architecture on landscape-level population dynamics. Ecology 77:375–394. Moloney, K. A., S. A. Levin, N. R. Chiariello, and L. Buttel. 1992. Pattern and scale in a serpentine grassland. Theoretical Population Biology 41:257–276. Montalvo, A. M., L. J. Feist-Alvey, and C. E. Koehler. 2002. The effect of fire and cold treatments on seed germination of annual and perennial populations of Eschscholzia californica (Papaveraceae) in southern California. Madroño 49:207–227. Montalvo, A. M., P. A. McMillan, and E. B. Allen. 2002. The relative importance of seeding method, soil ripping, and soil variables on seeding success. Restoration Ecology 10:52–67. Montalvo, A. M., S. L. Williams, K. J. Rice, S. L. Buchmann, C. Cory, S. N. Handel, G. P. Nabhan, R. Primack, and R. H. Robichaux. 1997. Restoration biology: A population biology perspective. Restoration Ecology 5:277–290. Moody, A., and J. A. Jones. 2000. Soil response to canopy position and feral pig disturbance beneath Quercus agrifolia on Santa Cruz Island, California. Applied Soil Ecology 14:269–281. Moody, M. E., and R. N. Mack. 1988. Controlling the spread of plant invasions: The importance of nascent foci. Journal of Applied Ecology 25:1009–1021. Mooney, H. A. 1977. Southern coastal scrub. Pages 471–490 in M. G. Barbour and J. Major, editors. Terrestrial vegetation of California. John Wiley and Sons, New York. Mooney, H. A., and J. A. Drake, editors. 1986. Ecology of biological invasions of North America and Hawaii. Springer, New York. Mooney, H. A., and R. J. Hobbs. 2000. Invasive species in a changing world. Island Press, Covelo, CA. Mooney, H. A., R. J. Hobbs, J. Gorham, and K. Williams. 1986. Biomass accumulation and resource utilization in co-occurring grassland annuals. Oecologia 70:555–558. Moore, C., M. Bastian, and H. Hunt. 2001. Final Report: Long term vegetation and faunal succession in an artificial northern California vernal pool system. Final report E99TL01. California Department of Transportation, Environmental Impact Mitigation Branch, Sacramento. Moore, J. C., E. L. Berlow, D. C. Coleman, P. C. Ruiter, Q. Dong, A. Hastings, N. C. Johnson, K. S. McCann, K. Melville, P. J. Morin, K. Nadelhoffer, A. D. Rosemond, D. M. Post, J. L. Sabo,

356

REFERENCES

K. M. Scow, M. J. Vanni, and D. H. Wall. 2004. Detritus, trophic dynamics and biodiversity. Ecology Letters 7: 584–600. Moore, L. A., and C. B. Field. 2006. The effects of elevated atmospheric CO2 on the amount and depth distribution of plant water uptake in a California annual grassland. Global Change Biology 12:578–587. Moorhead, D. L., and T. Callaghan. 1994. Effects of increasing ultraviolet B radiation on decomposition and soil organic matter dynamics: a synthesis and modelling study. Biology and Fertility of Soils 18:19– 26. Morgan, J. A., D. E. Pataki, C. Korner, H. Clark, S. J. Del Grosso, J. M. Grunzweig, A. K. Knapp, A. R. Mosier, P. C. D. Newton, P. A. Niklaus, J. B. Nippert, R. S. Nowak, W. J. Parton, H. W. Polley, and M. R. Shaw. 2004. Water relations in grassland and desert ecosystems exposed to elevated atmospheric CO2. Oecologia (Berlin) 140:11–25. Morison, J. I. L. 1985. Sensitivity of stomata and water use efficiency to high CO2. Plant, Cell and Environment 8: 467–474. Moritz, M. A. 2003. Spatiotemporal analysis of controls on shrubland fire regimes: Age dependency and fire hazard. Ecology 84:351–361. Morrison, S. A., N. MacDonald, K. J. Walker, L. Lozier, and M. R. Shaw. 2007. Facing the dilemma at eradication’s end: uncertainty of absence and the Lazarus effect. Frontiers in Ecology 5:271– 276. Morton, J. B., S. R. Bentivenga, and J. D. Bever. 1995. Discovery, measurement and interpretation of diversity in arbuscular end fungi (Glomales, Zygomycetes). Canadian Journal of Botany Suppl. 1:S25–S32. Mousseau, T. A., and C. W. Fox. 1998. Maternal effects as adaptations. Oxford University Press, New York. Moyes, A. B., M. S. Witter, and J. A. Gamon. 2005. Restoration of native perennials in a California annual grassland after prescribed spring burning and solarization. Restoration Ecology 13:659–666. Mueller-Dombois, D., and H. Ellenberg. 1974. Aims and methods of vegetation ecology. John Wiley and Sons, New York. Muick, P. C. 1995. Effects of shade and simulated herbivory on seedling emergence, survival and growth of two California oaks: Coast live oak (Quercus agrifolia Nee.) and blue oak (Quercus douglasii Hook. and Arn.). Ph.D. dissertation, University of California, Berkeley. Muick, P. C., and J. W. Bartolome. 1987. An assessment of natural regeneration of oaks in California. California Dept. of Forestry and Fire Protection, Sacramento. Muir, J. 1894. The mountains of California. 1977 reprint, Ten Speed Press, Berkeley, CA. ———. 1974. Rambles of a botanist among the plants and climates of California. Dawson’s Book Shop, Los Angeles. ———. 1986. Muir among the animals. Sierra Club Books, San Francisco. Muller, C. H. 1966. Role of chemical inhibition (allelopathy) in vegetational composition. Bulletin of the Torrey Botanical Club 93:332–351. Munz, P. A. 1959. California plant communities. Pages 11–18 in P. A. Munz and D. D. Keck, editors. A California flora with supplement. University of California Press, Berkeley. Munz, P. A., and D. D. Keck. 1959. A California flora with supplement. University of California Press, Berkeley.

Murphy, A. H. 1970. Predicted forage yield based on fall precipitation in California annual grasslands. Journal of Range Management 23:363–365. Murphy, A. H., and L. J. Berry. 1973. Range pasture benefits through tree removal. California Agriculture 27:8–10. Murphy, A. H., and B. Crampton. 1964. Quality and yield of forage as affected by chemical removal of blue oak (Quercus douglasii). Journal of Range Management 17:142–144. Murphy, D. D., and P. R. Ehrlich. 1989. Conservation biology of California’s remnant native grasslands. Pages 201–211 in L. F. Huenneke and H. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Dordrecht, The Netherlands. Myers, N., and J. Kent. 1998. Perverse subsidies: tax $$ undercutting our economies and environments alike. International Institute for Sustainable Development, Winnipeg, Manitoba. Nagy, E. S. 1997. Frequency-dependent seed production and hybridization rates: Implications for gene flow between locally adapted plant populations. Evolution 51:703–714. Nagy, E. S., and K. J. Rice. 1997. Local adaptation in two subspecies of an annual plant: implications for migration and gene flow. Evolution 51:1079–1089. Nagy, L., and J. Proctor. 1997a. Soil Mg and Ni as causal factors of plant occurrence and distribution at the Meikle Kilrannoch ultramafic site in Scotland. New Phytologist 135:561–566. ———. 1997b. Plant growth and reproduction on a toxic alpine ultramafic soil: Adaptation to nutrient limitation. New Phytologist 137:267–274. Naiman, R. J., and H. Decamps. 1997. The ecology of interfaces: Riparian zones. Annual Review of Ecology and Systematics 28: 621–658. Nambiar, E. K. S. 1987. Do nutrients retranslocate from fine roots? Canadian Journal of Forest Research–Revue Canadienne de Recherche Forestière 17:913–918. NASS. 2005. U. S. Department of Agriculture National Livestock Statistics Service. 2005 Data and statistics: Livestock. http://www. nass.usda.gov/Data_and_Statistics/Quick_Stats/. National Academy of Sciences. 2001. Compensating for wetland losses under the Clean Water Act. National Academies Press, Washington, DC. NatureServe. 2006. NatureServe explorer: An online encylopedia of life, Version 4.4. NatureServe, Arlington, VA. http://www. natureserve.org/explorer. NatureServe, California Native Plant Society, and California Department of Fish and Game. 2003. Classification of the vegetation of Point Reyes National Seashore, Golden Gate National Recreation Area, Samuel P. Taylor, Mount Tamalpais, and Tomales State Parks, Marin, San Francisco, and San Mateo Counties, CA. California Department of Fish and Game, Wildlife and Habitat Data Analysis Branch, Sacramento. Naveh, Z. 1967. Mediterranean ecosystems and vegetation types in California and Israel. Ecology 48:445–459. Naylor, R. L. 2000. The economics of alien species invasions. Pages 241–260 in H. A. Mooney and R. J. Hobbs, editors. Invasive species in a changing world. Island Press, Washington, DC. NCDC. 2007. Fossil and surface pollen data. National Climatic Data Center, U.S. Department of Commerce. http://www. ncdc.noaa.gov/paleo/pollen.html (accessed March 20, 2007). Negrini, R. M., P. E. Wigand, S. Draucker, K. Gobalet, J. K. Gardner, M. Q. Sutton, and R. M. Yohe. 2006. The Rambla highstand

shoreline and the Holocene lake-level history of Tulare Lake, California, USA. Quaternary Science Reviews 25:1599–1618. Nelson, L. L., and E. B. Allen. 1993. Restoration of Stipa pulchra grasslands: Effects of mycorrhizae and competition from Avena barbata. Restoration Ecology 1:40–50. Newmark, M. H., and M. R. Newmark, editors. 1916. Sixty years in southern California, 1853–1913, containing the reminiscences of Harris Newmark. 1930 reprint, Zeitlin and Ver Brugge, Los Angeles, CA. NGN. 2005. National Grassbank Network. http://www.grassbank. net (accessed June 8, 2007). Nicolai, N. 1992. Distribution of Gambelia silus, Ammospermophilus nelsoni, Dipodomys ingens, and Vulpes macrotis mutica in the Southern San Joaquin Valley, California. Pages 159–168 in D. F. Williams, S. Byrne, and T. A. Rado, editors. Endangered and sensitive species of the San Joaquin Valley: Their biology, management and conservation. California Energy Commission, Sacramento. Nomland, G. A. 1935. Sinkyone notes. University of California Publications in American Archaeology and Ethnology 36:149–178. Nowak, R. S., D. S. Ellsworth, and S. D. Smith. 2004. Functional responses of plants to elevated atmospheric CO2—do photosynthetic and productivity data from FACE experiments support early predictions? New Phytologist 162:253–280. Noy-Meir, I. 1973. Desert systems: Environment and producers. Annual Review of Ecology and Systematics 4:25–51. ———. 1988. Dominant grasses replaced by ruderal forbs in a vole year in undergrazed Mediterranean grasslands in Israel. Journal of Biogeography 15:579–587. Noy-Meir, I., M. Gutman, and Y. Kaplan. 1989. Responses of Mediterranean grassland plants to grazing and protection. Journal of Ecology 77:290–310. NPS. 2003. Annual Performance Plan for FY 2004. U.S. Department of Interior, National Park Service, Pt. Reyes, CA. http://www.nps. gov/pore/home_mngmntdocs_archives_performanceplans.htm. NRPI. 1997. The Natural Resource Projects Inventory. Information Center for the Environment, Davis, CA. http://www. ice.ucdavis.edu/nrpi/. Null, J. 2006. Mid-winter dry spells. http://ggweather.com/ enso/winter_dry_spells.htm (accessed May 22, 2006). Nuzum, R. C. 2005. Using livestock grazing as a resource management tool in California. Contra Costa Water District, Concord, CA. http://www.ccwater.com/files/LivestockGrazing Final72005.pdf. Odion, D. C. 2000. Seed banks of long-unburned stands of maritime chaparral: Composition, germination behavior and survival with fire. Madroño 47:195–203. Odion, D. C., and F. W. Davis. 2000. Fire, soil heating, and the formation of vegetation patterns in chaparral. Ecological Monographs 70:149–169. Odion, D. C., and K. A. Haubensak. 2002. Response of French broom to fire. Pages 296–307 in N. G. Sugihara, M. E. Morales, and T. J. Morales, editors. Proceedings of the symposium Fire in California ecosystems: Integrating ecology, prevention and management 2002. Miscellaneous Publication No. 2. Association for Fire Ecology, Berkeley, CA. Odion, D. C., and C. M. Tyler. 2002. Are long fire-free periods needed to maintain the rare, fire-recruiting shrub Arctostaphylos morroensis (Ericaceae)? Conservation Ecology 6:4. Oechel, W. C., S. J. Hastings, G. L. Vourlitis, M. A. Jenkins, and C. L. Hinkson. 1995. Direct effects of elevated CO2 in chaparral and

REFERENCES

357

Mediterranean-type ecosystems. Pages 58–75 in J. L. Moreno and W. C. Oechel, editors. Global change and Mediterraneantype ecosystems. Ecological Studies 117. Springer-Verlag, New York. Oenema, O., G. L. Velthof, S. Yamulki, and S. C. Jarvis. 1997. Nitrous oxide emissions from grazed grassland. Soil Use & Management 13:288–295. Ogle, S. M., F. J. Breidt, and K. Paustian. 2005. Agricultural management impacts on soil organic carbon storage under moist and dry climatic conditions of temperate and tropical regions. Biogeochemistry 72:87–121. Okano, Y., K. R. Hristova, C. M. Leutenetter, L. E. Jackson, R. F. Denison, B. Gebreyesus, D. LeBauer, and K. M. Scow. 2004. Application of real-time PCR to study effects of ammonium on population size of ammonia-oxidizing bacteria in soil. Applied and Environmental Microbiology 20:1008–1016. Olff, R. T., and M. E. Ritchie. 1998. Effects of herbivores on grassland plant diversity. Trends in Ecology and Evolution 13: 261–265. Olson, B. E. 1999. Grazing and weeds. Pages 85–96 in R. L. Sheley and J. K. Petroff, editors. Biology and management of noxious rangeland weeds. Oregon State University Press, Corvallis. Opperman, J. J., and A. M. Merenlender. 2000. Deer herbivory as an ecological constraint to restoration of degraded riparian corridors. Restoration Ecology 8:41–47. Orr, P. C. 1956. Pleistocene man in Fishbone Cave, Pershing County, Nevada. Nevada State Museum Bulletin 2:1–20. Orr, R. T. 1940. The rabbits of California. Occasional Papers of the California Academy of Sciences 19:1–208. Osono, T. 2002. Phyllosphere fungi on leaf litter of Fagus crenata: occurrence, colonization and succession. Canadian Journal of Botany–Revue canadienne de botanique 80: 460–469. Oswald, J. W., and B. R. Houston. 1951. A new virus disease of cereals, transmissible by aphids. Plant Disease Reporter 35: 471–475. Oswald, J. W., and B. R. Houston. 1953. The yellow-dwarf virus disease of cereal crops. Phytopathology 43:128–136. Overton, J. M., and S. A. Levin. 2003. Components of spatial patterning in a serpentine grassland. Ecological Research 18: 405–421. Owen-Smith, N. 1987. Pleistocene extinctions: The pivotal role of megaherbivores. Paleobiology 13:351–362. Packard, S., and C. F. Mutel, editors. 1996. The tallgrass restoration handbook for prairies, savannas, and woodlands. Island Press, Covelo, CA. Paddison, J., editor. 1999. A world transformed: Firsthand accounts of California before the gold rush. Heyday Books, Berkeley, CA. Padgett, P. E., E. B. Allen, A. Bytnerowicz, and R. A. Minnich. 1999. Changes in soil inorganic nitrogen as related to atmospheric nitrogenous pollutants in southern California. Atmospheric Environment 33:769–781. Page, J. Ranchers form a “radical center” to protect wide-open spaces. Smithsonian, June 1997, 50–61. Paine, L. K., D. Undersander, and M. D. Casler. 1999. Pasture growth, production, and quality under rotational and continuous grazing management. Journal of Production Agriculture 12:569–577. Painter, E. L. 1995. Threats to the California flora: Ungulate grazers and browsers. Madroño 42:180–188.

358

REFERENCES

Painter, E. L., and A. J. Belsky. 1993. Application of herbivore optimization theory to rangelands of the western United States. Ecological Applications 3:2–9. Pan, Z., J. H. Christensen, R. W. Arritt, W. J. Gutowski, E. S. Takle, and F. Otieno. 2001. Evaluation of uncertainties in regional climate change simulations. Journal of Geophysical ResearchAtmospheres 106:17735–17751. Palmer, R. E. 1982. Ecological and evolutionary patterns in Holocarpha (Compositae, Madiinae). PhD dissertation. University of California, Davis. Parker, I. M., and S. H. Reichard. 1998. Critical issues in invasion biology for conservation science. Pages 283–305 in P. L. Fiedler and P. M. Kareiva, editors. Conservation biology. Chapman & Hall, New York. Parker, J. D., D. E. Burkepile, and M. E. Hay. 2006. Opposing effects of native and exotic herbivores on plant invasions. Science 311:1459–1461. Parker, S. 2006. Nitrogen cycling dynamics over space and time in annual and perennial grasslands in California. Doctoral dissertation. University of California, Santa Barbara. Parker, V. T., and V. R. Kelly. 1989. Seed banks in California chaparral and other Mediterranean climate shrublands. Pages 231–255 in M. A. Leck, V. T. Parker, and R. L. Simpson, editors. Ecology of soil seed banks. Academic Press, San Diego, CA. Parker, V. T., and R. Kersnar. 1989. Regeneration potential in French broom, Cytisus monspessulanus. Unpublished report to the Land Management Division of the Marin Municipal Water District. Parker, V. T., and C. H. Muller. 1982. Vegetational and environmental-changes beneath isolated live oak trees (Quercus agrifolia) in a California annual grassland. American Midland Naturalist 107:69–81. Parker, W. H. 1992. Focal point seed zones site-specific seed zone delineation using geographic information systems. Canadian Journal of Forest Research 22:267–271. Parkman, E. B. 2002. Evidence of Rancholabrean rubbing rocks on California’s north coast. Society for California Archaeology Newsletter 36:26–31. Parmenter, R. R., and J. A. MacMahon. 1983. Factors determining the abundance and distribution of rodents in a shrub-steppe ecosystem: the role of shrubs. Oecologia 59:145–156. Parsons, D. J. 1981. The historical role of fire in foothill communities of Sequoia National Park. Madroño 46:112–120. Parsons, D. J., and S. DeBenedetti. 1984. Effects of prescribed fire on annual grasslands in Sequoia National Park. Bulletin of the Ecological Society of America 65:258 (abstract). Parsons, D. J., and T. J. Stohlgren. 1989. Effects of varying fire regimes on annual grasslands in the southern Sierra Nevada of California. Madroño 36:154–168. Pavlik, B. M., and M. G. Barbour. 1986a. Demography of endemic psammophytes, Eureka Valley, California. I. Seed production, dispersal, and herbivory. State of California, Department of Fish and Game, Office of Endangered Species, Sacramento. ———. 1986b. Demography of endemic psammophytes, Eureka Valley, California. II. Survivorship, seed bank dynamics, and frequency of establishment. State of California, Department of Fish and Game, Office of Endangered Species, Sacramento. Peart, D. R. 1980. Seed dispersal by grasses in a California coastal grassland. American Zoologist 20:947.

Peart, D. R. 1989a. Species interactions in a succesional grassland. I. Seed rain and seedling recruitment. Journal of Ecology 77:236–251. Peart, D. R. 1989b. Species interactions in a successional grassland. II. Colonization of vegetated sites. Journal of Ecology 77:252–266. Peart, D. R. 1989c. Species interactions in a successional grassland. III. Effects of canopy gaps, gopher mounds and grazing on colonization. Journal of Ecology 77:267–289. Peart, D. R. 1994. Impact of feral pig activity on vegetation patterns associated with Quercus agrifolia on Santa Cruz Island, California. Ph.D. dissertation, Arizona State University, Tempe, AZ. Pendelton, D. F., J. W. Menke, W. A. Williams, and R. G. Woodmansee. 1983. Annual grassland ecosystem model. Hilgardia 5:1–44. Perevolotsky, A., and N. G. Seligman. 1998. Role of grazing in Mediterranean rangeland ecosystems: Inversion of a paradigm. Bioscience 48:1007–1017. Peri, D. W., S. M. Patterson, and J. L. Goodrich. 1982. Ethnobotanical mitigation Warm Springs Dam—Lake Sonoma, California. Elgar Hill, Environmental Analysis & Planning, Penngrove, CA. Perlot, J. N. 1985. Gold seeker: Adventures of a Belgian Argonaut during the gold rush years. Edited and translated by H. H. Bretnor. Yale University Press, New Haven, CT. Perry, D. A., M. P. Amaranthus, J. G. Borchers, S. L. Borchers, and R. E. Brainerd. 1989. Bootstrapping in ecosystems. Bioscience 39:230–237. Peters, H. A., E. E. Cleland, H. A. Mooney, and C. B. Field. 2006. Herbivore control of annual grassland composition in current and future environments. Ecology Letters 9:86–94. Peters, S. 2002. Mycorrhizal inoculum: Evaluating need and performance in revegetation and reclamation projects. Grasslands 12:1–10. Peterson, D. A. 2005. Facilitating use of prescribed fire to manage invasive exotic grasses on California’s annual grasslands: Reseeding with annual clovers and grasses to mitigate first-year forage loss. M.S. thesis, University of California, Davis. Peterson, P. M. 1993. Muhlenbergia. Pages 1272–1274 in J. C. Hickman, editor. The Jepson manual: Higher plants of California. University of California Press, Berkeley. ———. 2000. Systematics of the Muhlenbergiinae (Chloridoideae: Eragrostideae). Pages 195–212 in S. W. L. Jacobs and J. Everett, editors. Grasses: Systematics and evolution. CSIRO, Melbourne. ———. 2003. Poaceae (Gramineae). Nature encyclopedia of life sciences. John Wiley & Sons, Ltd, London. http://www.els.net/ [doi:10.1038/npg.els.0003689] ———. In press. Muhlenbergia. In B. G. Baldwin, S. Boyd, B. J. Ertter, D. J. Keil, R. W. Patterson, T. J. Rosatti, and D. H. Wilken, editors. The Jepson manual of higher plants of California, 2nd edition. University of California Press, Berkeley. Peterson, P. M., J. Cayouette, Y. S. N. Ferdinandez, B. Coulman, and R. E. Chapman. 2002. Recognition of Bromus richardsonii and B. ciliatus: Evidence from morphology, cytology, and DNA fingerprinting (Poaceae: Bromeae). Aliso 20:21–36. Peterson, P. M., J. T. Columbus, and S. J. Pennington. In press. Classification and biogeography of New World grasses: Chloridoideae. In J. T. Columbus, E. A. Friar, J. M. Porter, L. M. Prince, and M. G. Simpson, editors. Monocots: comparative biology and evolution. Rancho Santa Ana Botanic Garden, Claremont, CA.

Peterson, P. M., and J. J. Ortíz-Diaz. 1998. Allelic variation in the amphitropical disjunct Muhlenbergia torreyi (Poaceae: Muhlenbergiinae). Brittonia 50:381–391. Peterson, P. M., and J. J. Saarela. In press. Calamagrostis. In B. G. Baldwin, S. Boyd, B. J. Ertter, D. J. Keil, R. W. Patterson, T. J. Rosatti, and D. H. Wilken, editors. The Jepson manual of higher plants of California, 2nd edition. University of California Press, Berkeley. Phillips, F. M., M. G. Zreda, L. V. Benson, M. A. Plummer, D. Elmore, and P. Sharma. 1996. Chronology for fluctuations in late Pleistocene Sierra Nevada glaciers and lakes. Science 274: 749–751. Phillips, K. 2001. Restoration of a California ranch benefits both wildlife and landowners. U.S. Environmental Protection Agency Bi-Weekly Restoration Information Update. Pickart, A. 2006. Vegetation of diked herbaceous wetlands of Humboldt Bay National Wildlife Refuge: classification, description, and ecology. U.S. Fish and Wildlife Service. Humboldt Bay National Wildlife Refuge, Arcata, CA. Pickart, A. J., and M. G. Barbour. In press. Beach and Dune. In M. G. Barbour, T. Keeler-Wolf, and A. Schoenherr, editors. Terrestrial vegetation of California, 3rd Edition. University of California Press, Berkeley. Pickart, A. J., and J. O. Sawyer. 1998. Ecology and restoration of northern California coastal dunes. California Native Plant Society Press, Sacramento. Pickett, S. T. A. 1976. Succession: an evolutionary interpretation. American Naturalist 110:107–119. Pickett, S. T. A., and P. S. White, editors. 1985. The Ecology of natural disturbance and patch dynamics. Academic Press, Orlando, FL. Pickford, G. G. 1932. The influence of continued heavy grazing and of promiscuous burning on spring-fall ranges in Utah. Ecology 13:159–171. Piemeisel, R. L., and F. R. Lawson. 1937. Types of vegetation in the San Joaquin Valley of California and their relation to the beet leafhopper. USDA Technical Bulletin no. 557, Washington, DC. Pietola, L., R. Horn, and M. Yli-Halla. 2005. Effects of trampling by cattle on the hydraulic and mechanical properties of soil. Soil and Tillage Research 82:99–108. Pilling, A. R. 1978. Yurok. Pages 137–154 in R. F. Heizer, editor. Handbook of North American Indians Vol. 8. California. Smithsonian Institution, Washington, DC. Pitcairn, M. J., E. M. Coombs, and G. L. Piper. 2004. Yellow starthistle. Pages 421–435 in E. M. Coombs, J. K. Clark, G. L. Piper, and A. F. Cofrancesco, editors. Biological control of invasive plants in the United States. Oregon State University Press, Corvallis. Pitcairn, M. J., B. Villegas, D. M. Woods, G. Wilber, A. Duffy, and M. El-Bawdri. 2003. Statewide survey of yellow starthistle biological control agents. Pages 45–49 in D. M. Woods, editor. Biological Control Program Annual Summary, 2002. California Department of Food and Agriculture, Plant Health and Pest Prevention Services, Sacramento. Pitcairn, M. J., D. M. Woods, D. Joley, and V. Popescu. 2002. Seven-year population buildup and combined impact of biological control insects on yellow starthistle. Pages 57–59 in D. M. Woods, editor. Biological Control Program Annual Summary, 2001. California Department of Food and Agriculture, Plant Health and Pest Prevention Services, Sacramento.

