Ecosystems of disturbed ground [1st ed] 0444824200, 9780444824202, 0444417028, 9780444417022

Citation preview

ECOSYSTEMS OF THE WORLD 16

ECOSYSTEMS OF DISTURBED GROUND

ECOSYSTEMS OF THE WORLD Editor in Chief: David W. Goodall Centre for Ecosystem Studies, Edith Cowan University, Joondalup, W.A. (Australia)

I. TERRESTRIAL ECOSYSTEMS

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16.

A. Natural Terrestrial Ecosystems Wet Coastal Ecosystems Dry Coastal Ecosystems Polar and Alpine Tundra Mires: Swamp, Bog, Fen and Moor Temperate Deserts and Semi-Deserts Coniferous Forests Temperate Deciduous Forests Natural Grasslands Heathlands and Related Shrublands Temperate Broad-Leaved Evergreen Forests Mediterranean-Type Shrublands Hot Deserts and Arid Shrublands Tropical Savannas Tropical Rain Forest Ecosystems Wetland Forests Ecosystems of Disturbed Ground

17. 18. 19. 20. 21.

B. Managed Terrestrial Ecosystems Managed Grasslands Field Crop Ecosystems Tree Crop Ecosystems Greenhouse Ecosystems Bioindustrial Ecosystems

II. AQUATIC ECOSYSTEMS

22. 23.

A. Inland Aquatic Ecosystems River and Stream Ecosystems Lakes and Reservoirs

24. 25. 26. 27. 28.

B. Marine Ecosystems Intertidal and Littoral Ecosystems Coral Reefs Estuaries and Enclosed Seas Ecosystems of the Continental Shelves Ecosystems of the Deep Ocean

29.

C. Managed Aquatic Ecosystems Managed Aquatic Ecosystems

III. UNDERGROUND ECOSYSTEMS 30.

Subterranean Ecosystems

ECOSYSTEMS OF THE WORLD 16

ECOSYSTEMS OF DISTURBED GROUND Edited by Lawrence R. Walker Department of Biological Sciences, University of Nevada, Las Vegas, 4505 Maryland Parkway, Box 454004, Las Vegas, NV 89154-4004, USA

1999 ELSEVIER Amsterdam – Lausanne – New York – Oxford – Shannon – Singapore – Tokyo

Elsevier Science B.V. Sara Burgerhartstraat 25 P.O. Box 211, 1000 AE Amsterdam, The Netherlands © 1999 Elsevier Science B.V. All rights reserved. This work is protected under copyright by Elsevier Science B.V., and the following terms and conditions apply to its use: Photocopying Single photocopies of single chapters may be made for personal use as allowed by national copyright laws. Permission of the publisher and payment of a fee is required for all other photocopying, including multiple or systematic copying, copying for advertising or promotional purposes, resale, and all forms of document delivery. Special rates are available for educational institutions that wish to make photocopies for non-profit educational classroom use. Permissions may be sought directly from Elsevier Science Rights & Permissions Department, PO Box 800, Oxford OX5 1DX, UK; phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail: [email protected]. You may also contact Rights & Permissions directly through Elsevier’s home page (http://www.elsevier.nl), selecting first ‘Customer Support’, then ‘General Information’, then ‘Permissions Query Form’. In the USA, users may clear permissions and make payments through the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; phone: (978) 7508400, fax: (978) 7504744, and in the UK through the Copyright Licensing Agency Rapid Clearance Service (CLARCS), 90 Tottenham Court Road, London W1P 0LP, UK; phone: (+44) 171 631 5555; fax: (+44) 171 631 5500. Other countries may have a local reprographic rights agency for payments. Derivative Works Tables of contents may be reproduced for internal circulation, but permission of Elsevier Science is required for external resale or distribution of such material. Permission of the publisher is required for all other derivative works, including compilations and translations. Electronic Storage or Usage Permission of the publisher is required to store or use electronically any material contained in this work, including any chapter or part of a chapter. Except as outlined above, no part of this work may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the publisher. Address permissions requests to: Elsevier Science Rights & Permissions Department, at the mail, fax and e-mail addresses noted above. Notice No responsibility is assumed by the Publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. First edition 1999 Library of Congress Cataloging-in-Publication Data Ecosystems of disturbed ground / edited by L.R. Walker. − − 1st ed. p.

cm. − − (Ecosystems of the world ; 16)

Includes bibliographical references. ISBN 0−444−82420−0 R.

1. Ecology. 2. Nature− −Effect of human beings on. II. Series.

QH545.A1E2824

I. Walker, L.

1999

577.27− −dc21

99-41984 CIP

ISBN 0 444 82420 0 (Volume) ISBN 0 444 41702 8 (Series) ∞ The paper used in this publication meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper). 

Printed in the Netherlands

PREFACE As the human population inexorably grows, its cumulative impacts on the earth’s resources are hard to ignore. The ability of the earth to support more humans is dependent on the ability of humans to manage natural resources wisely. Because disturbance alters resource levels, effective management requires understanding of the ecology of disturbance. Editorship of this book was undertaken with several goals in mind. First, I wanted to present an organized summary of the many types of disturbances that impact the earth, with as global a focus as the existing literature allowed. The book is organized into chapters that deal primarily with natural disturbances (Chapters 2–13), anthropogenic disturbances (Chapters 14–20), overviews of natural processes that occur across disturbance types (Chapters 21–27), and human interactions with and responses to disturbance (Chapters 28–30). Chapter 31 explores a hierarchical view of disturbance; Chapter 32 examines the concept that the consequences of growth of the human population themselves represent the ultimate disturbance, and suggests ways to ameliorate human impacts. “Natural” and “anthropogenic” disturbances generally are interrelated. The focus of this book is on disturbances that have a direct physical impact on terrestrial systems, excluding primarily atmospheric phenomena such as acid rain or increases in carbon dioxide and decreases in ozone (but not wind), and primarily aquatic phenomena such as cultural eutrophication or chemical, thermal, and bacterial pollution of waterways. A second purpose was to enhance understanding of the concept of disturbance in order to manage it better. The development of theory related to disturbance is in its infancy. One of the few book-length overviews of the topic was that provided by Pickett and White (1985), in which they focused on alterations of relatively pristine habitats. With the wealth of examples from around the world presented in this volume, perhaps we can make further progress toward a general theory of disturbance. Alternatively, one may recognize that site-specific characteristics do not allow such generalizations. To those ends, authors were given some freedom to interpret disturbance as they deemed appropriate. Authors addressing a particular

disturbance type or disturbance in a particular habitat (Chapters 2–20) were asked to address, as far as possible, the disturbance regime, the damage caused by the disturbance, the responses of the biota to the disturbance, and interactions between disturbance types and between disturbance and humans. Authors of process chapters (Chapters 21–27) were to seek generalizations from a global perspective of short- and longer-term responses across various types of disturbed ecosystems. In the final Chapter 33, Willig and I evaluate common threads in the previous chapters and make contributions toward the development of a theory about disturbance. A third goal was to provide insights for land managers on how to incorporate lessons about disturbance into their efforts. Some of the chapters (e.g., Chapters 11, 14–16, 18–20) explicitly address managed systems such as pasture-land, urban habitats, and agriculture; other chapters address management issues more generally. Any level of generalization that can be made about disturbance responses will aid managers of disturbed ecosystems. Human well-being on this increasingly crowded planet depends on the success to which land management policies (see Chapter 30) apply such lessons. I wish to thank the many chapter authors for their patience, hard work, excellent insights, and chapter reviews; the 36 non-author peer reviewers for their constructive criticisms that made my job easier; the series editor David Goodall for overall guidance and careful editing; Rachel Lawrence and Dorothy Dean for assistance with correspondence and proofing; the University of Nevada, Las Vegas (UNLV) for providing substantial logistical support; colleagues and graduate students of ecology at UNLV for engaging discussions; Michael Willig for helping me to synthesize the whole volume; and my wonderful wife, Elizabeth Powell, for her continual support. During the development of the ideas presented here, I was supported by NSF grants BSR-8811902 and DEB-9411973 to the Institute for Tropical Ecosystem Studies, University of Puerto Rico, and the International Institute of Tropical Forestry, as part of the Long-Term Ecological Research Program in v

vi

PREFACE

the Luquillo Experimental Forest. Additional support was provided by the U.S. Fish and Wildlife Service, the U.S. National Park Service, and the U.S. Forest Service. Finally, the completion of this book was facilitated by a sabbatical leave from UNLV. Lawrence R. Walker Editor

REFERENCES Pickett, S.T.A. and White, P.S., 1985. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, Florida, 472 pp.

LIST OF CONTRIBUTORS E.B. ALLEN Department of Botany and Plant Sciences University of California Riverside, CA 92521-0124, USA

M.L. CADENASSO Institute of Ecosystem Studies Box AB Millbrook, NY 12545-0129, USA

M.F. ALLEN Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA

C.R. CARROLL Institute of Ecology University of Georgia Athens, GA 30602, USA J.A. COOKE School of Life and Environmental Sciences University of Natal Durban 4041, South Africa

D. BAINBRIDGE Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA

C.M. D’ANTONIO Department of Integrative Biology University of California Berkeley Berkeley, CA 94720, USA

A.H. BALDWIN Natural Resources Management Program Department of Biological Resources Engineering University of Maryland College Park, MD 29742, USA

R. DEL MORAL Department of Botany University of Washington Box 355325 Seattle, WA 98195-5325, USA

C.J. BARROW Centre for Developmental Studies University College of Swansea University of Wales Swansea SA2 8PP, United Kingdom

S. DEMARAIS Department of Wildlife and Fisheries Mississippi State University Mississippi State, MS 39762, USA

D. BINKLEY Department of Forest Sciences Colorado State University Fort Collins, CO 80523, USA

M.B. DICKINSON Department of Biological Sciences Florida State University Tallahassee, FL 32306-2043, USA

I.K. BRADBURY Dept. of Geography University of Liverpool P.O. Box 147 Liverpool L69 3BX, United Kingdom

C. DOLJANIN Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA

N.V.L. BROKAW Manomet Center for Conservation Sciences P.O. Box 1770 Manomet, MA 02345, USA vii

viii

LIST OF CONTRIBUTORS

T.L. DUDLEY Department of Integrative Biology University of California Berkeley Berkeley, CA 94720, USA

G.S. HARTSHORN Organization for Tropical Studies Box 90630 Durham, NC 27708-0630, USA

G.E. ECKERT 1434 Pine Street Norristown, PA, USA

C. HARVEY Department of Entomology Cornell University 5126 Comstock Hall Ithaca, NY 14853-0901, USA

F. EDWARDS Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA O. ENGELMARK Swedish Centre for Ecological Sustainability (Swecol) S-901 87 Ume¨a, Sweden C.M. GHERSA Departamento de Ecolog´ıa Facultad de Agronom´ıa, Universidad de Buenos Aires Av. San Martin 4453 1417 Buenos Aires, Argentina M. GIAMPIETRO Istituto Nazionale della Nutrizione Unit of Special Food Technology Via Ardeatina 546 00178 Rome, Italy S.Yu. GRISHIN Institute of Biology and Pedology Russian Academy of Sciences Vladivostok 690022, Russia P.J. GUERTIN Environmental Division U.S. Army Construction and Engineering Research Lab Box 4005 Champaign, IL 61820, USA S. HARNEY Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA

C. HINKSON Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA R.J. HOBBS CSIRO Wildlife and Ecology Private Bag, PO Wembley WA 6014, Australia D.W. JOHNSON Biological Sciences Center Desert Research Institute P.O. Box 60220 Reno, NV 89506, USA E.E. JORGENSEN Department of Range, Wildlife and Fisheries Management Texas Tech University Lubbock, TX 79409, USA ´ ´ V. KOMARKOV A Villa Elisabeth 5 CH-1854 Leysin, Switzerland ´ R.J.C. LEON Departamento de Ecolog´ıa Facultad de Agronom´ıa, Universidad de Buenos Aires Av. San Martin 4453 1417 Buenos Aires, Argentina J. LORETI Facultad de Agronom´ıa, Universidad de Buenos Aires Av. San Martin 4453 1417 Buenos Aires, Argentina

LIST OF CONTRIBUTORS

ix

M.D. LOWMAN The Mary Selby Botanical Gardens 811 S. Palm Ave. Sarasota, FL 34236, USA

S.T.A. PICKETT Institute of Ecosystem Studies Box AB Millbrook, NY 12545-0129, USA

R. MACALLER Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA

D. PIMENTEL Department of Entomology Cornell University 5126 Comstock Hall Ithaca, NY 14853-0901, USA

M. MACK Department of Integrative Biology University of California Berkeley Berkeley, CA 94720, USA

M. RILLIG Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA

J.A. MACMAHON College of Science Utah State University UMC 4400 Logan, UT 84322-4400, USA

P.W. RUNDEL Department of Biology University of California Los Angeles Los Angeles, CA 90024, USA

J.A. MATTHEWS Department of Geography University of Wales Swansea Singleton Park Swansea SA2 8PP, United Kingdom

T.D. SCHOWALTER Entomology Department Oregon State University Corvallis, OR 97331-2907, USA

M.A. MCGINLEY Ecology Program Department of Biological Sciences Texas Tech University Lubbock, TX 79409-313, USA

B. SCHULTZ Biological Sciences Center Desert Research Institute P.O. Box 60220 Reno, NV 89506, USA

K.L. MCKEE National Wetlands Research Center 700 Cajundome Blvd. Lafayette, LA 70506, USA

M. SEMMARTIN Facultad de Agronom´ıa, Universidad de Buenos Aires Av. San Martin 4453 1417 Buenos Aires, Argentina

M. OESTERHELD Departamento de Ecolog´ıa Facultad de Agronom´ıa, Universidad de Buenos Aires Av. San Martin 4453 1417 Buenos Aires, Argentina

¨ C. SIGUENZA Department of Botany and Plant Sciences University of California Riverside, CA 92521-0124, USA

J.M. PARUELO Facultad de Agronom´ıa, Universidad de Buenos Aires Av. San Martin 4453 1417 Buenos Aires, Argentina

R.E. SOJKA USDA Agricultural Research Service Northwest Irrigation and Soils Research Lab 3793N-3600E Kimberley, ID 83341, USA

x

LIST OF CONTRIBUTORS

U. STARFINGER ¨ Institut f¨ur Okologie Technische Universit¨at Berlin Schmidt-Ott Strasse 1 D-12165 Berlin, Germany

F. WIELGOLASKI Department of Biology, University of Oslo, Box 1045, Blindern N-0316 Oslo, Norway

H. SUKOPP ¨ Institut f¨ur Okologie Technische Universit¨at Berlin Schmidt-Ott Strasse 1 D-12165 Berlin, Germany

M.R. WILLIG Ecology Program Department of Biological Sciences and the Museum Texas Tech University Lubbock, TX 79409-313, USA

D.J. TAZIK Environmental Division U.S. Army Construction and Engineering Research Lab Box 4005 Champaign, IL 61820, USA

S.D. WILSON Department of Biology University of Regina Regina, Saskatchewan S45 0A2, Canada

L.R. WALKER Department of Biological Sciences University of Nevada, Las Vegas 4505 Maryland Parkway Box 454004 Las Vegas, NV 89154-4004, USA S.L. WEBB Biology Department Drew University Madison, NJ 07940-4000, USA D.F. WHIGHAM Smithsonian Environmental Research Center Box 28 Edgewater, MD 21037, USA J.L. WHITMORE Vegetation Management and Protection Research USDA Forest Service, Box 96090 Washington, DC 220090-6090, USA

J. WU Department of Life Sciences Arizona State University West Box 37100 Phoenix, AZ 85069, USA L.C. YOSHIDA Department of Botany and Plant Sciences University of California Riverside, CA 92521-0124, USA T.A. ZINK Department of Biology Soil Ecology and Restoration Group San Diego State University San Diego, CA 92182-0057, USA

CONTENTS PREFACE

. . . . . . . . . . . . . . . . . . . . . . .

LIST OF CONTRIBUTORS

. . . . . . . . . . . . . .

Chapter 1. AN INTRODUCTION TO TERRESTRIAL DISTURBANCES by L.R. Walker and M.R. Willig . . . . . . Chapter 2. DISTURBANCE REGIMES AND ECOSYSTEM RESPONSE ON RECENTLY-DEGLACIATED SUBSTRATES by J.A. Matthews . . . . . . . . . . . . .

v vii

1

17

Chapter 3. STRESS AND DISTURBANCE IN COLD REGION ECOSYSTEMS by V. Kom´arkov´a and F.E. Wielgolaski . .

39

Chapter 4. ECOLOGICAL EFFECTS OF EROSION by D. Pimentel and C. Harvey . . . . . . .

123

Chapter 5. VOLCANIC DISTURBANCES AND ECOSYSTEM RECOVERY by R. del Moral and S.Yu. Grishin . . . . .

137

Chapter 6. BOREAL FOREST DISTURBANCES by O. Engelmark . . . . . . . . . . . . . .

161

Chapter 7. DISTURBANCE BY WIND IN TEMPERATEZONE FORESTS by S.L. Webb . . . . . . . . . . . . . . . Chapter 8. BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS by D.F. Whigham, M.B. Dickinson and N.V.L. Brokaw . . . . . . . . . . . . . . . Chapter 9. FOREST HERBIVORY: INSECTS by T.D. Schowalter and M.D. Lowman

. . .

Chapter 10. DISTURBANCE IN MEDITERRANEANCLIMATE SHRUBLANDS AND WOODLANDS by P.W. Rundel . . . . . . . . . . . . . . Chapter 11. GRAZING, FIRE, AND CLIMATE EFFECTS ON PRIMARY PRODUCTIVITY OF GRASSLANDS AND SAVANNAS by M. Oesterheld, J. Loreti, M. Semmartin and J.M. Paruelo . . . . . . . . . . . . . . . .

187

Chapter 12. DISTURBANCE IN DESERTS by J.A. MacMahon . . . . . . . . . . . .

307

Chapter 13. DISTURBANCE REGIMES IN NORTH AMERICAN WETLANDS by K.L. McKee and A.H. Baldwin . . . . .

331

Chapter 14. MINING by J.A. Cooke

. . . . . . . . . . . . . . .

365

Chapter 15. DISTURBANCE ASSOCIATED WITH MILITARY EXERCISES by S. Demarais, D.J. Tazik, P.J. Guertin and E.E. Jorgensen . . . . . . . . . . . . . . .

385

Chapter 16. DISTURBANCE IN URBAN ECOSYSTEMS by H. Sukopp and U. Starfinger . . . . . .

397

Chapter 17. DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS by C.M. D’Antonio, T.L. Dudley and M. Mack . . . . . . . . . . . . . . . . . .

413

Chapter 18. DISTURBANCE IN TEMPERATE FORESTS OF THE NORTHERN HEMISPHERE by D. Binkley . . . . . . . . . . . . . . .

453

Chapter 19. ANTHROPOGENIC DISTURBANCE AND TROPICAL FORESTRY: IMPLICATIONS FOR SUSTAINABLE MANAGEMENT by G.S. Hartshorn and J.L. Whitmore . . .

467

Chapter 20. SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA by C.M. Ghersa and R.J.C. Le´on

. . . . .

487

Chapter 21. PHYSICAL ASPECTS OF SOILS OF DISTURBED GROUND by R.E. Sojka . . . . . . . . . . . . . . .

503

Chapter 22. SOIL MICROORGANISMS by M.F. Allen, E.B. Allen, T.A. Zink, S. Harney, L.C. Yoshida, C. Sig¨uenza, F. Edwards, C. Hinkson, M. Rillig, D. Bainbridge, C. Doljanin and R. MacAller . . . . . . .

521

Chapter 23. RESPONSES OF CARBON AND NITROGEN CYCLES TO DISTURBANCE IN FORESTS AND RANGELANDS by D.W. Johnson and B. Schultz . . . . . .

545

223

253

271

287

xi

xii Chapter 24. DISTURBANCE AND PRIMARY PRODUCTION IN TERRESTRIAL ECOSYSTEMS by I.K. Bradbury . . . . . . . . . . . . . . Chapter 25. PATTERNS AND PROCESSES IN PRIMARY SUCCESSION by L.R. Walker . . . . . . . . . . . . . . Chapter 26. PLANT INTERACTIONS DURING SECONDARY SUCCESSION by S.D. Wilson . . . . . . . . . . . . . . Chapter 27. THE RESPONSE OF ANIMALS TO DISTURBANCE AND THEIR ROLES IN PATCH GENERATION by M.R. Willig and M.A. McGinley . . . . Chapter 28. HOW HUMANS RESPOND TO NATURAL OR ANTHROPOGENIC DISTURBANCE by C.J. Barrow . . . . . . . . . . . . . . .

CONTENTS

571

585

611

Chapter 31. PATCH DYNAMICS AND THE ECOLOGY OF DISTURBED GROUND: A FRAMEWORK FOR SYNTHESIS by S.T.A. Pickett, J. Wu and M.L. Cadenasso

707

Chapter 32. ECONOMIC GROWTH, HUMAN DISTURBANCE TO ECOLOGICAL SYSTEMS, AND SUSTAINABILITY by M. Giampietro . . . . . . . . . . . . .

723

Chapter 33. DISTURBANCE IN TERRESTRIAL ECOSYSTEMS: SALIENT THEMES, SYNTHESIS, AND FUTURE DIRECTIONS by M.R. Willig and L.R. Walker . . . . . .

747

GLOSSARY

769

. . . . . . . . . . . . . . . . . . . . . .

633 SYSTEMATIC LIST OF GENERA 659

AUTHOR INDEX

. . . . . . . . . .

773

. . . . . . . . . . . . . . . . . . .

777

SYSTEMATIC INDEX

Chapter 29. RESTORATION OF DISTURBED ECOSYSTEMS by R.J. Hobbs . . . . . . . . . . . . . . .