REFERENCES

359

Pitcairn, M. J., J. A. Young, C. D. Clements, and J. Balciunas. 2002. Purple starthistle (Centaurea calcitrapa) seed germination. Weed Technology 16:452–456. Pitt, M. D., and H. F. Heady. 1978. Responses of annual vegetation to temperature and rainfall patterns in northern California. Ecology 59:336–350. ———. 1979. The effects of grazing intensity on annual vegetation. Journal of Range Management 32:109–114. Platenkamp, G. A. 1990. Phenotypic plasticity and genetic differentiation in the demography of the grass Anthoxanthum odoratum. Journal of Ecology 78:772–788. ———. 1991. Phenotypic plasticity and population differentiation in seeds and seedlings of the grass Anthoxanthum odoratum. Oecologia 88:515–520. Platenkamp, G. A., and R. G. Shaw. 1993. Environmental and genetic maternal effects on seed characters in Nemophila menziesii. Evolution 47:540–555. Platenkamp, G. A. J., and T. C. Foin. 1990. Ecological and evolutionary importance of neighbors in the grass Anthoxanthum odoratum. Oecologia 83:201–208. Platt, W. J. 1975. The colonization and formation of equilibrium plant species associations on badger disturbances in tall-grass prairie. Ecological Monographs 45:285–305. Pollak, O., and T. Kan. 1998. The use of prescribed fire to control invasive weeds at Jepson Prairie Preserve. Pages 241–249 in C. W. Witham editors. Ecology, conservation and management of vernal pool ecosystems: Proceedings from a 1996 conference. California Native Plant Society. Pollock, J. F. 2006. Lassen Foothills controlled burn study 1996–2003, statistical modeling and analysis; a report to The Nature Conservancy, San Francisco, California. University of California, Santa Cruz. Poorter, H., and M. L. Navas. 2003. Plant growth and competition at elevated CO2: on winners, losers and functional groups. New Phytologist 157:175–198. Potthoff, M., J. Dyckmans, H. Flessa, A. Muhs, F. Beese, and R. G. Joergensen. 2005a. Dynamics of maize (Zea mays L.) leaf straw mineralization as affected by the presence of soil and the availablility of nitrogen. Soil Biology & Biochemistry 37: 1259–1266. Potthoff, M., L. E. Jackson, K. L. Steenwerth, I. Ramirez, M. R. Stromberg, and D. E. Rolston. 2005b. Soil biological and chemical properties in restored perennial grassland in California. Restoration Ecology 13:61–73. Potthoff, M., K. L. Steenwerth, L. E. Jackson, R. E. Drenovsky, K. M. Scow, and R. G. Joergensen. 2006. Soil microbial community composition as affected by restoration practices in California grassland. Soil Biology & Biochemistry 38: 1851–1860. Potts, D. L., T. E. Huxman, B. J. Enquist, J. F. Weltzin, and D. G. Williams. 2006. Resilience and resistance of ecosystem functional response to a precipitation pulse in a semi-arid grassland. Journal of Ecology 94:23–30. Powers, S. 1976. Tribes of California. University of California Press, Berkeley. Prasad, V., C. A. E. Stromberg, H. Alimohammadian, and A. Sahni. 2005. Dinosaur coprolites and the early evolution ofgrasses and grazers. Science 310:1177–1180. Preston, W. L. 1981. Vanishing landscapes: Land and life in the Tulare Lake basin. University of California Press, Berkeley. ———. 2002a. Post-Columbian wildlife irruptions in California: Implications for cultural and environmental understanding.

360

REFERENCES

Pages 111–140 in C. E. Kay and R. T. Simmons, editors. Wilderness and political ecology: Aboriginal influences and the original state of nature. University of Utah Press, Salt Lake City. ———. 2002b. Portents of plague from California’s protohistoric period. Ethnohistory 49:69–121. Price, M. V., and S. H. Jenkins. 1986. Rodents as seed consumers and dispersers. Pages 191–235 in D. R. Murray, editor. Seed dispersal. Academic Press, New York. Priestley, H. I., editor and translator. 1937. A historical, political, and natural description of California by Pedro Fages, soldier of Spain. University of California Press, Berkeley. Proctor, J., and S. R. J. Woodell. 1975. The ecology of serpentine soils. Advances in Ecological Research 9:255–366. Provenza, F. D., and D. F. Balph. 1988. Development of dietary choice in livestock on rangelands and its implications for management. Journal of Animal Science 66:2356–2368. Public Policy Institute of California. 2006. California’ futurepopulation. http://www.ppic.org/content/pubs/jtf/JTF_ FuturePopulationJTF.pdf. Pyke, C. R., and J. Marty. 2005. Cattle grazing mediates climate change impacts on ephemeral wetlands. Conservation Biology 19:1619–1625. Raab, L. M., and T. L. Jones, editors. 2004. Prehistoric California: Archeology and the myth of paradise. University of Utah Press, Salt Lake City. Rai, K. N., and S. K. Jain. 1982. Population biology of Avena. IX. Gene flow and neighborhood size in relation to microgeographic variation in Avena barbata. Oecologia 53:399–405. Rajakaruna, N. 2003. Edaphic differentiation in Lasthenia: A model for studies in evolutionary ecology. Madroño 50: 31–40. Rajakaruna, N., and B. A. Bohm. 1999. The edaphic factor and patterns of variation in Lasthenia californica. American Journal of Botany 86:1576–1596. Rajakaruna, N., G. E. Bradfield, B. A. Bohm, and J. Whitton. 2003a. Adaptive differentiation in response to water stress by edaphic races of Lasthenia californica (Asteraceae). International Journal of Plant Science 164:371–376. Rajakaruna, N., M. Y. Siddiqi, and J. Whitton. 2003b. Differential responses to Na
/K
and Ca2
/Mg2
in two edaphic races of the Lasthenia californica (Asteraceae) complex: A case for parallel evolution of physiological traits. New Phytologist 157:93–103. Rajakaruna, N., B. G. Baldwin, R. Chan, A. M. Desrochers, B. A. Bohm, and J. Whitton. 2003c. Edaphic races and phylogenetic taxa in the Lasthenia californica complex (Asteraceae: Heliantheae): and hypothesis of parallel evolution. Molecular Ecology 12(6):1675– 1679. Rastetter, E. B., and G. R. Shaver. 1992. A model of multiple-element limitation for acclimating vegetation. Ecology 73:1157–1174. Ratliff, R. D. 1986. Cattle responses to continuous and seasonal grazing of California USA annual grassland. Journal of Range Management 39:482–485. Ratliff, R. D., D. A. Duncan, and S. E. Westfall. 1991. California oak-woodland overstory species affect herbage understory: Management implications. Journal of Range Management 44:306–310. Raven, P. J., and C. Axelrod. 1978. Origin and relationships of the California flora. University of California Publications in Botany 72:1–144. Reeder, J. D., G. E. Schuman, and J. A. Morgan. 2004. Response of organic and inorganic carbon and nitrogen to long-term

grazing of the shortgrass steppe. Environmental Management 33:485–495. Reever Morghan, K. J., E. A. Leger, and K. J. Rice. 2003. Clopyralid effects on yellow starthistle (Centaurea solstitialis) and non-target species. Weed Science 51:596–600. Reever Morghan, K. J., and K. J. Rice. 2005. Centaurea solstitialis invasion success is influenced by Nassella pulchra size. Restoration Ecology 13:524–528. Reeves, K., and J. Morris. 2000. Biological monitoring report for the Hollister Hills State Vehicular Recreation Area. T.O. Cattle Company, San Juan Bautista, CA. ———. 2001. Holistic management and biological planning in California: You be the judge. California Native Grass Association Newsletter:4–7. Reichman, O. J., and E. W. Seabloom. 2002. The role of pocket gophers as subterranean ecosystem engineers. Trends in Ecology and Evolution 17:44–49. Reiner, R. J. 2001. Protecting biodiversity on grazed grasslands in California: A presentation to the Association for the Advancement of Science, January 24, 1999. Conserve Online. http:// conserveonline.org/coldocs/2001/05/AAAS_Talk_1999.doc. Reiner, R. J., P. Hujik, and J. Pollock. In press. Predicting vegetation response to fire in California annual grasslands. In R. Schlising and D. Alexander, editors. The vernal pool complexes of the northstate: Biology, conservation, and management conference, March 23, 2006. Butte Environmental Council, Chico, CA. Reiner, R. J., and R. Swenson. 2000. Saving vernal pools in the Cosumnes River watershed. Fremontia 27:4:33–37. Reinhart, K. O., and R. M. Callaway. 2006. Soil biota and invasive plants. New Phytologist 170:445–457. Rejmánek, M., and M. J. Pitcairn. 2002. When is eradication of exotic pest plants a realistic goal? Pages 249–253 in C. R. Veitch and M. N. Clout, editors. Turning the tide: The eradication of invasive species. IUCN SSC Invasive Species Specialist Group, Gland, Switzerland. Rejmánek, M., and J. M. Randall. 1994. Invasive alien plants in California: 1993 summary and comparison with other areas in North America. Madroño 41:161–177. ———. 2004. The total number of naturalized species can be a reliable predictor of the number of alien pest species. Diversity and Distributions 10:367–363. Rejmánek, M., D. M. Richardson, M. G. Barbour, M. J. Crawley, G. F. Hrusa, P. B. Moyle, J. M. Randall, D. Simberloff, and M. Williamson. 2002. Biological invasions: Politics and the discontinuity of ecological terminology. Bulletin of the Ecological Society of America 83:131–133. Reynolds, H. L., B. A. Hungate, F. S. Chapin, and C. M. D’Antonio. 1997. Soil heterogeneity and plant competition in an annual grassland. Ecology 78:2076–2090. Reynolds, R. D. 1959. The effect upon the forest of natural fire and aboriginal burning in the Sierra Nevada. M.S. thesis, University of California, Berkeley. Reynolds, S., J. D. Corbin, and C. M. D’Antonio. 2001. The effects of litter and temperature on the germination of native and exotic grasses in a coastal California grassland. Madroño 48:230–235. Riba, M., and J. Terradas. 1987. Characteristiques de la resposta als incendis en els ecosistemes mediterranis. Pages 63–75 in J. Terradas, editor. Ecosistemes terrestres. Diputacio de Barcelona, Spain. Rice, K. J. 1985. Responses of Erodium to varying microsites: The role of germination cueing. Ecology 66:1651–1657.

———. 1987. Interaction of disturbance patch size and herbivory in Erodium colonization. Ecology 68:1113–1115. ———. 1989a. Competitive interactions in California annual grasslands. Pages 59–71 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Press, Dordrecht, The Netherlands. ———. 1989b. Impacts of seed banks on grassland community structure and population dynamics. Pages 212–230 in M. A. Leck, V. T. Parker, and R. L. Simpson, editors. Ecology of soil seed banks. Academic Press, San Diego, CA. Rice, K. J., and A. R. Dyer. 2001. Seed aging, delayed germination and reduced competitive ability in Bromus tectorum. Plant Ecology 155:237–243. Rice, K. J., and N. C. Emery. 2003. Managing microevolution: Restoration in the face of global change. Frontiers in Ecology 1:469–478. Rice, K. J., and E. E. Knapp. 2000. Evolutionary factors affecting the probability of local adaption or should we expect to see ecotypes behind every rock? Pages 221–226 in J. E. Keeley, M. Baer-Keeley, and C. J. Fotheringham, editors. 2nd interface between ecology and land development in California. USGS Open-File Report 00-62. U.S. Geological Survey, Denver, CO. Rice, K. J., and E. S. Nagy. 2000. Oak canopy effects on the distribution patterns of two annual grasses: The role of competition and soil nutrients. American Journal of Botany 87: 1699–1706. Richards, J. H., and M. M. Caldwell. 1985. Soluble carbohydrates, concurrent photosynthesis and efficiency in regrowth following defoliation: A field study with Agropyron species. Journal of Applied Ecology 22:907–920. Richards, R. T., and M. R. George. 1996. Evaluating changes in ranch management practices through extension education. Journal of Range Management 49:76–80. Richardson, D. M., P. Pysek, M. Rejmanek, M. G. Barbour, F. D. Panetta, and C. J. West. 2000. Naturalization and invasion of alien plants: Concepts and definitions. Diversity and Distributions 6:93–107. Ridolfi, M. C., R. A. Dahlgren, W. R. Horwarth, and K. W. Tate. 1999. Changes in soil quality due to oak tree removal in California oak woodland. Pages 18–27 in A. Zabel and G. Sposito, editors. Soil quality in the California environment. Annual Report of Research Projects 1998–1999. M. Theo Kearney Foundation of Soil Science, Division of Agriculture and Natural Resources, University of California, Berkeley. Rikard, W. H., and J. F. Cline. 1980. Cheatgrass communities: Effects of plowing on species composition and productivity. Northwest Science 54:216–221. Rillig, M. C., M. F. Allen, J. N. Klironomos, N. R. Chiariello, and C. B. Field. 1998. Plant species-specific changes in rootinhabiting fungi in a California annual grassland: responses to elevated CO2 and nutrients. Oecologia 113:252–259. Rillig, M. C., C. B. Field, and M. F. Allen. 1999a. Soil biota responses to long-term atmospheric CO2 enrichment in two California annual grasslands. Oecologia 119:572–577. Rillig, M. C., S. F. Wright, M. F. Allen, and C. B. Field. 1999b. Rise in carbon dioxide changes soil structure. Nature 400:628. Rillig, M. C., S. F. Wright, M. R. Shaw, and C. B. Field. 2002. Artificial climate warming positively affects arbuscular mycorrhizae but decreases soil aggregate water stability in an annual grassland. Oikos 97:52–58.

REFERENCES

361

Ritchie, M. E., D. Tilman, and J. M. H. Knops. 1998. Herbivore effects on plant and nitrogen dynamics in oak savanna. Ecology 79:165–177. Rizzo, D. M., M. Garbelotto, and E. M. Hansen. 2005. Phytophthora ramorum: Integrative research and management of an emerging pathogen in California and Oregon forests. Annual Review of Phytopathology 43:309–435. Roberts, B. A., and J. Proctor. 1992. The ecology of areas with serpentinized rocks: A world view. Kluwer Academic Publishers, Dordrecht, The Netherlands. Robins, P. 2002a. Bring farm edges back to life. Yolo County Resource Conservation District, Woodland, CA. ———. 2002b. Cost share programs for resource conservation and wildlife habitat development. Pages 85–94 in P. Robins, editor. Bring farm edges back to life. Yolo County Resource Conservation District, Woodland, CA. Robinson, G. R., J. F. Quinn, and M. L. Stanton. 1995. Invasibility of experimental habitat islands in a California annual grassland. Ecology 76:786–794. Robinson, R. H. 1971. An analysis of ecological factors limiting the distribution of a group of Stipa pulchra associations. Korean Journal of Botany 14:61–80. Roché, B. F., C. T. Roché, and R. C. Chapman. 1994. Impacts of grassland habitat on yellow starthistle (Centaurea solsitialis L.) invasion. Northwest Science 68:86–96. Roché, C. T., R. Vilatersana, T. Garnatje, R. Gamarra, N. Garcia-Jacas, A. Susanna, and D. C. Thill. 2003. Tracking an invader to its origins: The invasion case history of Crupina vulgaris. Weed Research 43:177–189. Rodell, C. F. 1977. A grasshopper model for a grassland ecosystem. Ecology 58:227–245. Rodriguez-Rojo, M. P., D. Sanchez-Mata, R. G. Gavilan, S. RivasMartinez, and M. G. Barbour. 2001a. Syntaxonomical approach for classification of the Californian serpentine annual grasslands. Lazaroa 22:83–94. ———. 2001b. Typology and ecology of Californian serpentine annual grasslands. Journal of Vegetation Science 12:687–698. Rohrbaugh, R. W., D. L. Reinkeing, D. H. Wolfe, S. K. Sherrod, and M. L. Jenkins. 1999. Effects of prescribed burning and grazing on nesting and reproductive success of three grassland passerine species in tallgrass prairie. Studies in Avian Biology 19:165–170. Romañach, S. S., O. J. Reichman, and E. W. Seabloom. 2005. Seasonal influences on burrowing activity of a subterranean rodent, Thomomys bottae. Journal of Zoology 266:319–325. Rosenthal, R. 1979. The “file drawer problem” and tolerance for null results. Psychological Bulletin 86:638–641. Rosenzwieg, M. L. 1995. Species diversity in space and time. Cambridge University Press, Cambridge, UK. Rosiere, R. E. 1987. An evaluation of grazing intensity influences on California annual range. Journal of Range Management 40:160–165. Ross, M. A., and J. L. Harper. 1972. Occupation of biological space during seedling establishment. Journal of Ecology 60:77–88. Rossi, R. S. 1979. Land use and vegetation change in the oak woodland-savanna of northern San Luis Obispo County, California (l774–l978). Ph.D. dissertation, University of California, Berkeley. Rowe, H. I., E. T. Bartlett, and L. E. Swanson Jr. 2001. Ranching motivations in two Colorado counties. Journal of Range Management 54:314–321.

362

REFERENCES

Rowe, P. B., and L. F. Reimann. 1961. Water use by brush, grass, and grass-forb vegetation. Journal of Range Management 59:175–180. Rudaz, A. O., E. A. Davidson, and M. K. Firestone. 1991. Sources of nitrous oxide following wetting of a dry soil. FEMS Microbial Ecology 85:117–124. Ruddiman, W. F. 2001. Earth’s climate: Past and future. W. H. Freeman and Comapany, New York. Ruiz-Jaen, M. C., and T. Mitchell Aide. 2005. Restoration success: How is it being measured? Restoration Ecology 13:569–577. Russell, W. H., and J. R. McBride. 2003. Landscape scale vegetationtype conversion and fire hazard in the San Francisco Bay area open spaces. Landscape and Urban Planning 64:201–208. Russell-Smith, J., and D. M. J. S. Bowman. 1992. Conservation of monsoon rainforest isolates in the Northern Territory, Australia. Biological Conservation 59:51–63. Rustad, L. E., J. L. Campbell, G. M. Marion, R. J. Norby, M. J. Mitchell, A. E. Hartley, J. H. C. Cornelissen, and J. Gurevitch. 2001. A meta-analysis of the response of soil respiration, net nitrogen mineralization, and aboveground plant growth to experimental ecosystem warming. Oecologia 126:543–562. Saarela, J. J., and P. M. Peterson. In press. Bromus. In B. G. Baldwin, S. Boyd, B. J. Ertter, D. J. Keil, R. W. Patterson, T. J. Rosatti, and D. H. Wilken, editors. The Jepson Manual of higher plants of California, 2nd edition. University of California Press, Berkeley, CA. Sabater, F., A. Butturini, E. Marti, I. Munoz, A. Romani, J. Wray, and S. Sabater. 2000. Effects of riparian vegetation removal on nutrient retention in a Mediterranean stream. Journal of the North American Benthological Society 19:609–620. Saenz, L., and J. O. Sawyer. 1986. Grasslands as compared to adjacent Quercus garryana woodland understories exposed to different grazing regimes. Madroño 33:40–46. Safford, H. D., and S. P. Harrison. 2001. Grazing and substrate interact to affect native vs. exotic diversity in roadside grasslands. Ecological Applications 11:1112–1122. ———. 2004. Fire effects on plant diversity in serpentine vs. sandstone chaparral. Ecology 85:539–548. Safford, H. D., J. H. Viers, and S. Harrison. 2005. Serpentine endemism in the Calfornia flora: A database of serpentine affinity. Madroño 52:222–257. Sakai, A. K., F. W. Allendorf, J. S. Holt, D. M. Lodge, J. Molofsky, K. A. With, S. Baughman, R. J. Cabin, J. E. Cohen, N. C. Ellstran, D. E. McCauley, P. O’Neill, I. M. Parker, J. N. Thompson, and S. G. Weller. 2001. The population biology of invasive species. Annual Review of Ecology and Systematics 32:305–332. Sala, O. E., W. J. Parton, L. A. Joyce, and W. K. Lauenroth. 1988. Primary production of the central grassland region of the United States. Ecology 69:40–45. Sala, O. E., and J. M. Paruelo. 1997. Ecosystem services in grasslands. Pages 237–252 in G. C. Daily, editor. Nature’s services: Societal dependence on natural ecosystems. Island Press, Washington DC. Salathe, E. P. 2003. Comparison of various precipitation downscaling methods for the simulation of streamflow in a rainshadow river basin. International Journal of Climatology 23:887–901. Sampson, A. W. 1917. Succession as a factor in range management. Journal of Forestry 15:593–596. ———. 1944. Plant succession on burned chaparral lands in northern California. University of California, Agriculture Experiment Station Bulletin 685.

———. 1952. Range management principles and practices. John Wiley and Sons, New York. Sampson, A. W., A. Chase, and D. W. Hedrick. 1951. California grasslands and range forage grasses. University of California Agricultural Experiment Station Bulletin 724:2–131. Sampson, A. W., and B. S. Jespersen. 1963. California range brushlands and browse plants. Manual 33. University of California Agricultural Experiment Station, Extension Service 102B112, Berkeley, CA. Sampson, A. W., and E. C. McCarty. 1930. The carbohydrate metabolism of Stipa pulchra. Hilgardia 5:61–99. Sapir, E., and L. Spier. 1943. Notes on the culture of the Yana. University of California Anthropological Records 3: 239–298. Sauer, C. O. 1950. Grassland climax, fire and management. Journal of Range Management 3:16–21. Sauer, J. D. 1988. Plant migration: The dynamics of geographic patterning in seed plant species. University of California Press, Berkeley, CA. Saunders, G. 1993. Observations on the effectiveness of shooting feral pigs from helicopters. Wildlife Research 20:771–776. Savage, D. E. 1951. Late Cenozoic vertebrates of the San Francisco Bay region. Bulletin of the Department of Geological Science, University of California 28:215–314. Savelle, G. D. 1977. Comparative structure and function in a California annual and native bunchgrass community. Ph.D. Dissertation, University of California, Berkeley. Savelle, G. D., and H. F. Heady. 1970. Mediterranean annual species: Their response to defoliation. Pages 548–560 in Proceedings of the XI International Grassland Congress 1970. Savile, D. B. O. 1970. Some Eurasian Puccinia species attacking Cardueae. Canadian Journal of Botany 48:1553–1566. Savory, A., and J. Butterfield. 1999. Holistic management: A new framework for decision making. Island Press, Covelo, CA. Sawyer, J. O., and T. Keeler-Wolf. 1995. A manual of California vegetation. California Native Plant Society, Sacramento. Sawyer, J. O., T. Keeler-Wolf, and J. M. Evens, editors. 2007. A manual of California vegetation, 2nd ed. California Native Plant Society, Sacramento. Sawyer, J. O., D. A. Thornburgh, and J. R. Griffin. 1977. Mixed evergreen forest. Pages 359–382 in M. G. Barbour and J. Major, editors. Terrestrial vegetation of California. John Wiley and Sons, New York. Schiffman, P. M. 1994. Promotion of exotic weed establishment by endangered giant kangaroo rats (Dipodomys ingens) in a California grassland. Biodiversity and Conservation 3:524–537. ———. 1995. Impact of endangered giant kangaroo rat on grassland restoration. Bulletin of the Ecological Society of America 76:238. ———. 1997. Animal-mediated dispersal and disturbance: Driving forces behind alien plant naturalization. Pages 87–94 in J. O. Luken and J. W. Thieret, editors. Assessment and management of plant invasions. Springer-Verlag, New York. ———. 2000. Mammal burrowing, erratic rainfall and the annual lifestyle in the California prairie: Is it time for a paradigm shift? Pages 153–160 in J. E. Keeley, M. Baer-Keeley, and C. J. Fotheringham, editors. 2nd interface between ecology and land development in California. USGS Open-File Report 00-62. U.S. Geological Survey, Denver, CO. ———. 2005. The Los Angeles prairie. Pages 38–51 and 298–303 in W. Deverell and G. Hise, editors. Land of sunshine: An envi-

ronmental history of metropolitan Los Angeles. University of Pittsburgh Press, Pittsburgh, PA. Schimel, J. P., L. E. Jackson, and M. K. Firestone. 1989. Spatial and temporal effects of plant-microbial competition for inorganic nitrogen in a California annual grassland. Soil Biology and Biochemistry 21:1059–1066. Schlesinger, W. H., and A. M. Pilmanis. 1998. Plant-soil interactions in deserts. Biogeochemistry 42:169–187. Schlichting, C. D. 1986. The evolution of phenotypic plasticity in plants. Annual Review of Ecology and Systematics 17: 667–693. Schoenherr, A. A. 1992. A natural history of California. University of California Press, Berkeley. Scholes, R. J., and S. R. Archer. 1997. Tree-grass interactions in savannas. Annual Review of Ecology and Systematics 28: 517–544. Schultz, C. B., and E. C. Crone. 1998. Burning prairie to restore butterfly habitat: A modeling approach to management tradeoffs for the Fender’s blue. Restoration Ecology 6:244. Schultz, P. W., and L. Zelezney. 2003. Reframing environmental messages to be congruent with American values. Human Ecology Review 10:126–136. Schuur, E. A. G. 2003. Productivity and global climate revisited: The sensitivity of tropical forest growth to precipitation. Ecology 84:1165–1170. Scott, T. A., L. Fernandez, and M. Allen. 2006. Land use planning. in J. M. Scott, D. Goble, and F. W. Davis, editors. The Endangered Species Act at thirty. Volume 2: Conserving biodiversity in human-dominated landscapes. Island Press, Washington, DC. Seabloom, E. W., O. N. Bjornstad, B. M. Bolker, and O. J. Reichman. 2005. Spatial signature of environmental heterogeneity, dispersal, and competition in successional grasslands. Ecological Monographs 75:199–214. Seabloom, E. W., E. T. Borer, V. L. Boucher, R. S. Burton, K. L. Cottingham, L. Goldwasser, W. K. Gram, B. E. Kendall, and F. Micheli. 2003a. Competition, seed limitation, disturbance, and reestablishment of California native annual forbs. Ecological Applications 13:575–592. Seabloom, E. W., W. S. Harpole, O. J. Reichman, and D. Tilman. 2003b. Invasion, competitive dominance, and resource use by exotic and native California grassland species. Proceedings of the National Academy of Sciences USA 100: 13384–13389. Seabloom, E. W., and O. J. Reichman. 2001. Simulation models of the interactions between herbivore foraging strategies, social behavior, and plant community dynamics. American Naturalist 157:76–96. Seabloom, E. W., O. J. Reichman, and E. J. Gabet. 2000. The effect of hillslope angle on pocket gopher (Thomomys bottae) burrow geometry. Oecologia 125:26–34. Seabloom, E. W., and S. A. Richards. 2003. Multiple stable equilibria in grasslands mediated by herbivore population dynamics and foraging behavior. Ecology 84:2891 – 2904. Seabloom, E. W., J. W. Williams, D. Slayback, D. M. Stoms, J. H. Viers, and A. P. Dobson. 2006. Human impacts, plant invasion, and imperiled species in California. Ecological Applications 16:1338–1350. Seastedt, T. R., J. M. Briggs, and D. J. Gibson. 1991. Control on nitrogen limitation in tallgrass prairie. Oecologia 87:72–79.