673

Chapter 30. ENVIRONMENTAL POLICIES AS INCENTIVES AND DISINCENTIVES TO LAND DISTURBANCE by G.E. Eckert and C.R. Carroll . . . . . .

689

GENERAL INDEX

. . . . . . . . . . . . . . . .

821

. . . . . . . . . . . . . . . . . .

833

Chapter 1

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES Lawrence R. WALKER and Michael R. WILLIG

is now 5.8×109 and is projected to reach 10–12×109 by the year 2040. What are the consequences of such growth? What is the carrying capacity of the earth (Cohen, 1995)? Can human intelligence and technology prevent or even postpone a global collapse? Estimates of the ecological footprint (a concept that calculates how much arable land is needed to sustain a given level of energy consumption per member of a human population; Wackernagel and Rees, 1996) of those countries with the highest standards of living already are 15 times greater in area than the geographical space they occupy. Clearly, the world does not have the resources to sustain the entire human population at a standard of living similar to that in the more affluent nations of the world. Giampietro (Chapter 32, this volume) explores ways in which wise resource management and curtailment of resource abuse can improve the future prospects of humans and the biosphere.

WHY STUDY DISTURBANCE?

Dramatic, large-scale natural disturbances (e.g., volcanic eruptions, fires, hurricanes, floods) are important to understand because they destroy property, cause human injury, and disrupt emotional lives. Human interference with natural disturbances (e.g., fire suppression) may actually make them more destructive (e.g., larger, hotter fires: Bond and van Wilgen, 1996). Disturbances are also important to all living organisms because they have beneficial effects such as nutrient recycling, resetting of successional pathways, and maintenance of species diversity (Luken, 1990). The exponential increase in human population density guarantees that more people are affected by natural disturbances every year. It is clear that one needs to continue efforts to predict and avoid disturbances, minimize damage, and maximize the ability of human society to restore degraded systems. Some anthropogenic disturbances are well publicized (e.g., spills of oil or toxic waste, bomb explosions). Yet the more gradual disturbances that do not receive as much attention, such as urbanization, excavation of minerals, soil erosion as a result of agriculture, or logging of forests, may have far greater consequences. In fact, anthropogenic disturbances are ubiquitous and all ecosystems of the world are disturbed at least partially by human activities. Both natural and anthropogenic disturbances clearly impact the entire earth. Understanding how to live with or mitigate natural disturbances, and moderate the consequences of human actions, is imperative (Thomas, 1956; Botkin et al., 1989). The consequences of increased human population represent the ultimate disturbance. Humans currently consume or utilize 40% of the earth’s primary production (Vitousek et al., 1986). The human population

PERSPECTIVES ON DISTURBANCE

Disturbances have been the subject of many myths and legends. Gods have been associated with disturbances such as volcanoes (Ixtocewatl and Pococatepetl in Mexico; Vulcan in ancient Rome; Pele in Hawaii), windstorms (Luquillo in Puerto Rico; Hurakan in Mayan culture), floods (Janaina in Brazil; Poseidon in ancient Greece), and fire (Loki in Norse mythology; Prometheus in ancient Greece). The biblical Noah dealt with a flood, and Moses’ enemies were subjected to a herbivore (locust) outbreak. Disturbances have directly altered human history. Volcanoes have destroyed cities (e.g., Pompeii in Italy; St. Pierre in Martinique) and altered world climates (Krakatau in Indonesia) (Sheets and Grayson, 1979; 1

2

Simkin and Fiske, 1983). Hurricanes have repeatedly damaged buildings and biota (e.g., Hurricane Hugo in the Caribbean and the eastern United States: B´enitoEspinal and B´enito-Espinal, 1991; Finkl and Pilkey, 1991; Walker et al., 1991, 1996). Fertile soils along river floodplains (e.g., the Nile, Tigris, or Euphrates) have nurtured civilizations, but often at the cost of extensive losses of lives and property (Officer and Page, 1993). Famous fires have altered the histories of cities such as Chicago, Rome and San Francisco, and the vegetation of entire continents (Komarek, 1983). Biotic disturbances are perhaps most damaging. The Black Death killed one-third of all people in medieval Europe, and many Native Americans died from diseases such as smallpox and malaria introduced by Europeans (cf. Crosby, 1986; Officer and Page, 1993). Cultural and environmental concerns traditionally have been shaped by the interplay between resource availability and the local disturbance regime. Degradation of land caused by erosion and deforestation was noted by Greek and Roman writers, and Confucianism in China addressed environmental concerns (Barrow, 1991). Humans typically have responded to natural disturbances by management (use of fire by many native cultures), exploitation (use of early-successional plants for food), or avoidance (minimal use of deserts, lava fields, and glacial valleys). Attitudes toward natural resources can evolve from exploitation to conservation when human population densities reach local carrying capacities. However, the demise of some societies [e.g., the Maya in Central America, the Hohokam in Arizona (U.S.A.), and the Assyrians in Mesopotamia] has been attributed in part to the collapse of the local resource base from over-exploitation (Thomas, 1956). The remarkable ability of humans to accommodate to naturally or anthropogenically caused environmental change (or to migrate out of disturbed areas – as with the Dust Bowl in Oklahoma, U.S.A.: Worster, 1979) suggests that most disturbances modify but do not destroy cultures. Most landscapes are now the product of a long history of human land use (e.g., the Mediterranean basin: Rundel, Chapter 10, this volume). For the last 100–200 years, Western cultures have been systematically recording observations about various natural disturbances (e.g., volcanoes: Whittaker et al., 1989; glaciers: Chapin et al., 1994) and anthropogenic disturbances (e.g., changes in levels of atmospheric carbon dioxide: Vitousek, 1994) and ecosystem responses to disturbance (e.g., succession:

Lawrence R. WALKER and Michael R. WILLIG

Clements, 1928). Such long-term observations allow an examination of disturbance on various time scales with the partitioning of short-term fluctuations from longerterm cycles (Magnuson, 1990). They also facilitate the distinction of human impacts from natural fluctuations. Recognition of the role of humans in global warming or acid rain, and the growing impacts of mining, agriculture, and urbanization have increased environmental awareness in recent decades. This awareness has fostered the growth of environmental politics (e.g., the Green Parties in Europe), entrepreneurism (e.g., the purchase of natural areas by private agencies such as the Nature Conservancy operating from the United States), and cooperation at the local level (restoration activities), the regional level (credits to companies that reduce pollution), and the global level (relief of national debt in exchange for establishment of nature reserves). Interactions of culture and disturbance are further discussed in this volume by Ghersa and Leon (Chapter 20), Barrow (Chapter 28), Hobbs (Chapter 29), Eckert and Carroll (Chapter 30) and Giampietro (Chapter 32).

DEFINITIONS OF DISTURBANCE

As the literature on disturbance ecology has proliferated in the last two decades, so too has the lexicon. Nonetheless, maturation of the science requires a precise use of terminology along with straightforward clarification when terms are used in different ways. At the same time, terms should be sufficiently general so that they are useful to an appreciable segment of the practitioners in the discipline. On occasion, growth of a discipline can be stymied significantly by vague or illdefined terminology, in part because synthesis requires incisive understanding and in part because confusion over terminology can lead to division among practitioners who disagree about definitions. Such semantic differences can give the impression of disagreement over substantive or conceptual issues, lead to heated or senseless debate, and delay the maturation of a scientific discipline. We do not attempt to resolve such semantic and conceptual differences here. Indeed, authors contributing to this volume were given broad latitude in the use of terms so as to engender individual creativity. Nonetheless, we follow White and Pickett (1985) and provide an introduction to widely accepted meanings of selected terms in the lexicon of disturbance ecology,

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES

so that the general reader will have an appreciation of the scope of the discipline, and specialists will be motivated to provide more detailed definitions or alternate terminology as appropriate (see Pickett et al., Chapter 31, this volume). A disturbance is a relatively discrete event in time and space that alters the structure of populations, communities, and ecosystems. It can do so by altering the density, the biomass, or the spatial distribution of the biota, by affecting the availability and distribution of resources and substrate, or by otherwise altering the physical environment. It often results in the creation of patches and the modification of spatial heterogeneity. Disturbance is a relative term that requires explicit delineation of the system of concern, including the spatial and temporal scale of the components of interest. The cause of a disturbance may be thought of as the agent or entity initiating the changes in the structure of the ecological system of interest. For example, highspeed winds are agents of disturbance for hurricanes. If the cause originates outside the system of interest, as is the situation for hurricanes, the disturbance is considered to be exogenous, whereas if the cause of the disturbance originates inside the system of interest, as when a tree-fall results from natural senescence, the disturbance is considered to be endogenous. Clearly, definition of the system of interest is integral to such considerations, and a clear distinction is not always possible. The likelihood of an exogenous disturbance may be affected by the state of the system of interest and characteristics of endogenous disturbances may be affected by characteristics of previous exogenous disturbances. Indeed, the dichotomy between purely endogenous and exogenous disturbances might more appropriately be considered as a continuum of intermediate possibilities. Disturbances are most often characterized by the central tendency, variability, and distribution of three attributes: frequency, extent, and magnitude. Frequency measures the number of events per unit of time or the probability that an event will occur. Extent is the actual physical area affected by a disturbance. It can be estimated from the area of a single event (e.g., a tree-fall), or from the sum of the areas affected by equivalent events over a particular time period (e.g., gap area created by all tree-falls in a year). Extent is often reported as the proportion of an entire landscape in which a particular disturbance occurred

3

in a given time period. Magnitude includes two interrelated attributes: intensity and severity. Intensity is the physical force of an event (e.g., wind-speed for hurricanes), whereas the impact on or consequences to the system of interest is the severity (e.g., the biomass of trees that were killed by passage of a hurricane). Intensity and severity are usually correlated, and the terms often are used interchangeably, at least in part, because the physical forces of many disturbances, especially those generated by the biota (e.g., treefalls, rodent mounds, insect outbreaks) are difficult to quantify. Clearly, severity reflects the response of the biota to the disturbance and may not be fully documented until a considerable time has elapsed since the disturbance event impinged on the system of interest. Most systems are simultaneously subjected to a number of disturbances (e.g., hurricanes, landslides, tree-falls, herbivory, droughts, and human activities all affect the structure and function of Caribbean forests). The sum of all disturbances at a particular place and time is termed the disturbance regime. The different disturbance events enhance or diminish the frequency, extent, or magnitude of other disturbances. Such interactions are considered synergisms, and are important considerations to address in understanding disturbance and recovery in ecological systems.

TYPES OF DISTURBANCE

Because virtually every habitat experiences some level of disturbance, no book can easily cover the entire topic. This book focuses on disturbances that physically impact the ground. It does not address atmospheric or aquatic disturbances. Primarily natural disturbances (Chapters 2–13) can be categorized by the four classical elements: earth, air, water, and fire (Table 1.1). Disturbances linked to the earth are independent of all causal factors other than tectonic forces (del Moral and Grishin, Chapter 5, this volume). Disturbances involving air, water, and fire are primarily driven by an interplay of climatic, topographic, and soil factors. In addition, biotic variables influence fire and are represented by both non-human disturbances (e.g., herbivory) and human disturbances (Table 1.1). Disturbances often trigger other disturbances, so that there is an interlacing web of disturbance interactions (for a detailed example, see Fig. 33.2 below). For instance, volcanoes can trigger earthquakes, earthquakes

4

Lawrence R. WALKER and Michael R. WILLIG

Table 1.1 Examples of some of the major types of disturbance of the earth 1 Element

Primary disturbance 2

Earth (tectonic)

earthquake (1) erosion (>50) volcano (1)

Air

hurricane (15) tornado (50)

Biota – non-human

herbivory (nd) invasion (nd) other animal activity 3 (nd)

Biota – human

agriculture (45) forestry (10) mineral extraction (1) military activity 4 (1–40) transportation 5 (5) urban (3)

1

Data from many sources; nd = no data available. Approximate percent of earth’s terrestrial surface regularly affected by each disturbance is in parentheses. 3 Includes building, excavating, waste products, movement, death, diseases, parasites. 4 U.S.A., 1%; Vietnam, 40%. 5 Includes motorized and non-motorized transportation. 2

or hurricanes can trigger landslides, hurricanes or landslides can induce flooding, and flooding can cause landslides. These interactions may augment, diminish, or neutralize the interacting disturbances. Anthropogenic disturbances are, of course, always interacting with natural disturbances (e.g., road-building can trigger a landslide). A hierarchical view of disturbance types (cf. O’Neill et al., 1986; Pickett et al., 1987) may be most useful in examining disturbance interactions, and in making spatial and temporal scales explicit for each disturbance under consideration. When the common types of disturbance of the world (from Table 1.1) are compared by frequency, extent, and severity using a subjective ranking procedure (1, least; 5, most), several patterns emerge (Fig. 1.1). Primarily anthropogenic disturbances are usually greater in extent (mean score = 3.2) than

Fig. 1.1. The frequency, spatial extent, and severity of 19 types of disturbance throughout the world based on their subjectively ranked scores from 1 (least) to 5 (most). Intensity and severity scores were highly correlated and thus are represented on a single axis. Disturbances are: AG, agriculture; AN, animal activities; DR, drought; EA, earthquakes; ER, erosion; FI, fire; FL, flooding; FO, forestry; GL, glaciers; HE, herbivory; HU, hurricanes; IN, invasions; MI, mining; ML, military; TF, tree falls; TO, tornadoes; TR, transportation; UR, urban; VO, volcanoes. Anthropogenic disturbances are shaded. Uncircled letters occupy the same location as adjacent circles (VO, GL; EA, TO; IN, AG).

natural disturbances (mean score = 2.1), presumably because of the cosmopolitan distribution of humans. Anthropogenic disturbances are also slightly more severe (mean score = 3.8) than natural disturbances (mean score = 3.0), but similar in frequency (mean scores 3.1 and 2.8, respectively). Of the five most severe disturbance types (score = 5), natural disturbances (glaciers and volcanoes) were less extensive and frequent than anthropogenic disturbances (mining, transportation, urban development). Transportation was rated uniquely high in both extent and severity. Other outliers were herbivory, tree-falls, and animal activities, all of which received very low scores for severity, but high scores for frequency. Most important, perhaps, is the broad range of extent, severity, and frequency among the disturbance types, particularly those representing natural disturbances. At relatively large spatial scales (~104 –1010 m2 ) and long temporal scales (~102 –104 yr), many areas of the earth are dominated by only one or a few major disturbance types. Inside the front cover of this book, we have mapped areas where disturbances related to

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES

earth, air, water, and fire predominate on terrestrial surfaces of the earth. Volcanoes and earthquakes result from plate tectonics (earth element) and predominate around the rim of the Pacific Ocean and in central Asia. Hurricanes (air element) develop in the tropics, but occasionally reach latitudes >45º N or S. Tornadoes reach further inland than hurricanes. Less severe windstorms are nearly ubiquitous at smaller spatial scales and were not included in the map. Floods or ice (excess of the water element) are important disturbances along river corridors and in boreal and polar regions. Drought (deficiency of the water element) is primarily a factor in mid-latitude, hot deserts, but also in northeastern Brazil (Mares et al., 1985). Droughts and floods are dictated largely by ocean currents, global wind patterns, and regional topography, although human activities often influence both droughts (e.g., desertification) and flooding (river channelization). Fire is the most ubiquitous type of terrestrial disturbance after human urban and agricultural activities (Bond and van Wilgen, 1996). It is important in tundra, coniferous forests, temperate grasslands and shrublands, and tropical grasslands and savannas, although only the most flammable biomes (coniferous forests and Mediterranean-climate shrublands) are shown. Biotic disturbances can be considered a fifth category of disturbance. Non-human biotic disturbances include plant and animal invasions, herbivory, and other animal activities (e.g., excavating, building, movement, waste products, disease, and parasitism). These activities are too ubiquitous and small in scale to map globally. In contrast, anthropogenic disturbances, equally ubiquitous but occurring at larger spatial scales,

5

can more readily be mapped globally (Fig. 1.2). There is a strong similarity between the distributions of human population (Fig. 1.2A) and common human disturbances (Figs. 1.2B, 1.2C, 1.2D). Current anthropogenic disturbances reflect human land-use patterns that are a consequence of historical settlements based primarily on the presence of soils suitable for agriculture (Fig. 1.2B), and appropriate waterways or land routes for transportation. More recent urbanization reflects primarily transportation centers (Fig. 1.2C) that have excellent access to power sources or to agricultural products (cf. Cronon, 1991). Many humans (45%) now live in or near cities, and this trend is accelerating. Nevertheless, some human activities such as mineral extraction (Fig. 1.2D) and military installations may actually promote low human population densities, but still represent severe disturbance (e.g., northern Alaska, northern Venezuela, eastern Saudi Arabia). Inside the back cover of this book, we have mapped all human influences together, using four hemeroby classes (see Sukopp and Starfinger, Chapter 16, this volume) representing degrees of human influence: (1) minimal: mountains, tundra, undeveloped forest; (2) moderate: low human population densities, some agriculture; (3) major: moderate human population densities, intense agriculture (e.g., deep plowing, clearcutting, biocides); and (4) maximal: high urban population densities, sealed or poisoned land surfaces. This measure of combined influences of humans emphasizes that most damage occurs where population densities are high. Agriculture and resource extraction, although often locally severe, do not alter the environment as much as pavement and urban buildings.

6

Lawrence R. WALKER and Michael R. WILLIG

Fig. 1.2A. Global distribution of four aspects of anthropogenic disturbance. A. Human population distribution. Grey levels indicate different population densities: 100 km−2 . Various sources were used, including: Oxford World Atlas (1973) and The Times Atlas of the World (1990, 1995).

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES

Fig. 1.2A (continued).

7

8

Lawrence R. WALKER and Michael R. WILLIG

Fig. 1.2B. Agriculture excluding forestry. Grey levels indicate relative intensity:

sparse;

low;

moderate;

high.

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES

Fig. 1.2B (continued).

9

10

Lawrence R. WALKER and Michael R. WILLIG

Fig. 1.2C. Surface transportation (roads and railroads). Grey levels indicate relative intensity:

sparse;

low;

moderate;

high.

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES

Fig. 1.2C (continued).

11

12

Lawrence R. WALKER and Michael R. WILLIG

Fig. 1.2D. Mineral resource extraction (including oil, gas and coal). Grey levels indicate relative intensity: high.

sparse;

low;

moderate;

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES

Fig. 1.2D (continued).

13

14

Lawrence R. WALKER and Michael R. WILLIG

Table 1.2 Distribution of topics discussed in each chapter. Parentheses indicate minor topics Chapter

Element 1

Geographic region 2

Ecoregion 3

Trophic level 4

Theme 5

2

1,2,3

1,2,(3),(4),5,6

1,5

1,(2),3

1,3,(4),7,8

3

1,2,3,4,5

2,3,4,5,6,7,8

1,2,4,6

1,2,(3)

2,(4),5,6,7,9,12

4

1,2,3,5

(1),3,(5),6,7

(2),5,8

1,(2),3

1,5,6,10

5

1,(2),3

3,4,(5),6,8

4,5

1,2,3

1,2,3,8,10,12

6

2,4,5

3,5,6

5

1,2

2,3,4,7,(10)

7

2,5

3,4,5,6,7,8

5,6

1,2,3

1,2,3,(4),5,8,10,13

8

2,(4)

1,3,4,6,7,8

5,6

1

1,2,3,5,6,8,14

9

5

4,6

5

1,2

1,2,3,5,6,9,11

10

4,5

1,4,5,6,7

1,2,4,5,8

1

1,10,12

11

4,5

1,6,7

(1),2,3,(5)

1,2,(3)

1,5,6,7,9,14

12

2,3,4,5

1,3,4,5,6,7

1

1,2,(3)

1,2,3,(4),14

13

(2),3,4,5

3,6,8

6

1,2

1,2,3,4,5,6,(7),9,10,(14)

14

1,5

1,(3),4,5,6,(7)

1,2,3,5

1

1,3,5,6,13

15

5

4,5,6

1,2,4,5,6

1,2

1,10,13

16

5

(3),5,6

5,7

1,2

(2),3,(5),10,12

17

(1),(2),(3),4,5

1,4,5,6,8

1,2,3,4,5,6,8

1,2

1,3,4,5,(10),12

18

1,2,4,5

4,5,6

5,8

1,2

2,3,5,6,(10)

19

1,2,3,4,5

1,3,4,6,7,8

5

1,2

1,2,3,9,10,12,13

20

5

7

2,8

1,(3)

2,3,4,5,7,(8),10,12,14

21

1,3,5

6

(6),8

1,2

1,5

22

1,4,5

5,6

1,4,5

1,2,3

3,5,13,14

23

5

1,5,6

2,4,5

1,2,3

1,3,4,5,(6)

24

1,4,5

1,4,5,6

2,4,5

1,2

1,5,6

25

1,2,3,5

1,2,3,4,5,6,8

5

1,2,3

1,3,4,5,6

26

5

(4),6

2,5,(6),8

1

2,3,4,5,6

27

2,4,5

4,6,8

2,4,5,8

1,2,3

1,2,3,4,5,6,10,14

28

4,5

4,5,6,7

(2),(4),(5)

2

1,(2),5,7,9,13,14

29

5

4,6,7

2,4,5,(6),8

1,2,3

1,5,7,9,11,12,13,14

30

5

1,3,6,7,8

(1),2,5,6,8

1,2,3

3,7,9,10,13,14

31

1,2,3,4,5

6,7

5

2

2,(3),7,(13),14

32

5

1,3,4,5,6

8

1,2

5,6,8,14

1

Element: 1, earth; 2, air; 3, water; 4, fire; 5, biota. Geographic region: 1, Africa; 2, Antarctica; 3, Asia; 4, Australasia (Australia, New Zealand, Micronesia); 5, Europe; 6, North America; 7, South America; 8, Islands. 3 Ecoregion: 1, desert; 2, grassland; 3, savanna; 4, shrubland; 5, forest; 6, wetland; 7, urban; 8, agroecosystem. 4 Trophic level: 1, producer; 2, consumer; 3, decomposer. 5 Theme: 1, interactions; 2, spatial heterogeneity; 3, succession; 4, competition; 5, nutrient cycling; 6, productivity; 7, stability and resilience; 8, predictability; 9, thresholds; 10, biodiversity; 11, functional redundancy; 12, invasive species; 13, restoration and management; 14, modeling.

2

AN INTRODUCTION TO TERRESTRIAL DISTURBANCES DISTURBANCE THEMES

The chapters in this volume approach the topic of disturbance from many perspectives (Table 1.2). Each chapter addresses how at least one of the four basic elements or ethers (earth, air, water, fire) and the biota may be an agent of disturbance, and a few chapters address all of them. The most frequently covered type of disturbance is biotic (particularly human). The most frequently described geographical region is North America, but all regions of the world are discussed (more than simply a reference or brief mention) in the following rank order: North America ≫ Australasia = Europe > Africa = Asia = South America > islands ≫ Antarctica. This representation probably reflects both author bias and available literature, although all regions except Antarctica are discussed in at least ten chapters. Ecological regions were discussed in the order: forests ≫ grasslands > shrublands = agroecosystems > deserts = wetlands ≫ savannas > urban areas, again suggesting the distribution of available literature (and humans), and despite the global importance of urbanization. Most chapters address effects of disturbance on primary producers and consumers, but a substantial fraction also consider decomposers. Fourteen themes emerge in the following order: nutrient cycling > interactions = succession ≫ spatial heterogeneity > productivity = biodiversity > competition = modeling > stability and resilience > thresholds = invasive species = restoration and management > predictability ≫ functional redundancy. This order suggests that disturbances often interact and that there are intimate links between disturbance and nutrient cycling, succession, spatial heterogeneity, productivity, and biodiversity. In the last chapter of the volume, we examine the lessons learned from earlier chapters in this volume about the relationships between disturbance and these important themes. Ecologists have made great strides in understanding the role of disturbance in shaping natural systems. Successional responses and competitive interactions among species have generated particularly large numbers of papers. Other responses to disturbance (notably belowground processes) have received very little attention. Land managers have developed a broad base of knowledge about practical issues relating to intentional human disturbances such as agriculture. However, neither ecologists nor land managers have developed a

15

robust set of predictions about the consequences of disturbances. Much more integration of management and theory is needed in order to address the environmental challenges which humans face. Especially important to understand are the consequences of irregular natural disturbances such as hurricanes or volcanoes, the recent and overwhelming human impacts such as erosion or clear-cutting, and the interactions among them. This global compendium of examples of disturbed ground offers a sampling of the types of data that are available and some preliminary generalizations and conceptual models. We hope this book will stimulate more longterm monitoring of disturbed ground, experiments that address the mechanisms behind biotic responses to disturbance, and studies that compare responses within and among various types of disturbances (and ideally across gradients of disturbance severity). Such types of data are needed to provide the basis for predictions about disturbance. REFERENCES Barrow, C.J., 1991. Land Degradation. Cambridge University Press, Cambridge, 295 pp. B´enito-Espinal, D.B. and B´enito-Espinal, E. (Editors), 1991. L’Ouragan Hugo: genese, incidences g´eographiques et e´ cologiques sur la Guadeloupe. Co-editors: Parc National de la Guadeloupe, D´el´egation R´egionale a l’Action Culturelle, and Agence Guadeloup´eene de l’Environnement du Tourisme etdes Loisirs. Imprimerie D´esormeaux, Fort-de-France, Martinique. Bond, W.J. and van Wilgen, B.W., 1996. Fire and Plants. Chapman and Hall, London, 263 pp. Botkin, D.B., Caswell, M.F., Estes, J.E. and Orio, A.A. (Editors), 1989. Changing the Global Environment: Perspectives on Human Involvement. Academic Press, New York, 459 pp. Chapin III, F.S., Walker, L.R., Fastie, C.L. and Sharman, L.C., 1994. Mechanisms of primary succession following deglaciation at Glacier Bay, Alaska. Ecol. Monogr., 64: 149–175. Clements, F.E., 1928. Plant Succession and Indicators. H.W. Wilson, New York, 453 pp. Cohen, J.E., 1995. How Many People Can the Earth Support? Norton, New York, 532 pp. Cronon, W., 1991. Nature’s Metropolis: Chicago and the Great West. Norton, New York, 530 pp. Crosby, A.W., 1986. Ecological Imperialism – The Biological Expansion of Europe, 900–1900. Cambridge University Press, Cambridge, 368 pp. Finkl, C.W. and Pilkey, O.H. (Editors), 1991. Impacts of Hurricane Hugo: September 10–22, 1989. J. Coastal Res., 8: 356 pp. Special issue. Komarek, E.V., 1983. Fire as an anthropogenic factor in vegetation ecology. In: W. Holzner, M.J.A. Werger and I. Ikusima (Editors), Man’s Impact on Vegetation. Dr W. Junk, The Hague, pp. 77–82. Luken, J.O., 1990. Directing Ecological Succession. Chapman and Hall, London, 251 pp.

16 Magnuson, J.J., 1990. Long-term ecological research and the invisible present. BioScience, 40: 495–501. Mares, M.A., Willig, M.R. and Lacher Jr., T.E., 1985. The Brazilian caatinga in South American zoogeography: tropical mammals in a dry region. J. Biogeogr., 12: 57–69. Officer, C. and Page, J., 1993. Tales of the Earth. Oxford University Press, Oxford, 226 pp. O’Neill, R.V., DeAngelis, D.L., Waide, J.B. and Allen, T.F.H., 1986. A Hierarchical Concept of Ecosystems. Princeton University Press, Princeton, NJ, 253 pp. Oxford World Atlas, 1973. Saul B. Cohen (Geographical Editor), Oxford University Press, Oxford, 190 pp. Pickett, S.T.A., Collins, S.L. and Armesto, J.J., 1987. Models, mechanisms, and pathways of succession. Bot. Rev., 53: 335–371. Sheets, P.D. and Grayson, D.K., 1979. Volcanic Activity and Human Ecology. Academic Press, New York, 644 pp. Simkin, T. and Fiske, R.S., 1983. Krakatau 1883: The Volcanic Eruption and its Effects. Smithsonian Institution Press, Washington, DC, 464 pp. Thomas, W.L., 1956. Man’s Role in Changing the Face of the Earth. University of Chicago Press, Chicago, IL, 1193 pp. Times Atlas of the World, 1990. Times Books, London, 395 pp. Times Atlas of the World, 1995. Times Books, London, 374 pp.

Lawrence R. WALKER and Michael R. WILLIG Vitousek, P.M., 1994. Beyond global warming: ecology and global change. Ecology, 75: 1861–1876. Vitousek, P.M., Ehrlich, P.R., Ehrlich, A.H. and Matson, P.A., 1986. Human appropriation of the products of photosynthesis. BioScience, 36: 368–373. Wackernagel, M. and Rees, W., 1996. Our Ecological Footprint. New Society Publishers, Philadelphia, PA, 160 pp. Walker, L.R., Brokaw, N.V.L., Lodge, D.J. and Waide, R.B. (Editors), 1991. Ecosystem, plant and animal responses to hurricanes in the Caribbean. Biotropica, 23: 313–521. Walker, L.R., Silver, W.L., Willig, M.R. and Zimmerman, J.K. (Editors), 1996. Long term responses of Caribbean ecosystems to disturbance. Biotropica, 28: 414–614. White, P.S. and Pickett, S.T.A., 1985. Natural disturbance and patch dynamics: an introduction. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, Florida, pp. 3–13. Whittaker, R.J., Bush, M.B. and Richards, K., 1989. Plant recolonization and vegetation succession on the Krakatau Islands, Indonesia. Ecol. Monogr., 59: 59–123. Worster, D.E., 1979. Dustbowl: the Southern Plains in the 1930s. Oxford University Press, Oxford, 277 pp.

Chapter 2

DISTURBANCE REGIMES AND ECOSYSTEM RESPONSE ON RECENTLY-DEGLACIATED SUBSTRATES John A. MATTHEWS

Elven, 1978a, 1980; Matthews, 1978, 1979a; Paternoster, 1984). More recent developments include the ordination of communities (e.g., Matthews, 1979b–d; Matthews and Whittaker, 1987; Whittaker, 1989, 1991, 1993; Crouch, 1993; Vetaas, 1994) and autecological studies (St¨ocklin, 1990; St¨ocklin and B¨aumler, 1996). Studies outside of North America and Europe have been relatively few in number and, with the notable exception of work associated with the ice-free Antarctic landscapes (e.g., Smith, 1993; Walton, 1993; WynnWilliams, 1993; Lyons et al., 1997) and in New Zealand (e.g., Wardle, 1980, 1991; Sommerville et al., 1982; Burrows, 1990), generally less detailed. However, some investigations have also taken place in South America (Lawrence and Lawrence, 1959; Heusser, 1960, 1964; Rabassa et al., 1981; Veblen et al., 1989; Jordan, 1991), Africa (Coe, 1967; Spence, 1989; Mahaney, 1990), Irian Jaya (Hope, 1976) and Asia (e.g., Turmanina and Volodina, 1978; Solomina, 1989). Although recently-deglaciated terrain originates as a result of major disturbance (nudation) and is subjected to a wide range of disturbances, both the disturbances and their ecological effects are, paradoxically, poorly understood. Rather, emphasis has been placed on primary succession as ecosystem development following nudation. With few exceptions, most notably Oliver et al. (1985) and Whittaker (1991), development of these ecosystems has been viewed as a largely biological, deterministic process in a static physical environment, which is relatively immune from subsequent disturbance. This traditional, oversimplified view was challenged by Matthews (1992), who enlarged upon the concept of recently-deglaciated terrain as a geoecological landscape in which biological processes are coupled with abiotic processes of the physical

INTRODUCTION

Recently-deglaciated terrain holds a special place in the history of ecology, largely as a result of the classical studies on primary succession and soil development that have been carried out in these landscapes, in North America (e.g., Cooper, 1923a–c, 1939), the Alps (L¨udi, 1921, 1945, 1958; Braun-Blanquet and Jenny, 1926; Friedel, 1934, 1937, 1938; Negri, 1934) and Scandinavia (Fægri, 1933). These were some of the first detailed investigations of primary succession and soil development, and they took advantage of the spatial chronosequence of ecosystems that exists in front of a retreating glacier. Over the last 40 years, recently-deglaciated landscapes have continued to inspire important empirical and theoretical contributions to understanding ecological succession. Cooper’s early work at Glacier Bay, Alaska, has been followed by detailed research on the vegetation itself (including the long-term monitoring of change), on the soil changes that accompany vegetation change, and on the mechanisms of change (e.g., Crocker and Major, 1955; Lawrence, 1958, 1979; Mirsky, 1966; Lawrence et al., 1967; Reiners et al., 1971; Bormann and Sidle, 1990; Chapin et al., 1994; Fastie, 1995). Similar investigations have been made at other glaciers in North America (e.g., Scott, 1974; Birks, 1980; Sondheim and Standish, 1983; Fitter and Parsons, 1987; Blundon and Dale, 1990; Blundon et al., 1993; Helm and Allen, 1995). In Europe, studies of plant communities of recently-deglaciated substrates have generally involved phytosociological approaches, which have traditionally emphasized the classification and mapping of communities as a basis for understanding environmental relationships (e.g., Jochimsen, 1963, 1970; Persson, 1964; Richard, 1973; 17

18

environment. The term ‘geoecology’ was coined by Troll (1971) with the aim of placing the study of ecosystems in the broader context of the geographical landscape (German: ‘Landschaft’). It emphasizes the spatial organization of ecosystems in the landscape, and the broader framework of environmental processes, both natural and cultural, with which biological and ecological processes interact. Thus, successional changes on deglaciated substrates can be viewed as a result of the interaction of dynamic physical processes, including disturbance, with biotic processes. In this chapter, I focus on the importance of disturbance in the recently-deglaciated landscape, with particular attention to ecological effects on ecosystems. Of necessity, the chapter relies heavily on the studies reviewed in Matthews (1992) with the addition of the rather limited amount of more recent research. However, there has been no previous comprehensive review, ecological or otherwise, of disturbance in these environments. After summarizing the global distribution and general characteristics of recently-deglaciated terrain, this chapter develops three main themes: (1) the classification of disturbance types and their characteristics; (2) observed effects of disturbances on ecosystems, particularly the direct effects on the substrate and on plants; and (3) the role of disturbance as a driving force in primary succession.

GLOBAL DISTRIBUTION AND GENERAL CHARACTERISTICS OF DEGLACIATED SUBSTRATES

Although about 96% of Earth’s glacier ice is in Antarctica and Greenland, ice bodies – glaciers, ice caps and ice sheets – and hence deglaciated substrates occur on all continents in a remarkably wide range of polar and alpine environments. Glaciers accumulate wherever there is a surplus of snowfall over snowmelt for sufficient years to allow the consolidation of snow into ice; ice wastes away, glacier margins retreat and deglaciated substrates are exposed when melting (ablation) exceeds accumulation. On a global scale, the latitudinal and altitudinal distribution of glaciers is primarily dependent on low air temperatures: thus, glaciers occur at or close to sea level in some parts of the Arctic and Antarctic, but only above about 5000 m on tropical mountains. At a continental scale, glaciers are found at increasing altitudes from oceanic coasts towards continental interiors primarily

John A. MATTHEWS

in response to decreasing snowfall, and they may be absent from many cold but dry continental interiors because of precipitation starvation. Locally, other factors are influential, particularly wind and aspect through their effects on snow-drifting and topographic shading. A zone of recently-deglaciated terrain – the glacier foreland – currently occurs in front of most glaciers world wide. This has been produced by glacier retreat from the ‘Little Ice Age’ glacial maximum attained at various times during the last few centuries. In southern Norway, for example, most glaciers attained their ‘Little Ice Age’ maxima around the middle of the eighteenth century (Matthews, 1991; Bickerton and Matthews, 1993) (Fig. 2.1); in the Alps, some reached their maxima around AD 1600, others in the middle of the nineteenth century (Grove, 1988). Successively older terrain occurs with increasing distance from the glacier front, and a range of techniques are of potential use in dating glacier forelands with varying levels of accuracy (see Matthews, 1992). In some cases, older substrates may be available as a result of pre‘Little Ice Age’ (Neoglacial) glacier advances and retreats. The extent of the glacier foreland varies with the magnitude of glacier retreat. On the one hand, large valley glaciers in maritime climates tend to be relatively dynamic; glacier retreat rates may reach >100 m yr−1 , and the foreland may extend for several kilometres from the glacier; some small cirque glaciers in continental climates, on the other hand, are characterized by a narrow deglacierized zone 50% water when emplaced. Debris flows are less fluid (20 m2 ) had higher light levels and lower humidity than the surrounding subtropical broadleaf climax forest (Barik et al., 1992). When Canham et al. (1990) examined gaps (size 78 m2 ) in five different forest types, they showed that light penetrates to the understory in and around the gap, particularly along the north side of gaps in the northern hemisphere. Light zones increased with latitude, reflecting the angle of sunlight reaching the adjacent understory. This peak of light availability along the north edge of gaps was recognized earlier (Minckler and Woerheide, 1965; Minckler et al., 1973). Poulson and Platt (1989) reported strong growth responses along north edges of gaps, but little along south edges. However, the daily duration of direct sunlight is brief even in sizeable gaps. In taller deep-crowned Pseudotsuga menziesii forests in Oregon, a canopy gap produces a much smaller light zone than in four other forest types (Canham et al., 1990). Physical gap measurements are made in a variety of ways. Most widely used are Brokaw’s (1982) definition of a gap as an opening extending down to a height of 2 m, and Runkle’s (1982) similar definition of gap as the area beneath a canopy opening, used in tandem with the expanded gap concept which extends from the opening outward to the bases of surrounding trees. However, widely differing gap criteria hamper efforts to compare studies. Runkle (1992) reviewed the options and problems, and provided a gap-sampling protocol developed through workshops with participants who study diverse forests (see also Christensen and Franklin, 1987). One key point of difference is how tall the understory can be within a gap; another is the lower size limit for recognized gaps. When small gaps and understory-filled areas are excluded, this poses particular problems for attempts to utilize gap dynamics for insight into windstorms. The same features used to delineate gaps are also measured as gap consequences, an appropriate approach for scrutinizing gap dynamics but misleading and somewhat circular when all wind-related turnover is of interest. When light gaps favor shade-tolerant species Contrary to expectation, existing plants are the usual

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS

beneficiaries of light gap formation. Especially when small, gaps are filled by upward growth of understory trees and saplings (Watt, 1947; Bray, 1956; Brewer and Merritt, 1978; Runkle, 1984, 1990b; Canham, 1985, 1989, 1990; Spies et al., 1990a; Cho and Boerner, 1991) or simply by lateral growth of adjacent trees (Gysel, 1951; Hibbs, 1982; Runkle and Yetter, 1987; Frelich and Martin, 1988; Spies and Franklin, 1989). Even in tropical forests, new species have only sparse gap-related opportunities because so few gaps are large; the existing understory normally controls the response (Brokaw and Scheiner, 1989), as in temperate-zone forests. When a layer of shade-tolerant understory trees advances toward the canopy, succession is in effect accelerated by disturbance (Lorimer, 1980), especially if mortality is heaviest among shade-intolerant trees. Similarly, disturbance by logging advances succession in Michigan forests, despite the large size of openings (Abrams and Scott, 1989). In some forests the loss of individual trees does not create measurable light gaps at all (Lieberman et al., 1985, 1989; Spies and Franklin, 1989; Webb, 1989). The non-gap component of forest turnover and of windstorm dynamics is generally overlooked despite these aforementioned studies, but recent findings of Hubbell (1999) have drawn attention to the limits of gap dynamics at last. Discrete, measurable gaps are most likely to form in forests with a monolayered canopy (Fig. 7.2), broad tree crowns, relatively short trees, (Spies et al., 1990a), and an open or weak-wooded subcanopy (Webb, 1989).

Fig. 7.2. Diagram showing how a monolayered forest (a and b) forms larger, more discrete light-gaps than a multilayered forest with understory layers (c and d).

In both old-growth and successional forests of western Oregon and Washington, most tree deaths are not followed by gap formation (Spies et al., 1990a). If gaps form, they are often restricted to the canopy

203

layer. The tall narrow conifers form small gaps if any, well below the gap-size threshold (estimated at 700– 1000 m2 ) for shade-intolerant Pseudotsuga menziesii (Douglas-fir) to regenerate (Spies and Franklin, 1989). However, gaps in earlier stages of succession have larger effective sizes because of the lower stature of the canopy. Even in successional forests, however, the principal response to tree-falls is release of a shadetolerant understory (Spies and Franklin, 1989; Spies et al., 1990a). Even large-area, high-intensity windstorms can fail to set back succession in cleared patches. Glitzenstein et al. (1986) in Texas concluded that disturbance by a tornado caused accelerated succession in at least one (Pinus–Quercus) of the two forest types studied. Peterson and Pickett (1995) in Pennsylvania found that average species richness increased over a six-year period following tornado disturbance, but that this increase occurred both in openings and in surrounding forest. Despite the large blow-down size, shade-intolerant tree species did not predominate, perhaps because of an absence of propagules (Populus and Prunus might be expected) in a large old-growth tract; instead the shade-tolerant taxa were the chief beneficiaries (Acer saccharum, Fagus grandifolia, Tsuga canadensis). However, yellow birch (Betula alleghaniensis) also benefited; this intermediate-tolerance species probably would not regenerate without openings. In some cases, shade-intolerant species that initially appear in light gaps do not survive long and are replaced quickly by shade-tolerant species (Clinton et al., 1994; Kupfer and Runkle, 1996). Resprouting of damaged trees can constitute a major component of the windstorm response, as seen after hurricane disturbance in South Carolina (Putz and Sharitz, 1991). Likewise, rapid and dense resprouting of Populus tremuloides minimized establishment opportunities for other species in a recent blow-down (1995) in northwestern Minnesota (S. Webb, pers. observ.). Many temperate-zone forests have one or more shade-tolerant tree species that establish beneath an intact canopy and respond to light gaps. In the original vision of gap dynamics, gaps are filled not by patches of light-demanding species but clumps of shade-tolerant saplings (Bray, 1956). In eastern North America’s deciduous forests, such shade-tolerant gap fillers include Acer saccharum and Fagus grandifolia (Brewer and Merritt, 1978; Barden, 1980, 1981; Canham, 1990; Runkle, 1990b; Poulson and Platt,

204

1996), and also Nyssa sylvatica (Orwig and Abrams, 1994), all of which grow slowly under the canopy and respond to release when canopy trees die overhead or nearby. Runkle found that gap regeneration was dominated by shade-tolerant Acer saccharum and Fagus grandifolia in forests both in Ohio (Runkle, 1990b) and in North Carolina (Runkle and Yetter, 1987). Brewer and Merritt (1978) reported the same situation in Warren Woods, southeastern Michigan [but see Poulson and Platt (1996) who anticipated shadeintolerant trees in large gaps within this same forest]. Canham (1985) showed that Acer saccharum grows into the canopy slowly, having an average of three periods of suppression interspersed with periods of release corresponding to increased light from nearby canopy gaps. Acer saccharum appears in study after study as a major beneficiary of windstorms (Held and Bryant, 1989; Webb, 1989; Mladenoff, 1990; Cho and Boerner, 1991), on account of its shade tolerance and advanced regeneration poised to capture newly available resources, and also due to its wind-firmness and flexibility (King, 1986). Many gap studies in other parts of the world reveal similar dynamics that favor shade-tolerant trees established before gap formation. In Chilean rainforests, gaps were filled not by the existing canopy species of disturbance origin (Nothofagus dombeyi) but by shade-tolerant Aextoxicon punctatum (Veblen, 1985a,b). Existing emergent canopy species were not reproducing and will likely disappear if only small gaps form (Veblen, 1985a). In old-growth forests of Argentina, Nothofagus betuloides formed a welldeveloped understory capable of responding to release, except where excluded by an understory of the subcanopy tree Drimys winteri; where Nothofagus pumilio was present, its seedlings and saplings also expanded upward into gaps from pre-gap establishment (Rebertus and Veblen, 1993b). In one of two xeric localities in Chile also examined by Veblen (1989a), Nothofagus dombeyi was regenerating in the understory and thus increasing at the expense of a canopy species, Nothofagus antarctica, that does not regenerate in tree-falls. The second Chilean site also had a predisturbance understory, but in this case the understory more closely matched the canopy, suggesting that gap dynamics maintain the forest’s existing composition (Veblen, 1989a). In extensive light gaps within a cool temperate Fagus crenata forest in Japan, only 2.5% of all stems within 36 gaps were of pioneer species (Nakashizuka, 1984); suppressed saplings of Fagus