REFERENCES

363

Seligman, N. G. 1996. Management of Mediterranean grasslands. Pages 359–391 in J. Hodgson and A. W. Illius, editors. The ecology and mangement of grazing systems. CAB International, Oxfordshire, UK. SER. 2004. The SER primer on ecological restoration. Society for Ecological Restoration International Science & Policy Working Group, Tucson, AZ. http://www.ser.org/content/ecological_ restoration_primer.asp. Sharp, L. A., M. Hironaka, and E. W. Tisdale. 1957. Viability of medusahead (Elymus caput-medusae L.) seed collected in Idaho. Journal of Range Management 10:123–126. Shaw, C. A., and S. M. Cox. 1993. The giant short-faced bear. Pages 22–23 in I. Sterling, editor. Bears, the majestic creatures of the wild. Rodale Press, Emmaus, Pennsylvania. Shaw, R. M., E. S. Zavaleta, N. R. Chiariello, E. E. Cleland, H. A. Mooney, and C. B. Field. 2002. Grassland responses to global environmental changes suppressed by elevated CO2. Science 298:1987–1990. Shaw, W. T. 1934. The ability of the giant kangaroo rat as a harvestor and storer of seeds. Journal of Mammalogy 15:275–286. Shea, K., and P. Chesson. 2002. Community ecology theory as a framework for biological invasions. Trends in Ecology and Evolution 17:170–176. Sheley, R. L., J. S. Jacobs, and M. F. Carpinelli. 1998. Distribution, biology, and management of diffuse knapweed (Centaurea diffusa) and spotted knapweed (Centaurea maculosa). Weed Technology 12:353–362. Sheley, R. L., S. Kedzie-Webb, and B. D. Maxwell. 1999. Integrated weed management on rangeland. Pages 57–68 in R. L. Sheley and J. K. Petroff, editors. Biology and management of noxious rangeland weeds. Oregon State University Press, Corvallis. Sheley, R. L., and L. L. Larson. 1995. Interference between cheatgrass and yellow starthistle at 3 soil depths. Journal of Range Management 48:392–397. Sheley, R. L., L. L. Larson, and D. E. Johnson. 1993. Germination and root dynamics of range weeds and forage species. Weed Technology 7:234–237. Sheley, R. L., B. Olson, and C. Hoopes. 1998. What is so dangerous about the impacts of noxious weeds on the ecology and economy of Montana? Extension Bulletin 152, Montana State University, Bozeman. Sheley, R. L., and J. K. Petroff, editors. 1999. Biology and management of noxious rangeland weeds. Oregon State University Press, Corvallis. Shepard, P. 1996. A Paul Shepard reader: The only world we’ve got. Sierra Club Books, San Francisco. Sher, A. A., D. E. Goldberg, and A. Novoplansky. 2004. The effect of mean and variance in resource supply on survival of annuals from Mediterranean and desert environments. Oecologia 141:353–362. Shipek, F. C. 1989. An example of intensive plant husbandry: the Kumeyaay of southern California. Pages 159–170 in D. R. Harris and G. C. Hilman, editors. Foraging and farming: The evolution of plant exploitation. Unwin-Hyman Publishers, London. Shipek, F. C. 1991. Delfina Cuero. Ballena Press, Menlo Park, California. Shishkoff, N., and W. Bruckart. 1993. Evaluation of infection of target and nontarget hosts by isolates of the potential biocontrol agent Puccinia jaceae that infect Centaurea spp. Phytopathology 83:894–898.

364

REFERENCES

Shock, C. C., M. B. Jones, W. A. Williams, and D. M. Center. 1984. Competition for S and N by associations of three annual range species in lysimeters. Plant and Soil 81:311–321. Shoulders, C. L. 1994. Methods of restoring Nassella pulchra (purple needlegrass) at Jepson Prairie Preserve, Solano County, California. M.S. thesis, University of Wisconsin, Madison. Shuford, W. D., and I. C. Timossi. 1989. Plant communities of Marin County. California Native Plant Society, Sacramento. Shurin, J. B., E. T. Borer, E. W. Seabloom, K. Anderson, C. A. Blanchette, B. Broitman, S. D. Cooper, and B. S. Halpern. 2002. A cross-ecosystem comparison of the strength of trophic cascades. Ecology Letters 5:785–791. Siemann, E., J. Haarstad, and D. Dilman. 1997. Short-term and long-term effects of burning on oak savanna arthropods. The American Midland Naturalist 137:349–361. Silveira, J. G. 2000. Vernal pools and relict dune land at Arena Plains. Fremontia 27/28:38–47. Silver, S. 1978. Shastan peoples. Pages 211–224 in R. F. Heizer, editor. Handbook of North American Indians. Vol. 8: California. Smithsonian Institution, Washington, DC. Simberloff, D., and B. Van Holle. 1999. Positive interactions of nonindigenous species: Invasional meltdown? Biological Invasions 1:21–32. Simpson, G. G. 1980. Splendid isolation: The curious history of South American mammals. Yale University Press, New Haven, CT. Sims, P. L., and P. G. Risser. 2000. Grasslands. Pages 323–356 in M. G. Barbour and W. D. Billings, editors. North American terrestrial vegetation. Cambridge University Press, New York. Singer, F. J., and K. A. Schoenecker. 2003. Do ungulates accelerate or decelerate nitrogen cycling? Forest Ecology and Management 181:189–204. Singer, F. J., W. T. Swank, and E. E. C. Clebsch. 1984. Effects of wild pig rooting in a deciduous forest. Journal of Wildlife Management 48:464–473. Singer, M. J., and D. N. Munns. 1999. Soils: An introduction. 4th ed. Prentice Hall, Upper Saddle River, NJ. Six, J., S. D. Frey, R. K. Thiet, and K. M. Batten. 2006. Bacterial and fungal contributions to carbon sequestration in agroecosystems. Soil Science Society of America Journal 70:555–569. Skinner, C. N., and C. Chang. 1996. Fire regimes, past and present. Pages 1041–1070 in Sierra Nevada Ecosystem Project: Final report to Congress, Vol. 2. Assessments and scientific basis for management options. Report 36. University of California Centers for Water and Wildland Resources. Skinner, M. W., and B. A. Pavlik. 1994. California Native Plant Society’s inventory of rare and endangered plants in California, 5th edition. Special Publication 1 California Native Plant Society, Sacramento. Smith, A. H., and W. E. Martin. 1972. Socioeconomic behavior of cattle ranchers, with implications for rural community development in the West. American Journal of Agricultural Economics 54:217–225. Smith, B. D. 1995. The emergence of agriculture. Scientific American Library, New York. Smith, D. W., and W. L. Verrill. 1998. Vernal pool landforms and soils of the Central Valley, California. Proceedings of the conference on the ecology, conservation, and management of vernal pool ecosystems. California Native Plant Society, Sacramento California.

Smith, J. K. 2000. Wildland fire in ecosystems: effects of fire on fauna. General Technical Report RMRS-GTR-42-Volume 1. USD A Rocky Mountain Experiment Station, Ogden, UT. Smith, J. P. 2006. The grasses of California. HSU Herbarium Miscellaneous Publications 5. Humboldt State University, Arcata, CA. Smith, T. A. 1970. Effects of disturbance on seed germination in some annual plants. Ecology 51:1106–1108. Snow, G. E. 1972. Some factors controlling the establishment and distribution of Quercus agrifolia Nee and Q. engelmannii Greene in certain southern California oak woodlands. Ph.D. dissertation, Oregon State University, Corvallis. Snyder, M. A., J. L. Bell, L. C. Sloan, P. B. Duffy, and B. Govindasamy. 2002. Climate responses to a doubling of atmospheric carbon dioxide for a climatically vulnerable region. Geophysical Research Letters 29:9–1. Snyder, R. L., and R. O. Pruitt. 1992. Evapotranspiration data management in California. Pages 128–133 in Irrigation & Drainage Session Proceedings/Water Forum 9. American Society of Civil Engineering, Baltimore, MD. Snyder, S. R. 2003. Managing the recovery of mycorrhizal population in restoration. Grasslands 13:1–7. Solomeshch, A. I., and M. G. Barbour. 2006. Defining restoration targets for the Great Valley Grassland State Park (San Joaquin Valley). Grasslands 16(2):1–17. Soreng, R. J. 1993. Poa. Pages 1284–1290 in J. C. Hickman, editor. The Jepson manual: Higher plants of California. University of California Press, Berkeley. ———. 2007. Poa L. Pages 486–601 in M. E. Barkworth, K. M. Capels, S. Long, and M. B. Piep, editors. Magnoliophyta: Commelinidae (in part): Poaceae, part 1. Flora of North America North of Mexico. 24. Oxford University Press, New York. ———. In press. Poa. In B. G. Baldwin, S. Boyd, B. J. Ertter, D. J. Keil, R. W. Patterson, T. J. Rosatti, and D. H. Wilken, editors. The Jepson manual of higher plants of California, 2nd edition. University of California Press, Berkeley. Soreng, R. J., G. Davidse, P.M. Peterson, F.O. Zuloaga, E.J. Judziewicz, T.S. Filgueiras, and O. Morrone. 2005. Catalogue of New World grasses. Suprageneric classification. Missouri Botanic Garden, St. Louis, MO. http://mobot.mobot.org/W3T/ Search/nwgclass.html. Soreng, R. J., and J. I. Davis. 1998. Phylogenetics and character evolution in the grass family (Poaceae): Simultaneous analysis of morphological and chloroplast DNA restriction site character sets. Botanical Review 64:1–90. ———. 2000. Phylogenetic structure in Poaceae subfamily Pooideae as inferred from molecular and morphological characters: misclassification verses reticulation. Pages 61–74 in W. L. Jacobs and J. Everett, editors. Grasses: Systematics and evolution. CSIRO, Collingwood, Australia. Sork, V. L., F. W. Davis, P. E. Smouse, V. J. Apsit, R. J. Dyer, J. F. Fernandez, and B. Kuhn. 2002. Pollen movement in declining populations of California Valley oak, Quercus lobata: Where have all the fathers gone? Molecular Ecology 11: 1657–1668. Sousa, W. P. 1984. The role of disturbance in natural communities. Annual Review of Ecology and Systematics 15:353–391. Sparkman, P. S. 1908. The culture of the Luiseño Indians. University of California Publications in American Archaeology and Ethnology 8:187–234.

Sparling, G. P., D. Wheeler, and E. T. Vesely. 2006. What is soil organic matter worth? Journal of Environmental Quality 35:548–557. Spatz, G., and D. Mueller-Dombois. 1975. Succession patterns after pig digging in grassland communities on Mauna Loa, Hawaii. Phytocoenologia 3:346–373. Sperry, T. M. 1994. The Curtis Prairie restoration, using the singlespecies planting method. Natural Areas Journal 14:124–127. Spira, T. P., and L. K. Wagner. 1983. Viability of seeds up to 211 years old extracted from adobe brick buildings in California and northern Mexico. Amercan Journal of Botany 70:303–307. Standiford, R. B., N. McDougald, W. Frost, and R. Phillips. 1997. Factors influencing the probability of oak regeneration on southern Sierra Nevada woodlands in California. Madroño 44:170–183. Standiford, R. B., and P. Tinnin. 1996. Guidelines for managing California’s hardwood rangelands. Publication 3368. Division of Agriculture and Natural Resources, University of California, Berkeley. Stark, J. M. 1991. Environmental factors versus ammonia-oxidizer population characteristics as dominant controllers of nitrification in an oak woodland-annual grassland soil. Ph.D. dissertation, University of California, Berkeley. Stark, J. M., and M. K. Firestone. 1995. Mechanisms for soil moisture effects on activity of nitrifying bacteria. Applied and Environmental Microbiology 61:218–221. Stebbins, G. L. 1942. The genetic approach to problems of rare and endemic species. Madroño 6:241–272. ———. 1957. Self-fertilization and population variability in the higher plants. American Naturalist 91:337–354. Steenwerth, K. L., L. E. Jackson, F. J. Calderón, K. M. Scow, and D. E. Rolston. 2005. Response of microbial community composition and activity in agricultural grassland soils after a simulated rainfall. Soil Biology & Biochemistry 37:2249–2262. Steenwerth, K. L., L. E. Jackson, F. J. Calderón, M. R. Stromberg, and K. M. Scow. 2002. Soil microbial community composition and land use history in cultivated and grassland ecosystems of coastal California. Soil Biology and Biochemistry 34: 1599–1611. Steenwerth, K. L., L. E. Jackson, F. J. Calderón, M. R. Stromberg, and K. M. Scow. 2003. Soil microbial community composition and land use history in cultivated and grassland ecosystems of coastal California. Erratum. Soil Biology and Biochemistry 35:489–500. Steenwerth, K. L., L. E. Jackson, E. A. Carlisle, and K. M. Scow. 2006. Microbial communities of a native perennial bunchgrass do not respond consistently across a gradient of land-use intensification. Soil Biology & Biochemistry 38:1797–1811. Stephens, S. L. 1997. Fire history of a mixed oak-pine forest in the foothills of the Sierra Nevada, El Dorado County, California. Pages 191–198 in N. H. Pillsbury, J. Verner, and W. D. Tietje, editors. Proceedings of a symposium on oak woodlands: ecology, management, and urban interface issues. General Technical Report. PSW-GTR-160. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Stephens, S. L., and L. W. Ruth. 2005. Federal forest fire policy in the United States. Ecological Applications 15:532–542. Stephenson, B. A. 2002. Carmel Mountain and Del Mar Mesa Preserves management plan. City of San Diego, CA. Sterner, J. D., and R. H. Barrett. 1991. Removing feral pigs from Santa Cruz Island, California. Transactions of the Western Section of the Wildlife Society 27:47–53.

REFERENCES

365

Sterner, R. W., and J. J. Elser. 2002. Ecological stoichiometry: The biology of elements from molecules to the biosphere. Princeton University Press, Princeton, NJ. Steward, J. H. 1933. Ethnography of the Owens Valley Paiute. University of California Publications in American Archaeology and Ethnology 33:423–438. Stewart, O. C. 1935. Pomo unpublished field notes in the possession of D. Theodoratus. Stock, C. S., and J. M. Harris. 1992. Rancho La Brea, a record of Pleistocene life in California. Science Series 37. Los Angeles County Museum. Stoll, S. 1998. The fruits of natural advantage: Making the industrial countryside in California. University of California Press, Berkeley. Stone, C. P. 1985. Alien animals in Hawaii’s native ecosystems: Toward controlling the adverse effects of introduced vertebrates. Pages 251–584 in C. P. Stone and J. M. Scott, editors. Hawaii’s terrestrial ecosystems: Preservation and management. Cooperative National Park Resources Studies Unit. University of Hawaii, Honolulu. Stone, E. C. 1951. The stimulative effect of fire on the flowering of the golden brodiaea (Brodiaea ixiodes Wats. var. lugens Jeps.). Ecology 32:534–537. Storer, T. I., and L. P. Tevis. 1996. California grizzly. University of California Press, Berkeley. Strauss, S. Y., C. O. Webb, and N. Salamin. 2006. Exotic taxa less related to native species are more invasive. Proceedings of the National Academy of Sciences 103:5841–5845. Strohecker, H. F., W. W. Middlekauff, and D. C. Rentz. 1968. The grasshoppers of California (Orthoptera: Acridoidea). Bulletin of the California Insect Survey. Vol. 10. University of California Press, Berkeley. Stromberg, C. A. E. 2002. The origin and spread of grass-dominated ecosystems in the late Tertiary of North America: Preliminary results concerning the evolution of hypsodonty. Paleoclimatology, Paleogeography, Paleoecology 177:59–75. Stromberg, M. R., and J. R. Griffin. 1996. Long-term patterns in coastal California grasslands in relation to cultivation, gophers and grazing. Ecological Applications 6:1189–1211. Stromberg, M. R., and P. Kephart. 1996. Restoring native grasses to California old fields. Restoration and Management Notes 14:102–111. Stromberg, M. R., P. Kephart, and M. Sicular-Mertens. 2002. Tillage: Cheap weed and gopher control before planting native grasses. Grasslands 12:1–7. Stromberg, M. R., P. Kephart, and V. Yadon. 2001. Composition, invasibility, and diversity in coastal California grasslands. Madroño 48:236–252. Stulen, I., and J. Den Hertog. 1993. Root-growth and functioning under atmospheric CO2 enrichment. Vegetatio 104: 99–115. Suarez, A. V., D. T. Bolger, and T. J. Case. 1998. Effects of fragmentation and invasion on native ant communities in coastal southern California. Ecology 79:2041–2056. Sugihara, N. G., and L. J. Reed. 1987. Vegetation ecology of the Bald Hills oak woodlands of Redwood National Park. Redwood National Park Research and Development Technical Report 21, Redwood National Park, Orick, CA. Sugihara, N. G., L. J. Reed, and J. M. Lenihan. 1987. Vegetation of the Bald Hills oak woodlands, Redwood National Park, California. Madroño 34:193–208.

366

REFERENCES

Sulak, A., and L. Huntsinger. 2002. Sierra Nevada grazing in transition: The role of Forest Service grazing in the foothill ranches of California. A report to the Sierra Nevada Alliance, the California Cattlemen’s Association, and the California Rangeland Trust, Sierra Nevada Alliance, University of California, Berkeley. http://www.sierranevadaalliance.org/publications. ———. 2007. Public land grazing in California: untapped conservation potential for private lands? Rangelands 29(3):9–12. Sulak, A., L. Huntsinger, R. Standiford, A. M. Merenlender, and S. K. Fairfax. 2004. The agricultural conservation easement: a strategy for oak woodland conservation in California. Advances in Geoecology 37:309–326. Sultan, S. E. 1987. Evolutionary implications of phenotypic plasticity in plants. Evolutionary Biology 21:127–178. Suneson, C. A. 1956. An evolutionary plant breeding method. Agronomy Journal 48:188–191. Suttle, K. B., and M. A. Thomsen. In press. Climate change and grassland restoration in California: lessons from six years of rainfall manipulation in a northern grassland. Madroño. Suttle, K. B., M. A. Thomsen, and M. E. Power. 2007. Species interactions reverse species responses to climate change. Science 315:640– 642. Sweet, S. B. 2005. Effect of seed maturity on the resistance of weed weeds to simulated fire: implications for timing prescribed burns. M.S. thesis, University of California, Davis. Swetzer, R. A., and D. H. Van Vuren. 2002. Rooting and foraging effects of wild pigs on tree regeneration and acorn survival in California’s oak woodland ecosystems. Proceedings of the 5th symposium on oak woodlands: Oaks in California’s changing landscape. General Technical Report PSW-GTR-184. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Swetnam, T. W., C. H. Baisan, K. Morino, and A. C. Caprio. 1998. Fire history along elevational transects in the Sierra Nevada, California. USGS Biological Resources Division, Sequoia–Kings Canyon Field Station, Three Rivers, CA. Swiecki, T. J., and E. A. Bernhardt. 2002. Effects of fire on naturally occurring blue oak (Quercus douglasii) saplings. Pages 251–259 in R. B. Standiford, D. McCreary, and K. L. Purcell, editors. Proceedings of the fifth symposium on oak woodlands: Oaks in California’s changing landscape. General Technical Report PSW-GTR-184. USDA Forest Service, Pacific Southwest Research Station. Swiecki, T. J., E. A. Bernhardt, and C. Drake. 1997. Factors affecting blue oak sapling recruitment. Pages 157–161 in N. H. Pillsbury, J. Verner, and W. D. Tietje, editors. Proceedings of a symposium on oak woodlands: Ecology, management, and urban-interface issues. General Technical Report PSW-GTR160. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Swift, M. J., O. W. Heal, and J. M. Anderson. 1979. Decomposition in terrestrial ecosystems. University of California Press, Berkeley. Sylvia, D. M., and S. E. Williams. 1992. Vesicual-arbuscular mycorrhizae and environmental stress. Pages 101–124 in G. J. Bethlenfalvay and R. G. Linderman, editors. Mycorrhizae in sustainable agriculture. American Society of Agronomy, Madison, WI. Takemoto, B. K., B. E. Croes, S. M. Brown, N. Motallebi, F. D. Westerdahl, H. G. Margolis, B. T. Cahill, M. D. Mueller, and J. R. Holmes. 1995. Acidic deposition in California: Findings

from a program of monitoring and effects research. Water, Air, and Soil Pollution 85:261–272. Talbot, M. W., H. H. Biswell, and A. L. Hormay. 1939. Fluctuations in the annual vegetation of California. Ecology 20:394 –402. Tanaka, Y., and K. Yano. 2005. Nitrogen delivery to maize via mycorrhizal hyphae depends on the form of N supplied. Plant, Cell, and Environment 28:1247–1254. Tardiff, S. E., and J. A. Stanford. 1998. Grizzly bear digging: Effects on subalpine meadow plants in relation to mineral nitrogen availability. Ecology 79:2219–2228. Tate, K. 2003. Riparian friendly grazing project. Western Region Sustainable Agriculture Research and Education. Project Report SW01-044. Tate, K. W., E. R. Atwill, J. W. Bartolome, and G. Nader. 2006. Significant Escherichia coli attenuation by vegetative buffers on annual grasslands. Journal of Environmental Quality 35:795–805. Tate, K. W., E. R. Atwill, N. K. McDougald, and M. R. George. 2003. Spatial and temporal patterns of cattle feces deposition on rangeland. Journal of Range Management 56:432–438. Tate, K. W., E. R. Atwill, N. K. McDougald, M. R. George, and D. Witt. 2000. A method for estimating cattle fecal loading on rangeland watersheds. Journal of Range Management 53: 506–510. Tate, K. W., R. A. Dahlgren, M. J. Singer, B. Allen-Diaz, and E. R. Atwill. 1999. On California rangeland watersheds: timing, frequency of sampling affect accuracy of water quality monitoring. California Agriculture 53:44–48. Tate, K. W., D. M. Dudley, N. K. McDougald, and M. R. George. 2004. Effect of canopy and grazing on soil bulk density. Journal of Range Management 57:411–417. Tate, K. W., M. D. C. Pereira, and E. R. Atwill. 2004. Efficacy of vegetated buffer strips for retaining Cryptosporidium parvum. Journal of Environmental Quality 33:2243–2251. Taylor, B. 1951. New pictures from California. Biobooks, Oakland, CA. Taylor, D. W., and W. Davilla. 1986. Vegetation of Site 300, Lawrence Livermore National Laboratory, San Joaquin County, California. UCRL-15873 ( J-262). BioSystems Analysis, Inc., Santa Cruz, CA. Templeton, B. C. 1964. The fruits and seeds of the Rancho La Brea Pleistocene deposits. Ph.D. dissertation, Oregon State University, Corvallis. ter Braak, C. J. F. 1987. Ordination. Pages 91 – 169 in R. H. A. Songman, C. J. F. ter Braak, and O. F. R. van Tengeren, editors. Data analysis in community and landscape ecology. Centre for Agricultural Publishing and Documentation, Wageningen, The Netherlands. ter Braak, C. J. F., and P. Smilauer. 1998. CANOCO reference manual and user’s guide to CANOCO for Windows: software for canonical community ordination (version 4). Microcomputer Power, Ithaca, NY. Thomas, F. 1996. Using annual legumes to manage weeds. Proceedings of California Weed Science Society 48:120–122. Thomas, F. 1997. Selecting cover crops to suppress weeds. Proceedings of California Weed Science Society 49:68–71. Thomas, K. A., T. Keeler-Wolf, J. Franklin, and P. Stine. 2004. Mojave Desert Ecosystem Program: Central Mojave vegetation database. Technical Report 08-2004, United States Geological Survey, Western Regional Science Center, Sacramento, CA.