Sara L. WEBB

crenata were abundant and likely to replace most windthrown canopy trees (Yamamoto, 1989). In a more species-rich warm temperate Japanese forest, nearly all predicted gap successors were from five species widely present as suppressed saplings (Yamamoto, 1992). In subalpine old-growth forests of southwestern British Columbia, the many small, persistent gaps are filled primarily by lateral-growth of neighboring trees or by shade-tolerant Abies amabilis (Lertzman and Krebs, 1991; Lertzman, 1992). Likewise, in central Sweden, even the largest gaps filled with the shadetolerant Picea abies, while little regeneration was seen for more light-demanding species: Betula spp., Pinus sylvestris and Populus tremula (Qinghong and Hytteborn, 1991). Light gaps can play a key structural and developmental role in the forest without admitting shade-intolerant species, as in several coniferous forest examples. In a Colorado subalpine forest, Abies lasiocarpa and Picea engelmannii are both shade-tolerant dominants, but Abies lasiocarpa reproduces under the closed canopy whereas Picea engelmannii only begins to regenerate when canopy gaps form (Aplet et al., 1988). In this case, new colonization occurs in gaps but the colonist is a shade-tolerant species already present; here gaps interact with other life-history characteristics (such as greater longevity for Picea engelmannii; Veblen, 1986) to reorganize and maintain the community. Similarly, in old-growth Picea abies forests in Sweden (Leemans, 1991), saplings and understory trees were concentrated in gaps but scarce under closed canopies, demonstrating the importance of gaps for regeneration; but the species of gap saplings and trees was the same (Picea abies) as that dominating the canopy. A similar dynamic apparently operates in subalpine forest of central Japan, where advanced regeneration by shadetolerant Abies mariesii might be accompanied into canopy gaps by occasional shade-intolerant codominants that sustain less mortality (Yamamoto, 1995). Shade-intolerant trees lacked colonization opportunities in a Nothofagus forest in New Zealand, where gap beneficiaries were trees and saplings already present in the understory (Stewart and Rose, 1990; Stewart et al., 1991). The two dominant species differed in life history traits, but both responded favorably to windstorms, to the exclusion of other trees. Shade-intolerant plants in gaps In contrast, there are examples of gaps (created by windstorms and other causes) where shade-intolerant

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS

trees are supported in a diversity-enriching patchdynamics pathway. Large clearings from hurricanes, tornadoes, and blow-downs seem most likely to favor shade-intolerant species, but smaller light gaps can also have this effect under some (but, as has been shown, not all) circumstances. One component of diversity enrichment comes from the buried pool of viable seeds, whose germination is triggered by light or other environmental changes when the canopy opens up. In eastern North America, Prunus spp. (Marks, 1974; Dunn et al., 1983; Held and Bryant, 1989) and Rubus spp. account for the majority of the woody plant seed bank. A dense thicket of Rubus brambles is common in large blow-downs shortly after the storm. Figure 7.1 shows such a thicket in the Flambeau Tract in Wisconsin eight years after a 1977 down-burst struck an old-growth forest. Rubus allegheniensis increased for three years following one tornado disturbance in Pennsylvania but then declined as tree saplings closed in (Peterson and Pickett, 1995). However, the same locality had little growth of Prunus, apparently because its seeds were sparse in the soils of this isolated old-growth forest (Peterson and Pickett, 1995). Most of the wind-dispersed Betula species (birches), which are light-demanding to various degrees, depend upon disturbances to regenerate in sizeable blowdowns. Six years after the Pennsylvania tornado, a large clearing of 400 ha2 supported high densities of Betula alleghaniensis and B. lenta, which rarely regenerate in closed forest (Peterson and Pickett, 1995). A similar increase for Betula alleghaniensis developed throughout the large Wisconsin blow-down (Fig. 7.1) (Dunn et al., 1983). Following the 1938 hurricane in New England, birches germinated on many wind-throw mounds (including Betula alleghaniensis, B. lenta, B. papyrifera, B. populifolia; Spurr, 1956; Henry and Swan, 1974), where they still persist today (S. Webb, pers. obs.). Both B. alleghaniensis and B. lenta show rapid lateral growth into small gaps (Hibbs, 1982). In a Vermont montane spruce–fir forest, Betula papyrifera thrived in gaps (of unknown cause) but was scarce under closed canopies (Perkins et al., 1992). A spruce–fir forest in the southern Appalachian Mountains (Tennessee) had ten times more Betula alleghaniensis in gaps than in background forest plots (White et al., 1985a,b). However, small tree-fall gaps in a Michigan forest were not large enough to benefit Betula alleghaniensis (Mladenoff, 1990). Other trees that depend upon light can thrive

205

in areas of wind-throw. By eleven years after a Kentucky tornado, several shade-intolerant trees had increased in density, including Prunus serotina, Quercus muehlenbergii and Q. rubra (Held and Bryant, 1989). A decline in density but an increase in basal area were seen for Fraxinus americana, another light-demanding species generally expected to benefit from canopy openings. However, other shade-intolerant and midtolerant species had declined (Carya cordiformis, Celtis occidentalis, Gleditsia triacanthos and Juglans nigra) while the shade-tolerant Acer saccharum increased its numbers from 35% to 53% of relative density (Held and Bryant, 1989). Beyond such studies linked with known windstorms, one can draw inferences, with caution, from the forest light-gap literature. Even though gap-oriented studies tend to oversample large gaps, and sometimes use the presence of shade-intolerant trees as evidence for past gaps, there are surprisingly few cases in which shade-intolerant species predominate in gaps. In most investigations, shade-tolerant trees are the most important respondents in light gaps, as already discussed. Gap size is one key factor in controlling whether the existing flora is simply restructured, or whether instead the gap patch supports a distinctive assemblage. The sizes of light gaps in one Indiana beech–maple (Fagus–Acer) forest corresponded to sizes of patches of seral species (Fraxinus spp., Liriodendron tulipifera, Quercus spp., Sassafras spp., etc.), suggesting that these species got established in light gaps (Williamson, 1975). The light-demanding tree Liriodendron tulipifera showed no pattern of suppression and release in its tree rings, like that seen in more shadetolerant trees (Orwig and Abrams, 1994). L. tulipifera depends on sizeable openings and is more common following agriculture than following disturbances that form small gaps (Clebsch and Busing, 1989). Only in the largest clear-cut patches was Liriodendron tulipifera reproducing ten years after logging in Illinois (Minckler and Woerheide, 1965). Nevertheless, this species is an important codominant within southern Appalachian forests, along with other shadeintolerant (Fraxinus americana) and shade-tolerant (Tsuga canadensis) trees, suggesting regular gapforming disturbance events of some magnitude in the past (Lorimer, 1980; Barden, 1981). In Warren Woods, southwest Michigan, Poulson and Platt (1989) concluded that Liriodendron tulipifera needed less light than Fraxinus americana but more than Prunus

206

serotina; all three of these shade-intolerant trees were expected to reproduce in sufficiently large gaps. The abundance of oaks in eastern North America has long suggested a disturbance origin for the forests, given their lack of shade tolerance. Quercus rubra might maintain dominance after large-scale wind-throw if vigorous oak seedlings are already present (Hibbs, 1982), an uncommon situation today. In smaller gaps, Quercus rubra does not reproduce (Bray, 1956; Webb, 1989; Cho and Boerner, 1991). Other studies also suggest that light-demanding trees enter into the forest only in relatively large gaps. In an Ohio forest, Runkle (1990b) found shadeintolerant Liriodendron tulipifera in only the four largest of 36 gaps, and found another shade-intolerant tree, Fraxinus americana, at high densities in large, young gaps but not in older gaps, suggesting inability to persist until forest closure. However, this 12-year resurvey showed most gap takeover by pre-existing shade-tolerant stems (Acer saccharum and Fagus grandifolia), as was also true for gaps in more speciesrich hardwood forests in the Southern Appalachian Mountains (Runkle and Yetter, 1987). These Southern Appalachian cove forests have received much scrutiny, often with conflicting conclusions about the role of gap dynamics in maintaining shade-intolerant trees. Clinton et al. (1994) reported a general increase in richness with increasing gap size, and Barden (1981) found shade-intolerant trees more prevalent in multiple-tree gaps than in single-tree gaps. Although 97% of all predicted gap successors were shade-tolerant Acer saccharum and Fagus grandifolia, shade-intolerant trees were expected to persist through occasional capture of large gaps (Barden, 1981). A similar correlation was found between gap area and tree species richness in a subtropical broadleaf climax forest in India (Barik et al., 1992). These gap size/diversity correlations reflect in part an increase in light availability as gap size increases (Canham et al., 1990). One mechanism by which light gaps theoretically could enrich diversity is through gap partitioning, where gaps of different sizes, or different zones within a gap, favor different trees (Denslow, 1980). Runkle et al. (1995) sought but did not find differences suggesting gap partitioning by two coexisting species of Nothofagus in southern New Zealand. Sipe and Bazzaz (1994, 1995) examined three species of maple (Acer pensylvanicum, A. rubrum and A. saccharum) for gap partitioning, by transplanting seedlings into controlled experimental gaps. After three years, the

Sara L. WEBB

three species showed differences in growth responses to large gaps but not to small gaps (nor in control areas). Linkages with environmental measurements showed little support for the gap-partitioning concept. Heterogeneity within gaps has not been widely examined in temperate forests; more studies with control areas like those of Sipe and Bazzaz (1995) would also be useful to elucidate dispersal constraints on patchy gap vegetation. In temperate forests in general, evidence for gap partitioning has not been forthcoming aside from specialized utilization of wind-generated substrates (Yamamoto, 1992; see also section “Mounds, pits, and woody debris”, pp. 208–211 below). Gap enrichment depends strongly upon what species are part of the system. For example, in Chile Nothofagus dombeyi has greater gap importance at high elevations where shade-tolerant competitors are absent (Veblen, 1985b). In a Minnesota Pinus–Abies forest, opportunities for shade-intolerant species result from the absence of the shade-tolerant layer of understory Acer saccharum so prevalent on other soils nearby (Webb, 1989). The importance of understory development is further discussed below. In search of more insights into wind disturbance, the analogy between harvest openings and windstorm gaps must be drawn with caution, because of the very different substrates that result when logged trees are removed, often by heavy machinery; wood substrates are removed, and wind-throw mounds do not form. In Maine, gaps created by forest harvest differed from natural gaps, the former having more herbaceous plants and Rubus, in part because of larger gap size but also because of soil disturbance during the harvest (Kimball et al., 1995). Gaps from logging were examined in a New Hampshire hardwoods forest; these large openings (292–1032 m2 ) had shade-tolerant trees dominating the smaller range, but admitted shade-intolerant and mid-tolerant trees in the larger openings, including Acer rubrum, Betula alleghaniensis, B. papyrifera and Prunus pensylvanica (McClure and Lee, 1993). The importance of the understory The forest understory is a major part of the equation for windstorm responses. A well-developed understory can preclude gap formation and can inhibit new colonists where gaps do form (Fig. 7.2). As already emphasized, many temperate forests include understory layers of shade-tolerant trees, saplings, and seedlings that are likely to benefit from death of canopy trees (Canham,

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS

207

Fig. 7.3. Photographs illustrating well-developed understory layers of saplings (Acer saccharum) and shrubs (Corylus cornuta and others) in Minnesota Pinus forest (Webb, 1989). Similar shade-tolerant understories are the major beneficiaries of individual tree-falls in many forests around the world.

1985, 1990) and can continue to shade the forest floor against establishment of light-demanding species. The case of Acer saccharum was described above and is illustrated by Fig. 7.3. In one Japanese beech forest, an extensive pre-disturbance seedling bank of Acer mono and Fagus crenata might help to explain the absence of pioneer trees such as Betula spp. in gaps (Hara, 1983, 1985, 1987). If the understory comprises wind-firm species, it will benefit from canopy damage; however, if understory species are damage-prone, then canopy gaps are more likely to extend deeper toward the forest floor (Webb, 1989). The importance of the understory flora is illustrated by the contrasting effects of a moderately severe 1983 windstorm in two Minnesota stands (Webb, 1989). Discrete light gaps formed in a Pinus–Abies community because the shade-tolerant understory firs (Abies spp.) had weak wood and a growth form rendering them vulnerable to snagging by a falling tree; the firs rarely survived the fall of overstory trees. Meanwhile, in a nearby Pinus–Acer community, few tree-falls produced measurable, discrete light gaps. Here the understory was dominated by shade-tolerant species with very strong wood (Acer saccharum and

Ostrya virginiana), which survived and often benefited from the fall of overtopping canopy trees. Thus the wind-firmness of shade-tolerant understory trees helps to explain why these two forests responded differently to the same disturbance event. Tall understory trees, a thicket of shrubs, or a growth of bamboo can all dominate the dynamics of a wind-damaged forest patch. Understory bamboos influence tree regeneration in some South American and Asian forests. Veblen (1989b) found that tall bamboos (Chusquea spp.) inhibited tree regeneration in gaps in Chilean and Argentinian Nothofagus forests, but that trees did regenerate in gaps farther south where bamboos were absent. In China, Taylor and Zisheng (1988) found different gap-replacement trends in two old-growth Abies faxoniana/Betula utilis stands, one with more bamboo (Sinarundinaria fangiana) and the other with regeneration by gap-favored Betula trees. In temperate broadleaf forests of Japan, a dense undergrowth of bamboo (Sasa nipponica) apparently inhibits seedlings of trees that elsewhere establish a seedling bank (Acer mono, Fagus crenata) but does not inhibit reproduction by sprouting for some other trees (Acer japonicum, A. sciadophylloides, Magnolia obovata;

208

Nakashizuka et al., 1992; Yamamoto, 1995). In a dense matrix of bamboo, tree-fall gaps are generally not colonized by shade-intolerant trees, except where windthrow mounds and rotting logs provide bamboo-free substrates (Nakashizuka, 1989). Small trees that never play a canopy role are sometimes the major gap beneficiaries. These include several species in Japanese broadleaved-evergreen forests (Yamamoto, 1992), the understory tree Cornus florida in a New Jersey hardwood forest (Ehrenfeld, 1980), Acer pensylvanicum in a northern hardwoods forest (Hibbs et al., 1980), the understory tree Ostrya virginiana in northwestern Minnesota hardwood forests (Webb, 1989), and Drimys winteri in Argentina (Rebertus and Veblen, 1993b). Other types of understory vegetation can also play a role in windstorm dynamics. Tall shrubs characterize the hardwood and mixed forests of northwestern Minnesota (Fig. 7.3), contributing alongside Acer saccharum understories to a lack of light on the forest floor where trees blow down. Many shrubs can sprout in high densities in response to light gaps (Gysel, 1951; Dunn et al., 1983). In some areas of the southern Appalachians (North Carolina), an understory of Rhododendron, a tall broadleaved evergreen shrub, inhibits tree regeneration before and after gaps form (Runkle, 1985; Clinton et al., 1994). At one New Zealand site, heavy fern cover combined with scattered shrubs to limit densities of tree seedlings and saplings, compared with two otherwise similar stands of oldgrowth Nothofagus forest (Stewart et al., 1991). Few studies of post-storm colonization examine herbaceous plants (Dunn et al., 1983; Peterson and Campbell, 1993). Several surveys of light gaps and other microsites (mounds, pits, logs) not associated with windstorms have demonstrated diverse herbaceous response patterns across a range of forest types. Distinctive assemblages appeared in light gaps (Goldblum, 1997) and on old mounds and pits in eastern New York (Beatty, 1984) and on mounds and pits in Pennsylvania (Peterson and Campbell, 1993). These studies had large sample sizes and careful statistical analyses. Other studies have found patterns that do not suggest species enrichment within light gaps or on microsites. In an Illinois forest, the herbaceous plants on logs, mounds, and pits were the same species found within one meter on the undisturbed forest floor (Thompson, 1980). In a Tennessee forest, fallen logs supported lower total diversity than the soil, but those plants present had more vigorous growth on logs (Bratton, 1976).

Sara L. WEBB

Increases in cover but not richness are sometimes seen in the understory layer (Moore and Vankat, 1986). For example, three years after experimental gaps were created, the fern Thelypteris novaboracensis had increased in density within gaps in northwestern Pennsylvania (Collins and Pickett, 1988b); the vernal Erythronium americanum responded more immediately with more stems in gaps than non-gap areas, but did not subsequently increase in cover (Collins and Pickett, 1988a). Large gaps but not small gaps had higher densities of seedlings of the light-demanding small tree Prunus serotina. Thus, as with trees, the understory plants of winddisturbed landscapes are often rearranged and restructured without changes to the overall composition and richness of the forest. However, well-designed timetransgressive studies suggest that some windstormtype disturbances can help to explain the complex patterning observed in the understory of the temperatezone forest. Herbivory Herbivory can also influence gap dynamics and responses to windstorms. Many forested areas of eastern North America are subject to anthropogenically elevated levels of herbivory by white-tailed deer (Odocoileus virginianus). The composition and open structure of the shrub and seedling layer thus may be an artifact of human settlement and predator eradication. Deer browsing is thought to explain the disappearance, over 50 years, of a Pennsylvania forest understory (Whitney, 1986) and also to explain the scarcity of Tsuga canadensis regeneration in another Pennsylvania understory, with consequences for tornado responses (Peterson and Pickett, 1995). Herbivory also contributes to the paucity of tree establishment in Patagonian blow-downs (Veblen et al., 1996), while introduced herbivores in New Zealand are thought to influence forest regeneration there as well (Jane, 1986). Thus, the all-important understory with its strong influence on windstorm consequences may itself look very different from the understory from which today’s canopy trees emerged. Today’s deep light gaps in some temperate forests would have been broken up vertically by understory layers in the past. Mounds, pits, and coarse woody debris In addition to light gaps, another mechanism by which

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS

windstorm disturbance might modify the community is by generating substrate heterogeneity: wind-throw mounds and pits when trees uproot, stumps when trees snap in the wind, and rotting logs in either case (Fig. 7.4). Such microsites differ from the surrounding forest floor in such features as temperature, moisture retention, nutrients (Stark, 1994), and freedom from competing plant roots; and thus might support different biotic assemblages. As with light gaps, such microsites do not always enrich diversity but instead play quite different roles in different forests.

Fig. 7.4. Illustration of microsites formed when trees are uprooted or snapped by windstorms. (a) Uprooted trees typically form a mound and a pit, although (b) pits are sometimes absent; (c) some tree-falls form no basal microsites; (d) snapped trees leave behind stumps and, as in all cases without salvage logging, a rotting log.