Thomasson, J. R. 1987. Fossil grasses: 1820-1986 and beyond. Pp. 159–167 in T. R. Soderstrom, K. W. Hilu, C. S. Campbell, and M. E. Barkworth, editors. Grass systematics and evolution. Smithsonian Institution Press, Washington, DC. Thomasson, J. R., M. E. Nelson, and R. J. Zakrzewski. 1986. A fossil grass (Gramineae: Chloridoideae) from the Miocene with Kranz anatomy. Science 233:876–878. Thompson, R., T. R. Van Devender, P. S. Martin, T. Foppe, and A. Long. 1980. Shasta ground sloth at Shelter Cave, New Mexico: Environment, diet, and extinction. Quaternary Research 14: 360–376. Thomsen, C. D., M. P. Vayssières, and W. A. Williams. 1997. Mowing and subclover plantings suppress yellow starthistle. California Agriculture 51:15–20. Thomsen, C. D., W. A. Williams, and M. R. George. 1990. Managing yellow starthistle on annual range with cattle. Knapweed 4:3–4. Thomsen, C. D., W. A. Williams, M. R. George, W. B. McHenry, F. L. Bell, and R. S. Knight. 1989. Managing yellow starthistle on rangeland. California Agriculture 43:4–7. Thomsen, C.D., W.A. Williams, M. Vayssières, and F.L. Bell. 1994. Yellow starthistle control. Range Science Report 33. University of California, Davis. Thomsen, C. D., W. A. Williams, M. Vayssières, F. L. Bell, and M. R. George. 1993. Controlled grazing on annual grassland decreases yellow starthistle. California Agriculture 47:36–40. Thomsen, M. A. 2005. Ecological resistance, propagule supply and the invasion of California coastal prairie by the European grass Holcus lanatus L. PhD dissertation. University of California, Berkeley. Thomsen, M. A., J. C. Corbin, and C. M. D’Antonio. 2006a. The effect of soil nitrogen on competition between native and exotic perennial grasses from northern coastal California. Plant Ecology 186:23–35. Thomsen, M. A., C. M. D’Antonio, K. B. Suttle, and W. P. Souza. 2006b. Ecological resistance, seed density and their interactions determine patterns of invasion in a California coastal grassland. Ecology Letters 9:160–170. Thurber, G. 1880. Gramineae. Pages 253–328 in S. Watson, editor. Botany, Vol 2. U.S. Geological Survey of California, Washington, DC. Tierney, T., and J. H. Cushman. 2006. Temporal changes in native and exotic vegetation and soil characteristics following disturbances by feral pigs in a California grassland. Biological Invasions 8:1073–1089. Tietje, W., and R. H. Barrett. 1993. The wild pig in California oak woodland: Ecology and economics, Workshop Proceedings May 25 and 26, 1993. Integrated Hardwood and Range Management Program, University of California, Berkeley, San Luis Obispo. Tietje, W. D., S. L. Nives, J. A. Honig, and W. H. Weitkamp. 1991. Effect of acorn planting depth on depredation, emergence, and survival of valley and blue oak. Pages 14–20 in R. B. Standiford, editor. Proceedings of the symposium on oak woodlands and hardwood rangeland management. General Technical Report PSW-GTR-126. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Tilman, D. 1980. Resources: a graphical-mechanistic approach to competition and predation. American Naturalist 116: 362–393.

REFERENCES

367

———. 1982. Resource competition and community structure. Princeton University Press, Princeton, NJ. ———. 1983. Plant succession and gopher disturbance along an experimental gradient. Oecologia 60:285–292. ———. 1988. Dynamics and structure of plant communities. Princeton University Press, Princeton, NJ. ———. 1993. Species richness of experimental productivity gradients: How important is colonization limitation? Ecology 74:2179–2191. Timbrook, J. 1990. Ethnobotany of Chumash Indians, California, based on collections by John P. Harrington. Economic Botany 44:236–253. Timbrook, J., J. R. Johnson, and D. D. Earle. 1993. Vegetation burning by the Chumash. Pages 117–149 in T. C. Blackburn and M. K. Anderson, editors. Before the wilderness: Environmental management by Native Californians. Ballena Press, Menlo Park, CA. Timmerman, A., J. Oberhuber, A. Bacher, M. Esch, M. Latif, and E. Roeckner. 1999. Increasing El Niño frequency in a climate model forced by future greenhouse warming. Nature 398: 694–697. Tinnin, R. O., and C. H. Muller. 1971. The allelopathic potential of Avena fatua: Influence of herb distribution. Bulletin of the Torrey Botanical Club 98:243–250. ———. 1972. The allelopathic influence of Avena fatua: The allelopathic mechanism. Bulletin of the Torrey Botanical Club 99:287–292. Tisdall, J. M., and J. M. Oades. 1982. Organic matter and waterstable aggregates in soils. Journal of Soil Science 33:141–163. Tissue, D. T., R. B. Thomas, and B. R. Strain. 1993. Long-term effects of elevated CO2 and nutrients on photosynthesis and Rubisco in loblolly pine seedlings. Plant, Cell, and Environment 16:859–865. TNC. 2000. Effects of prescribed fire and cattle grazing on a vernal pool grassland landscape: Recommendations for monitoring. The Nature Conservancy, San Francisco. Torell, L. A., N. R. Rimbey, O. A. Ramirez, and D. W. McCollum. 2005. Income earning potential versus consumptive amenities in determining ranchland values. Journal of Agricultural and Resource Economics 30:537–560. Torn, M. S., E. Mills, and J. Fried. 1998. Will climate change spark more wildfire damage? LBNL-42592, Lawrence Berkeley National Laboratory, Berkeley, CA. Torne, E. 1990. Assessing feeding activities of soil-living animals. I. Bait-lamina-tests. Pedobiologia 34:89–101. Torrence, J. R., J. M. DiTomaso, and A. Hazebrook. 2003a. Largescale integrated management of yellow starthistle (Centaurea solstitialis) at Fort Hunter Liggett, Monterey County, California. Page 63 in M. Kelly, editor. Proceedings, California Exotic Pest Plant Council Symposiums [sic].Vol. 6. California Exotic Pest Plant Council, Berkeley. ———. 2003b. Long-term, large-scale, integrated management of yellow starthistle (Centaurea solstitialis) at Fort Hunter Liggett, Monterey County, California: Lower cost follow-up maintenance is vital for successful control program. Page 126 in M. Kelly, editor. Proceedings, California Exotic Pest Plant Council Symposiums [sic]. Vol. 6 California Exotic Pest Plant Council, Berkeley. Treseder, K. K., and P. M. Vitousek. 2001. Effects of soil nutrient availability on investment in acquisition of N and P in Hawaiian rain forests. Ecology 82:946–954.

368

REFERENCES

Trust for Public Land. 2005. Over $1.7 billion approved for conservation. Trust for Public Land. http://www.tpl.org/tier3_ cd.cfm?content_item_id20217&folder_id186. Tsiouvaras, C. N., N. A. Havlik, and J. W. Bartolome. 1989. Effects of goats on understory vegetation and fire hazard reduction in a coastal forest in California. Forest Science 35:1125– 1131. Turitzin, S. N. 1982. Nutrient limitations to plant growth in a California serpentine grassland. American Midland Naturalist 107:95–99. Turner, A. 1997. The big cats and their fossil relatives. Columbia University Press, New York. Turner, C. L., T. R. Seastedt, and M. I. Dyer. 1993. Maximization of aboveground grassland production-the role of defoliation frequency, intensity, and history. Ecological Applications 3:175–186. Tuskes, P. M., and J. F. Emmel. 1981. The life history and behavior of Euproserpinus euterpe (Sphingidae). Journal of the Lepidopterists’ Society 35:27–33. Twigg, L. E., T. Lowe, G. Martin, and M. Everett. 2005. Feral pigs in the north-western Australia: Basic biology, bait consumption, and the efficacy of 1080 baits. Wildlife Research 32: 281–296. Twisselman, E. C. 1967. A flora of Kern County, California. Wasmann Journal of Biology 25:1–395. Tyler, C. M. 1995. Factors contributing to postfire seedling establishment in chaparral: Direct and indirect effects of fire. Journal of Ecology 83:1009–1020. ———. 1996. Relative importance of factors contributing to postfire seedling establishment in maritime chaparral. Ecology 77(7):2182– 2195. Tyler, C. M., B. Kuhn, and F. W. Davis. 2006. Demography and recruitment limitations of three oak species in California. Quarterly Review of Biology 81:127–152. Tyler, C. M., B. E. Mahall, F. W. Davis, and M. Hall. 2002. Factors limiting recruitment in valley and coast live oak. Pages 565–572 in R. B. Standiford, D. McCreary, and K. L. Purcell, editors. Proceedings of the fifth symposium on oak woodlands: Oaks in California’s changing landscape. General Technical Report PSW-GTR-184. USDA Forest Service, Pacific Southwest Research Station, Albany, CA. Tyler, C. M., D. C. Odion, and D. Meade. 1998. Reproductive biology and the conservation of Morro Manzanita (Arctostaphylos morroensis). California Department of Fish and Game, Sacramento. Tyndall, R. W., and J. C. Hull. 1999. Vegetation, flora and plant physiological ecology of serpentine barrens of eastern North America. Pages 67–82 in J. S. Anderson, J. S. Fralish, and J. M. Baskin, editors. Savannas, barrens and rock outcrop communities of North America. Cambridge University Press, New York. Tyser, R. W., and C. H. Key. 1988. Spotted knapweed in natural area fescue grassland: An ecological assessment. Northwest Science 62:151–160. Umbanhowar, C. E. J. 1992. Reanalysis of the Wisconsin prairie continuum. American Midland Naturalist 127:268–275. Underwood, E. C. 2004. Consequences of ecological disturbance on ground dwelling invertebrates: Implications for conservation. Ph.D. dissertation, University of California, Davis. Updike, D., and J. Waithman. 1996. Dealing with wild pig depredation in California: The strategic plan. Pages 40–43 in A. C.

Crabb, editor. Proceedings of the 17th Vertebrate Pest Conference 1996. University of California, Davis. USAF. 1997. Secretary of the Air Force, Air Force Instruction 327064, Integrated Natural Resource Management. Accession Number ADA329388. http://handle.dtic.mil/100.2/ ADA329388 (accessed July 5, 2007). USDA. 1985. Prescribed fire smoke management guide. Prescribed Fire and Fire Effects Working Team National Interagency Fire Center, NFES#1279, Boise, ID. USDA Forest Service. 2007. Fire Effects Information System. Forb list. http://www.fs.fed.us/database/feis/plants/forb/. USDA-NASS. 2006. California historical data—Statistics by state. USD A National Agriculture Statistics Service, http://www.nass. usda.gov/Statistics_by_State/California/Historical_Data/. USDA NRCS. 2006. The PLANTS Database. NRCS, National Plant Data Center, Baton Rouge, LA. http://plants.usda.gov. USDA-NRCS Soil Survey Staff. 1999. Soil taxonomy: a basic system of soil classification for making and interpreting soil surveys, 2nd ed. USDA-NRCS Agricultural Handbook no. 436. Government Printing Office, Washington, DC. USDA-NRCS STATSGO. 2005. State Soil Geographic Database. http://www.ncgc.nrcs.usda.gov/products/datasets/statsgo. USDA-NRI. 2000. NRCS Natural Resource Inventory (NRI). USDA Natural Resources Conservation Service. http://www.ca.nrcs. usda.gov/technical/nri/. USDA-SCS. 1978. Soil survey of Monterey County, California. Cartographic Division, Soil Conservation Service, USDA, Washington, DC. USDA-USFS. 2005a. About rangelands. USDA Forest Service, Washington, DC. http://www.fs.fed.us/rangelands/whoweare/ index.shtml. ———. 2005b. About us—mission. USDA Forest Service, Washington, DC. http://www.fs.fed.us/aboutus/mission.shtml. ———. 2006. Annual grazing statistical summary. USDA Forest Service, Government Printing Office, Washington, DC, http:// www.fs.fed.us/rangelands/reports/index.shtml. USFWS. 1997. Draft recovery plan for the Stephens kangaroo rat, Region 1. U.S. Fish and Wildlife Service, Portland, OR. ———. 1998. Recovery plan for serpentine soil species of the San Francisco Bay Area. U.S. Fish and Wildlife Service, Portland, OR. ———. 2004. Yreka phlox recovery plan. U.S. Fish and Wildlife Service, Portland, OR. USMC. 2005. United States Marine Corps Impact Guide: Environmental Stewardship. Marine Corps Base Camp Pendleton, CA. http://www.pendleton.usmc.mil/impact/stewardship.asp (accessed July 5, 2007). Valentini, R., J. A. Gamon, and C. B. Field. 1995. Ecosystem gas exchange in a Californian grassland: Seasonal patterns and implications for scaling. Ecology 76:1940–1952. Van Auken, O. W. 2000. Shrub invasions on North America semiarid grasslands. Annual Review of Ecology and Systematics 31:197–215. Vance, C. P., C. Uhde-Stone, and D. L. Allan. 2003. Phosphorus acquisition and use: Critical adaptations by plants for securing a nonrenewable resource. New Phytologist 157: 423–447. van der Heijden, M. G. A., J. N. Klironomos, M. Ursic, P. Moutoglis, R. Streitwolf-Engel, T. Boller, A. Wiemken, and I. R. Sanders. 1998. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396: 69–72.

Van Vuren, D., and B. E. Coblentz. 1987. Some ecological effects of feral sheep on Santa Cruz Island, California, U.S.A. Biological Conservation 41:253–268. Vaughn, C. E., and A. H. Murphy. 1982. Long-term effects of fertilization and subclover seeding on northern California annual range. Journal of Range Management 35:92– 95. Vaughn, C. E., D. M. Center, and M. B. Jones. 1986. Seasonal fluctuations in nutrient availability in some northern california annual range soils. Soil Science 141:43–51. Venable, D. L., and J. S. Brown. 1988. The selective interactions of dispersal, dormancy, and seed size as adaptations for reducing risk in variable environments. American Naturalist 131: 360–384. Verville, J. H. 2000. Biological nitrogen fixation in California annual grassland: responses to elevated carbon dioxide and effects on community niotrogen status and productivity. Ph.D. dissertation, Stanford University, Stanford, CA. Via, S., R. Gomulkiewicz, G. De Jong, S. M. Scheiner, C. D. Schlichting, and P. H. van Tienderen. 1995. Adaptive phenotypic plasticity: Consensus and controversy. Trends in Ecology and Evolution 10:212–217. Vickery, J. A., J. R. Tallowin, R. E. Feber, E. J. Asteraki, P. W. Atkinson, R. J. Fuller, and V. K. Brown. 2001. The management of lowland neutral grasslands in Britain: Effects of agricultural practices on birds and their food resources. Journal of Applied Ecology 38:647–664. Vickery, P. D., M. L. Hunter, and J. V. Wells. 1999. Effects of fire and herbicide treatment on habitat selection in grassland birds in southern Maine. Studies in Avian Biology 19:149–159. Vitousek, P. M. 1986. Biological invasions and ecosystem properties: Can species make a difference? Pages 163–176 in H. A. Mooney and J. A. Drake (editors). Ecology of biological invasions of North America and Hawaii. Springer-Verlag, New York. Vitousek, P. M., and R. W. Howarth. 1991. Nitrogen limitation on land and in the sea: How can it occur? Biogeochemistry 13:87–115. Vivrette, N., and C. H. Miller. 1977. Mechanisms of invasion and dominance of coastal grasslands by Mesembryanthemum crystallinum. Ecological Monographs 47:301– 318. Voegelin, E. W. 1938. Tubatulabal ethnography. Anthropological Records 2:1–84. ———. 1942. Culture element distributions XX: Northeast California. Anthropological Records 7:47–251. Vogl, R. J. 1974. Effects of fire on grasslands. Pages 139–182 in T. T. Kozlowski and C. E. Ahlgren, editors. Fire and ecosystems. Academic Press, New York. Volk, M., P.A. Niklaus, and C. Körner. 2000. Soil moisture effects determine CO2 responses of grassland species. Oecologia 125:380–388. von Lützow M, Kögel-Knabner I, Ekschmitt K, Matzner, E, Guggenberger, G. 2006. Stabilization of organic matter in temperate soils: Mechanisms and their relevance under different soil conditions—a review. European Journal of Soil Science 57: 426–444. Vreeland, J. K., and W. D. Tietje. 2002. Numerical response of small vertebrates to prescribed fire in California oak woodland. Pages 100–110 in Prescribed fire in California oak woodland. General Technical Report NE-GTR-288. USDA Forest Service, Northeastern Research Station, Newton Square, PA.

REFERENCES

369

Wacker, M., and N. M. Kelly. 2004. Changes in vernal pool edaphic settings through mitigation at the project and landscape scale. Wetlands Ecology and Management 12:165–178. Wagnon, H. K. 1952. A revision of the genus Bromus, section Bromopsis, of North America. Brittonia 7:415–480. Waithman, J. D., R. A. Sweitzer, D. Van Vuren, J. D. Drew, A. J. Brinkhaus, I. A. Gardner, and W. M. Boyce. 1999. Range expansion, population sizes, and management of wild pigs in California. Journal of Wildlife Management 63:298–308. Waldrop, M. P., and M. K. Firestone. 2004. Microbial community utilization of recalcitrant and simple carbon compounds: impact of oak-woodland plant communities. Oecologia 138:275–284. Walker, C. F., and W. A. Williams. 1963. Responses of annual-type range vegetation to sulfur fertilization. Journal of Range Management 16:64–69. Walker, J. W. 1994. Multispecies grazing: the ecological advantage. Sheep and Goat Research Journal Special Issue: 1994:52–64. Walker, R. B. 1954. The ecology of serpentine soils: a symposium. III. Factors affecting plant growth on serpentine soils. Ecology 35:259–266. Walker, R. B., H. M. Walker, and P. R. Ashworth. 1955. Calciummagnesium nutrition with special reference to serpentine soils. Plant Physiology 30:214–221. Wallace, N. M., J. A. Leitch, and F. L. Leistritz. 1992. Economic impact of leafy spurge on North Dakota wildland. North Dakota Farm Research 49:9–13. WallisDe Vries [sic], M. F., J. P. Bakker, and S. E. Van Wieren. 1998. Grazing and conservation management. Kluwer Academic Publishers, Dordrecht, The Netherlands. Walters, C. J. 1993. Dynamic models and large-scale field experiments in environmental impact assessment and management. Australian Journal of Ecology 18:53–61. ———. 1986. Adaptive management of renewable resources. Macmillan, New York. Walters, C. J., and C. S. Holling. 1990. Large-scale management experiments and learning by doing. Ecology 71:2060–2068. Wan, S., D. Hiui, and Y. Lou. 2001. Fire effects on nitrogen pools and dynamics in terrestrial ecosystems; a meta-analysis. Ecological Applications 11:1349–1356. Ward, P. 1968. Fire in relation to waterfowl habitat of the delta marshes. Pages 255–267 in E. V. Komarek, Sr., editor. Proceedings, 8th Tall Timbers Fire Ecology Conference 1968. Tall Timbers Research Station, Tallahassee, FL. Wardle, D. A. 2006. The influence of biotic interactions on soil biodiversity. Ecology Letters 9:870–886. Wardle, D. A., R. D. Bardgett, J. N. Klironomos, H. Setala, W. H. van der Putten, and D. H. Wall. 2004. Ecological linkages between aboveground and belowground biota. Science 304: 1629–1633. Warter, J. K. 1976. Late Pleistocene plant communities—evidence from the Rancho La Brea tarpits. Pages 32–39 in J. Letting, editor. Plant communities of southern California. California Native Plant Society Special Publication 2, Berkeley. Waterhouse, D. M., and J. E. Mahoney. 1983. Effect of pre-fallowing application of glyphosate on hoary cress (Cardaria draba (L.) Desv.). Australian Weeds 2:141–143. Watson, L., and M. J. Dallwitz. 1992. The grass genera of the world. CAB International, Wallingford, UK. ———. 1999. Grass genera of the world: Descriptions, identification, and information retrieval; including synonyms, morphology, anatomy, physiology, phytochemistry, cytology,

370

REFERENCES

classification, pathogens, world and local distribution, and references. http://biodiversity.uno.edu/delta/. Weaver, J. E., and N. W. Rowland. 1952. Effects of excessive natural mulch on development, yield, and structure of native grassland. Botanical Gazette 114:1–19. Webb, S. D. 1965. The osteology of Camelops. Los Angeles County Museum Bulletin no. 1. Weber, E. 1985. Grazing management controls weeds. Soil and Water Conservation News 6:4–5. Wedin, D. A., and D. Tilman. 1996. Influence of nitrogen loading and species composition on the carbon balance of grasslands. Science 274:1720–1723. Weil, J., and R. W. Allard. 1964. The mating system and genetic variability in natural populations of Collinsia heterophylla. Evolution 18:515–525. Weiner, J., and S. C. Thomas. 1986. Size variability and competition in plant monocultures. Oikos 47:211–222. Weiss, S. B. 1999. Cars, cows, and checkerspot butterflies: Nitrogen deposition and management of nutrient-poor grasslands for a threatened species. Conservation Biology 13:1478–1486. ———. 2003. Serpentine grassland restoration at Edgewood Natural Preserve, San Mateo County. Grasslands 13:7. ———. 2006. Impacts of nitrogen deposition on California ecosystems and biodiversity. CEC-500-2005-165, California Energy Commission, PIER Energy-Related Environmental Research, Sacramento. Welker, J. M., D. R. Gordon, and K. J. Rice. 1991. Capture and allocation of nitrogen by Quercus douglasii seedlings in competition with annual and perennial grasses. Oecologia 87:459–466. Wells, P. V. 1962. Vegetation in relation to geological substrate and fire in the San Luis Obispo quadrangle, California. Ecological Monographs 32:79–103. ———. 1965. Scarp woodlands, transported grassland soils, and concept of grassland climate in the Great Plains region. Science 148:246–249. ———. 1970. Historical factors controlling vegetation patterns and floristic distributions in the Central Plains region of North America. Pages 211–221 in W. Dort and J. K. Jones, editors. Pleistocene and Recent Environments of the Central Great Plains. Department of Geology, University of Kansas Special Publication, 3. University Press of Kansas, Lawrence. Weltzin, J. F., R. T. Belote, and N. J. Sanders. 2003. Biological invaders in a greenhouse world: will elevated CO2 fuel plant invasions? Frontiers in Ecology and the Environment 1:146–153. West, N. E., and J. A. Young. 2000. Intermountain valleys and lower mountain slopes. Pages 255–284 in M. G. Barbour and W. D. Billings, editors. North American terrestrial vegetation. Cambridge University Press, New York. Wester, L. 1981. Composition of native grasslands in the San Joaquin Valley, California. Madroño 28:231–241. Westmann, W. E. 1976. Vegetation conversion for fire control in Los Angeles. Urban Ecology 2:119–137. Westoby, M., B. H. Walker, and I. Noy-Meir. 1989. Opportunistic management for rangelands not at equilibrium. Journal of Range Management 42:266–274. WFAS. 2005. Wildlands fire assessment system. USDA Forest Service. http://www.fs.fed.us/land/wfas/welcome.htm. Whelan, J. R. 1995. The ecology of fire. Cambridge University Press, Cambridge, UK. Whicker, A., D., and J. K. Detling. 1988. Ecological consequences of prairie dog disturbances. Bioscience 38:778–785.