Mounds and pits The wind-throw mound (or knoll) and its associated pit (also called a cradle or crater) form only when trees are uprooted, a mode of damage nearly always caused by windstorms. In many old-growth forests the ground is distinctly uneven, a sign of past windstorms. Mounds and pits can cover 1.6–48% of the forest floor (Beatty, 1984; Webb, 1989). Long after the upturned roots have rotted away, mounds of soil and rock will persist, perhaps for 300–500 years (Denny and Goodlett, 1956; Stephens, 1956), if not leveled by agricultural or silvicultural activities. Soil profiles are rearranged by the uprooting process (Lutz and Griswold, 1939; Lutz, 1940; Lyford and MacLean, 1966; Vasenev and Targulyan, 1994) and various other soil features are modified (Stone, 1975; Beatty and

209

Stone, 1986; Norton, 1989; Habecker et al., 1990; Semmel, 1993). In many cases, mounds support more trees than either pits or surrounding forest (Denny and Goodlett, 1956; Lyford and MacLean, 1966; Henry and Swan, 1974; Raup, 1981; Collins and Pickett, 1982; Nakashizuka, 1989). The exposed mineral soil on mounds is an ideal seed bed for small seeds (Hutnick, 1952), and mounds are free of competition from the extensive living roots elsewhere. The same species that proliferate in sizeable light gaps are also particularly successful on wind-throw mounds (Hutnick, 1952; Henry and Swan, 1974; Dunn et al., 1983), as is true for tropical pioneer species observed by Putz et al. (1983). Acer saccharum and Prunus serotina, both trees, were most abundant on mound tops in a Pennsylvania forest (Collins and Pickett, 1982). Several herbaceous plant species also had highest importance on mounds in this Pennsylvania forest, and the same was true in a New York forest, although here the distinctiveness of mound assemblages was greater in one forest type (without hemlock) than in another (Beatty, 1984). Tree ferns (Cyathea smithii and Dicksonia squarrosa) are specialists on wind-throwmounds in a New Zealand podocarp forest, where the dominant podocarps (Dacrydium cupressinum and Prumnopitys ferruginoides) are scarce on mounds (Adams and Norton, 1991). Conversely, plant diversity and cover can be low on wind-throw mounds in some forests (Denny and Goodlett, 1956; Peterson et al., 1990). Alongside the attraction of exposed soil, mounds pose challenges to would-be colonists: higher temperature extremes and lower levels of organic matter, litter cover, cation exchange capacity, nutrient content, and snow cover than pits (Beatty, 1984). Within a few years of a Pennsylvania tornado, Peterson et al. (1990) found lower species richness on mounds than in pits. In a Pinus/Acer forest in northwestern Minnesota, old wind-throw mounds supported lower densities and lower richness than the surrounding forest floor, with no unique species or assemblages on mounds (Webb, 1988). In a nearby Pinus/Abies forest, the characteristically small mounds were also floristically indistinguishable from control areas. This lack of significance for microsite patches might result from the small size of light gaps and the dense pre-windstorm understory. In this location, most mounds have not yet been colonized 13 years after they formed in 1983 (S. Webb, pers. observ.); if this time lag is typical, then

210

light gaps may close before the mound or pit substrates are suitable for colonization. Mounds also can differ from undisturbed microsites by impeding ingrowth by vegetative sprouts. Both mounds and pits had densities of Fagus grandifolia lower than the background values in two different Pennsylvania forests where this tree normally reproduces vegetatively (Collins and Pickett, 1982; Peterson and Campbell, 1993). However, root sprouts of Populus tremuloides were slightly more numerous on mounds than elsewhere in the Minnesota Pinus/Acer forest (Webb, 1988), perhaps deriving from the wind-thrown Populus trees themselves. Wind-throw mounds and pits also differ in size and configuration depending upon the species of windthrown tree and the details of the uprooting event (Beatty and Stone, 1986; Mattheck et al., 1995). Colonization patterns can vary with the size of mounds (Webb, 1988) or pits (Peterson et al., 1990) and can also vary within a mound or pit (Hutnick, 1952; Collins and Pickett, 1982; Peterson et al., 1990). Pits (“craters”; Falinski, 1978) can be hostile or favorable microhabitats, depending upon the depth to the water table and the size of the wind-cleared openings; the role of pits also apparently shifts as they age over time. In many eastern North American forests, pits support low diversity and low densities of plants (Thompson, 1980; Raup, 1981; Beatty, 1984), with assemblages different from those on mounds or elsewhere (Beatty, 1984). A New Zealand temperate forest also had pits with low diversity, gradually filling with silt and then with a dense growth of bryophytes (Adams and Norton, 1991). The inhibiting factor in a central New York forest was the deep litter that accumulates in pits; when litter was removed, germination and establishment were enhanced for plants previously growing only on mounds (Beatty and Sholes, 1988). Litter depth was also negatively correlated with species richness in a northwestern Pennsylvania forest (Peterson and Campbell, 1993). Another problem in pits is moisture accumulation (Hutnick, 1952). In a Polish forest dominated by Picea abies, a shallow water table apparently caused pits to fill with short-lived plants of aquatic and mud habitats (Falinski, 1978). Hydrophilic plants also colonized pits in a northern Pennsylvania forest, but in this case the pits had higher species richness than mounds (Denny and Goodlett, 1956). The higher moisture level in pits, if not excessive, can promote seedling survivorship whereas the exposed

Sara L. WEBB

mound might be a microhabitat subject to drought. Acer rubrum, a tree of both uplands and swamps, was most common in pits following hurricane damage to a New Hampshire forest (Henry and Swan, 1974). After a Pennsylvania tornado (Peterson and Pickett, 1990), new pits were more rapidly colonized than mounds, and supported higher species richness. Even Betula alleghaniensis, common in other forests on mounds only, was abundant in pits at this site. Perhaps pits are favorable immediately following windstorms before litter accumulates, whereas on mounds mineral soil is exposed for a longer period of time. However, tree seedlings were more abundant as pits aged in Minnesota forests with a moderate-windstorm disturbance regime (Webb, 1988). Here pits of all ages supported lower plant density and diversity than mounds or control areas, with a flora distinguished by an absence of otherwise ubiquitous Acer saccharum and the presence of otherwise uncommon seedlings of Quercus rubra and Tilia americana. These geographic differences in vegetation on pits and mounds probably result from the complex interplay of local climate, soil type, depth to the water table, and disturbance magnitude and intensity as they influence the size and duration of light gap conditions. Stumps and rotting logs Not all wind-thrown trees are uprooted; thus mound and pit microsites are not always created (Fig. 7.4). When Peterson and Pickett (1991) reviewed 17 windstorm studies, they found a wide range in the prevalence of uprooting. From 0% to 91% of wind-damaged trees were uprooted, but in the majority of examples uprooting accounted for less than 50% of the damage. In two Minnesota forests, 37% and 80% of wind-throws following a moderate thunderstorm resulted in snapped trees (excluding bent trees; Webb, 1988), and 33% of wind-thrown trees following a tornado in Pennsylvania were snapped (Peterson and Pickett, 1991). Twice as many trees were snapped as uprooted by a hurricane in South Carolina sloughs (Putz and Sharitz, 1991); 50% of damaged trees were snapped by the same hurricane elsewhere in South Carolina on a floodplain (Duever and McCollom, 1993). The stump formed by tree breakage represents a substrate different from mounds or pits but similar to rotting logs. Of course, stumps and logs are formed whenever trees die, not only as a result of windstorms. Undisturbed old-growth forests can have 45% (Maser and Trappe, 1984) or more of the surface covered with

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS

dead wood. A widespread policy of salvage logging hampers ability to probe the importance of such coarse woody debris in forests. Coarse woody debris can serve to conserve nutrients through gradual nutrient release during the decay process (Harmon et al., 1986), but can immobilize nitrogen in other forests (Zimmerman et al., 1995). These woody substrates can also be important sites for germination and establishment. Best known in this regard are the so-called nurse logs that support tree seedlings unable to establish elsewhere in forests of the Pacific Northwest [Washington and Oregon (U.S.A.) and adjacent Canada; Minore, 1972; Franklin and Dyearsness, 1973; Christy and Mack, 1984; Harmon and Franklin, 1989]. Tree seedlings predominate on rotting wood in other regions as well, particularly for conifers in North American coniferous forests (Harmon et al., 1986; Webb, 1989; Pauley and Clebsch, 1990; Gibson and Brown, 1991; Hofgaard, 1993), but also in Nothofagus forests in Chile (Veblen, 1985b) and New Zealand (Stewart, 1986), and in an oldgrowth Fagus–Abies forest in Japan where bamboo inhibits seedlings on soil (Nakashizuka, 1989). In the New Zealand forests, the two dominant, shade-tolerant tree species (Nothofagus menziesii and Weinmannia racemosa) constituted most regeneration on fallen trees in both small tree-falls and within a larger area of windthrow (Stewart, 1986). Experimental work in the Pacific Northwest indicates that these woody microsites are favorable primarily because of freedom from competition from herbs and mosses on the forest floor; these experiments ruled out waterlogging (on the forest floor), litter shedding (on logs), nutrient differences, and differential seed predation as alternative explanations (Harmon and Franklin, 1989). In other less moist environments than the temperate rainforests of that research, rotting wood might have other significant advantages for tree reproduction: constancy of moisture availability, abundant mycorrhizae (Harvey et al., 1979), and thin layers of germination-promoting mosses (Place, 1955; St. Hilaire and Leopold, 1995). Stumps can differ somewhat from rotting logs in their ecological role. Only on stumps did Tsuga heterophylla become established within gaps that were otherwise filled by Abies amabilis (Minore, 1972). Stumps sometimes resprout, perpetuating the occupation of a site by the same species. Resprouting is common for some species (particularly angiosperms) but not possible for others (particularly conifers). Immediately

211

after a Pennsylvania tornado, 25% of all snapped trees resprouted, but only 68% of these were still alive four years later (Peterson and Pickett, 1991). In a Japanese forest, Castanopsis cuspidata resprouts from stumps and thus can persist, despite its absence from the predisturbance understory (Yamamoto, 1992). Another contribution of coarse woody debris is as habitat for diverse fungal species including those involved with decomposition. Old-growth forests preserve a high level of fungal diversity that is not paralleled in second-growth forests or even in virgin forests where wind-thrown trees are removed (Hood et al., 1989). In experimental gaps within a Pseudotsuga menziesii forest in Oregon, the “skirt” zone at the bases of stumps, snags, and even living trees supported higher densities of mycorrhizal root tips than fallen logs (Vogt et al., 1995). A key question about the importance of rotting wood is how quickly the stump or log will become suitable for seedling establishment. Wood decomposition rates vary widely among localities [from decades in subalpine forests (Lambert et al., 1980) to within a few years in tropical rain forests (Lieberman et al., 1985)], and also among tree species (Hood et al., 1989) and sizes (larger trees decay more slowly; Maser and Trappe, 1984). Where portions of fallen trees are propped above the ground, decomposition is slowed in eastern and midwestern forests in the U.S.A. (pers. observ.), but in temperate rainforests a standing dead tree will decompose more quickly than a fallen tree (Cline et al., 1980). Below-ground gaps: soil moisture and nutrient changes As foresters have long recognized, light is not the only resource for which plants compete in forests. When a tree blows down, the pulse of new resources previously usurped by that tree includes not only light but also below-ground resources: water, nutrients, and even space itself (McConnaughay and Bazzaz, 1991). Thus, a below-ground “gap” is created, an opportunity like the above-ground light gap for increased growth by existing plants or possibly for establishment by new colonists. Here I review studies of change at and below the soil level, beyond those changes involved with formation of microsites (mound, pit, stump, log) as already discussed. Evidence for below-ground gaps comes from several

212

types of research. Trenching experiments have demonstrated release from competition when neighboring roots were excluded, even without canopy clearance (Toumey and Kienholz, 1931; Korstian and Coile, 1938; Horn, 1985). Comparative photographs show a lush growth of tall shrubs and grasses eight years after trenching, in contrast with unvegetated untrenched plots (Toumey and Kienholz, 1931). In a more manipulative study, soil moisture levels were higher and transplanted seedlings of Acer rubrum and Cornus florida had better survivorship or growth where neighboring roots were severed within a North Carolina forest (Horn, 1985). Low root densities can also suggest the presence of below-ground gaps. The biomass of fine roots was significantly less in small tree-fall gaps than in surrounding forest soils in a Pennsylvania study (Wilczynski and Pickett, 1993). In contrast, small gaps in a Pinus contorta forest in southeastern Wyoming differed little from undisturbed areas in abundance of fine root tips and ectomycorrhizal roots; however, increasing size of the clearing up to a 30-tree gap was correlated with a decrease in root growth and ectomycorrhizae. Thus, the large gaps had detectable underground components (Parsons et al., 1994a,b). Total root biomass, and especially biomass of small roots, were also lower in openings within New Jersey Pinus rigida forests (Ehrenfeld et al., 1995). Measurements of fine-root biomass and turnover are increasingly utilized as indicators of environmental variation and change (Vogt et al., 1993). Nutrient availability has also been examined in gaps. In a Pinus contorta forest in Wyoming (Parsons et al., 1994b), the gradient of increasing gap size was paralleled by increasing nitrogen mobilization, suggesting a below-ground gap with respect to the consumption of nitrogen. These findings might be different in windstorm-caused gaps where, unlike the logged openings of that project, fallen trees remain on the ground. Nitrogen cycling was also modified by small tree-fall gaps in Acer saccharum/Tsuga canadensis forest in Michigan, but in different directions under the two dominant trees (Mladenoff, 1987). Nitrogen mineralization and nitrification were greater in gaps within evergreen, acid-soil Tsuga areas but were less in gaps in Acer saccharum areas, as compared with paired intact control (non-gap) areas. In the soils of New Jersey pine barrens with very low nitrogen content, openings of apparent fire origin had virtually no organic matter and five times less litter compared

Sara L. WEBB

with the forested matrix, perhaps because of a scarcity of ericaceous shrubs with their abundant surface roots. Nitrogen availability in these openings was low because of low inputs, while nitrogen availability in the forest was low because of rapid cycling (Ehrenfeld et al., 1995). With the recent explosion of inquiry into belowground ecological processes, there is a growing realization that the forest floor is highly heterogeneous and that analysis should be stratified by patch type (Vogt et al., 1995). Windstorms contribute to spatial patterning via light gaps, coarse woody debris, mound/pit microtopography, and diminished root competition and thus enhanced resource availability where trees are killed. Seed dispersal, seed banks, and seed predation The future of a wind-disturbed area depends not only upon resource pulses and understory development but also upon seed ecology: seed banks, seed dispersal, seed predation, and seed pathogens. Pathogenicity is lower in light gaps for a tropical forest (Augspurger, 1984), while the other seed-related interactions can be stronger or weaker after windstorms under various circumstances. Because, as has been seen, windstorm responses are usually controlled by pre-existing vegetation, seed dynamics may not be of universal importance. Seed ecology is thus perhaps most important in large blow-downs, on mounds and other wind-created microsites, and for shade-tolerant forest species in areas with small gaps and/or well developed understory vegetation. Seed availability itself depends upon the history of the site, the size of the clearing as it influences distances to seed sources, and the behavior of seed dispersers and seed predators as modified by the presence of fallen trees and light-gaps. The seed bank is a pool of buried, viable, but dormant seeds. For most dormant forest seeds, germination is triggered by environmental signals linked to light (V´asquez-Yanes and Orozco-Segovia, 1994). As already noted, in North America the seed bank is most conspicuously the source of light-demanding Prunus and Rubus, genera known for long persistence as seeds and for abundant growth in large blow-downs. However, the seed bank is actually a surprisingly diverse collection of propagules (Leck et al., 1989; Beatty, 1991). A high degree of spatial variability in seed pools (V´asquez-Yanes and Orozco-Segovia, 1994) reflects in part the history of the forest. Peterson and

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS

Carson (1996) found a minimal seed-bank response in a Pennsylvania old-growth forest without previous history of disturbance, whereas nearby second-growth forests responded to the same tornado with extensive germination from a rich seed bank of plants presumably present at an earlier stage of succession. Peterson and Carson reviewed other studies that also support the importance of stand history to the abundance of buried seeds. In contrast, Beatty (1991) found that buried seed pools were similar in composition to the existing vegetation in a wide range of habitats in central New York, thus reflecting recent seed rain rather than stand history. This was also true of seed banks of mounds and pits: most species present as seeds were already established in the vegetation. Mladenoff (1990) also found that seed banks were compositionally similar to the existing forest understory in small gaps in western Upper Michigan; interestingly, gap seed banks had larger densities of seeds than the seed bank of the undisturbed forest (Mladenoff, 1990). Seed dispersal can also influence windstorm responses (Schupp et al., 1989). If seed sources are too distant, then a plant that is perfectly capable of growing on disturbed ground will be excluded by the dispersal constraint. In tropical rainforests, vertebrates disperse seeds abundantly and over substantial distances (Denslow and Gomez Diaz, 1990). Dispersal distances seem more limited for many temperate-zone species. In a New Hampshire northern hardwoods forest, the quantity of wind-dispersed seeds of both Acer saccharum and Betula alleghaniensis declined exponentially with distance from a forest edge, while larger nut-like seeds of Fagus grandifolia were virtually absent from the open area (Hughes and Fahey, 1988). The seed rain might extend farther into tree-fall gaps than into the logged openings of that study, because coarse woody debris would provide cover and perches for vertebrate dispersal agents (McDonnell and Stiles, 1983) and could also introduce roughness and thus more wind turbulence which would promote deposition of wind-dispersed propagules. Other studies document substantial dispersal distances for beechnuts and acorns (Johnson and Adkisson, 1986), and paleoecological tracking of postglacial tree migrations also shows that most temperate forest trees have a surprising capacity for dispersal over long distances (Davis, 1981b, 1983; Webb, 1986, 1987). Seed predation rates can also change when trees fall, because neither vertebrate nor invertebrate seed predators utilize all microhabitats equally. For example,

213

uprooted trees influence the movements of rodents in a Polish forest (Olszewski, 1968) and presumably elsewhere. Ant diversity and abundance were enriched in wind-fall areas in a Picea/Fagus forest in Bavaria (Theobald-Ley and Horstmann, 1990), including some species of seed-consuming ants. In a Minnesota study, post-dispersal predation on seeds of the tall shrub Prunus virginiana was much heavier in closed forest than in adjacent open fields or small tree-fall gaps (Webb and Willson, 1985), perhaps reflecting avoidance of exposed sites by vertebrate foragers. Such predators caught in traps baited with P. virginiana seeds included Clethrionomys gaperi, Eutamias minimus, Peromyscus leucopus, Spermophilus tridecemlineatus, Tamias striatus, and Zapus hudsonius. Meanwhile, smaller ant-dispersed seeds of the woodland herb Uvularia grandiflora faced heavier post-dispersal predation (after elaisome removal) in open fields than in small tree-fall gaps or closed forest (Webb and Willson, 1985). Similar studies by Whelan et al. (1991) with Cornus drummondi and Prunus serotina also showed spatial heterogeneity in predation intensity depending upon the microhabitat, but furthermore demonstrated high variance in predation patterns between years. In both disturbed and intact forests, a better understanding is needed of seed predation, which could help explain the rarity of some seedlings and the abundance of others.

CONCLUSIONS

Generalizations are elusive when one considers the myriad scenarios that have been documented for windstorm consequences in temperate forests. Windstorm disturbance can enrich diversity or deplete it at the landscape scale, and can set back succession or accelerate it within patches where trees have blown down. Most temperate forests around the world have understory layers of shade-tolerant species that are capable of responding to canopy openings. No general predictions about windstorm consequences can be made on the basis of forest type (conifer, evergreen, deciduous, xeric, humid) or successional stage. Each tree species has its own individualistic combination of shade tolerance, wind-firmness, and other life-history features that dictate its vulnerability to wind damage and capacity to regenerate following the disturbance. The type of response depends upon the details of the forest itself and can hinge upon understory

214

composition, sometimes controlled by the presence or absence of a single key taxon (such as bamboo or Acer saccharum). A pre-storm understory of shade-tolerant saplings, shrubs, or forbs may inhibit new regeneration by light-demanding species, even where large canopy openings are created. A post-storm flush of resprouting by trees and shrubs can have a similar effect. Thus, strong autogenic forces operate in the wake of a windstorm. Human activity has modified forest structure and understory development indirectly by changing patterns of herbivore abundance and distribution; deer in eastern North America and introduced mammals in New Zealand exert so much influence that it is difficult to understand natural forest dynamics. The disturbance response also depends upon allogenic aspects of individual windstorm events and of the disturbance regime in a given locality. Windstorms do not vary along a single-dimensional continuum; for example, hurricanes disturb larger areas (higher magnitude) but with lower wind-speeds (lower intensity) compared with tornadoes. Blow-downs and tornadoes that create large clearings also have fringe zones with scattered tree-falls. Surveys of mortality and damage show that tree species differ in vulnerability to wind damage but with site-specific influence of confounded factors such as topography, tree size, and fungal infection. Some trees survive even the most severe windstorms. Two strong concerns hamper the utility of much winddamage research: methods of assessing damage must be standardized (Everham, 1995), and the undamaged component of the forest must be sampled as context for interpreting the profile of mortality. Wind-throw mounds, pits, stumps, and rotting logs each play important roles in windstorm responses in some but not all forests. Thus, large tracts need to be protected from salvage logging for the sake of tree regeneration, which is dependent on dead wood in many temperate forests. The natural patchiness of vegetation makes causal patterns difficult to tease apart without long-term study over large areas. Unfortunately, fallen and damaged trees are usually removed following windstorms, leaving only small, fragmented nature preserves where dynamics of coarse woody debris can be studied. More generally, our understanding of windstorm disturbance regimes is hampered by extensive forest clearance and conversion to agriculture of most portions of the temperate forest biome. For example, of virgin forest present in the United States before European

Sara L. WEBB

settlement, only a tiny fraction remains (0.4%–1.6% for eastern regions; Frelich, 1995; Davis, 1996). Yet nearly all sizeable remnants in eastern North America have been impacted by catastrophic windstorms in recent decades. Knowledge of windstorm consequences necessitates the preservation of extensive areas so that the full range of patch types and the full range of disturbance responses are preserved (White, 1987; Foster et al., 1996). Long-term studies are needed, for several reasons: to monitor windstorm return times, damage patterns, and responses; and to examine interactions among windstorm events. Recent projects that began after specific windstorms should be continued not only to track changes in blow-down areas but also to track subsequent windstorm occurrences and consequences. Remarkably few studies actually examine dynamics that are known to relate to windstorms rather than to unspecified gap-forming causes and chronologies. Such gap-dynamics studies provide only limited elucidation of windstorm consequences, because they exclude turnover not occurring in gaps but include gaps created by non-wind causes. In the future, every effort should be made to identify causal agents and timing of gap formation, not only for insights into disturbance but also to understand the extent to which the present gap configuration is in equilibrium, and to assign ecological mechanisms to the dynamics. Future gap research should also incorporate “control” nongap patches in which parallel sampling is done, to permit distinguishing aspects of the gap composition and structure which are related to the gap and others which simply mirror the background context. The extensive literature cited in this chapter provides a framework in which to place additional research on gap dynamics in temperate forests. Early detailed studies of forests in the eastern United States led to generalizations that patch dynamics promote landscape-level diversity, generalizations which turned out to be contradicted by studies in other regions of the United States and of the world. Further progress in disturbance ecology requires a global perspective and more research outside of North America. ACKNOWLEDGEMENTS

This chapter was supported in part by research grants from the Minnesota Department of Natural Resources and from Drew University. For helpful suggestions I thank Charlie Cogbill, Sara Cooper-Ellis, David Foster, Lawrence Walker and Dennis Whigham.

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS REFERENCES Abe, S., Masaki, T. and Nakashizuka, T., 1995. Factors influencing sapling composition in canopy gaps of a temperate deciduous forest. Vegetatio, 120: 21–32. Abrams, M.D. and Scott, M.L., 1989. Disturbance-mediated succession in two Michigan forest types. For. Sci., 35: 42–49. Adams, J.A. and Norton, M.L., 1991. Soil and vegetation characteristics of some tree wind-throw features in a South Westland rimu forest. J. R. Soc. N. Z., 21: 33–42. Akachuka, A.E., 1993. Recovery and morphology of Pinus resinosa trees 50 years after they were displaced by a hurricane. For. Ecol. Manage., 56: 113–129. Allen, B.P., Pauley, E.F. and Sharitz, R.R., 1997. Hurricane impacts on liana populations in an old-growth southeastern bottomland forest. J. Torrey Bot. Soc., 124: 34–42. Allison, T., Moeller, R.E. and Davis, M.B., 1986. Pollen in laminated sediments provides evidence for a mid-Holocene forest pathogen outbreak. Ecology, 67: 1101–1105. Aplet, G.H., Laven, R.D. and Smith, F.W., 1988. Patterns of community dynamics in Colorado USA Englemann spruce– subalpine fir forests. Ecology, 69: 312–319. Armesto, J.J., Casasso, I. and Dollenz, O., 1992. Age structure and dynamics of Patagonian beech forests in Torres del Paine National Park, Chile. Vegetatio, 98: 13–22. Arriaga, L., 1988. Natural disturbance and tree-falls in a pine–oak forest on the peninsula of Baja California, Mexico. Vegetatio, 78: 73–80. Augspurger, C.K., 1984. Seedling survival among tropical tree species: interactions of dispersal distance, light gaps, and pathogens. Ecology, 65: 1705–1712. Barden, L.S., 1979. Tree replacement in small canopy gaps of a Tsuga canadensis forest in the southern Appalachians, Tennessee. Oecologia, 44: 141–142. Barden, L.S., 1980. Tree replacement in a cove hardwood forest of the southern Appalachians. Oikos, 35: 16–19. Barden, L.S., 1981. Forest development in canopy gaps of a diverse hardwood forest of the southern Appalachian Mountains. Oikos, 37: 205–209. Barden, L.S., 1989. Repeatability in forest gap research studies in the Great Smoky Mountains. Ecology, 70: 558–559. Barik, S.K., Pandey, H.N., Tripathi, R.S. and Rao, P., 1992. Microenvironmental variability and species diversity in tree-fall gaps in a sub-tropical broadleaved forest. Vegetatio, 103: 31–40. Battaglia, L.L., Sharitz, R.R. and Minchin, P.R., 1999. Patterns of seedling and overstory composition along a gradient of hurricane disturbance in an old-growth bottomland hardwood community. Can. J. For. Res., 29: 144-156. Battan, L.J., 1961. The Nature of Violent Storms. Doubleday, Garden City, New York, 158 pp. Beatty, S.W., 1984. Influence of microtopography and canopy species on spatial patterns of forest understory plants. Ecology, 65: 1406–1419. Beatty, S.W., 1991. Colonization dynamics in a mosaic landscape: the buried seed pool. J. Biogeogr., 18: 553–563. Beatty, S.W. and Sholes, O.D.V., 1988. Leaf litter effect on plant species composition of deciduous forest pits. Can. J. For. Res., 18: 553–559.

215

Beatty, S.W. and Stone, E.L., 1986. The variety of soil microsites created by tree-falls. Can. J. For. Res., 16: 539–548. Becquey, J. and Riou-Nivert, P., 1987. The existence of areas of stability in stands: consequences for management. Rev. For. Fr. (Nancy), 139: 323–334 (in French). Boose, E.R., Foster, D.R. and Fluet, M., 1994. Hurricane impacts to tropical and temperate forest landscapes. Ecol. Monogr., 64: 369–400. Bormann, F.H. and Likens, G.E., 1979. Catastrophic disturbance and the steady-state in northern hardwood forests. Am. Sci., 67: 660–669. Bratton, S.P., 1976. Resource division in an understory herb community: response to temporal and micro-topographical gradients. Am. Nat., 110: 679–693. Bray, J.R., 1956. Gap phase replacement in a maple-basswood forest. Ecology, 37: 598–600. Brewer, R. and Merritt, P.G., 1978. Wind throw and tree replacement in a climax beech–maple forest. Oikos, 30: 149–152. Brokaw, N.V.L., 1982. The definition of tree-fall gap and its effect on measures of forest dynamics. Biotropica, 14: 158–160. Brokaw, N.V.L. and Scheiner, S.M., 1989. Species composition in gaps and structure of a tropical forest. Ecology, 70: 538–541. Brokaw, N.V.L. and Walker, L.R., 1991. Summary of the effects of Caribbean hurricanes on vegetation. Biotropica, 23: 442–447. Burrows, C.J., 1990. Processes of Vegetation Change. Unwin-Hyman, London, 551 pp. Caldwell, M.M. and Pearcy, R.W., 1994. Exploitation of Environmental Heterogeneity by Plants: Ecophysiological Processes Above and Belowground. Academic Press, New York, 429 pp. Canham, C.D., 1985. Suppression and release during canopy recruitment in Acer saccharum. Bull. Torrey Bot. Club, 112: 134–145. Canham, C.D., 1988. An index for understory light levels in and around canopy gaps. Ecology, 69: 1634–1638. Canham, C.D., 1989. Different responses to gaps among shadetolerant tree species. Ecology, 70: 548–550. Canham, C.D., 1990. Suppression and release during canopy recruitment in Fagus grandifolia. Bull. Torrey Bot. Club, 117: 1–7. Canham, C.D. and Loucks, O.L., 1984. Catastrophic wind-throw in the presettlement forests of Wisconsin. Ecology, 65: 803–809. Canham, C.D. and Marks, P.L., 1985. The response of woody plants to disturbance: patterns of establishment and growth. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, pp. 197–216. Canham, C.D., Denslow, J.S., Platt, W.J., Runkle, J.R., Spies, T.A. and White, P.S., 1990. Light regimes beneath closed canopies and tree-fall gaps in temperate and tropical forests. Can. J. For. Res., 20: 620–631. Canham, C.D., Finzi, A.C., Pacala, S.W. and Burbank, D.H., 1994. Causes and consequences of resource heterogeneity in forests: interspecific variation in light transmission by canopy trees. Can. J. For. Res., 24: 337–349. Carlton, G.C. and Bazzaz, F.A., 1998. Resource congruence and forest regeneration following an experimental hurricane blowdown. Ecology, 79: 1305–1319. Chazdon, R.L., 1988. Sunflecks and their importance to forest understorey plants. Adv. Ecol. Res., 18: 1–63.

216 Chen, T.-H. and Horng, F.W., 1993. Survey on the present conditions of Peng Hu coastal forests and the recovery of trees damaged from salt-mist brought by typhoon. Bull. Taiwan For. Res. Inst. N. Ser., 8: 129–142 (in Chinese). Cho, D.-S. and Boerner, R.E.J., 1991. Canopy disturbance patterns and regeneration of Quercus species in two Ohio old-growth forests. Vegetatio, 93: 9–18. Christensen, N.L. and Franklin, J.F., 1987. Small-scale disturbance in forest ecosystems. Bull. Ecol. Soc. Am., 68: 51–53. Christy, E.J. and Mack, R.N., 1984. Variation in demography of juvenile Tsuga heterophylla across the substratum mosaic. J. Ecol., 72: 75–91. Clark, J.S., 1988. Effect of climate change on fire regimes in northwestern Minnesota. Nature, 334: 233–235. Clark, J.S., 1989a. Ecological disturbance as a renewal process: theory and application to fire history. Oikos, 56: 17–30. Clark, J.S., 1989b. Effects of long-term water balances on fire regime, north-western Minnesota. J. Ecol., 77: 989–1004. Clark, J.S., 1991. Forest tree growth rates and probability of gap origin – a comment. Ecology, 70: 1166–1169. Clark, J.S., 1992. Disturbance and tree life history on the shifting mosaic landscape. Ecology, 72: 1102–1118. Clebsch, E.E.L. and Busing, R.T., 1989. Secondary succession, gap dynamics, and community structure in a southern Appalachian cove forest. Ecology, 70: 728–735. Cline, S.P., Berg, A.B. and Wight, H.M., 1980. Snag characteristics and dynamics in Douglas fir forests, western Oregon. J. Wildl. Manage., 44: 773–780. Clinton, B.D., Boring, L.R. and Swank, W.T., 1994. Regeneration patterns in canopy gaps of mixed-oak forests of the southern Appalachians: influences of topographic position and evergreen understory. Am. Midl. Nat., 132: 308–319. Cogbill, C.V., 1996. Black growth and fiddlebutts: the nature of oldgrowth red spruce. In: M.B. Davis (Editor), Eastern Old-Growth Forests: Prospects for Rediscovery and Recovery. Island Press, Washington, D.C., pp. 113–125. Collins, B.S. and Pickett, S.T.A., 1982. Vegetation composition and relation to environment in an Allegheny hardwoods forest. Am. Midl. Nat., 108: 117–123. Collins, B.S. and Pickett, S.T.A., 1987. Influence of canopy opening on the environment and herb layer in a northern hardwoods forest. Vegetatio, 70: 3–10. Collins, B.S. and Pickett, S.T.A., 1988a. Demographic responses of herb layer species to experimental canopy gaps in a northern hardwoods forest. J. Ecol., 76: 437–450. Collins, B.S. and Pickett, S.T.A., 1988b. Response of herb layer cover to experimental canopy gaps. Am. Midl. Nat., 119: 282–290. Collins, B.S., Dunne, K.P. and Pickett, S.T.A., 1985. Responses of forest herbs to canopy gaps. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, pp. 218–234. Collins, S.L., 1961. Benefits to understory from canopy defoliation by gypsy moth larvae. Ecology, 42: 836–838. Coutts, M.P. and Grace, J. (Editors), 1995. Wind and Trees. Cambridge University Press, Cambridge, 585 pp. Dahir, S.E. and Lorimer, C.G., 1996. Variation in canopy gap formation among developmental stages of northern hardwood stands. Can. J. For. Res., 26: 1875–1892. Davis, M.B., 1981a. Outbreaks of Forest Pathogens in Quaternary

Sara L. WEBB History. Proc. 4th Int. Palynology Conf., Lucknow, India, 3: 216–227. Davis, M.B., 1981b. Quaternary history and the stability of deciduous forests. In: D.C. West, H.H. Shugart and D.B. Botkin (Editors), Forest Succession. Springer-Verlag, New York, pp. 132–177. Davis, M.B., 1983. Holocene vegetation history of the eastern deciduous forest. In: H.E. Wright Jr. (Editor), Late Quaternary Environments of the United States, Vol. 2, the Holocene. University of Minnesota Press, Minneapolis, pp. 166–181. Davis, M.B., 1986. Climatic instability, time lags, and community disequilibrium. In: J. Diamond and T.J. Case (Editors), Community Ecology. Harper and Row, New York, pp. 269–284. Davis, M.B., 1996. Eastern Old-Growth Forests: Prospects for Rediscovery and Recovery. Island Press, Washington, D.C., 383 pp. De Seze, E., 1987. Inquiry into the wind-throws of 1982 in the Limousin, France. Rev. For. Fr. (Nancy), 139: 276–280 (in French). Debourdieu, J., 1987. Adaptation of management plans and techniques in the public forests damaged by wind in the Auvergne and the Limousin, France. Rev. For. Fr.(Nancy), 139: 335–337 (in French). Denny, C.S. and Goodlett, J.C., 1956. Microrelief resulting from fallen trees. In: C.S. Denny (Editor), Surficial Geology and Geomorphology of Potter County, Pennsylvania. U.S. Geological Survey, Professional Paper 288, pp. 59–66. Denslow, J.S., 1980. Gap partitioning among tropical rainforest trees. Biotropica, 12: 47–55 (supplement). Denslow, J.S. and Gomez Diaz, A.E., 1990. Seed rain to tree-fall gaps in a neotropical rain forest. Can. J. For. Res., 20: 642–648. Denslow, J.S. and Spies, T., 1990. Canopy gaps in forest ecosystems: an introduction. Can. J. For. Res., 20: 619. Despres, F., 1987. The wind-throws of November 6 and 7, 1982, in the Auvergne, France. Rev. For. Fr. (Nancy), 139: 281–287 (in French). Dessens, J., 1972. Influence of ground roughness on tornadoes: a laboratory simulation. J. Appl. Meteorol., 11: 72–75. Duever, M.J. and McCollom, J.M., 1993. Hurricane Hugo effects on old-growth floodplain forest communities at Four Holes Swamp, South Carolina. In: J.C. Brissette (Editor), Proceedings of the Seventh Biennial Southern Silvicultural Research Conference. USDA Forest Service General Technical Report SO-93, pp. 197– 202. Dunn, C.P., Guntenspergen, G.R. and Dorney, J.R., 1983. Catastrophic wind disturbance in an old-growth hemlock-hardwood forest. Can. J. Bot., 61: 211–217. Dynesius, M. and Jonsson, B.G., 1991. Dating uprooted trees: comparison and application of eight methods in a boreal forest. Can. J. For. Res., 21: 655–665. Easter, M.J. and Spies, T.A., 1994. Using hemispherical photography for estimating photosynthetic photon flux density under canopies and in gaps in Douglas fir forests of the Pacific northwest. Can. J. For. Res., 24: 2050–2058. Ehrenfeld, J.G., 1980. Understory response to canopy gaps of varying size in a mature oak forest. Bull. Torrey Bot. Club, 107: 29–41. Ehrenfeld, J.G., Zhu, W. and Parsons, W.F.J., 1995. Above- and belowground characteristics of persistent forest openings in New Jersey pinelands. Bull. Torrey Bot. Club, 122: 298–305.

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS Ennos, A.R., 1997. Wind as an ecological factor. Trends Ecol. Evol., 12: 108–111. Everham III, E.M., 1995. A comparison of methods for quantifying catastrophic wind damage to forests. In: M.P. Grace and J. Grace (Editors), Wind and Trees. Cambridge University Press, Cambridge, England, pp. 340–357. Falinski, J.B., 1978. Uprooted trees, their distribution and influence in the primeval forest biotope. Vegetatio, 38: 175–183. Fischer, A., 1992. Long term vegetation development in Bavarian Mountain forest ecosystems following a natural destruction. Vegetatio, 103: 93–104. Foster, D.R., 1988a. Disturbance history, community organization, and vegetation dynamics of the old-growth Pisgah forest, southwestern New Hampshire USA. J. Ecol., 76: 105–134. Foster, D.R., 1988b. Species and stand response to catastrophic wind in central New England USA. J. Ecol., 76: 135–151. Foster, D.R., 1995. Land-use history and four hundred years of vegetation change in New England. In: B.L. Turner II, A.G. Sal, F.G. Bernaldez and F. di Castri (Editors), Global Land Use Change: a Perspective from the Columbian Encounter. Consejo Superior de Investigaciones Cientificas, Madrid, Spain, pp. 253–319. Foster, D.R. and Boose, E.R., 1992. Patterns of forest damage resulting from catastrophic wind in central New England, USA. J. Ecol., 80: 79–98. Foster, D.R. and Boose, E.R., 1995. Hurricane disturbance regimes in temperate and tropical forest ecosystems. In: M.P. Coutts and J. Grace (Editors), Wind and Trees. Cambridge University Press, Cambridge, England, pp. 305–339. Foster, D.R., Schoonmaker, P.K. and Pickett, S.T.A., 1990. Insights from paleoecology to community ecology. Trends Ecol. Evol., 5: 119–122. Foster, D.R., Zebryk, T., Schoonmaker, P. and Lezberg, A., 1992. Postsettlement history of human land-use and vegetation dynamics of a Tsuga canadensis (hemlock) woodlot in central New England. J. Ecol., 80: 773–386. Foster, D.R., Orwig, D.A. and McLachlan, J.S., 1996. Ecological and conservation insights from reconstructive studies of temperate old-growth forests. Trends Ecol. Evol., 11: 419–424. Foster, D.R., Aber, J.D., Melillo, J.M., Bowden, R.D. and Bazzaz, F.A., 1997. Temperate forest response to natural catastrophic disturbance and chronic anthropogenic stress. Bioscience, 47: 437–445. Foster, D.R., Knight, D. and Franklin, J., 1999. Landscape patterns and legacies resulting from large infrequent forest disturbances. Ecology, in press. Foster, J.R. and Reiners, W.A., 1986. Size distribution and expansion of canopy gaps in a northern Appalachian USA spruce–fir forest. Vegetatio, 68: 109–114. Franklin, J.F. and Dyearsness, C.T., 1973. Natural vegetation of Oregon and Washington. U.S. Forest Service General Technical Report PNW-8. Pacific Northwest Forest and Range Experiment Station, Portland, Oregon, 417 pp. Frelich, L.E., 1995. Old forest in the lake states today and before European settlement. Nat. Areas J., 15: 157–167. Frelich, L.E. and Lorimer, C.G., 1991. Natural disturbance regimes in hemlock–hardwood forest of the Upper Great Lakes region. Ecol. Monogr., 61: 145–164. Frelich, L.E. and Martin, G.L., 1988. Effects of crown expansion into gaps on evaluation of disturbance intensity in northern hardwood forests. For. Sci., 34: 530–536.

217

Frelich, L.E. and Reich, P.B., 1995. Neighborhood effects, disturbance, and succession in forests of the western Great Lakes region. Ecoscience, 2: 148–158. Frissell Jr., S.S., 1973. The importance of fire as a natural ecological factor in Itasca State Park, Minnesota. Quat. Res., 3: 397–407. Fujita, T.T., 1978. Manual of down-burst identification. Satellite and Mesometeorology Research Project Research Paper No. 156, University of Chicago, Chicago, Illinois. Gallagher, G.J., 1975. Wind-thrown in state forests in the republic of Ireland. Ir. For., 32: 154–167. Gallimore, R.G. and Lettau, H.H., 1970. Topographic influence on tornado tracks and frequencies in Wisconsin and Arkansas. Trans. Wisconsin Acad. Sci. Arts Lett., 58: 101–127. Gardner, L.R., Michener, W.K., Williams, T.M., Blood, E.R., Kjerve, B., Smock, L.A., Lipscomb, D.J. and Gresham, C.A., 1992. Disturbance effects of Hurricane Hugo on a pristine coastal landscape: North Inlet, South Carolina, USA. Neth. J. Sea Res., 30: 249–263. Gibson, N. and Brown, M.J., 1991. The ecology of Lago strobosfrankinii (Podocarpaceae) in Tasmania: 2. Population structure and spatial pattern. Aust. J. Ecol., 16: 223–230. Glenn-Lewin, D.C., Peet, R.K. and Veblen, T.T. (Editors), 1992. Plant Succession: Theory and Prediction. Chapman and Hall, London, 352 pp. Glitzenstein, J.S. and Harcombe, P.A., 1988. Effects of the December 1993 tornado on forest vegetation of the Big Thicket, southeast Texas, USA. For. Ecol. Manage., 25: 269–290. Glitzenstein, J.S., Harcombe, P.A. and Streng, D.R., 1986. Disturbance, succession, and maintenance of species diversity in an east Texas forest. Ecol. Monogr., 56: 243–258. Gloyne, R.W., 1968. The structure of wind; and its relevance to forestry. In: Society of Foresters of Great Britain: Wind Effects on the Forest. Oxford University Press, pp. 7–19 (or In: R.M.V. Palmer (Editor), Report of the Eight Discussion Meeting, Society of Foresters of Great Britain. Forestry (suppl), 41: 7–19.) Goldblum, D., 1997. The effects of tree-fall gaps on understory vegetation in New York State. J. Vegetation Sci., 8: 125–132. Goodlett, J.C., 1954. Vegetation adjacent to the border of the Wisconsin drift in Potter County, Pennsylvania. Harv. For. Bull., 25. Grace, J., 1977. Plant Responses to Wind. Academic Press, London, 204 pp. Gray, A.N. and Spies, T.A., 1992. Tree establishment in canopy openings in mature Douglas-fir forests. Northwest Environ. J., 8: 207–208. Gresham, C.A., Williams, T.M. and Lipscomb, D.J., 1991. Hurricane Hugo: wind damage to southeastern U.S. coastal forest tree species. Biotropica, 23: 420–426. Gysel, L.W., 1951. Borders and openings of beech–maple woodlands in southern Michigan. J. For., 49: 13–19. Habecker, M.S., McSweeney, K. and Meyers, N.L., 1990. Variability of a fragiochrept with wind-throw microtopography in north central Wisconsin, USA. Soil Sci. Soc. Am. J., 54: 483–488. Hara, M., 1983. A study of the regeneration process of a Japanese beech forest. Ecol. Rev., 20: 115–129. Hara, M., 1985. Forest response to gap formation in a climax beech forest. Jpn. J. Ecol., 35: 337–343. Hara, M., 1987. Analysis of seedling banks of a climax beech forest: ecological importance of seedling sprouts. Vegetatio, 71: 67–74.