White, K. L. 1966. Structure and composition of foothill woodland in central coastal California. Ecology 47:229–237. ———. 1967. Native bunchgrass (Stipa pulchra) on Hasting Reservation, California. Ecology 48:949–955. White, P. S. 1979. Pattern, process, and natural disturbance in vegetation. Botanical Review 45:220–299. White, P. S., and J. L. Walker. 1997. Approximating nature’s variation: Selecting and using reference information in restoration ecology. Restoration Ecology 5:338–349. White, R. 1975. Indian land use and environmental change: Island County, Washington, a case study. Arizona and the West 17:327–338. White, R., S. Murray, and M. Rohweder. 2000. Pilot analysis of global ecosystems: Grassland ecosystems. Research Report 1-56973-461-5, World Resources Institute, Washington, DC. Whitford, W. G., and F. R. Kay. 1999. Bioturbation by mammals in deserts: A review. Journal of Arid Environments 41:203–230. Whitson, T. D., A. Mooney, and M. Griswold. 1989. Control of hoary cress (Cardaria draba) with various herbicides. Proceedings Western Society of Weed Science 42:39. Whittaker, R. H. 1975. Communities and ecosystems. Collier MacMillan, London. Wiens, J. A. 1984. On understanding a non-equilibrium world: Myth and reality in community patterns and processes. Pages 439–457 in D. R. Strong, D. Simberloff, K. G. Abele, and A. B. Thistle, editors. Ecological communities: Conceptual issues and the evidence. Princeton University Press, Princeton, NJ. Wigand, P. E. 1987. Diamond Pond, Harney County, Oregon: Vegetation history and water table in the eastern Oregon Desert. Great Basin Naturalist 47:427–458. ———. In press, a. Estuarine evolution, chaparral history, and changing plant resource availability. in J. A. Homburg and M. Palacios-Fest, editors. Paleoenvironments of the Ballona SRI Technical Series, Statistical Research, Tucson, AZ. ———. In press, b. Sites CAL-LAN-63, CAL-LAN-64, CAL-LAN-206 (Del Rey Bluffs, Southern California): pollen and macrobotanical analysis. In J. Douglass, editor. SRI Technical Series, Statistical Research, Tucson, AZ. Wigand, P. E., and D. Rhode. 2002. Great Basin vegetation history and aquatic systems: The last 150,000 years. Pages 309–367 in R. Hershler, D. B. Madsen, and D. R. Currey, editor. Great Basin aquatic systems history. Smithsonian Contributions to Earth Sciences 33. Smithsonian Institution Press, Washington, DC. Willett, I. R., and T. Batey. 1977. The effects of metal ions on the root surface phosphatase activity of grasses differing in tolerance to serpentine soil. Plant and Soil 48:213–221. Williams, D. F., S. Byrne, and T. A. Rado, editors. 1992. Endangered and sensitive species of the San Joaquin Valley, California. California Energy Commission, Sacramento. Williams, D. F., and K. S. Kilburn. 1991. Dipodomys ingens. Mammalian Species 377:1–7. Williams, J. W., E. W. Seabloom, D. Slayback, D. M. Stoms, and J. H. Viers. 2005. Anthropogenic impacts upon plant species richness and net primary productivity in California. Ecology Letters 8:127–137. Williams, K., and R. J. Hobbs. 1989. Control of shrub establishment by springtime soil water availability in an annual grassland. Oecologia 81:62–66. Williams, K., R. J. Hobbs, and S. P. Hamburg. 1987. Invasion of an annual grassland in northern California by Baccharis pilularis ssp. consanguinea. Oecologia 72:461–465.

Williamson, J., and S. Harrison. 2002. Biotic and abiotic limits to the spread of exotic revegetation species. Ecological Applications 12:40–51. Williamson, M. H., and A. Fitter. 1996. The characters of successful invaders. Biological Conservation 78:163–170. Williamson, S. C., J. K. Detling, J. L. Dodd, and M. I. Dyer. 1989. Experimental evaluation of the grazing optimization hypothesis. Journal of Range Management 42:149–152. Wills, R. 2000. Effective fire planning for California grasslands. Pages 75–78 in J. E. Keeley, M. Baer-Keeley, and C. J. Fotheringham, editors. 2nd interface between ecology and land development in California. USGS Open-File Report 00-62. U.S. Geological Survey, Denver, CO. Wilson, C. G. 1955. Alice Eastwood’s wonderland: The adventures of a botanist. California Academy of Sciences, San Francisco. Wilson, M. V., and D. L. Clark. 2001. Controlling invasive Arrhenatherum elatius and promoting native prairie grasses through mowing. Applied Vegetation Science 4:129–138. Winsome, T., and P. Hendrix. 2000. Earthworm ecology in California. Oaks ’n’ Folks 15. Winsome, T., W. R. Horwath, L. Epstein, and P. F. Hendrix. 2003. Functional differences between exotic and native earthworm species in a California grassland. Ecological Society of America Annual Meeting Abstracts 88:361. Wohlgemuth, E. 2004. Nine thousand years of plant use in native central California: Implications of the archaeobotanical record for archaeologists, native peoples, and restoration practitioners. Paper presented at the 27th Annual Ethnobiology Conference 2004. University of California, Davis. Wolfe, L. M. 1938. John of the mountains: the unpublished journals of John Muir. University of Wisconsin Press, Madison. Wolters, G. L., and G. P. Eberlein. 1986. Sulfur or sulfur plus nitrogen increases beef production on California annual range. Journal of Range Management 39:125–129. Wood, A. W., L. R. Leung, V. Sridhar, and D. P. Lettenmaier. 2004. Hydrologic implications of dynamical and statistical approaches to downscaling climate model outputs. Climatic Change 62:189–216. Wood, Y. A., T. Meixner, P. J. Shouse, and E. B. Allen. 2006. Altered ecohydrologic response drives native shrub loss under conditions of elevated nitrogen deposition. Journal of Environmental Quality 35:76–92. Woodburne, M. O., and C. S. Swisher. 1995. Land mammal highresolution geochronology, intercontinental overland dispersals, sea level, climate, and vicariance. Pages 335–364 in W. A. Berggren, D. V. Kent, M.-P. Aubry, and J. Hardenbol. Geochronology, time scales, and global stratigraphic correlation. SEPM Special Publication 40054 2073946.Society for Sedimentary Geology, Tulsa, OK. Woodmansee, R. G. 1978. Additions and losses of nitrogen in grassland ecosystems. Bioscience 28:448–453. Woodmansee, R. G., and D. A. Duncan. 1980. Nitrogen and phosphorus dynamics and budgets in annual grasslands. Ecology 61:893–904. Woods, D. M., D. B. Joley, M. J. Pitcairn, and V. Popescu. 2004. Impact of biological control insects on yellow starthistle at one site in Yolo County. Pages 35–37 in D. M. Woods, editor. Biological Control Program annual summary, 2003. California Department of Food and Agriculture, Plant Health and Pest Prevention Services, Sacramento.

REFERENCES

371

Woods, D. M., and B. Villegas. 2004. Continued success with biological control agents of squarrose knapweed. Pages 27–28 in D. M. Woods, editor. Biological Control Program annual summary, 2003. California Department of Food and Agriculture, Plant Health and Pest Prevention Services, Sacramento. Woods, L. E. 1989. Active Organic matter distribution in the surface 15 cm of undisturbed and cultivated soil. Biology and Fertility of Soils 8:272–278. Woodward, F. I., M. R. Lomas, and C. K. Kelly. 2004. Global climate and the distribution of plant biomes. Philosophical Transactions of the Royal Society of London Series B Biological Sciences 359:1465–1476. Woolfenden, W. B. 1993. Pollen present in cores OL-92-1, 2 and 3. Pages 312, 314–332, in G. I. Smith and J. L. Bischoff, editors. Core OL-92 from Owens Lake, Southeast California. USGS Open-File Report 93-683. United States Geological Survey, Denver, CO. ———. 2003. A 180,000-year pollen record from Owens Lake, CA: terrestrial vegetation change on orbital scales. Quaternary Research 59:430–444. Woolfolk, J. J., and D. A. Duncan. 1962. Fertilizers increase range production. Journal of Range Management 15:42–45. WRCC. 2004. Classification of El Niño and La Niña winters. http:// www.wrcc.dri.edu/enso/ensodef.html (accessed November 6, 2005). Wright, H. A., and A. W. Bailey. 1982. Fire ecology of the United States and southern Canada. John Wiley and Sons, New York. Wright, H. A., and J. O. Klemmedson. 1965. Effects of fire on bunchgrasses of the sagebrush region of southern Idaho. Ecology 46:680–688. Wrysinski, J. 2002. Roadside establishment of native perennial grasses. Pages 19–27 in P. Robins, editor. Bring farm edges back to life. Yolo County Resource Conservation District, Woodland, CA. Wu, L., and S. Jain. 1980. Self-fertility and seed set in naturalpopulations of Anthoxanthum odoratum L. Botanical Gazette 141(3):300 – 304. Wu, J., and S. A. Levin. 1994. A spatial patch dynamic modeling approach to pattern and process in an annual grassland. Ecological Monographs 64:447–464. Wu, K. K., and S. K. Jain. 1978. Genetic and plastic responses in geographic differentiation of Bromus rubens populations. Canadian Journal of Botany 56:873–879. Wu, Z. L., and P. M. Peterson. 2006. Muhlenbergia. Pages 486–487 in Z. Y. Wu, P. H. Raven, and D. Y. Hong, editors. Flora of China: Poaceae, volume 22. Science Press, Beijing, and Missouri Botanic Garden, and St. Louis, MO. Wyland, L. J., L. E. Jackson, W. E. Chaney, K. Klonsky, S. T. Koike, and B. Kimple. 1996. Altering surface soil dynamics with winter cover crops in a vegetable cropping system: impacts on yield, nitrate leaching, pests and management costs. Agriculture, Ecosystems and Environment 59:1–17. Xiong, S., and C. Nilsson. 1997. Dynamics of leaf litter accumulation and its effects on riparian vegetation: A review. Botanical Review 63:240–264. ———. 1999. The effects of plant litter on vegetation: A metaanalysis. Journal of Ecology 87:984–994. Yazdi-Samadi, B., K. K. Wu, and S. K. Jain. 1978. Population biology of Avena. VI. Role of genetic polymorphisms in outcome of interspecific competition. Genetica 48:188–191.

372

REFERENCES

Yoakum, J. 1980. Habitat management guides for the American pronghorn antelope. Technical Note 347, US Department of Interior, Bureau of Land Management, Denver, CO. Yoakum, J. D., and B. W. O’Gara. 2000. Pronghorn. Pages 559–577 in S. Demaris and P. Karusman, editors. Ecology and management of large mammals in North America. Prentice-Hall, Upper Saddle River, NJ. York, A. 2000. Long-term effects of frequent low-intensity burning on ant communities in coastal blackbutt forests of southeastern Australia. Austral Ecology 25:83–98. Yoshida, L. C., and E. B. Allen. 2001. Response of ammonium and nitrate by a mycorrhizal annual invasive grass and a native shrub in southern California. American Journal of Botany 88:1430–1436. ———. 2004. N-15 uptake by mycorrhizal native and invasive plants from a N-eutrophied shrubland: A greenhouse experiment. Biology and Fertility of Soils 39:243–248. Young, J. A. 1992. Ecology and management of medusahead (Taeniatherum caput-medusae ssp. asperum [Simk.] Melderis). Great Basin Naturalist 52:245–252. Young, J. A., and C. D. Clements. 1999. Native perennial grass communities of the Carson Desert of northwestern Nevada. Pages 133–136 in E. D. McArthur, W. K. Ostler, and C. L. Wamboldt, editors. Proceedings: Shrubland ecotones 1999. RMRS-P-11. USDA Forest Service, Rocky Mountain Research Station, Ogden, UT. Young, J. A., C. D. Clements, and H. C. Jansen. In press. Sagebrush steppe. Pages in press in M. G. Barbour, T. KeelerWolf, and A. Schoenherr, editors. Terrestrial vegetation of California. 3rd edition. University of California Press, Berkeley. Young, J. A., and R. A. Evans. 1978. Population dynamics after wildfires in sagebrush grasslands. Journal of Range Management 31:283–289. ———. 1989. Seed production and germination dynamics in California annual grasslands. Pages 39–45 in L. F. Huenneke and H. A. Mooney, editors. Grassland structure and function: California annual grassland. Kluwer Academic Publishers, Dordrecht, The Netherlands. Young, J. A., R. A. Evans, and B. L. Kay. 1970. Phenology of reproduction of medusahead. Weed Science 18:451–454. ———. 1971. Germination of annual grasses in simulated litter. Agronomy Journal 63:656–659. Young, J. A., R. A. Evans, C. A. Raguse, and J. R. Larson. 1981. Germinable seeds and periodicity of germination in annual grasslands. Hilgardia 49:1–37. Young, J. A., R. A. Evans, and J. Robison. 1972. Influence of repeated annual burning on a medusahead community. Journal of Range Management 25:372–375. Young, J. A., D. E. Palmquist, and R. R. Blank. 1998. The ecology and control of perennial pepperweed (Lepidium latifolium L.). Weed Technology 12:402–405. Young, T. P. 2000. Restoration ecology and conservation biology. Biological Conservation 92:73–83. Young, T. P., D. A. Petersen, and J. J. Clary. 2005. The ecology of restoration: Historical links, emerging issues and unexplored realms. Ecology Letters 8:662–673. Zammit, C. A., and P. H. Zedler. 1988. The Influence of dominant shrubs, fire, and time since rire on soil seed banks in mixed chaparral. Vegetatio 75:175–187.

———. 1993. Size structure and seed production in even-aged populations of Ceanothus greggii in mixed chaparral. Journal of Ecology 81:499–511. ———. 1994. Organization of the soil seed bank in mixed chaparral. Vegetatio 111:1–16. Zavaleta, E. S. 2006. Shrub establishment under experimental global changes in a California grassland. Plant Ecology 184:53–63. Zavaleta, E. S., M. R. Shaw, N. R. Chiariello, H. A. Mooney, and C. B. Field. 2003a. Additive effects of simulated climate changes, elevated CO2, and nitrogen deposition on grassland diversity. Proceedings of the National Academy of Sciences 100:7650–7654. Zavaleta, E. S., M. R. Shaw, N. R. Chiariello, B. D. Thomas, E. E. Cleland, C. B. Field, and H. A. Mooney. 2003b. Grassland responses to three years of elevated temperature, CO2, precipitation, and N deposition. Ecological Monographs 73:585–604. Zavaleta, E. S., B. D. Thomas, N. R. Chiariello, G. P. Asner, M. R. Shaw, and C. B. Field. 2003c. Plants reverse warming effect on ecosystem water balance. Proceedings of the National Academy of Sciences USA 100:9892–9893. Zavon, J. A. 1982. Grazing and fire effects on annual grassland composition and sheep diet selectivity. M.S. thesis, University of California, Davis. Zedler, P. H., and C. Black. 1992. Seed dispersal by a generalized herbivore: rabbits as dispersal vectors in a semiarid California vernal pool landscape. American Midland Naturalist 128:1–10.

Zedler, P. H., C. R. Gautier, and G. S. MacMaster. 1983. Vegetation change in response to extreme events: the effect of a short interval between fires in California chaparral and coastal scrub. Ecology 64:809–818. Zeiner, D. C., W. F. Laudenslayer, and K. E. Mayer, editors. 1988. California’s wildlife. Vol. 1. Amphibians and reptiles. California Department of Fish and Game, Sacramento. Zeiner, D. C., W. F. Laudenslayer, K. E. Mayer, and M. White, editors. 1990. California’s wildlife. Vol. 2. Birds. California Department of Fish and Game, Sacramento. Zelles, L. 1997. Phospholipid fatty acid profiles in selected members of soil microbial communities. Chemosphere 35:275–294. Zigmond, M. L. 1978. Kawaiisu basketry. Journal of California Anthropology 5:199–215. ———. 1981. Kawaiisu ethnobotany. University of Utah Press, Salt Lake City. Zimmerman, J. L. 1997. Avian community responses to fire, grazing, and drought in the tallgrass prairie. Pages 167–180 in F. L. Knoph and F. B. Samson, editors. Ecology and conservation of Great Plains vertebrates. Springer-Verlag, New York. Ziska, L. H. 2003. Evaluation of the growth response of six invasive species to past, present, and future atmospheric carbon dioxide. Journal of Experimental Botany 54: 395–404. Zohary, D., and M. Hopf. 2000. Domestication of plants in the Old World. 3rd edition. Oxford University Press, Oxford.

REFERENCES

373

INDEX

abiotic conditions, 156 abiotic gradients, adaptation to, 142–143 abundance and climate shifts, 46 decreases in, 39 influence of precipitation on, 37, 44, 46 long- and short-term scales of, 42 and rising CO2 concentrations, 224–225 of seedlings, 92 species shifts in, 26 Acanthomintha duttoni, 153 Achnatherum speciosum alliance, 32–33 acorns dispersal of, 173–174 feral pigs and, 194 predators, 173 production, 173 active restoration strategy, 271 adaptation, 132 to abiotic gradients, 142–143 to biotic selection, 143–144 coastal versus inland climatic regimes, 143 local, 138, 139, 141, 273, 278, 280 coadapted gene complexes and, 139–141 genotypic variation paralleling, 139 and molecular genetic markers, 138 phenotypic plasticity and, 141–142 and serpentine soils, 142–143 adaptive management, 240, 251 adaptive phenotypic plasticity, 141 adaptive weed management, 284 Adenostoma fasciculatum, 170, 172 aesthetic improvements, 268 agave, 61 aggressiveness, defined, 81

Agoseris heterophylla, 151 agriculture, 107 conversion of land to, 3, 154 crop rotations, 75 history of, 75 and invasive non-native species, 72, 82 pests, 189 Agropyron desertorum, 164 Agrostis stolonifera, 274 Aira cayrophyllea, 157, 194, 201 air quality requirements, 216 alfisols, 108, 109 alien species, 67 Allard, Robert, 131 allee effects, 132 allelopathic compounds, 172 alliance, defined, 23 allozyme heterozygosity, 139 allozyme markers, 133, 142 American cheetah (Miracinonyx trumani), 50 American Farmland Trust, 308 American lion (Panthera leo atrox), 50 American mastodon (Mammut americanum), 51 aminopyralid, 290, 291, 292 Ammophila aernaria, 276 Amsinckia tessellate, 185 Andropogoneae, 12 angelica (Angelica tomentosa), 63 Animal and Plant Health Inspection Service (APHIS), U.S. Department of Agriculture, 286 annual grasses, 156–157 density of, 157 dry-down, 244 early observations of, 53 exotics, domination by, 72, 76

germination, 79 interaction of with perennial species, 1 longevity, 164 microbial communities, 256 rainfall and, 87 replacement of with perennial grasses, 256 root mass, 79 self-thinning, 157 anthropogenic ignitions, conversion of shrubland to grassland, 170–171 anthropogenic nitrogen deposition, 320 ants Argentine ants (Linepithema humile), 153, 186 seed-harvesting ants (Messor Andrei), 151–152, 186 arbuscular mycorrhizae (AM) fungi, 104–105, 224, 272 Arctostaphylos spp., 170, 172 Arctostaphylos bakeri bakeri, 153 Argentine ants (Linepithema humile), 153, 186 Aristida spp. Aristida oligantha, 91 Aristida purpurea, 33 Arrhenatherum elatisu, 30 arroyo lupine (Lupinus succulentus), 135, 138 Artemisia californica, 171, 172 artichoke thistle (Cynara cardunculus), 281 association, 23 Astragalus, 211 atmosphere and climate, 218–229 global warming, 218, 222–223 human activities, influence of, 218 modeling grassland responses, 227–229 multiple global changes community responses to, 225–226 ecosystem responses to, 226–227

375

potential of invasive species to interact with environmental changes, 220–221 predicted changes in fire regime, 219–220 in nitrogen deposition, 220 in temperature and precipitation, 218–219 responses to global environmental changes, 221–227 biomass and allocation changes, 224 community composition, abundance, and diversity, 224–225 expected in California valley-type grasslands, 228 Jasper Ridge Global Change Experiment, 225–227 leaf-level responses, 223–224 microbial communities, 224 nitrogen deposition, 221–222 precipitation change, 222 rising atmospheric CO2 concentrations, 223–225 warming, 222–223 axial precession, 38 baby blue-eyes (Nemophila menziesii), 136, 139, 142 back fires, 208, 211 badger (Taxidea taxus), 183 Baeria chrysotoma, 201 Baker, Herbert, 131 baling, 275 barbed goatgrass (Aegilops triumcialis), 101, 133–134, 149, 152, 154, 155, 163, 203, 212, 281, 293, 320 barley yellow dwarf virus (BYDV), 82, 256 basketry, 59–60, 209 bast fibers, 61 Bay checkerspot butterfly (Euphydryas editha bayensis), 153, 221, 222, 310 bears black (Ursus americanus), 49 grizzly (Ursus arctos), 49, 55, 184–185 short-faced (Arctodus simus), 49 bees, in serpentine grasslands, 152 beet leafhopper (Eutettix tenellus), 189 beetles Eremosaprinus spp., 185 Geomysaprinus spp., 185 BEHAVE/BEHAVE-Plus, 216 bighorn sheep (Oreamnos americanus), 51 big tarweed (Blepharizonia plumose), 139 biodiversity credits, 309 effects of human activities on, 107 native, restoration of, 1, 320 biological control control agents approved for use in California, list of, 288–289 of exotic plants, 286–290 insects, 296 plants targeted for, list of, 287 rationale for, 82

376

INDEX

biomass accumulation, 240 changes, and rising CO2 concentrations, 224 removal, responses to, 245 peak aboveground, 122 peak plant, 99, 100 peak standing, 204 biomes, world types in relation to precipitation and temperature, 178 biotic interactions, 156 biotic resistance hypothesis, 81–82 biotic selection, adaptation to, 143–144 birds. See also raptors effects of fire on, 213 horned lark (Eremophila alpestris), 186, 213 mountain plover (Charadrius montanus), 188, 213 mourning dove (Zenaida macroura), 186 western kingbird (Tyrannus verticalis), 189 western meadowlark (Sturnella neglecta), 189 western scrub jay (Aphelocoma coerulescens), 173 white-tailed kite (Elanus leucurus), 184 bison, 44, 48 giant, (Bison latifrons), 51 Ice Age (Bison antiquus), 51 black bear (Ursus americanus), 49 black mustard (Brassica nigra), 214 blessed milk thistle (Silybum marianum), 178, 289 blue-eyed grass (Sisyrinchium bellum), 61 blue oak (Quercus douglasii), 78–79, 173, 176–178, 222 blue wild rye (Elymus glaucus), 62, 135, 149, 167, 245, 279, 273 blunt-nosed leopard lizards (Gambelia silus), 185, 188, 302 bobcat (Lynx rufus), 50 Botta’s pocket gopher (Thomomys bottae), 180, 211 boundary-crossing habitat management, 252 Bouverie Preserve, 239 Brachypodium distachyon, 150 breeding systems, 132 gene flow and, 135–136 genetics of inbreeding, 136–138 inbreeding effects, 135 variable outcrossing, 138 Brewer, William, 73 Briza minor, 194 broadcast seeding, 275 Brodiaea spp., 152 Brodiaea coronaria, 27 Bromeae, 13 bromes (Bromus spp.), 141, 152 California (Bromus carinatus), 62, 167 downy (Bromus tectorum), 1, 281, 283, 293, 294 eradicating, 179 Japanese (Bromus japonicus), 212 red (Bromus madritensis), 149, 187, 199, 212, 281, 293

ripgut (Bromus diandrus), 149, 156, 164, 175, 177, 178, 199, 211, 214, 286, 293 soft chess (Bromus hordeaceous), 27, 30, 125, 137, 141, 142, 143, 146, 147, 149, 150, 152, 154, 156, 177, 178, 201, 222 brooms French (Genista monspessulana), 80, 171, 215, 250, 281, 285 Scotch (Cystisus scoparius), 80, 171, 172, 215, 250, 281, 285 browsing by megafauna, effects of, 51–52 and oak recruitment, 174 brush clearing, 170 bulbs, corms, and tubers, 63 burning, 66, 110, 159, 170. See also fire(s) for grassland management, 215–216 intervals between, 171 for invasive plant control, 292–293 permits, 209 prescribed, 166, 167, 168, 203, 216, 276, 277, 292 for rangeland improvement, 209 reburn, 170, 171 in serpentine grasslands, 151 species most susceptible to, 293 for weed control, 272 burrowing, 53, 75, 201, 212 animals, effects of fire on, 213 effects of on vegetation, 185–186 by feral pigs, 185 by foxes, 184 by grizzly bears, 184–185 historical accounts of, 181–182 by raptors, 184 by reptiles, 184 by rodents, 180–184, 186 and oak recruitment, 174 recolonization, 213 secondary users, 185 burrowing owl (Athene cunicularia), 184, 185, 213 buttercups (Ramunculus californicus), 62, 245 butterflies, 277 Bay checkerspot (Euphydryas editha bayensis), 153, 221, 222, 310 effects of fires on, 213–214 Fender’s blue (Icaricia icarioides fenderi), 214 Calcium (Ca), 100 -magnesium ratio (Ca:Mg), 146, 150, 154 California brome (Bromus carinatus), 62, 167 California Coastal Act, 304, 306 California condor (Gymnogyps californianus), 302, 310 California Crop Improvement Association seed source and founder population certification by, 273 California Department of Fish and Game (CDFG), 304

California Department of Transportation, native grass database and restoration guidelines, 279 California Endangered Species Act (CESA), 200, 304 grassland-associated species listed under, 304, 312–314 Natural Community Conservation Plans (NCCPs), 302, 304 California Environmental Quality Act (CEQA), 215, 271, 304 county implementation of, 299, 306–307 California grasslands annual grass-dominated valley grassland, 198 classification of, 198 components of, 156–157 annual and perennial forbs, 157 annual grasses, 156–157 distribution of, 235, 297, 298, 319 perennial grasses, 157 human migration to, 57 list of, by county, 315 and North American grasslands compared, 1 ownership and management profile, 316–317 perennial grass-dominated coastal prairie, 198 subdivisions, 68 California Invasive Plant Council (CAL-IPC), 67, 68, 83 Invasive Plant Inventory, 14 Invasive Plants of Greatest Ecological Concern in California’s Wildlands, 281 California Land Conservation Act of 1965. See Williamson Act California native grasses, database of, 273 California Native Grasslands Association (CNGA), 255, 320 California Native Plant Society (CNPS), 307 Inventory of Rare and Endangered Plants, 305 California Natural Diversity Database, 271 California oak woodlands, distribution of, 235 California oatgrass (Danthonia californica), 30, 157, 193, 202, 210, 212, 245, 249 California Partners in Flight Grassland Bird Conservation Plan, 213 California poppy (Eschscholzia californica), 136, 139, 143, 145,149, 157 California Rangeland Trust, 239, 308, 311 Calmagrostis ophiditis, 149 Calycadenia multiglandulosa, 152, 186 Cammisonia campestris, 189 Canada thistle (Cirsium arvense), 286, 290 canopy structure, impact of exotic plants on, 283 carbon (C) credits, 309 and fire, effects of, 208 fixation, reduced, 165 gain, via photosynthesis, 199

limitation, 121 -nitrogen ratio (C:N), 276 sequestration, 107–118, 218 supplementation, 276 uptake, 153 carbon dioxide (CO2) concentration, 218 emissions, 216 enrichment experiments, 153 and productivity, 92 Carmel Valley, 111 Carrizo Plain Ecological Reserve, 304 caryopsis, 7 Castilleja spp. Castilleja affinis neglecta, 153 Castilleja attenuate, 27 Castilleja densiflora, 153. 186 cattle, 171, 174, 244, 250, 277 grazing, 239, 241 ranches, land grants to, 198 Ceanothus spp., 170 Ceanothus ferrisiae, 153 Ceanothus greggii, 172 Centaurium venustum, 27 cereal yellow dwarf virus (CYDV), 82 Cerocarpus betuloides, 171 Chaenactis glabriuscula, 150 chance colonization, 133 chia (Salvia columbariae), 62 Chinese houses (Collinsia hetrophylla), 138 Chloridoideae, 9, 11, 13–14 chlorsulfuron, 290, 291 chromosome counts, 132 Cicendia quadrangularis, 27 Cirsium spp. (thistles), 157 Canada thistle (Cirsium arvense), 286, 290 classification and regression tree (CART) analysis, 201 Clean Water Act (CWA), 299, 302 Clements, Frederic E., 53, 55, 72, 198 climate. See also atmosphere and climate coastal versus inland regimes, 143 conditions, influence of, 1, 26 and grassland species distribution, hypothesis of, 53, 55 as a predictor of grassland productivity, 91 regional differences in, 45–46 variations in, year-to-year, 88 climate change, 3, 320 effects of on community composition and productivity, 92, 93 climax community concept, 198 clipping, to control invasive plants, 285–286 clopyralid, 290, 291, 292, 294–295 clovers (Trifolium spp.), 63, 125, 157, 158, 211 rose clover (Trifolium hirtum), 134, 135 showy Indian clover (Trifolium amoenum), 139 Trifolium bifidum, 27 Trifolium gracilentum, 214 Trifolium incarnatum, 293 Trifolium microcephalum, 212 Trifolium subterraneum, 293 Trifolium willdenovii, 27

coadapted gene complexes, 139–141 coast live oak (Quercus agrifolia), 173, 175, 178 cold desert grassland, 31–32 co-limitation, defined, 120 colonization by chance, 133 community responses to, 79 genetic bottlenecks, role of, 133–134 genetics of, 131–135 invasive species, origins of, 131–134 recolonization, 170, 213 Columbian mastodon (Mammuthus columbi), 51 community composition changing, 281 and climate shifts, 46 disturbance, effects of, 185 effects of fire on, 211–212 litter, influence of, 166 pre-cultivation, 114 pre-fire and diversity of, 214 responses to multiple global changes, 225–226 and rising CO2 concentrations, 224–225 soil moisture patterns, influence of, 92, 93 community types, 33 compaction, 199 compensatory growth, 199 competition, 119, 157–168 aboveground productivity, 165–166 annual grasses versus annual and perennial forbs, 158 defined, 156 edaphic sites serpentine soils, 163 vernal pools, 163 exotic annual grasses versus native perennial grasses, 158–162 interior grasslands, 158–159, 162 coastal grasslands, 162 from exotic annual species, 276 fire, effects of, 167, 168, 217 genetic polymorphism, effects of, 143 grazing, 167, 168 incumbency advantage, 164 from invasive non-native species, 77–78 life history characteristics, 163–166 longevity, 164 management strategies to alter competitive outcomes, 166–168 negative versus positive trees, 176 nitrogen use, 166 oaks and grasslands, 176–178 oaks with grasses, 175 perennial grasses versus annual and perennial forbs perennial grasses versus perennial grasses, 162–163 perennial lifestyle as an advantage, 159 physical barriers, 166 and reproduction, 144 resource, 127, 163 rooting patterns, 164–165 seed addition, 167–168

INDEX

377

seedlings, 168 in serpentine grasslands, 150 summer activity, 164 competitive dominance, 156 competitive interactions, evolution of, 143–144 competitive outcomes management strategies to alter, 166–168 measurement of, 156 competitive superiority, 156 condiments, from plants, 61 conservation. See also regulation and conservation biodiversity credits, 309 carbon credits, 309 direct funding for, 309 easements, 238, 239, 244, 299 ecosystem services incentives, 309 grass banking, 311 grassland reserves, establishment of, 310 incentives, increasing awareness of, 310 Natural Resource Conservation Service (NRCS), 309 permits, 309 projected development in California, 310 public financing, 309 socioeconomic benefits of, 308–309 tax incentives for, 309–310 technical assistance, 309 zoning for, 305 Conservation Reserve Program (CRP), 111, 302 consumer-controlled ecosystems, 179 controlled experimentation, 239 cordage, 61 Cordylanthus tenuis capillaries, 153 “coup de grass,” 189 coyote (Canis latrans), 49, 183 coyote brush (Baccharis pilularis), 78, 171, 222, 250 Crespi, Juan, 52, 53, 57, 66 crop rotations, 75 cultivation, 55. See also agriculture; tillage Cynodonteae, 15 Cynosurus echinatus, 30, 194 damaging species, 67 Daucus pusillus, 27 Dechampsia caespitosa, 144 decomposition, rates of, 189 deep soil water, 79–80 deergrass (Muhlenbergia rigens), 60 deer mice (Peromyscus spp.), 213 defoliation, 198–199, 202, 244 determining the effects of, 199 frequency and timing of, 199 intensity, 198, 202 responses to, 198–199 selective, 205 under-/over-compensation for, 199 Delphininum nudicaule, 61 diagnostic species 23 Dichelostemma capitata, 151, 152

378

INDEX

diffuse knapweed (Centaurea diffusa), 288 disease defined, 81 facilitators, 75, 82 response to, 144 viruses, 82, 256 disturbance, 26, 180–186, 191 anthropogenic, 169 burrowing animals, 53, 75, 180–184, 185–186 and conversion from shrubland to grassland, 169 gophers, 26, 111, 152, 171, 172 feral pigs, 185 foxes, 184 grizzly bears, 184–185 hawks and owls, 184 importance of in ecological system structure, 191 invertebrates, 115 irrigation as, 114 natural, 66 organisms responsible for, 105 rainfall as, 114 road construction, 256 secondary burrow users, 185 snakes, 184 soil microbial community response to, 113–114 soil minimization before restoration, 272 and spread of exotic species, 191 systematic, 66 diversity California Natural Diversity Database, 271 impact of exotic plants on, 283 influence of precipitation on, 37 measures of, 278 remnant California grasslands, 255 and rising CO2 concentrations, 224–225 dogbane (Apocynum cannabinum), 61 domesticated crops, 64 dominant species, 149 dormancy, 91, 141–142, 211 downy brome (Bromus tectorum), 1, 281, 283, 293, 294 Drake, Sir Francis, 190 drill seeding, 275 drought, 72, 77, 82, 92, 158, 200, 219 during the growing season, 92 mid-winter, 88 multiyear, 89 stress, 204 summer, 87, 90, 119, 127, 164, 174, 200, 222 susceptibility to, 165 Drucker, Philip, 59 dry climate, 90 Dudleya setchelli, 153 dust, 286 Dust Bowl, 37 dwarf pronghorn (Capromeryx minor), 50 dynamic global vegetation models (DGVM), 227 MAPSS-CENTURY (MC1), 227

early vegetation, observers’ accounts of, 52 earthworms, 105 easements, 238, 239, 244, 299 eccentricity, 38 ecological interactions, 3 Ecological Society of America, Vegetation Classification Panel, 23 ecological succession, 256 ecosystem-level productivity, predictive modeling of, 153 ecosystems consumer-controlled hypothesis, 179 developing healthy and functional, 296 structure, impact of exotic plants on, 283 eddy covariance technique, 153 El Niño/Southern Oscillation (ENSO) events, 89, 219, 276 elk (Cervus elaphus), 50, 183 Elymus spp., 152 Elymus elymoides, 149–150 Elymus glaucus, 62, 91, 135, 149, 167, 245, 279, 273 Elymus multisetus, 149, 275 endangered species. See rare and endangered species Endangered Species Act (EDA), 298, 299, 301–302, 303 exemptions, permits, and exceptions, 301 habitat conservation plans (HCPs), 301–302, 303 plant protection, 301 endemic plants, serpentine grasslands, 147, 153, 154 enemy release hypothesis, 81 entisols, 108, 109 Environmentally Sensitive Habitat Areas (ESHAs), 306 Equus spp. Equus conversidens, 51 feral burro (Equus asinus), 191 feral horse (Equus caballus), 191 western horse (Equus cf. occidentalis), 51 Eremocarpus setigerus, 91 Eriogonum fasciculatum, 150 Erodium spp. (filaree), 125, 157, 158, 212 Erodium botrys, 27, 142, 188, 211 Erodium brachycarpum, 142, 143, 211 Erodium cicutarium, 143, 165, 185, 186, 189, 245 erosion, 94, 199, 310 control, 268, 275–276, 279 following tillage, 286 impact of exotic plants on, 284 Eryngium aristulatum, 27 European settlement invasive plants following, 52 species composition at, 52–56 vegetation, observers’ accounts of, 52 evaporation rate, 37, 38, 41

evapotranspiration, 88, 108, 108, 223 exotic plants, 67, 281–296. See also invasive non-native species annual grasses, average pre- and post-fire cover, 212 biological control, 286–290 control agents approved for use in California, list of, 288–289 plants targeted for, list of, 287 biomass, 193 chemical control, 290–292 commonly used herbicides, list of, 291 cultural control, 292–293, 295 grazing, 292 prescribed burning, 292–293 revegetation, 293, 295 detection and rapid response (EDRR) concept, 284 developing a management strategy, 295 fire as a means to suppress, 214 high concern category, 281 impacts of, 3, 283–284 canopy structure, 283 on ecosystem structure and functional processes, 283–284 on humans, 283 on livestock production, 283 on soil erosion, 284 species richness, 283 integrated management strategies, 294–295 herbicides and revegetation combined, 294 long-term, prescribed burning and clopyralid combined, 294–295 invasion by, 171, 281 low concern category, 281 mechanical control, 285–286 hand labor, 285 mowing and clipping, 285–286 thatch removal, 286 tillage, 286 moderate concern category, 281 most common in California valley and foothill grasslands, 282 techniques for managing, 284–285 weeds, integrated approaches to managing, 295–296 fail-safe number, 123 farewell-to-spring (Clarkia spp.), 149, 151 farmed fields, abandonment of, 256 Farmland Security Zones (FSZs), 307 Federal Land Policy and Management Act of 1976, 234 fencing, 240, 242, 292 Fender’s blue butterfly (Icaricia icarioides fenderi), 214 feral burro (Equus asinus), 191 feral goat (Capra hircus), 191 feral horse (Equus caballus), 191 feral pig (Sus scrofa), 105, 185, 191–196 acorns and, 194 contact with humans, 192

distribution of, 192 disturbances by and ecosystem processes, 194–195 grassland vegetation responses to, 193–194 eradication of, 195–196 human contact, 195 hunting, ground- and aerial-based, 195 origins of, 192 overview, 192–193 population control, 195–196 in promotion of invasive exotic plants, 194 reproductive capacity, 192 rooting by, 192–193 success of, outside native range, 192 feral sheep (Ovis aires), 191 fertilizers minimizing flow of, 310 use of, 220 fescues Festuca spp. Festuca californica, 149 Festuca idahoensis, 31, 149 red fescue (Festuca rubra) 30, 146, 157, 164 tall fescue (Festuca arundinacea), 157, 276 Vulpia spp., 131, 152, 156, 158 small fescue (Vulpia microstachys), 137–138, 142, 146, 149, 150, 151, 157, 202 Vulpia bromoides, 27, 199 Vulpia myuros 149, 194 fiddleneck (Amsinckia douglasiana), 101 file drawer problem, 122 fire(s), 37, 167, 207–217 accidental ignitions, 209 animal adaptations to, 48 behavior factors affecting, 207–208 predictive models of, 216 burn intervals, 171 in California grasslands overview, 207–209 specifics, 209–210 combustion, three stages of, 207 community responses to, 79 and conversion from shrubland to grassland, 169 in determining grassland distribution, 207 effects of, 49 on birds, 213 on community composition, 211–212 direct versus indirect, 212–213, 217 on individual plants, 210–211 on insects and other invertebrates, 213–214 on invasive and undesirable plants, 214 on mammals, 213 on seeds, 211, 277 soils and nutrients, 208–209

species composition, 207 temporary community reorganization, 213 frequency, 207, 212, 217, 219, 229, 257 hazard reduction programs, 234, 235 historic fire regime, establishment of, 209 human-set, 58 ignition sources and patterns, 208, 217 intensity, 207, 210, 212 as landscape management, 320 lightning and, 35, 175, 209 monitoring effects of, 216–217 oaks and, 175–176 patterns, human alterations of, 220 preheating phase, 207 prescribed burn, 151, 166, 167, 168, 203, 215–216, 217, 276, 277, 292–293 reburn, 170, 171 as a restoration tool, 207, 214, 265–267 return interval, 175, 257 seasonal variations in, 209, 212 seedling resistance to, 211 in serpentine grasslands, 151, 154 set by Native Americans, 57, 58, 59, 64, 65, 75, 175, 209, 219, 256 shrub communities, loss of, 209 size and pattern of, 207 smoke management, 216 suppression, 2, 55 wind speed and direction, influence of, 208 fire-adapted plant species, 215 fire followers, 212 Fire Management Plans, 215–216 fire prescription, defined, 216 fire regime(s), 217 components of, 207 and the invasion of introduced plants, 209–210 predicted changes in, 219–220 Fire Regime Condition Class system, 209 Fire Unit Plan, 215–216 flank fires, 208 flat-headed peccary (Platygonus compressus), 50 flies false peacock fly (Chaetorellia succinea), 287 fruit fly (Chaetorellia australis), 287 gall fly (Urophora jaculata), (Urophora sinnaseva), 287 flora annualization of, 3 species domination in, 1 florets, 7 fluctuating resource theory of invasion, 93 fog, 164, 220 forage quality improvement, 268 forage yield, and timing/amount of annual precipitation, 222 forbs, 83, 92, 125, 157 annual, frequency distribution of, 55 early observations of, 53

INDEX

379

effects of fire on, 211, 212, 214 grazing, effects of, 202 in serpentine grasslands, 149 fossil fuel burning, 218 founder effect, 134–135 fragmentation, 320 habitat, 55, 186, 213, 252 plant, 249 Frankenia salina, 27 Free-Air CO2 Enrichment (FACE) system, 225 French broom (Genista monspessulana), 80, 171, 215, 250, 281, 285 full ecological restoration, 254 fustucoid grasses, 9 gall fly Urophora jaculata, 287 Urophora sinnaseva, 287 gene flow, 135–136, 273 among populations, 138 general circulation models (GCMs), 219 genetic bottlenecks, 133–134 inbreeding species, 133 genetic differentiation, selective, 141 genetic diversity, within a population, 136 genetic drift, 133, 134–135, 136, 139 outcrossing species, 135 versus selection, 134 genetic integrity, preserving, 273 The Genetics of Colonizing Species, 131 genetic swamping, 274 genetic variability, within mixed mating systems, 136 genetic variation intraspecific heritable, 273 in peripheral populations, 134–135 in restoration projects, 273–274, 280 within-population, 133 genotypes home versus nonlocal, 139 mesic, 133, 140 xeric, 133, 140 year specialist, 134 genotypic variation paralleling local adaptation, 139 germination in exotic versus native perennial grasses, 78 plant-soil relationships to, 142 timing of, 141 seed, 92, 211 giant bison, (Bison latifrons), 51 Giant Redwood (Sequoia gigantean), 2 gilia (Gilia spp.), 149 birdseye gilia (Gilia tricolor), 145 dune gilia (Gilia capitata ssp. chamissonis), 136 Gilia achilleifolia, 150 Gilia capitata, 136, 142, 143 glacial maxima, 41 global biome ordination, 179 global climate change, 39, 89, 119, 218 expected responses of California valley-type grasslands, 228 impacts of, 320 global warming, 218, 222–223

380

INDEX

glomalin, 224 glumes, 7 glyphosate, 294 goats, 244, 250 feral (Capra hircus), 191 golden eagle (Aquila chysaetos), 184 goldfields (Lasthenia spp.), 212 Lasthenia californica, 27, 120, 142–143, 145, 146, 149, 150, 152, 185, 186 Lasthenia glabrata ssp. glabrata, 27 Lasthenia gracilis, 146 Lasthenia platycarpha, 27 gophers, 26, 111, 126, 171, 172, 188, 201 Botta’s pocket (Thomomys bottae), 180, 211 exclusion experiments, 152 herbivory by, 162 and oak recruitment, 174, 175 pocket, 183, 184, 194 seed burial by, 211 in serpentine grasslands, 152 gopher snake (Pituophis melanoleucus), 184 gorse (Ulex europaea), 171, 281 Government Code, California, 304 grains and seeds, as food, 62–63 granivory and seed dispersal, 186–188 burrowing rodents and, 186–187 grazing and, 187 native plants, 186 non-native plants, 186–188 grass banks, 244, 311 grass-clover-filaree years, 26, 158 grasses (Poaceae) alert category, 14–15 Aristidoideae, 9, 11 Arundinoideae, 9, 11 Bambusoideae, 9, 10 in California, classification of, 10–11 Chloridoideae, 9, 11, 13–14 common ancestor, 9 comprehensive list of, 7 Cynodonteae, 15 Danthonoideae, 9, 11 diagnostic features of, 8 ecology, 8–9 Ehrhartoideae, 9, 10 evolution toward specialization, 15 festucoid, 9 fires and, 9 herbivorous animals, 9 introductions, 14–15 morphology, 7–8 Panicoideae, 9, 11, 12 Andropogoneae, 12 Paniceae, 12 phylogeny, 9 pollen, first documented appearance of, 37 Pooideae, 9, 10, 12–13 Bromeae, 13 Meliceae, 12 Poaeae, 12–13 Stipese, 12 Triticeae, 13

species known to occur in California, list of, 15–20 subfamilies, 9 systematics of, 7–20 worldwide distribution of, 8 grasshoppers (Melanoplus spp.), 58, 185, 188–189 increases in following fires, 213 and weather conditions, 189 grasshopper sparrows (Ammodrammus savannarum), 189 grassland, defined, 21, 33, 55 grassland reserves, establishment of, 310 grasslands. See also California grasslands adjacent to wetlands, 26 associations and alliances, classification of, 33 climatic diversity of, 21 climatic variations, responses to, 38 community classification and nomenclature, 21–34 conversion, 1, 3 to agriculture, 76, 92, 110 of shrublands to, 169–171 co-occurring woody vegetation, 169, 170 distribution of in California, 1, 22, 169 ecology climatic changes, 2 fire suppression and, 2 historical factors influencing, 2–3 hydrological changes, 2–3 indigenous cultures, influence of, 2 large herbivores, loss of, 2 non-native species, invasion by, 3 ecosystems defined, 4 ethnobotany and ethnozoology of, 59–64 Pleistocene and pre-European, 37–56 evolution of, 37 exotic annual domination, 76 fire, effects of, 210–212 grazing effects on, 203–205 history Holocene, 41–45 Pleistocene, 38–41 pre-European, 73–74 human influences on, 1, 66 invasive non-native species, 67–84 and land use change, 110–118 managing, to foster growth/abundance of edible grains and seeds, 62 Native American uses and management of, 57–66 non-native perennial, 30 oaks, effects of, 176–178 prehistoric, descriptions of, 198 relict grasslands, 54, 55, 56, 113 remnant grasslands, 255, 278–280 on roofs, 269–270 shrubs, establishment into, 171–172

species distribution, Clements’ hypothesis of, 53, 55 successional trajectory of, 179 types of, 21, 33 cold desert grassland, 31–32 within Great Valley Grasslands State Park, 25 north coastal grassland, 29–31 serpentine grassland, 28 valley/south coastal grassland, 23–28 warm desert grasslands, 32–33 vegetation and community type definitions, 21–23 grass years, 120, 165 gravimetric water content, 140 gray fox (Urocyon cineroargenteus), 183, 184 grazing, 31, 37, 55, 159, 162, 166, 167, 197–206 brooms, resistance to, 250 cattle, 105, 241, 244, 250, 277 climate variability and, 202 community responses to, 200–202 continuous, 245, 249 defined, 197 defoliation, 198–199, 202, 205 dietary preferences, 241–242, 244 for endangered species, 239 to enhance native grasses, 206 for fire hazard reduction, 209, 239, 250–251 and fires, effects of, 212 forbs and, 214 goats, 241, 244, 250 and grasslands, effects on, 2, 197, 198, 203–205 history of in California grasslands, 197–198 horses, 250 influence of on biomass, 200 and mediation of global climate change effects, 251 by megafauna, effects of, 51–52 to mimic natural disturbance, 241 native grasses and forbs, effects on, 202–203, 253 and non-native species, 202–203 nutrient redistribution, 199–200 overgrazing, 27, 49, 92, 234, 310 patterns, determining, 240–241 plant species richness, 201–202 and plant viral infections, 256 on Point Reyes National Seashore, 238 of poison oak, 250 prescribed, 242 in presence of oak canopy, 203–204 process and management, 198–200 and productivity, 200, 233, 253, 292 by Rancholabrean mammals, 48 removal of, 30 as a restoration tool, 257, 268, 277 riparian zones, 204–205 rotational, 236, 237, 243, 245, 249, 292 seasonality of, 249 seed dispersal and, 187, 203 and serpentine soils, 201 sheep, 105, 244, 250

species composition and, 200–201, 240 and species diversity, increasing, 241 species responses to, 245 systems, defined, 243 thistles, 277 timing of, 253, 292 tolerance, 48, 245 trampling, 199 and understory, effects on, 197 and vegetation, effects on, 233 vernal pools and, 204 Warm-Springs Seasonal Wetland Unit, 239 for weed control, 292 and wildlife, effects on, 205 grazing intensity, 201, 244 defined, 198 manipulation of, 197 measurement of, 198 grazing management, 233–253 adaptive approach to, 240, 244, 251 boundary-crossing, 252 climatic conditions, 249 conservation easements, 238, 239, 244, 299 controlled grazing, 236, 239 to control yellow starthistle and noxious weeds, 244–245 deferment, 243 defined, 233 examples of, 244–251 exclusion, 242–243 fencing, 240, 242, 245, 292 fenceline and road verge studies, 249 goals of, 233–239, 252 grass banks, 244 grazier constraints and needs, 243–244 grazing animals, 241–242 grazing plans, 243 grazing prescriptions, 240–241 grazing systems, 243 herding, 240 holistic principles, 249 incentives to private landowners, 237 integrated programs, 244 land conservation incentive program, 237 land trusts, 239 manager information needs, 251–252 natural resource regulation, 238 natural systems, 243 objectives of, variations in, 239 patterns, determining, 240–241 prescribed burning, 236 principles of, 239–243 public acquisition of grasslands, 237 public lands, 238–239 for resource sustainability, 241 “rest,” 243 in restoration of native grasses and forbs, 245, 249–250 rotational practices, 236, 237, 243, 245, 292 studies of livestock grazing effects on native plants, summary of, 246–248

sustainable use, 234 vernal pool and small-wetland biodiversity, 251 Great Valley Grasslands State Park, 24, 25 greenhouse gas emissions, 219 “green roofs,” 269–270 greens, in Native American diets, 63–64 green sloping sides, 260 Grinnell, Joseph, 173 grizzly bear (Ursus arctos), 49, 55, 184–185 ground sloths Megalonyx jeffersoni, 49 Nothrotheriops shastensis, 49 ground squirrels San Joaquin antelope (Ammospermophilus nelsoni), 185 Spermophilus beecheyi, 180, 183, 184, 188 groundwater, 279 growing season, 2, 91 growth cycle, 91 growth management, 306 growth rate, in exotic versus native perennial grasses compared, 78 gypsum-loving larkspur (Delphinium gypsophilum), 135, 138 habitat conversion, 153, 154 designated critical, 298 Environmentally Sensitive Habitat Areas (ESHAs), 306 fragmentation, 55, 186, 213, 252 improvement and protection of, 238, 268, 310 habitat conservation plans (HCPs), 301–302 affecting grassland species, 303 hand-clearing, 66 harding grass (Phalaris aquatica), 157 hares, 188 harmful species, 67 hawks ferruginous (Buteo regalis, 18)4 northern harrier (Circus cyaneus), 184 rough-legged (Buteo lagopus), 184 Hazelton, John, 59 head fires, 208 Heady, Harold, 64, 200 Heperolinon congestum, 153 herbicides, 168, 170, 217, 272, 276, 290–292 aminopyralid, 290, 291, 292 application of, 290–291 chlorsulfuron, 290, 291 clopyralid, 290, 291, 292, 294–295 commonly used, list of, 291 drift, 290 effectiveness of, 292 glyphosate, 294 imazapic, 290, 291, 292 post-emergent, 277, 278, 290 pre-emergent, 277 species variety and, 295 herbivore(s) large, loss of, 2 native, 203 selectivity, 240