218 Harmon, M.E. and Franklin, J.F., 1989. Tree seedlings on logs in Picea–Tsuga forests of Oregon and Washington. Ecology, 70: 48–59. Harmon, M.E., Bratton, S.P. and White, P.S., 1983. Disturbance and vegetation response in relation to environmental gradients in the Great Smoky Mountains. Vegetatio, 55: 129–139. Harmon, M.E., Franklin, J.F., Swanson, F.J., Sollins, P., Gregory, S.V., Latkin, J.D., Anderson, N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper, G.W., Cromack Jr., K. and Cummins, K.W., 1986. Ecology of coarse woody debris in temperate ecosystems. Adv. Ecol. Res., 15: 133–302. Harris, A.S., 1989. Wind in the forests of southeast Alaska USA and guides for reducing damage. General Technical Report PNWGTR-244. U.S. Forest Service, Pacific Northwest Research Station, 63 pp. Harvey, A.E., Larsen, M.J. and Jurgensen, M.F., 1979. Comparative distribution of ectomycorrhizae in soils of three western Montana forest habitat types. For. Sci., 25: 350–358. Heinrich, J., 1991. Causes of wind damage in beech forests (Fagus sylvatica) from a geoscientific viewpoint. Allg. Forst Jagdztg., 162(8): 145–149 (in German). Heinselman, M.L., 1973. Fire in the virgin forests of the Boundary Waters Canoe Area, Minnesota. Quat. Res., 3: 329–382. Heinselman, M.L., 1981. Fire and succession in the conifer forests of North America. In: D.C. West, H.H. Shugart and D.B. Botkin (Editors), Forest Succession. Springer-Verlag, New York, pp. 374– 405. Held, M.E. and Bryant, W.S., 1989. Recovery in a tornado-damaged hardwood forest in northern Kentucky. Trans. Kentucky Acad. Sci., 50: 16–21. Henry, J.D. and Swan, J.M.A., 1974. Reconstructing forest history from live and dead material – an approach to the study of forest succession in southwest New Hampshire. Ecology, 55: 772–783. Hibbs, D.E., 1982. Gap dynamics in a hemlock–hardwood forest. Can. J. For. Res., 12: 522–527. Hibbs, D.E., 1983. Forty years of succession in central New England. Ecology, 64: 1394–1401. Hibbs, D.E., Wilson, B.F. and Fischer, B.C., 1980. Habitat requirements and growth of striped maple (Acer pensylvanicum L.). Ecology, 61: 490–496. Hofgaard, A., 1993. Fifty years of change in a Swedish boreal oldgrowth Picea abies forest. J. Vegetation Sci., 4: 773–782. Hood, I.A., Sandberg, C.J. and Kimberley, M.O., 1989. A decay study of wind-thrown indigenous trees. N. Z. J. Bot., 27: 281–298. Horn, J.C., 1985. Responses of understory tree seedlings to trenching. Am. Midl. Nat., 114: 252–258. Hubbell, S.P., 1999. Light-gap disturbances, recruitment limitation, and tree diversity in a neotropical forest. Science, 283: 554–557. Hubert, E.E., 1918. Fungi as contributory causes of wind-fall in the Northwest. J. For., 16: 696–714. Huenneke, L.F., 1983. Understory response to gaps caused by the death of Ulmus americana in central New York. Bull. Torrey Bot. Club, 110: 170–175. Hughes, J.W. and Fahey, T.J., 1988. Seed dispersal and colonization in a disturbed northern hardwood forest. Bull. Torrey Bot. Club, 115: 89–99. Hunter, J.C. and Parker, V.T., 1993. The disturbance regime of an oldgrowth forest in coastal California. J. Vegetation Sci., 4: 19–24.

Sara L. WEBB Hutnick, R.J., 1952. Reproduction on wind-falls in a northern hardwood stand. J. For., 50: 693–694. Hutte, P., 1968. Experiments on windflow and wind damage in Germany. In: Society of Foresters of Great Britain: Wind Effects on the Forest. Oxford University Press, Oxford, England, pp. 20–25 (or In: R.M.V. Palmer (Editor), Report of the Eight Discussion Meeting, Society of Foresters of Great Britain. Forestry (suppl), 41: 7–19.) Hytteborn, H., Liu, Q.-H. and Verwijst, T., 1991. Natural disturbance and gap dynamics in a Swedish boreal spruce forest. In: N. Nakagoshi and F.B. Golley (Editors), Coniferous Forest Ecology from an International Perspective. SPB Academic Publishing SV., The Hague, Netherlands, pp. 93–108. Iwaki, H. and Totsuka, T., 1959. Ecological and physiological studies on the vegetation of Mt. Shimagare. II. On the crescent-shaped “dead-tree-strips” in the Yatasugatake and Chicubu Mountains. Bot. Mag., 72: 255–260. Jane, G.T., 1986. Wind damage as an ecological process in mountain beech forests of Canterbury, New Zealand. N. Z. J. Ecol., 9: 25–39. Jenkins, J., 1996. Notes on the Adirondack blow-down of July 15, 1995: scientific background, observations, responses, and policy issues. Report to the Wildlife Conservation Society. White Creek Field School, White Creek, New York. Johnson, W.C. and Adkisson, C.S., 1986. Dispersal of beechnuts by blue jays in fragmented landscapes. Am. Midl. Nat., 13: 319–324. Jonsson, B.G. and Dynesius, M., 1993. Uprooting in boreal spruce forests: long-term variation in disturbance rate. Can. J. For. Res., 23: 2383–2388. Kalinin, M.I., 1991. Assessment of wind-fall danger in the spruce stands of the Carpathian Mountains, Ukrainian SSR, USSR. Lesovedenie: 59–66 (in Russian). Kauhanen, H., 1991. Effects of the October 1985 wind-throw on forest structure in eastern Lapland. Luonnon Tutkija, 95: 45–48 (in Finnish). Kelty, M.J., 1986. Development patterns in two hemlock–hardwood stands in southern New England. Can. J. For. Res., 16: 885–891. Kessler, E. and White, G.F., 1983. Thunderstorms in a social context. In: E. Kessler (Editor), The Thunderstorm in Human Affairs. University of Oklahoma Press, Norman, Oklahoma, pp. 3–18. Kimball, A.J., Witham, J.W., Rudnicky, J.L., White, A.S. and Hunter Jr., M.L., 1995. Harvest-created and natural canopy gaps in an oak–pine forest in Maine. Bull. Torrey Bot. Club, 122: 115–123. King, D.A., 1986. Tree form, height growth, and susceptibility to wind damage in Acer saccharum. Ecology, 67: 980–990. Kohyama, T. and Fujita, N., 1981. Studies on the Abies populations of Mt. Shimigare. I. Survivorship curve. Bot. Mag., 94: 55–68. Konopka, J., 1985. Occurrence of winds mechanically damaging forest stands in Slovakia, Czechoslovakia. Ekologica-CSSR, 4: 267–288. Korstian, C.F. and Coile, T.S., 1938. Plant competition in forest stands. Duke Univ. Sch. For. Bull., 3: 1–125. Kupfer, J.A. and Runkle, J.R., 1996. Early gap successional pathways in a Fagus–Acer forest preserve: pattern and determinants. J. Vegetation Sci., 7: 247–256. Lambert, L., Lang, G.E. and Reiners, W.A., 1980. Loss of mass and chemical change in decaying boles of a subalpine balsam fir forest. Ecology, 61: 1460–1473.

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS Leck, M.A., Parker, V.T. and Simpson, R.L., 1989. Ecology of Soil Seed Banks. Academic Press, London, 462 pp. Leemans, R., 1991. Canopy gaps and establishment patterns of spruce (Picea abies L.) in two old-growth coniferous forests in central Sweden. Vegetatio, 103: 31–40. Lertzman, K.P., 1992. Patterns of gap-phase replacement in a sublpine, old-growth forest. Ecology, 73: 657–669. Lertzman, K.P. and Krebs, C.J., 1991. Gap-phase structure of a subalpine old-growth forest. Can. J. For. Res., 21: 1730–1741. Lertzman, K.P., Sutherland, G.D., Inselberg, A. and Saunders, S.C., 1996. Canopy gaps and the landscape mosaic in a coastal temperate rain forest. Ecology, 77: 1254–1270. Lieberman, M., Lieberman, D., Peralta, R. and Hartshorn, G.S., 1985. Mortality patterns and stand turnover rates in a wet tropical forest in Costa Rica. J. Ecol., 73: 915–924. Lieberman, M., Lieberman, D. and Peralta, R., 1989. Forests are not just swiss cheese: canopy sterogeometry of nongaps in tropical forests. Ecology, 70: 550–552. Lorimer, C.G., 1977. The presettlement forest and natural disturbance cycle of northeastern Maine. Ecology, 58: 139–148. Lorimer, C.G., 1980. Age structure and disturbance history of a southern Appalachian virgin forest. Ecology, 61: 1169–1184. Lorimer, C.G., 1985. Methodological considerations in the analysis of forest disturbance history. Can. J. For. Res., 15: 200–213. Lorimer, C.G. and Frelich, L.E., 1989. A methodology for estimating canopy disturbance frequency and intensity in dense temperate forests. Can. J. For. Res., 19: 651–663. Lorimer, C.G., Frelich, L.E. and Nordheim, E.V., 1988. Estimating gap origin probabilities for canopy trees. Ecology, 69: 778–785. Lorimer, C.G., Frelich, L.E. and Nordheim, E.V., 1992. Forest-tree growth rates and probability of gap origin: a reply to Clark. Ecology, 73: 1124–1128. Lutz, H.J., 1940. Disturbance of forest soil resulting from the uprooting of trees. Yale Univ. Sch. For. Bull., 45. Lutz, H.J. and Griswold, F.S., 1939. The influence of tree roots on soil morphology. Am. J. Sci., 237: 389–400. Lyford, W.H. and MacLean, D.W., 1966. Mound and pit microrelief in relation to soil disturbance and tree distribution in New Brunswick, Canada. Harv. For. Pap., 15: 18 pp. Maloney, K.A., 1986. Wave and non-wave regeneration processes in a subalpine Abies balsamea forest. Can. J. Bot., 64: 341–349. Mansikkaniemi, H. and Laitenen, T., 1991. Pattern of local wind changes in a fell region, northern Finland. Ann. Univ. Turku. Ser. A, 76: 11–20. Marchand, P.J., Goulet, F.L. and Harrington, T.C., 1986. Death by attrition: a hypothesis for wave mortality of subalpine Abies balsamea. Can. J. For. Res., 16: 591–596. Marks, P.L., 1974. The role of pin cherry (Prunus pensylvanica) in the maintenance of stability in northern hardwood ecosystems. Ecol. Monogr., 44: 73–88. Maser, C. and Trappe, J., 1984. The seen and unseen world of the fallen tree. General Technical Report PNW-164. U.S. Department of Agriculture, Forest Service, Pacific Northwest Forest and Range Experiment Station, 56 pp. Matlack, G.R., Gleeson, S.K. and Good, R.E., 1993. Tree-fall in a mixed oak–pine coastal plain forest: immediate and historical causation. Ecology, 74: 1559–1566. Mattheck, C., Bethge, K. and Albrecht, W., 1995. Failure modes of trees and related failure critera. In: M.P. Grace and

219

J. Grace (Editors), Wind and Trees. Cambridge University Press, Cambridge, England, pp. 195–203. McClure, J.W. and Lee, T.D., 1993. Small-scale disturbance in a northern hardwoods forest: effects on tree species abundance and distribution. Can. J. For. Res., 23: 1347–1360. McConnaughay, K.D.M. and Bazzaz, F.A., 1991. Is physical space a soil resource? Ecology, 72: 94–102. McDonnell, M.J. and Stiles, E.W., 1983. The structural complexity of old field vegetation and the recruitment of bird-dispersed plant species. Oecologia, 56: 109–115. Merrens, E.J. and Peart, D.R., 1992. Effects of hurricane damage on individual growth and stand structure in a hardwood forest in New Hampshire, USA. J. Ecol., 80: 787–795. Milne, R., 1992. Extreme wind speeds over a sitka spruce plantation in Scotland. Agric. For. Meteorol., 61: 39–53. Minckler, L.S. and Woerheide, J.D., 1965. Reproduction of hardwoods ten years after cutting as affected by site and opening size. J. For., February: 103–107. Minckler, L.S., Woerheide, J.D. and Schleshinger, R.C., 1973. Light, soil moisture, and tree reproduction in hardwood forest openings. Research Paper NC-89. U.S.D.A. Forest Service, North Central Forest Experiment Station, St. Paul, Minnesota, 6 pp. Minore, D., 1972. Germination and early growth of coastal tree species on organic seed beds. Research Paper PNW-135. U.S.D.A. Forest Service, Northwest Forest and Range Experiment Staion, Portland, Oregon, 18 pp. Mladenoff, D.J., 1987. Dynamics of nitrogen mineralization and nitrification in hemlock and hardwood tree-fall gaps. Ecology, 68: 1171–1180. Mladenoff, D.J., 1990. The relationship of the soil seed bank and understory vegetation in old-growth northern hardwood–hemlock tree-fall gaps. Can. J. Bot., 68: 2714–2721. Moore, M.R. and Vankat, J.L., 1986. Responses of the herb layer to the gap dynamics of a mature beech–maple forest. Am. Midl. Nat., 115: 336–347. Naka, K., 1982. Community dynamics of evergreen broadleaf forests in southwestern Japan. I. Wind damaged trees and canopy gaps in an evergreen oak forest. Bot. Mag. (Tokyo), 95: 385–399. Nakashizuka, T., 1984. Regeneration process of climax beech (Fagus crenata Blume) forests. IV. Gap formation. Jpn. J. Ecol., 34: 75–85. Nakashizuka, T., 1989. Role of uprooting in composition and dynamics of an old-growth forest in Japan. Ecology, 70: 1273–1278. Nakashizuka, T., 1991. Population dynamics of coniferous and broadleaved trees in a Japanese temperate mixed forest. J. Vegetation Sci., 2: 413–418. Nakashizuka, T., Iida, S., Tanaka, H., Shibata, M., Abe, S., Masaki, T. and Niiyama, K., 1992. Community dynamics of Ogawa Forest Reserve, a species-rich deciduous forest, central Japan. Vegetatio, 103: 105–112. Norton, D.A., 1989. Tree wind-throw and forest soil turnover. Can. J. For. Res., 19: 386–389. O’Cinn´eide, M.S., 1975. Aspect and wind direction as factors in forest stability: the case of Northern Ireland. J. Biogeogr., 2: 137–140. Oliver, C.D. and Larson, B.C., 1990. Forest Stand Dynamics. McGraw Hill, New York, 467 pp. Oliver, C.D. and Stephens, E.P., 1977. Reconstruction of a mixedspecies forest in central New England. Ecology, 58: 562–572.

220 Olszewski, J.L., 1968. Role of uprooted trees in the movements of rodents in forests. Oikos, 19: 99–104. Orwig, D.A. and Abrams, M.D., 1994. Contrasting radial growth and canopy recruitment patterns in Liriodendron tulipifera and Nyssa sylvatica: gap-obligate versus gap-facultative tree species. Can. J. For. Res., 24: 2141–2149. Parsons, W.F.J., Knight, D.H. and Miller, S.L., 1994a. Root gap dynamics in lodgepole pine forest: nitrogen transformations in gaps of different size. Ecol. Appl., 4: 354–362. Parsons, W.F.J., Miller, S.L. and Knight, D.H., 1994b. Rootgap dynamics in a lodgepole pine forest: ectomycorrhizal and nonmycorrhizal fine root activity after experimental gap formation. Can. J. For. Res., 24: 1531–1538. Pauley, E.F. and Clebsch, E.E.C., 1990. Patterns of Abies fraseri regeneration in a Great Smoky Mountains spruce–fir forest. Bull. Torrey Bot. Club, 117: 375–381. Payette, S., Filion, L. and Delwaide, A., 1990. Disturbance regime of a cold temperate forest as deduced from tree-ring patterns: the Tantare Ecological Reserve, Quebec, Canada. Can. J. For. Res., 20: 1228–1241. Peart, D.R., Cogbill, C.V. and Palmiotto, P.A., 1992. Effects of logging history and hurricane damage on canopy structure in a northern hardwood forest. Bull. Torrey Bot. Club, 119: 29–38. Percy, R.W., Chazdon, R.L., Gross, L.J. and Mott, K.A., 1994. Photosynthetic utilization of sunflecks: a temporally patchy resource on a time scale of seconds to minutes. In: M.M. Caldwell and R.W. Pearcy (Editors), Exploitation of Environmental Heterogeneity by Plants: Ecophysiological Processes Above- and Belowground. Academic Press, New York, pp. 175–208. Perkins, R.D., Klein, R.M., Badger, G.J. and Easter, M.J., 1992. Spruce–fir decline and gap dynamics on Camels Hump, Vermont. Can. J. For. Res., 22: 413–422. Peterson, C.J. and Campbell, J.E., 1993. Microsite differences and temporal change in plant communities of tree-fall pits and mounds in an old growth forest. Bull. Torrey Bot. Club, 120: 451–460. Peterson, C.J. and Carson, W.P., 1996. Generalizing forest regeneration models: the dependence of propagule availability on disturbance history and stand size. Can. J. For. Res., 26: 45–52. Peterson, C.J. and Pickett, S.T.A., 1990. Microsite and elevational influences on early forest regeneration after catastrophic windthrow. J. Vegetation Sci., 1(5): 657–662. Peterson, C.J. and Pickett, S.T.A., 1991. Tree-fall and resprouting following catastrophic wind-throw in an old-growth hemlock– hardwoods forest. For. Ecol. Manage., 42: 205–218. Peterson, C.J. and Pickett, S.T.A., 1995. Forest reorganization: a case study in an old-growth forest catastrophic blow-down. Ecology, 76: 763–774. Peterson, C.J. and Rebertus, A.J., 1997. Tornado damage and initial recovery in three adjacent, lowland temperate forests in Missouri. J. Vegetation Sci., 8: 559–564. Peterson, C.J., Carson, W.P., McCarthy, B.C. and Pickett, S.T.A., 1990. Microsite variation and soil dynamics within newly created tree-fall pits and mounds. Oikos, 58: 39–46. Pickett, S.T.A. and White, P.S. (Editors), 1985. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, 472 pp. Pimm, S.L., Davis, G.E., Loope, L., Roman, C.T., Smith III, T.J. and Tilmant, J.T., 1994. Hurricane Andrew. Bioscience, 44: 224–229. Place, I.C.M., 1955. The influence of seed-bed conditions on the

Sara L. WEBB regeneration of spruce and balsam fir, Bulletin 117. Canada Department of Northern Affairs and National Resources, Forestry Branch, Forest Research Division, Ottawa, Canada. Platt, W.J. and Strong, D.M. (Editors), 1989. Special feature: tree-fall gaps and forest dynamics. Ecology, 70: 536–576. Poulson, T.L. and Platt, W.J., 1989. Gap light regimes influence canopy tree diversity. Ecology, 70: 553–555. Poulson, T.L. and Platt, W.J., 1996. Replacement patterns of beech and sugar maple in Warren Woods, Michigan. Ecology, 77: 1234–1254. Purvis, J.C., 1973. Hurricanes. Disaster Preparedness Agency, Colombia, South Carolina. Putz, F.E. and Sharitz, R.R., 1991. Hurricane damage to old-growth forest in Congaree Swamp National Monument, South Carolina, USA. Can. J. For. Res., 21: 1765–1770. Putz, F.E., Coley, P.D., Lu, K., Montalvo, A. and Aiello, A., 1983. Uprooting and snapping of trees: structural determinants and ecological consequences. Can. J. For. Res., 13: 1011–1020. Qinghong, L. and Hytteborn, H., 1991. Gap structure, disturbance, and regeneration in a primeval Picea abies forest. J. Vegetation Sci., 2: 391–402. Raup, H.M., 1981. Forests in the Here and Now: a Collection of Writings of Hugh Miller Raup. Montana Conservation and Experiment Station, Missoula, Montana. Rebertus, A.J. and Veblen, T.T., 1993a. Partial wave formation in old-growth Nothofagus forests on Tierra del Fuego, Argentina. Bull. Torrey Bot. Club, 120: 461–470. Rebertus, A.J. and Veblen, T.T., 1993b. Structure and tree-fall gap dynamics in old-growth Nothofagus forests on Tierra del Fuego, Argentina. J. Vegetation Sci., 4: 641–654. Rebertus, A.J., Kitzberger, T., Veblen, T.T. and Roovers, L.M., 1997. Blow-down history and landscape patterns in the Andes of Tierra del Fuego, Argentina. Ecology, 78: 678–692. Romme, W.H. and Martin, W.H., 1982. Natural disturbance by tree falls in old-growth mixed mesophytic forest: Lilley Cornett Woods, Kentucky. In: R.N. Miller (Editor), Central Hardwood Forest Conference IV Proceedings. University of Kentucky, Lexington, pp. 367–383. Runkle, J.R., 1981. Gap regeneration in some old-growth forests of the eastern United States. Ecology, 62: 1041–1051. Runkle, J.R., 1982. Patterns of disturbance in some old-growth mesic forests of eastern North America. Ecology, 63: 1533–1546. Runkle, J.R., 1984. Development of woody vegetation in tree-fall gaps in a beech–sugar maple forest. Holarct. Ecol., 7: 157–164. Runkle, J.R., 1985. Disturbance regimes in temperate forests. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, Florida, USA, pp. 17–33. Runkle, J.R., 1990a. Eight years change in an old Tsuga canadensis woods affected by beech bark disease. Bull. Torrey Bot. Club, 117: 409–419. Runkle, J.R., 1990b. Gap dynamics in an Ohio USA Acer–Fagus forest and speculations on the geography of disturbance. Can. J. For. Res., 20: 632–641. Runkle, J.R., 1992. Guidelines and sample protocol for sampling forest gaps. General Technical Report PNW-GTR-283, USDA Forest Service, Pacific Northwest Research Station, 44 pp. Runkle, J.R., 1998. Changes in southern Appalachian canopy tree gaps sampled thrice. Ecology, 79: 1768–1780.