INDEX

381

herbivory, 9, 119 gophers, 162 grasshoppers, 188–189 history of, 2 miscellaneous insects, 189–190 native herbivores, 203 and oak recruitment, 174, 175 predicted effects of, 188 in serpentine grasslands, 151 small mammals, 188 Hermizonia congesta, 91 Hespereva sparsiflora, 185 Hesperevax caulescens, 27 Hesperolinon spp., 151 heterozygosity, 133, 134, 135, 137, 138, 139 highway iceplant (Carpobrotus edulis), 80, 186 highway rights-of-way, 279 hilltop grasslands, 30 Himalayan blackberry, 250, 281 Holistic Resource Management, 243 Holocarpha virgata, 27, 91 Holocene, 41–45 home genotypes, 139 honeysuckle (Lonicera hispidula), 250 Hordeum spp. Hordeum brachyantherum, 210, 275 Hordeum leporinum, 212 Hordeum murinum, 150 horned lark (Eremophila alpestris), 186, 213 horses, 250, 283 feral (Equus caballus), 191 western (Equus cf. occidentalis), 51 houndstongue, 283 house mouse (Mus musculus), 184 human activities, 55, 257 changes in California grasslands associated with, 218, 254 in conversion of shrubland to grassland, 170 effects of on grassland environments, 218, 320 nitrogen deposition and, 79 and soil biological processes, 107 human-plant interaction continuum, 64 hydraulic lift, 177 hydrological changes, 2–3,102 hydroseeding, 275 Hypochaeris glabra, 27, 210 Ice Age (Bison antiquus), 51 ignition pattern, 208 imazapic, 290, 291, 292 inbreeding, 134, 273 among-population variation/withinpopulation genetic diversity, 137 depression, 132, 135, 136 effects of, 135 genetic bottlenecks in, 133 genetics of, 136–138 impact of on genetic structure, 136 inceptisols, 108, 109 incidental take permit, 301 incumbency, advantage of, 164 Indian ricegrass (Achnatherum hymenoides), 62

382

INDEX

indicator species, 149 indigenous cultures, influence of on grassland ecology, 2. See also Native Americans inherited morphological traits, 133 insects Argentine ants (Linepithema humile), 153, 186 bees, in serpentine grasslands, 152 beet leafhopper (Eutettix tenellus), 189 beetles Eremosaprinus spp., 185 Geomysaprinus spp., 185 for biological control, 296 effects of fire on, 213–214 false peacock fly (Chaetorellia succinea), 287 fruit fly (Chaetorellia australis), 287 gall fly (Urophora jaculata), (Urophora sinnaseva), 287 seed-harvesting ants (Messor Andrei), 151–152, 186 intercalary meristems, 7 interior live oak (Quercus wizlezenii), 173 interpulses, 165 intraspecific genetic swamping, 274 intraspecific heritable genetic variation, 273 introduced species, 67 fire regime and, 209–210 introduction event, 133 invasion(s), 119 by exotic species, 171 fluctuating resource theory of, 93 following removal of livestock grazing, 178, 179 of serpentine grasslands by exotic annual grasses, 154 success from serpentine grasslands, 150 invasion biologists, 67 invasion meltdown, 93 invasive, defined, 67 invasive non-native species, 67–84. See also exotic plants annuals, 55, 72 community responses to, 79, 320 competition, 77–78 control of, 77, 234, 268 costs of, 77 as disease facilitators, 82 domination by, 67, 72 factors facilitating, 72, 75, 77, 82–83 harmful, 67 impacts of, 77–82, 83 with limited effects, 68 list of, 69–71 in the modern California grassland, 68–72, 282 and native species compared, 78 origins of, 68, 110, 131–134 and plant pathogens, 81–82 potential of to interact with environmental changes, 220–221

presettlement vegetation, 72 soil, effects on microbes, 80–81 moisture, 79–80 nutrients, 80 terminology, 67 transient species, 68 and water availability, 72 woody invaders, effects of, 78–79 invertebrates, disturbance by, 115 Iris spp., 245 irrigation, 111, 114, 272, 276, 277, 279 isolation by distance pattern, 136 isozyme markers, 134 Italian ryegrass (Lolium multiflorum), 91, 146, 149, 151, 153, 154, 155, 163, 178 Italian thistle (Carduus pycnocephalus), 178, 289 jaguar (Panthera onca), 50 Jain, Subodh, 131 Japanese brome (Bromus japonicus), 212 Jasper Ridge Global Change Experiment, 223, 225–227 Jepson, Willis Linn, 65 jewelweed (Streptanthus spp.) 136, 142, 146, 153 Juncus bufonius, 27 junipers (Juniperus spp.), 281 kangaroo rats (Dipodomys spp.), 180, 183, 184, 185, 188 giant (Dipodomys ingens), 187, 188 Stephens (Dipodomys stephens), 239, 310 Kern primrose sphinx moths (Euproserpinus euterpe), 185, 189 keystone species, 55, 185 kindling temperature, 207 knapweed diffuse (Centaurea diffusa), 288 Russian, 283 spotted (Centaurea maculosa), 284, 288 squarrose (Centaurea virgata var. squarrosa), 287, 289 Koeleria macrantha, 150 Komarek, E. V., 64 land appreciation, 236 land clearing, 110 land killing, 75 land trusts, 239, 308, 311 land use change, 110–118 ecological responses to, 111–114 history of in Carmel and Salinas Valleys, 113 long-term annual grassland and restored perennial grassland compared, 117 and plant communities in Monterey County grasslands, 111–112 response of soil microbial community to disturbance, 113–114

restoration of native perennials, impact of on soil biology, 116–118 soil biological activity and litter decomposition, 114–116 soil microbial communities, Carmel and Salinas Valleys, 112–113 county goals and policies for, 304 general plans for, 306 open space restrictions, 305 large-headed llama (Hemiauchenia macrocephala), 50 larkspur (Delphinium spp.) Delphininum nudicaule, 61 gypsum-loving (Delphinium gypsophilum), 135, 138 valley (Delphinium recurvatum), 135 western (Delphinium hesperium), 135 Layia ssp., 149 Layia chrysanthemoides, 27 Layia platyglossa, 150, 186 leaching, 94, 98 Lepidium spp. Lepidium latipes var. latipes, 27 pepperweed (Lepidium latifolium), 286 Lepus californicus, 187 Lessingia spp. Lessingia arachnoides, 153 Lessingia micradenia glabrata, 153 Lessingia micradenia micradenia, 153 Leymus spp. Leymus cinereus, 31, 62 Leymus condensatus, 61, 62, 65 Leymus triticoides, 62, 198 Lianthus spp., 149 Lianthus bicolor, 214 light, availability of 165 lightning, 37, 175, 209 lilies (Calochortus spp.), 149, 157 mariposa (Calochortus amabilis), 63 limitations to growth, defined, 120 litter, 189 and biomass removal, 240 decomposition, 114–116, 227 height and abundance, 116 influence of on community composition, 166 mass, 99 mass loss, seasonal patterns of, 104 perennial grass, decomposition of, 102 quality of, 226 Little Ice Age, 45 livestock cattle, 171, 174, 198, 239, 241, 244, 250, 277 enclosures, 242 exclusion, as shrubland stabilizer, 172 goats, 244, 250 grazing, 55, 170, 217 cessation of, 171, 178, 179 community responses to, 79 flow of waste into surface water, 310 and insect infestation, 189 and invasive non-native species, 72, 73, 83

and presence/absence of oak canopies, 250 to promote native species, 253 on public lands, 238, 239 as restoration, 268 role of in exotic annual domination, 76 on serpentine grasslands, 154 timed, 276 horses, 250, 283 production, 234, 236 impact of exotic plants on, 283 sheep, 244, 250 local adaptation, 138, 273, 278, 280 coadapted gene complexes and, 139–141 genotype variation paralleling, 139 local community types, detection of in the Sacramento Valley, 27 local molecular variation, 138–139 local topography, impact of, 42 lodicules, 7 Lolium spp. Italian ryegrass (Lolium multiflorum), 91, 146, 149, 151, 153, 154, 155, 163, 178 Lolium perenne ssp. multiflorum, 27 longevity, 164 Lotus spp., 157, 211 Lotus strigosus, 150 Lotus subpinnatus, 185 Lotus wrangelianus, 27, 152, 214, 224 lupines (Lupinus spp.), 63, 144, 157, 211 arroyo lupine (Lupinus succulentus), 135, 138 Lupinus arboerus, 80, 171, 222 Lupinus bicolor, 27 Lupinus nanus, 135, 138, 214 Madia elegans, 62 madrone (Arbutus menziesii), 250 magnesium (Mg), 121, 146 calcium-magnesium ratio (Ca:Mg), 146, 150, 154 Malpais Borderlands Group, 244, 311 mammals, effects of fire on, 213 MAPSS-CENTURY (MC1), dynamic global vegetation models (DGVM), 227 mariposa lily (Calochortus amabilis), 63 Mayfield, Thomas, 73 McLaughlin Natural Reserve, 304 meadow(s), 22, 149 meadowfoam (Limnanthes spp.), 135, 136, 245 wooly (Limnanthes floccose), 135, 245 meadow vole (Microtus californicus) 184, 188 mechanical raking, 256 Medicago polymorpha, 186 medicinal plants, 57, 61 Mediterranean rust (Puccinia jaceae), 82 Mediterranean sage (Salvia aethiopis), 288 medusahead (Taeniatherum caput-medusae), 101, 149, 154, 203, 212, 214, 215, 249, 253, 272, 277, 281, 283, 286, 293 megafauna, loss of, 257 Melanoplus devastator, 189 Melica californica, 149 Melica imperfecta, 149, 150

Meliceae, 12 Mesembryanthemum crystallinum, 80 mesic genotype, 133, 140 mesic grasslands, 159 meta-analysis, 119, 122 metapopulation, Wright’s model of, 134 mice deer mice (Peromyscus spp.), 213 house mouse (Mus musculus), 184 microbial communities annual grassland, 256 responses to rising CO2 concentrations, 224 microbial decomposition pathway, 199–200 Microseris douglasii, 150, 151, 152, 186 mid-winter drought, 88 milkweed (Asclepias fasicularis), 61 Minuartia spp. Minuartia californica, 150, 214 Minuartia douglasii, 150 moist native grassland, 30 molecular genetic markers, and adaptation, 138 molecular markers, 139 moles (Scapanus latimanus), 184 mollisols, 108 Montana de Oro State Park, 304 Monterey Pine (Pinus radiate), 2 morphological traits, 134 inherited, 133 moths Kern primrose sphinx (Euproserpinus euterpe), 185, 189 Opler’s longhorn (Adela oplerella), 153 mountain plover (Charadrius montanus), 188, 213 mourning dove (Zenaida macroura), 186 mowing, 201, 245, 271, 277 to control invasive plants, 285–286 as a fire hazard, 286 safety recovery zones, 285 on steep terrain, 286 timed, 276 Muir Heritage Trust, 239 Muir, John, 52, 53, 72, 73–74, 76 mulch, 276–277 mulch monitoring, 253 mule deer (Odocoileus hemionus), 50, 183 mule ears (Wyethia angustifolia), 62 multiple resource limitation, defined, 120 mycorrhizal infection, 126 Nassella cernua, 167, 249 Nassella lepida, 149, 202 planting from seed, 275 Nassella pulchra. See purple needlegrass National Marine Fisheries Service (NMFS), 298 National Park System, 238 Native Americans, 57–66 adornments, clothing and regalia, 59 animal products used by, 59 diets, diversity of, 62 disease epidemics among, 183 fire-making kit, 58 fires set by, 57, 58, 59, 64, 65, 75, 175, 209, 219, 256

INDEX

383

games, 57–58 hunting by, 183, 209 influences of on grasslands, 2, 64–66 selective harvesting, 64 vegetation changes, 64–66 insect food, 58 language groups, California territories associated with, 58 in missionary journals, 57 uses of grassland plants, 57 for basketry, 59–60, 209 for construction materials, 60–61 for cordage, 61 for face paint dyes, 59 for food and medicine, 57, 61–64 for thatch, 57, 60061 native animals, 180–190 burrowing, 180–184, 185–186 dietary preferences of, 188, 205 explicit inclusion of, 255 feral pigs, 185 foxes, 184 granivory and seed dispersal, 186–188 grizzly bears, 184–185 hawks and owls, 184 herbivory by insects, 188–190 small mammals, 188 and pre-European population patterns, 183 secondary burrow users, 185 snakes, 184 soil disturbance, 180–186 native grasses effects of grazing on, 202 extinction of, 21 reduced abundance of, 21 in the Sacramento Valley, 27 Natural Community Conservation Plans (NCCPs), 302, 304 affecting grassland species, 303 natural grazing systems, 213, 243 naturalization hypothesis, Darwin, 14 naturalized annual and perennial grasses, invasion of, 21 Natural Reserve System, University of California, 304 Natural Resources Conservation Service (NRCS), U.S. Department of Agriculture, 274, 309 natural selection, capacity of plant populations to respond to, 136–137 natural successional processes, 271 The Nature Conservancy (TNC), 239, 299, 311 Navarretia spp., 149, 151 Navarretia intertexta, 27 Navarretia leucoephala, 136 negative trees, 176 neighborhood structure, 135 neoendemism, 147 Neopluvial period, 43 net nitrogen mineralization, 195

384

INDEX

net primary productivity (NPP), 107, 223, 226 nitrogen (N) addition of, 276 availability, 223 carbon-nitrogen ratio (C:N), 276 decomposition, 95 fire effects on, 208 cycling, 95, 101, 223 grasshopper herbivory and, 189 in oak-grassland systems, 104 species’ effects on, 80 deposition, 94, 215, 218, 220, 227, 320 anthropogenic, 171, 320 community responses to, 79 environmental responses to, 221 dynamics, effects of other organisms on, 104–105 fixation, 68, 72, 78 80, 171, 172, 276 gas losses, 94, 98 immobilization, 102, 276 limitation, 120–121 124, 127 pools and fluxes in early spring, 99 during fall wet-up, 98 in fall/early winter, 98–99 general patterns of, 94–95 in late spring/peak biomass, 100 in perennials and annuals compared, 103 spring/peak physiology, 99 in summer, 96–98 variations across sites and years, 100 in winter, 99 translocation, 95 turnover, 95 uptake, 166, 251, 276 nitrous oxide (N2O) concentration, 218 noise, reduction of, 269 nongrass years, 158 non-indigenous, 67 nonlocal genotypes, 139 non-native species, 67 domination by, 1 effects of grazing on, 202–203 effects of on perennial grasses, 92–93 fire regime and, 209–210 invasive plants, 3, 67–84. See also exotic plants most common in California valley and foothill grasslands, 282 potential of to interact with environmental changes, 220–221 perennial grassland, 30 north coastal grassland, 29–31 alliances along the immediate coast, 29–30 interior grassland away from the immediate coast, 30–31 northern harrier (Circus cyaneus), 184 numerical/functional superiority, as indicator of competitive dominance, 156 nutrient(s) availability, 94, 121, 163 deficiencies, 146 fire effects on, 208–209

management of as a restoration strategy, 214 redistribution, 199–200 nutrient dynamics, 94–106 climate and plant growth, 96 effects of vegetation composition on, 101–104 future research directions, 105–106 nitrogen cycling rates, 95 nitrogen deposition, 94 nitrogen dynamics, effects of other organisms on, 104–105 nitrogen fixation, 95 nitrogen inputs and outputs, variability of, 94–95 nitrogen losses, 94 nitrogen pools and fluxes general patterns of, 94–96 of nutrients other than nitrogen, 100–101 seasonal cycling patterns, 96–100 seasonality of, 96–100 nutrient uptake, in exotic versus native perennial grasses compared, 78 oak canopy covering, 169 effects of on species composition, 250 grazing and, 203–204 oak-grassland systems, nitrogen cycling in, 104 oak recruitment, 1, 169, 173–176, 256 oaks acorn production, predation, and dispersal, 173–174 animal impact on, 174 blue oak (Quercus douglasii), 78–79, 173, 176–178, 222 coast live oak (Quercus agrifolia), 173, 175, 178 competition with grasses, 175 effects of on grasslands, 176–178 fires and, 175–176 interior live oak (Quercus wizlezenii), 173 Oregon white oak (Quercus garryana), 173 Quercus dumosa, 171 seedlings and shrub canopies, 174–175 shoot predation, herbivory, and browsing, 174–175 sudden death, 81 and understory productivity, 173, 176–177 and understory species composition, 173, 177–178 valley oak (Quercus lobata), 173 oak savanna, 173–178 defined, 169 oak woodland, 110 defined, 169 obliquity, 38 open space zoning, 305 Opler’s longhorn moth (Adela oplerella), 153 orchard grass (Dactylis glomerata), 150, 154, 157 Orcuttia spp., 245 Oregon white oak (Quercus garryana), 173

Orthocarpus spp., 245 Orthocarpus attenuatus, 212 outbreeding depression, 136 outcrossing, 144, 273 rates, 136 species, 132, 133, 134 genetic drift in, 135 self-fertilization in, 138 variable, causes and consequences of, 138 outsiring events, 138 overgrazing, 27, 49, 92, 234, 310 overhunting, 257 owls burrowing owl (Athene cunicularia), 184, 185, 213 Paniceae, 12 Panicoideae, 9, 11, 12 Panthera spp. American lion (Panthera leo atrox) (American lion), 50 jaguar (Panthera onca), 50 Paradise Urban Reserve area, Butte County, 306 parks, 238 passive restoration strategy, 271 pastoral zones, 238 paternal exclusion analysis, 138 pathogen, defined, 81 peak aboveground biomass, 122 peak plant biomass, 99, 100 peak standing biomass, 204 Pentachaeta belliflora, 153 pepperweed (Lepidium latifolium), 286 percent live seed (PLS) rating, 274 perennial grasses, 157 fire, effects of, 210–211 interaction of with annual species, 1 longevity, 164 phenology of, 157 population growth rates, calculating, 159 perioditite, 145 permanent wilting point, 91 permits, 271, 301 pesticides, minimizing flow of, 310 phenotypic plasticity, 141–142, 163, 273 phospholipid fatty acid (PLFA) analysis, 80–81, 113 phosphorus (P), 121 limitation, 121, 124, 127, 226 photosynthesis carbon gain via, 199 and rising CO2 concentrations, 223 phytoliths, 37, 38 Phytophthora ramorum, 81 plant(s) allocation patterns, application of economic theory to, 121–122 foods derived from, 61–64 bulbs, corms, and tubers, 62, 63 condiments, 61 grains and seeds, 62–63 greens, 62, 63–64 fragmentation, 249 macrofossils, 37, 38

materials, selecting and obtaining, 273–274 mortality, after fires, 210 pathogens alteration of by non-native invaders, 82 biotic resistance hypothesis, 81–82 enemy release hypothesis, 81 impacts of, 81–82, 83 invasive plants benefiting directly from, 82 potential invaders limited by, 81–82 terminology, 81 protection of, under federal Endangered Species Act, 301 species composition exotic and native-dominated grasslands compared, 101–102, 104 and nitrogen cycling, 101 and nutrient dynamics, 104–104 shifts in, 101 Plantago spp. Plantago coronopus, 27 Plantago erecta, 120, 150, 152, 153, 186 planting “plugs,” 274–275 from seed, 274, 275 techniques, 274–275 timing of, 274 plant-mutualist interactions, 126 plant-soil relations, 1, 142 plasticity evolution of, 142 phenotypic, 141–142, 163, 273 trans-generational, 142 Platystemon californicus, 153 Pleistocene and pre-European grassland ecosystems, 37–56 as animal habitats, 37 appearance of, 37 evidence of, 37–38 general characteristics, late quaternary history, 38–45 late quaternary paleoecology, 37–48 regional climate differences, 45–46 Rancholabrean mammal and, 48–52 species composition at time of first European settlement, 52–56 Pleuraphis jamesii alliance, 32 Pleuraphis rigida alliance, 32, 33 ploidy levels, 132, 138, 273 plowing, 256 plug planting, 274–275 Poa spp. Poa pratensis, 274 Poa secunda, 32, 111, 149, 150, 167, 245 Poaeae, 12–13 Pogogyne spp. Pogogyne douglasii, 27 Pogogyne ziziphoroides, 27 Point Pinole Park, Contra Costa County, 304 Point Reyes National Seashore, 29, 234, 238 poison oak (Toxicodendron diversilobum), 250

pollen contamination, 274 pollen ratios grass-to-sagebrush, 40, 42, 43, 48 grass-to-total terrestrial, 40, 42, 43, 48 pollination, 38, 144, 213 polymorphism, 134, 135, 139, 142, 143 Pooideae, 9, 10, 12–13 popcornflower (Plagiobothyrs nothofulvus), 62 population biology, ecological restoration and, 255 population differentiation, 138–142 coadapted gene complexes, 139–141 local adaptation, 138, 139, 141, 273, 278, 280 local molecular variation, 138–139 phenotypic plasticity and adaptation to heterogeneous environments, 141–142 population genetics, 1, 131–144 abiotic gradients, adaptation to, 142–143 adaptation and restoration, 132 biotic interactions as agents of selection, 144 biotic selection, adaptation to, 143–144 breeding systems, 135–138 chromosome counts, 132 colonization, 131–135 founder effect, 134–135 genetic bottlenecks, 133–134 genetic drift, 134–135 genetic variation, in peripheral populations, 134–135 introduction event, 133 invasive species, 131–134 ploidy levels, 132, 138, 273 population differentiation, 138–142 and restoration success, 132 population, increases in, 170, 320 positive trees, 176 potassium (K), 100, 121 prairie, 33, 149 defined, 22 precipitation and abundance, influence of, 37, 44, 46 altered patterns of, 320 annual, 90 average monthly, 47 changes, environmental responses to, 222 effects of on California grasslands, 222 and forage yield, 222 during glacial maxima, 41 increases, Holocene, 42 patterns during growing season, 119 productivity and, 120 seasonality of, 45 precipitation-to-evaporation ratio (P:E), 21–22 pre-cultivation composition, 114 predation, 172, 184, 189. See also raptors pre-European grasslands. See Pleistocene and pre-European grassland ecosystems, 37–56 prescribed burning, 166, 167, 168, 203, 276, 277

INDEX

385

for grassland management, 215–216 for invasive plant control, 292–293 justification for, 216 purposes of, 292 in serpentine grasslands, 151 species most susceptible to, 293 prescribed grazing, 242 pre-settlement vegetation, 72 priming effect, 104 pristine grassland, 242 proclimax community, 53 productivity aboveground, 165–166 climate as predictor of, 92 ecosystem-level, 153 effects of grazing on, 200 latitude and, 92 net primary (NPP), 107, 223, 226 precipitation amounts and timing, importance of, 120 responses, resource manipulation studies, 124–125 pronghorn antelope (Antilocapra Americana), 50, 183 Pseudoroegneria spicata, 31 Pteridium aquillnum var. pubescens, 60 public lands, 238–239 Public Rangelands Improvement Act of 1978), 234 Puccinia jaceae (Mediterranean rust), 82 var. solstitialis, 287, 290 pulses, 165 puma (Felis concolor), 49 puncturevine (Tribulus terrestris), 288 purple needlegrass (Nasella pulchra) for basketry, by Native Americans, 60 competition from exotic annuals, 78 dominance of, 26, 72, 157, 255 effects of fire on, 210, 212 germination, 164 grazing and, 77 growth rates of, 158 herbivory by gophers, 162 modeling population growth of, 159–161 negative effects of exotic annual grasses on, 159 nitrogen uptake, 166 persistence of, 111 planting from seed, 275 population differentiation, 139 response of to grazing, 253 restoration of, 268 root growth, 93, 102 seeds addition, 167 dispersal, 135, 273 effects of fire on, 211 in serpentine grasslands, 28 soil, 256 soil depth, 90, 91, 149–150 soil ripping depth, 272 viral infections, 256 pyrrolizidine alkaloids, 101

386

INDEX

purple starthistle (Centaurea calcitrapa), 135, 139, 149, 150, 157, 163, 165, 202, 249, 289 purslane speedwell (Veronica perigrina), 139 Quercus spp. blue oak (Quercus douglasii), 78–79, 173, 176–178, 222 coast live oak (Quercus agrifolia), 173, 175, 178 interior live oak (Quercus wizlezenii), 173 Oregon white oak (Quercus garryana), 173 Quercus dumosa, 171 valley oak (Quercus lobata), 173 rabbits, 188 rainfall. See also precipitation annual patterns, 255 in serpentine grasslands, 150–151 as disturbance, 114 favorable patterns and grass cover, 158 in early grassland ecology, 2 seasonal variations in, 1, 87 timing and amount of, 26, 87, 91, 93 Ramona Grassland Preserve, San Diego County, 304 ranching, 236 Endangered Species Act and, 238 negative perceptions of, 252 ranches, land grants to, 198 Rancholabrean mammals, 48–52 antilocapridae (pronghorns), 50 bovidae (cattle and sheep), 51 camelidae (camels), 50 candidae (dogs), 49 cervidae (deer and elk), 50 defined, 48 dietary habits, 48–49 edentate (ground sloths), 49 effects of on California grasslands, 51–52 equidae (horses), 51 felidae (cats), 49–50 grazing, 48 human hunting and, 48 population estimates, 49 proboscidea (elephants and mastodons), 51 tapiridae (tapirs), 51 tayassuidae (peccaries), 50 ursidae (bears), 49 Rancho San Luis de Gonzaga, 73 raptors, 184 burrowing owl (Athene cunicularia), 184, 185, 213 California condor (Gymnogyps californianus), 302, 310 ferruginous hawk (Buteo regalis), 184 golden eagle (Aquila chysaetos), 184 rough-legged hawk (Buteo lagopus), 184 rare and endangered species Endangered Species Act (EDA), 298, 299, 301–302, 303 and projected development, 310 serpentine grasslands, in 153–154, 155

reburn, 170, 171 reclamation seed mixes, 274 recolonization, 170, 213 red brome (Bromus madritensis), 149, 187, 199, 212, 281, 293 red fescue (Festuca rubra) 30, 146, 157, 164 red-legged frog (Rana aurora draytonii), 251, 302 red maids (Calandrinia ciliate), 62 Redwood Regional Park, Alameda County, 304 reed grass (Calamagrostis nutkaensis), 29, 245 reference ecosystem defined, 255 selection of, 255 regulation and conservation, 297–318 county authority agricultural zoning, 300, 305 coastal zoning, 305 conservation zoning, 305 county ownership and management of grasslands, 304 general plan element profile, 318 general plans and policy documents, 300, 304–306 general plan zones, area of grassland in, 305 land use goals and policies, 304 miscellaneous optional zoning, 305 open space zoning, 300, 305 readily available documents, 305 and state legislation, implementation of California Environmental Quality Act (CEQA), 299, 306–307 Williamson Act, 307–308 zoning ordinances, general, 304, 305, 306 easements, 238, 239, 244, 299 federal and state owned lands by grassland type, 301 endangered species associated with, 301 federal authority Clean Water Act (CWA), 299, 302 Conservation Reserve Program (CRP), 302 Endangered Species Act (EDA), 298, 299, 301–302, 303 federal ownership and management of grasslands, 298, 301–303 federal, state and county policies/regulations affecting grasslands, list of, 299–300 financial incentives for conservation, 309–310 interactions between federal, state, and county policy, 308 permits, 309 public policies and regulations, summary of, 308 socioeconomic benefits of conservation, 308–309 state authority, California Coastal Act, 304, 306 California Endangered Species Act (CSA), 299, 302, 304, 312–314

California Environmental Quality Act (CEQA), 215, 271, 299, 304, 306–307 Government Code, 304 state legislation, 304 state ownership and management of grasslands, 304 Williamson Act, 237, 300, 304, 307–308, 311 relative humidity, 207–208 relict analysis, 53, 54, 55–56, 198 relict grasslands characteristics of, 54, 55, 56 non-native plant species in, 56 perennial, 113 remnant grasslands conservation importance of, 255 diversity of, 255 managing restoration of, 278–280 remote sensing, 319 reproductive allocation, in exotic versus native perennial grasses compared, 78 reptiles, effects of fire on, 214 residual dry matter (RDM), 241, 245 manipulation of, 205, 249 monitoring of, 253 resource(s) defined, 120 competition, 127, 163 resource limitation, 119–127 aggressive competition and, 126 bottom-up controls, 119 carbon, 121 co-limitation, 120 defined, 119 experimental manipulation of water amounts, 119 magnesium, 121 multiple limitations and reactions, 120, 121–122, 127 nitrogen, 120–121, 124, 127 phosphorus, 121, 124, 127, 226 potassium, 121 resource addition experiments, 123 resource competition theory, 127 resource manipulation studies, metaanalysis of, 122–127 resource ratio theory, 127 sulfur, 121, 127 temporal and spatial patterns, 122 terminology, 120 testing for, 120 water, 119–120 resource manipulation studies functional group responses, 125–126 productivity responses, 124–125 trophic responses, 126–127 resource partitioning evolution of, 143 in serpentine grasslands, 150 resource ratio theory, 127 resource sustainability, 241 restoration, 131, 132, 254–280 abandoned farm fields, 256 after large-scale grading and soil loss, 256 constraints on, 255–268

fires, 256–257 grazing, 257, 268 invasive non-native species, 255–256 plant viruses, 256 road construction, 256 soils and land use, 256 definitions of, 254 detailed management strategy, need for, 268 ecological and genetic studies, 320 ecological succession, 256 effects of soil processes on, 280 erosion control, 275–276 establishing goals and implementation/management plan, 268–271 farm edge and roadside, 279 fire as a tool for, 214–215, 217 full ecological, 254 future challenges to, 280 genetic integrity, preserving, 273 genetic variation, 273, 280 grazing as a tool for, 245, 249–250, 277 green roofs, 269–270 initial site survey, 271 for landscaping purposes, 277 local adaptation, 278, 280 long-term site management, 276–278 low available plant nutrients, managing for, 214 microbial legacy, importance of, 320 monitoring, 278 mowing or spraying to support, 271 mulch, addition of, 276–277 of native biodiversity, 1, 320 native grass projects in northern California, 258–267 oak recruitment, 256 perennial grasses, focusing on, 255 permits, 271 photographic monitoring, 278 planting technique, 274–275 population biology and, 255 population genetics and, 132 productivity, maintenance of, 254 re-creation of original California grasslands, possibility of, 268 reference ecosystem, selection of, 255 reference sites, comparison to, 278 remnants, managing, 278–280 replacement of annual with perennial grasses, 256 restored sites, evaluating characteristics of, 280 rice straw, addition of, 277 road construction and, 277 roadsides, 278 roof surfaces, 278 seed/plant material selection, 273–274 serpentine grasslands, 154 site preparation, 271–273 soil biology, impact of native perennials on, 116–118 soil impoverishment as tool for, 276 speed, need for, 277 strategies for, 166–167 active, 271

passive, 271 stress tolerance and, 254 timeline for, 268 tracking projects, 280 urban development and, 277 vegetation-type conversion, 256 watershed level of organization, 203 weed-resistance as, 254 Restoration Ecology, 278 revegetation, 293, 295 reverse fertilization, 126 rhizomes, 7 rhizospheres, 152 rice straw, 277 riparian zones, 204–205, 251 ripgut brome (Bromus diandrus), 149, 156, 164, 175, 177, 178, 199, 211, 214, 286, 293 road construction, 76, 256 “road effect,” 154 root crops, 63. See also bulbs, corms, and tubers rooting depth, 90 rooting patterns, 164–165 root length, 153 root:shoot ratios, 199 root systems, development of, 164 rotational grazing practices, 236, 237, 243, 245, 292 Ruiz, Francisco Maria, 76 Rumex crispus, 186 rural economic development, 234 rush skeletonweed (Chondrilla juncea), 289 Russian knapweed, 283 rusts Mediterranean (Puccinia jaceae), 82 resistance to, 144 sabrecat (Smilodon fatalis), 50 sacaton (Sporobolis airoides), 24, 60, 186 safety recovery zones, 285 Salinas Valley, 111 saltgrass (Distichlis spicata), 27, 61 salts, accumulation of, 3 Salvia spp., 171 chia (Salvia columbariae), 62 Mediterranean sage (Salvia aethiopis), 288 sand dunes, 33 San Diego Multiple Species Conservation Program (MSCP), 302 sanicle (Sanicula arguta), 61 San Joaquin antelope ground squirrel (Ammospermophilus nelsoni), 185 San Joaquin kit fox (Vulpes macrotis mutica), 184, 302, 310 Santa Ana winds, 220 Santa Cruz tarplant (Holocarpha macradenia), 136 savanna, defined, 173 Sawday-Sexton Ranch, 76 sawdust, for soil enrichment, 276 scenic highways, 306 scimitar cat (Homtherium serum), 49–50 Scotch broom (Cystisus scoparius), 80, 171, 172, 215, 250, 281, 285

INDEX

387

Scotch thistle (Onopordum acanthium), 281, 289 scrub grassland, 23 seasonal variations fire, 209, 212 grazing, 249 in nitrogen pools and fluxes, 96–100 nutrient cycling patterns, 98 precipitation, 1, 45, 87 in soil moisture, calculating, 89 Sedgwick Reserve, 304 seed(s) addition, 167, 168 beating, 62, 65 collection sites, 132 dispersal, 134, 186–188, 211 dormancy, 141–142, 171 and fire, effects of, 211 formation, timing of, 90 germination, 92, 211 limitation, 167, 168 in Native American diets, 62–63 planting from, 274, 275 PLS (percent live seed) rating, 274 reclamation mixes, 274 selecting and obtaining, 273–274 zones, 273 seed bank alteration, 199 annual exotics, 277 dormancy, 171 flushing, 75 in management of exotic plants, 284 seed bed preparation, 272 seed-harvesting ants (Messor Andrei), 151–152, 186 seedling(s) abundance, 92 competition by, 168 densities, 92, 210 and fire, effects of, 210 survival, 134 thinning of, 95, 100 “seeps,” 251 selection, 133 biotic, adaptation to, 143–144 versus genetic drift, 134 natural, 136–137 versus random genetic drift, 134 of reference ecosystem, 255 of seed/plant materials for restoration, 273–274 unconscious, 132 selective genetic differentiation, 141 selective harvesting, 64 self-fertilization, 138 self-similarity, 152 serpentine grasslands, 28, 145–155 adaptation in, 142–143, 146–147 agricultural conversion, 154 annual rainfall, 150–151 barrens, 148, 151 bees in, 152 calcium-magnesium ratio (Ca:Mg), 146, 150, 154 competition in, 163 composition of, 149

388

INDEX

conservation of, 153–154 defined, 145 distribution of, 147–149, 154 dominant species, 149 ecosystem and belowground processes, 152–153 endemic plants, 147, 153, 154 exclusion of plants from, reasons for, 146 fire, 151, 154 forbs, 149 geology and soils, 145–146 gopher disturbance in, 152 growing wine grapes on, 154 habitat conversion, 153, 154 habitat loss and degradation, 153 harvester ants in, 151–152 herbivory, 151 indicator species, 149 invasions by exotic annual grasses, 154 ongoing invasion, 154 success from, 150 livestock grazing on, 154 low soil water capacity, 146 map of, 148 overview, 145–147 prescribed burns, 151 rare and endangered species in, 147, 153–154, 155, 320 resistance of to exotic dominance, 154 resource of, 150, 154 “road effect,” 154 slopes, soil depth, and species composition, 149–150 soil, 26 species richness in, 150, 151, 214 tolerance, 146 urban development and, 154 sheep, 244, 250 feral (Ovis aires), 191 Shepard, Paul, 57 short-faced bear (Arctodus simus), 49 shoot production, predicting, 223 shrub canopy, as shrubland stabilizer, 172 shrubland types, 171–172 shrub ox (Euceratherium collinum), 51 shrubs, establishment of into grassland, 171 shrub-steppe, 23, 33 Sidalcea calycosa, 214 Siesta Valley, East Bay Municipal Utility District, 304 silt deposition, 3 site management, long-term, 276–278 site preparation, for restoration, 271–273 site survey, for restoration, 271 site water management, 268 slenderflower thistle (Carduus tenuiflorus), 289 slope, 150 small fescue (Vulpia microstachys), 137–138, 142, 146, 149, 150, 151, 157, 202 Smith, Bruce, 65 smoke management, 216

Society for Ecological Restoration (SER), 254 California chapter (SERCAL), 255 soft chess (Bromus hordeaceous), 27, 30, 125, 137, 141, 142, 143, 146, 147, 149, 150,152, 154, 156, 177, 178, 201, 222 soil(s) adding activated carbon (charcoal) to, 276 adding sawdust to, 276 carbon sequestration, 107–118 cation exchange capacity, 3 compaction, 9 depletion rate, 89 depth, 90, 152 disturbance, 75, 105, 105, 180–186, 272 with Duripans, 109 erosion control, 268, 275–276, 279 invertebrates in, 105, 115 microbes community composition, 80–81 effects of invasive non-native species on, 80–81 mineralization rates in, 223 moisture, 223 availability, 195 deep soil water, 79–80 and fires, 207 gravimetric water content, 140 impact of exotic plants on, 79–80, 283–284 influences on, 88 low water capacity, 146 patterns, community conversion and, 92–93 potential, 140 rooting depth and phenology, 90–91 seasonal availability, calculating, 89 surface retention, 166 and timing of mowing, 285 water deficit/surplus, 88 wet and dry seasons, influence of, 87 nutrients, 3 availability, 94, 121, 163 deficiencies, 146 fire, effects of, 208–209 invasive non-native species, effects of, 80 in restoration efforts, 256 in serpentine grasslands, 26, 145–146 spatial extent of, map depicting, 108 stabilization, 268 storage capacity, by type, 89 texture, 90, 102 types of, 26, 108–110 soil microbial communities, 144 community composition, 104, 113, 319–320 legacy effect and, 112 response to disturbance, 113–114 soil organic matter (SOM), 107, 108 soil seed bank(s) alteration, 199 annual exotics, 277

dormancy, 171 flushing, 75 in management of exotic plants, 284 solarization, 168 solar output, 46 sowing, 62, 65 Spartium junceum, 281 spatial patterning, gophers as cause of, 152 species-abundance distribution, 152 species-area relationship, 152 species composition fires, effects of, 207, 217 grazing and, 200–201, 240 oak canopy, effects of, 250 shifts in caused by mowing and clipping, 285–286 site-specific factors affecting, 243, 250 species distribution, fractal theories of, 152 species diversity, 268 disturbances in, 191 fire, effects of, 214 grazing to increase, 241 species richness, 193, 243, 277 diversity and, 278 fires, effects of, 214 grazing and, 201–202 impact of exotic plants on, 283 in serpentine grasslands, 150, 151 unimodal distribution of, 201 spikelet, 7 spotted knapweed (Centaurea maculosa), 284, 288 spring-fed wetlands, 251 squarrose knapweed (Centaurea virgata var. squarrosa), 287, 289 starthistles. See thistles, starthistles State Responsibility Areas (SRA), 215 Stephanomeria virgata, 91 Stephens kangaroo rat (Dipodomys stephens), 239, 310 steppe, 33 defined, 22–23 ratio of herb-to-woody cover, 22–23 shrub-steppe, 23, 33 tree-steppe, 23, 33 Stipese, 12 St. John’s wort, common (Hypericum perforatum), 283, 286, 288, 296 stoichiometry, application of, 127 stolons, 7 stomatal conductance, 223–224 Streptanthus spp. (jewelweed), 142 Streptanthus albidus albidus, 153 Streptanthus albidus peramoenus, 153 Streptanthus glandulosus, 136 Streptanthus polygaloides, 146 stress tolerance, 163 succession, 53 sudden oak death, 81 sulfur (S), 100, 121, 127 summer drought, 87, 90, 119, 127, 164, 174, 200, 222 sun cups (Camissonia ovata), 63 surface water EPA standards for, 251 flow of livestock waste into, 310 runoff, 284

sustainability, 241, 309 swathing, 275, 277 Sylvilagus spp. Sylvilagus audubonii, 187 Sylvilagus bachmain, 187 tall fescue (Festuca arundinacea), 157, 276 tansy ragwort (Senecio jacobaeae), 283, 286, 288 tapir (Tapirus), 51 tarantulas (Aphonopelma spp.), 185 tarweed big tarweed (Blepharizonia plumose), 139 Hemizonia spp., 157 Hemizonia congesta, 27, 62, 150, 293 Taylor Grazing Act of 1934, 234 technological advances, 319 temperature(s), 207, 320. See also global warming Tending the Wild, 320 Terrestrial Vegetation of California, 67 testing for resource limitation, defined, 120 thatch, 57, 60–61, 65, 245 accumulation of, 210, 286 pros and cons of, 211 removal of, 286 The Nature Conservancy (TNC), 239, 299, 311 thermic soil temperature regime, 108 thistles artichoke (Cynara cardunculus), 281 blessed milk (Silybum marianum), 178, 289 Canada (Cirsium arvense), 286, 290 Cirsium spp., 157 effects of grazing on, 277 Italian (Carduus pycnocephalus), 178, 289 Scotch thistle (Onopordum acanthium), 281, 289 slenderflower (Carduus tenuiflorus), 289 starthistles as invasive plants, 256 purple (Centaurea calcitrapa), 289 yellow (Centaurea solstitalis), 77, 93, 101, 152, 157, 163, 165, 203, 211, 214, 215, 220, 244–245, 249, 253, 272, 277, 279, 281, 283, 285, 286, 289, 290, 292, 293, 294–295, 320 Thlaspi montanum, 146 tiger salamander (Ambystoma tiginum), 185, 251 tillage, 286 tolerance, defined, 81 topsoil, removal of, 276 toyon (Heteromeles arbutifolia), 250 trampling, 51–52, 188, 194, 199, 245 transformer species, 67 trans-generational plasticity, 142 transient species, 68 tree-steppe, 23, 33 Triphysaria eriantha, 27, 201 Triteleia hyacinthine, 27 Triticeae, 13 trophic levels, 119 Tropidocarpum gracile, 185

Trust for Public Land, 311 tufted hairgrass (Deschampsia caespitosa), 29, 157, 193 two-phase resource dynamics hypothesis, 165 Typha, 251 umbrella species, 302 unconscious selection, 132 understory biomass, 176 effects of grazing on, 197 productivity, 204 oaks, influence of, 173, 176–177 soils, fertility of, 177 species composition, oaks, influence of, 173, 177–178 undesirable species, 67 urban development, 154 urban heat island effect, 269 urbanization, 3, 107 U.S. Bureau of Land Management, lands owned by, 238 U.S. Department of Agriculture Animal and Plant Health Inspection Service (APHIS), 286 Natural Resources Conservation Service (NRCA), 274 U.S. Department of Defense, 298 lands owned by, 238–239 U.S. Environmental Protection Agency (EPA), 251 U.S. Fish and Wildlife Service, 298 U.S. Forest Service, lands owned by, 238 valley grassland systems, defined, 281 valley larkspur (Delphinium recurvatum), 135 valley oak (Quercus lobata), 173 valley/south coastal grassland, 23–28 climate, influences of, 26 disturbance, 26 “new natives,” 26 perennial versus annual types, 26 rainfall, timing and amount of, 26 species abundance shifts, 26 variable outcrossing, causes and consequences of, 138 variation local molecular, 138–139 within-population, 137 Vegetation Classification Panel, Ecological Society of America, 23 Vegetation Dynamics Development Tool (VDDT), 215 vegetation dynamics hypothesis, 215 vegetation management, 235 Vegetation Management Program, California Department of Forestry and Fire Protection, 215–216 vegetation units, detailed mapping of, 319 velvet grass (Holcus lanatus), 119, 126, 157, 164, 205, 222 verbal blindness, 52 vernal grass (Anthoxanthum odoration), 134, 141, 144, 186, 194

INDEX

389

vernal pools, 26, 277 adaptation in, 138–139 burrowing activity near, 184 competition in, 163 defined, 163 grazing and, 204, 253 grazing management for, 251 resistance to invasion by alien species, 163 vertisols, 26, 108, 109–110 vesicular-arbuscular mycorrhizaei, 152 virulence, defined, 81 viruses barley yellow dwarf virus (BYDV), 82, 256 cereal yellow dwarf virus (CYDV), 82 Vogl, Richard, 64 voles (Microtus spp.), 174, 213 meadow vole (Microtus californicus), 184, 188 warm desert grasslands, 32–33 invasive species in, 32 six types of, 32 water, 87–93 availability, 72, 119–120 in coastal habitats, 90 and distance from the coast, 90, 127 effects of on vegetation composition and productivity, 91–92 elevation and, 127 experimental manipulations, 119 latitude and, 127 patterns of, 87–91 rooting patterns and phenology, 90–91 soil texture and soil depth, influence of, 90 spatial variations in, 87, 90, 93, 122 temporal variations in, 87–90, 93, 122 topography and, 127 deficit/surplus, 88 fog as a source of, 164 potential, diurnal fluctuations in, 177 quality, 268, 279 retention, 268 uptake, in exotic versus native perennial grasses compared, 78 water balance diagram, 88 water districts, 238, 239 waterfowl, 213 watershed, 273 watershed level of organization, 203 weather. See also atmosphere and climate fronts, role of in fire control, 216 influence of on fires, 207 weed(s) adaptive management of, 284 control, 271–272 agents approved for release in California, list of, 288–289 biological, 286–290 burns, 272 postplanting, 272

390

INDEX

weeds targeted for, list of, 287 developing a management strategy for, 295 hand pulling, 285 integrated approaches to managing, 295–296 invasion by, 76 land use objectives and, 295 preventing reinvasion, 295 weevils Bangastemus orientalis, 287 ustenopus villosus E, 287 Larinus curtus, 287 western camel (Camelops hestermus), 50 western horse (Equus cf. occidentalis), 51 western kingbird (Tyrannus verticalis), 189 western larkspur (Delphinium hesperium), 135 western meadowlark (Sturnella neglecta), 189 western scrub jay (Aphelocoma coerulescens), 173 wet climate, 90 wetlands, 22 conservation zoning and, 305 grasslands adjacent to, 26 grazing management for, 251 spring-fed, 251 wheatgrass (Thinopyrum intermedium), 295 white-tailed kite (Elanus leucurus), 184 wild barley (Hordeum vulgare), 139 Wilder Ranch State Park, 304 wildfires, 31 wildflowers, 52, 53, 59, 62, 73, 145, 212, 276 wildlife grazing, 205 wild oats (Avena spp.), 131, 156, 186, 211 slender wild oats (Avena barbata) 133, 136, 139, 141, 143, 149, 158, 199, 221, 224 wild oats (Avena fatua), 91, 134, 137, 141, 143, 147, 154, 163, 178, 187, 256 wild parsley (Ligusticum grayi), 63 wild radish (Raphanus sativus), 136 Williamson Act (California Land Conservation Act of 1965), 237, 300, 304, 307–308 Farmland Security Zones (FSZs), 307 grassland area under contracts, 307, 311 wind direction, 207, 208 wind dispersal, 186 wind speed, 207, 208, 219 within-population genetic variation, 133, 137 wolves (Canis spp.) dire (Canis dirus), 49 gray (Canis lupus), 49 woodland musk ox (Symbos cavifrons), 51 woody invaders, effects of on ecosystem structure, 78–79 woody species, 167–179 acorn production, predation, and dispersal, 173–174 competition with grasses, 175 conversion of shrubland to grassland, 169–171

establishment of shrubs into grassland, 171–172 differences among shrubland types, 171–172 intentional conversion, 170 unintentional conversion, 170–171 effects of oaks on grasslands, 176–178 frequent fire, 175–176 grassland–oak savanna dynamics, factors affecting, 173–178 oak canopy covering, 169 oak recruitment, 169, 173–176 oak savanna, defined, 169 oak woodland, defined, 169 shoot predation, herbivory, and browsing, 174–175 shrubland-stabilizing feedbacks, 172 understory productivity, complex interactions affecting, 176–177 understory species composition, oak effects on, 177–178 woolen breeches (Hydrophyllum capitatum var. alpinum), 63 wooly meadowfoam (Limnanthes floccose), 135, 245 Xánus, Janos, 185 xeric genotypes, 133, 140 xeric grasslands, 159 xeric soil moisture regime, 108 yampah (Perideridia bolanderi), 63 yarrow Achillea borealis, 142 Achillea millefolium, 61 year specialist genotypes, 134 yellow-billed magpies (Pica nuttalli), 173 yellow starthistle (Centaurea solstitalis) competition, sensitivity to, 203 distribution of, 157 ecosystem alterations by, 220 effects of fire on, 211, 214, 215, 272 efforts to control, 77, 285, 286, 289, 290, 294–295 grazing management to control, 244–245, 249, 253, 292 growth, effects of Nassella pulchra on, 165 “high concern” classification, 281 microbial community, alterations in, 152 mowing, 277 as poison to horses, 283 potential to increase, 320 revegetation and, 293 along roadsides, 279 soil moisture and, 93, 101 Yreka phlox (Phlox hirsute), 153 yucca, 61 zero effect studies, 122 Zigadenius fremontil, 214 zoning ordinances, 306 variances, 306