DISTURBANCE BY WIND IN TEMPERATE-ZONE FORESTS Runkle, J.R. and Yetter, T.C., 1987. Tree-falls revisited: gap dynamics in the southern Appalachians, USA. Ecology, 68: 417–424. Runkle, J.R., Stewart, G.H. and Veblen, T.T., 1995. Sapling diameter growth in gaps for two Nothofagus species in New Zealand. Ecology, 76: 2107–2117. Schaetzl, R.J., Burns, S.F., Johnson, D.J. and Small, T.W., 1989. Tree uprooting: review of impacts on forest ecology. Vegetatio, 79: 165–176. Schupp, E.W., Howe, H.F., Augspurger, C.K. and Levey, D.J., 1989. Arrival and survival in tropical tree-fall gaps. Ecology, 70: 562–564. Scott, P.A., Hansell, R.I.C. and Erickson, W.R., 1993. Influences of wind and snow on northern tree-line environments at Churchill, Manitoba, Canada. Arctic, 46: 316–323. Seischab, F.K. and Orwig, D., 1991. Catastrophic disturbances in the presettlement forest of western New York, USA. Bull. Torrey Bot. Club, 118: 117–122. Semmel, A., 1993. Soils and wind-throw damage in forests of the Taunus foreland. Geol. Jahrb. Jessen, 121: 141–150 (in German). Sharitz, R.R., Vaitkus, M.R. and Cook, A.E., 1992. Hurricane damage to an old-growth floodplain forest in the southeast. In: J.C. Brissette (Editor), General Technical Report SO-93, Proc. 7th Biennial Southern Silvicultural Res. Conf. United States Department of Agriculture, Forest Service, pp. 203–210. Sharitz, R.R., Allen, B.P., Pauley, E.F. and Pederson, N., 1994. PostHurricane Tree Recruitment in the Congaree Swamp National Monument. Savannah River Ecology Lab, Aiken, South Carolina, 59 pp. Sheffield, R.M. and Thompson, M.T., 1992. Hurricane Hugo: effects on South Carolina’s forest resource, Research Paper SE284. United States Department of Agriculture, Forest Service, Southeastern Forest Experiment Station, 51 pp. Shiyatov, S.G., 1990. The use of dendrochronological methods in determining tree wind-throw time. Lesovedenie, 2: 72–81 (in Russian). Silen, R.R., Olson, D.L. and Weber, J.C., 1993. Genetic variation in susceptibility to wind-throw in young Douglas fir. For. Ecol. Manage., 61: 17–28. Simpson, R.H. and Lawrence, M.B., 1971. Atlantic hurricane frequencies along the U.S. coastline. National Oceanic and Atmospheric Administration, Technical Memo, NWS SR-58. Southern Region Headquarters, National Weather Service, Fort Worth, Texas, 14 pp. Sipe, T.W. and Bazzaz, F.A., 1994. Gap partitioning among maples (Acer) in central New England: shoot architecture and photosynthesis. Ecology, 75: 2318–2332. Sipe, T.W. and Bazzaz, F.A., 1995. Gap partitioning among maples (Acer) in central New England: survival and growth. Ecology, 76: 1587–1602. Spies, T.A. and Franklin, J.F., 1989. Gap characteristics and vegetation response in coniferous forests of the Pacific Northwest. Ecology, 70: 543–545. Spies, T.A., Franklin, J.F. and Klopsch, M., 1990a. Canopy gaps in Douglas-fir forests of the Cascade Mountains. Can. J. For. Res., 20: 649–658. Spies, T.A., Vogt, K.A., Franklin, J.F. and van Pelt, R., 1990b. Aboveand below-ground response of coniferous forest ecosystems to treefall gaps. Research Note. Northwest Environ. J., 6: 435–436. Sprugel, D.G., 1976. Dynamic structure of wave-regenerated Abies

221

balsamea forests in the northeastern United States. J. Ecol., 64: 889–911. Sprugel, D.G., 1984. Density, biomass, productivity, and nutrient cycling changes during stand development in wave-generated balsam fir forests. Ecol. Monogr., 54: 395–412. Spurr, S.H., 1956. Natural restocking of forests following the 1938 hurricane in central New England. Ecology, 37: 443–451. St. Hilaire, L.R. and Leopold, D.J., 1995. Conifer seedling distribution in relation to microsite conditions in a central New York forested minerotrophic peatland. Can. J. For. Res., 25: 261–269. Stark, J.M., 1994. Causes of soil nutrient heterogeneity at different scales. In: M.M. Caldwell and R.W. Pearcy (Editors), Exploitation of Environmental Heterogeneity by Plants: Ecophysiological Processes Above and Belowground. Academic Press, New York, pp. 255–284. Stearns, F.W., 1949. Ninety years change in a northern hardwood forest in Wisconsin. Ecology, 30: 350–358. Stephens, E.P., 1956. Uprooting: a forest process. Soil Sci. Soc. Am. Proc., 20: 113–116. Stewart, G.D. and Rose, A.B., 1990. The significance of life history strategies in the developmental history of mixed beech (Nothofagus) forests, New Zealand. Vegetatio, 87: 101–114. Stewart, G.H., 1986. Forest dynamics and disturbance in a beech–hardwood forest, Fiordland, New Zealand. Vegetatio, 68: 115–126. Stewart, G.H., Rose, A.B. and Veblen, T.T., 1991. Forest development in canopy gaps in old-growth beech (Nothofagus) forests, New Zealand. J. Vegetation Sci., 2: 679–690. Stone, E.L., 1975. Wind-throw influences on spatial heterogeneity in a forest soil. Sonderdruck aus Mitteilungen, Eidgenossische Anstalt f¨ur das forstliche Versuchswesen, 51: 77–87. Strong Jr., D.R., 1977. Epiphyte loads, tree falls, and perennial forest disruption: a mechanism for maintaining higher tree species richness in the tropics without animals. J. Biogeogr., 4: 215–218. Swain, A.M., 1973. A history of fire and vegetation in northeastern Minnesota as recorded in lake sediments. Quat. Res., 3: 383–396. Takahaski, F., 1979. Wave-regenerated Abies vetchii forests in Mt. Shimigare – movement of the “dead tree strips” during 28 years. Shinmrin-Kosoku, 122: 10–17. Taylor, A.H. and Zisheng, Q., 1988. Tree replacement patterns in subalpine Abies and Betula forests, Wolong Natural Reserve, China. Vegetatio, 78: 141–150. Theobald-Ley, S. and Horstmann, K., 1990. The ants (Hymnoptera, Formicidae) of wind-falls and adjacent forest habitats in the Bavarian Forest National Park, Germany. Waldhygiene, 18(3–4): 93–118 (in German). Thompson, J.N., 1980. Tree-falls and colonization patterns of temperate forest herbs. Am. Midl. Nat., 104: 176–184. Toumey, J.W. and Kienholz, R., 1931. Trenched plots under forest canopies. Yale Univ. Sch. For. Bull., 30. Toyooka, H., Ishizuka, M., Osawa, A., Kushima, H., Kanazawa, Y. and Soto, A., 1992. Forest succession over a 34-year period following a catastrophic windstorm in the headwaters of the River Ishikari, Hokkaido: Research Note. Bull. For. Prod. Res. Inst., 363: 59–151 (in Japanese). Trenchard, M.H., 1977. Baseline Climatological Data for the Big Thicket National Preserve. Report to the National Park Service, Big Thicket National Preserve, Beaumont, Texas, 200 pp. Tufty, B., 1987. 1001 Questions Answered about Hurricanes,

222 Tornadoes, and Other Natural Air Disasters. Dover Publications, New York, 381 pp. Valinger, E. and Lundquist, L., 1992. The influence of thinning and nitrogen fertilisation on the frequency of snow and wind-induced stand damage in forests. Scott. For., 46: 311–320. Valinger, E., Lundquist, L. and Brandel, G., 1994. Wind and snow damage in a thinning and fertilisation experiment in Pinus sylvestris. Scand. J. For. Res., 9(2): 129–134. Vasenev, I.I. and Targulyan, V.O., 1994. Wind-throw model of taiga sod-podzolic soil development. Pochvovedenie, 12: 5–16 (in Russian). V´asquez-Yanes, C. and Orozco-Segovia, A., 1994. Signals for seeds to sense and respond to gaps. In: M.M. Caldwell and R.W. Pearcy (Editors), Exploitation of Environmental Heterogeneity by Plants: Ecophysiological Processes Above and Belowground. Academic Press, New York, pp. 209–236. Veblen, T.T., 1985a. Forest development in tree-fall gaps in the temperate rain forests of Chile. Natl. Geogr. Res., 1: 162–183. Veblen, T.T., 1985b. Stand dynamics in Chilean Nothofagus forests. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, pp. 35–51. Veblen, T.T., 1986. Tree-falls and the coexistence of conifers in subalpine forests of the central Rockies. Ecology, 67: 644–649. Veblen, T.T., 1989a. Nothofagus regeneration in tree-fall gaps in northern Patagonia, Chile and Argentina. Can. J. For. Res., 19: 365–371. Veblen, T.T., 1989b. Tree regeneration responses to gaps along a transandean gradient. Ecology, 70: 541–543. Veblen, T.T. and Alaback, P.B., 1995. A comparative review of forest dynamics and disturbance in the temperate rainforests of North and South America. In: R.G. Lawton, P. Alaback and E.R. Fuentes (Editors), High Latitude Rain Forests of West Coasts of the Americas. Springer, New York, pp. 173–213. Veblen, T.T., Donoso, C., Kitzberger, T. and Rebertus, A.J., 1996. Ecology of southern Chilean and Argentinian Nothofagus forests. In: T.T. Veblen, R.S. Hill and J. Read (Editors), The Ecology and Biogeography of Nothofagus Forests. Yale University Press, New Haven, Connecticut, pp. 293–353. Vogt, K.A., Publicover, D.A., Bloomfield, J., Perez, J.M., Vogt, D.J. and Silver, W.L., 1993. Belowground responses as indicators of environmental change. Environ. Exp. Bot., 33: 189–205. Vogt, K.A., Vogt, D.J., Asbjornsen, H. and Dahlgren, R.A., 1995. Roots, nutrients, and their relationship to spatial patterns. Plant Soil, 168–169: 113–123. Watanabe, R., Nakashizuka, T., Honma, S., Hara, M. and Yoda, S., 1985. Studies of the Fagus crenata forest in Kayanodaira Japan. 2. Fallen trees by wind of the typhoon number 10 in 1982. Bull. Inst. Nat. Educ. Shiga Heights, 22: 15–18 (in Japanese). Watt, A.S., 1947. Pattern and process in the plant community. J. Ecol., 35: 1–265. Webb, S.L., 1986. Potential role of Passenger Pigeons and other vertebrates in rapid Holocene migrations of nut trees. Quat. Res., 26: 367–375. Webb, S.L., 1987. Beech range extension and vegetation history: pollen stratigraphy of two Wisconsin lakes. Ecology, 68: 1993–2005.

Sara L. WEBB Webb, S.L., 1988. Windstorm damage and microsite colonization in two Minnesota forests. Can. J. For. Res., 18: 1186–1195. Webb, S.L., 1989. Contrasting windstorm consequences in two forests, Itasca State Park, Minnesota. Ecology, 70: 1167–1180. Webb, S.L. and Willson, M.F., 1985. Spatial heterogeneity in postdispersal predation on Prunus and Uvularia seeds. Oecologia (Berlin), 67: 150–153. Whelan, C.J., Willson, M.F., Tuma, C.A. and Souza-Pinto, I., 1991. Spatial and temporal patterns of post-dispersal seed predation. Can. J. Bot., 69: 428–436. White, P.S., 1979. Pattern, process, and natural disturbance in vegetation. Bot. Rev., 45: 229–299. White, P.S., 1987. Natural disturbance, patch dynamics, and landscape pattern in natural areas. Nat. Areas J., 7: 14–22. White, P.S., MacKenzie, M.D. and Busing, R.T., 1985a. A critique on overstory–understory comparisons based on transition probability analysis of an old growth spruce–fir stand in the Appalachians. Vegetatio, 64: 37–45. White, P.S., MacKenzie, M.D. and Busing, R.T., 1985b. Natural disturbance and gap phase dynamics in southern Appalachian spruce–fir forests. Can. J. For. Res., 15: 233–240. Whitney, G.G., 1986. Relation of Michigan’s presettlement pine forests to substrate and disturbance history. Ecology, 67: 1548–1559. Wilczynski, C.J. and Pickett, S.T.A., 1993. Fine root biomass within experimental canopy gaps: evidence for a below-ground gap. J. Vegetation Sci., 4: 571–574. Williamson, G.B., 1975. Pattern and seral composition in an oldgrowth beech–maple forest. Ecology, 56: 727–731. Wilson, B.F. and Archer, R.R., 1979. Tree design: some biological solutions to mechanical problems. BioScience, 29: 293–298. Wooldridge, G., Musselman, R. and Massman, W., 1995. Windthrow and airflow in a subalpine forest. In: M.P. Grace and J. Grace (Editors), Wind and Trees. Cambridge University Press, Cambridge, England, pp. 358–375. Worrall, J.J. and Harrington, T.C., 1988. Etiology of canopy gaps in spruce–fir forests, Crawford Notch, New Hampshire, USA. Can. J. For. Res., 18: 1463–1469. Wright, H.E. and Heinselman, M.L., 1973. The ecological role of fire in natural conifer forests of western and northern North America. Quat. Res., 3: 319–328. Yamamoto, S.-I., 1989. Gap dynamics in climax Fagus crenata forests. Bot. Mag. (Tokyo), 102(1065): 93–114. Yamamoto, S.-I., 1992. Gap characteristics and gap regeneration in primary evergreen broad-leaved forests of western Japan. Bot. Mag. (Tokyo), 105(1077): 29–45. Yamamoto, S.-I., 1993. Gap characteristics and gap regeneration in a subalpine coniferous forest on Mt. Ontake, central Honshu, Japan. Ecol. Res., 8(3): 277–285. Yamamoto, S.-I., 1995. Natural disturbance and tree species coexistence in an old-growth beech–dwarf bamboo forest, southwestern Japan. J. Vegetation Sci., 6: 875–886. Zimmerman, J.K., Pulliam, W.M., Lodge, D.J., Qui˜nones-Orfila, V., Fletcher, N., Guzm´an-Grajales, S., Parrotta, J.A., Ashbury, C.E., Walker, L.R. and Waide, R.B., 1995. Nitrogen immobilization, decomposing woody debris, and the recovery of tropical wet forest from hurricane damage. Oikos, 72: 314–322.

Chapter 8

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

INTRODUCTION

Disturbance has been defined in various ways (e.g., Sousa, 1984; Rykiel, 1985; van Andel and van den Bergh, 1987; Pickett et al., 1989), in part because disturbance theory must deal with a wide variety of phenomena under that name (Pickett et al., 1989). A useful definition of disturbance for forests is “a relatively discrete event causing a change in the physical structure of the environment” (Clark, 1990). In forests, this change in physical structure refers primarily to damage and removal of aboveground biomass (Grime, 1979), and removal of the litter layer and mixing of surface soil layers (e.g., Putz, 1983). Changes in the physical structure of the environment (Clark, 1990) are concomitant with changes in environmental variables below the main canopy (e.g., soil resources, light quantity and quality, and temperature: Chazdon and Fetcher, 1984; Bellingham et al., 1996). For individual disturbance events, the size of the canopy opening is often correlated with most of the direct and indirect changes in environmental variables and resource levels (Brokaw, 1985b). Even small gaps (Canham and Marks, 1985) provide opportunities for regeneration and adult growth (Sousa, 1984; Oliver and Larson, 1990). Consistent with the above definition, in this chapter we focus on forest-canopy disturbances that occur as discrete events in time. We are primarily concerned with those relatively small canopy disturbances (background canopy gap disturbance) caused by a variety of agents and larger disturbances caused by major windstorms (catastrophic wind disturbance). A modification of Clark’s (1990) definition of the scale of wind-generated disturbances, which represents a continuum of disturbance (Everham and Brokaw, 1996; Lugo and Scatena, 1996), is useful because it

can be used to divide disturbances into those that we will consider to be background canopy disturbances (103 m2 : Table 8.1). We only indirectly deal with major disturbance events (e.g., fire, floods, river meanders, landslides, shifting agriculture) that both open the canopy and more substantially disturb the soil, understory, and litter layer (see Uhl, 1982a; Lugo et al., 1983; Johns, 1986; Oliver and Larson, 1990; Lugo and Scatena, 1996). Some types of canopy disturbances have been described by the concept of “patch dynamics” (White and Pickett, 1985), in which the disturbance effects are relatively discrete spatially, and where there is a “shifting mosaic” of patches of different ages. After disturbance, the patch changes through time and regains characteristics of the older patches. This concept applies well to canopy gaps formed by the fall of one to a few trees and blow-downs where many trees are felled together (see Sousa, 1984; White and Pickett, 1985; Whitmore, 1989; Clark, 1990). The concept is problematic in that gap edges are not always Table 8.1 Scale of wind-generated disturbances in neotropical moist forests 1 Type of disturbance

Approximate area of effect (m2 )

Hurricanes/typhoons

105 –107

Blow-downs/wind-throws

103 –105

Single tree-falls

102 –103

Branch falls

101 –102

1

223

Modified from Clark (1990).

224

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

obvious, openings do not always extend through all canopy layers (Lieberman et al., 1989), small gaps may be closed from the side (Runkle and Yetter, 1987), and patches may also be internally heterogeneous (White and Pickett, 1985). In contrast to discrete gaps, hurricane disturbance may often, though not always, be better described as “diffuse” (Pickett and White, 1985), particularly where the canopy is thinned and no discrete patches are formed (Everham and Brokaw, 1996). In this chapter we focus on background canopy disturbances that create canopy gaps extending to the understory [see Brokaw (1982a) and Popma et al. (1988) for definitions of canopy gaps] and catastrophic wind disturbances of both discrete and diffuse nature. One can view disturbances as having causes and effects (Rykiel, 1985) to which ecological systems respond (van Andel and van den Bergh, 1987). Three largely similar dichotomous groupings of causes of disturbance have been used (but see Lugo and Scatena, 1996): endogenous versus exogenous (White and Pickett, 1985), internal versus external (Vooren, 1986; van der Meer and Bongers, 1996b) and biotic versus abiotic. Endogenous causes (e.g., disease, competition, epiphyte and vine loading of canopies, and herbivory) originate from within the biotic community while exogenous causes (e.g., drought, wind, lightning, and rainfall) originate from without (White and Pickett, 1985). Endogenously caused disturbances may largely occur through the gradual fall of trees that died standing, hereafter referred to as standing mortality (Vooren, 1986; Krasny and Whitmore, 1992; van der Meer and Bongers, 1996b). Exogenously caused disturbances can occur gradually (e.g., from standing mortality owing to drought) or suddenly (e.g., from wind-throw). There is interaction between these two groups of causes (White and Pickett, 1985). For instance, though major wind disturbance obviously has a largely exogenous component, trees weakened by disease are more prone to damage (Putz and Sharitz, 1991; see also Everham and Brokaw, 1996). Wind is an important cause of canopy disturbance, and probably the primary cause of canopy disturbance in forests where large tropical storms or other strong, though localized, wind storms are common occurrences (see Lugo et al., 1983; Shaw, 1983; Brokaw, 1985b; Walker et al., 1991; Lugo and Scatena, 1996) and where other causes of major disturbance are not operative (Leighton and Wirawan, 1986; Foster, 1990). Vegetative response to disturbance depends on characteristics of the disturbed site (e.g., severity of damage), species availability

(e.g., propagule availability and survival through the event), and post-disturbance species performance such as growth rates, sprouting ability, and survival (Pickett et al., 1987; Pickett and McDonnell, 1989; Lugo and Scatena, 1996). Our objectives in this chapter are to build on the work of Everham and Brokaw (1996) and other reviewers by describing the causes and effects of background canopy disturbance (Denslow, 1980; Hartshorn, 1980; Brokaw, 1985a,b) and catastrophic wind disturbance (Lugo et al., 1983; Glitzenstein and Harcombe, 1988; Brokaw and Walker, 1991; Tanner et al., 1991; Lugo and Scatena, 1996; Zimmerman et al., 1996) in tropical and subtropical forests, and comparing vegetative response to catastrophic wind events with response to background canopy disturbance. Finally, we describe ecosystem response to catastrophic wind disturbance.

CAUSES AND EFFECTS OF CANOPY DISTURBANCE

Wind storms in the tropics – distribution and patterns Within the tropics, winds that are not associated with catastrophic events usually impact forests during periods of rainfall that occur chiefly in areas of atmospheric disturbance and persist for days, moving irregularly around the landscape, but often toward the west. The distribution, size, and intensity of tropical rainstorms varies temporally and spatially (Schwerdtfeger, 1976; Lauer, 1983), but storms that are potentially strong enough to generate winds causing canopy disturbances can generally be organized into three types: (1) The Intertropical Convergence Zone produces winds, that are rarely strong, over the oceans and widely over continents during rainy seasons. Forest disturbances associated with these types of windstorms are most often associated with individual or small groups of rain squalls. (2) Northern cold fronts (nortes) in certain longitudes (e.g., in the northern Caribbean) penetrate into the tropics and become part of the trade-wind system. Strong winds and showers or thunderstorms are commonly associated with these systems. (3) Larger-scale easterly-winds produce squall-line systems. These occur in West Africa, across the tropical Atlantic to the Caribbean, and infrequently in the central and western Pacific. Wind velocities

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

of more than 20 m s−1 are associated with these systems. These three types of wind disturbances occur throughout the tropics, but their occurrence and intensity are difficult to predict and their impacts are poorly known. Richards (1996) suggested that all tropical areas are subject to wind disturbances before and during convectional rainstorms and thunderstorms, and he cited several examples to demonstrate the range of impacts that wind disturbances can have. In Sarawak, winds associated with squalls created canopy gaps as small as 0.04 ha (equivalent to felling a single tree) but they can also cause catastrophic damage; areas as large as 80 ha, for example, were damaged by wind in swamp forests dominated by Shorea albida. Other authors have also documented the importance of wind as an agent of background canopy disturbance (e.g., Whitmore, 1989). Winds that cause background canopy disturbance can also create catastrophic disturbances that would be difficult to predict under any circumstances. Uhl et al. (1988a), for example, found that local rain and windstorms mostly accounted for the background rate of canopy disturbance in the S˜ao Carlos terra firme forest in Venezuela, but that windstorms also “occasionally knock down whole sections of forest”. Kellman and Tackaberry (1993) found that “extreme winds of atypical orientation” caused trees to fall in a direction that was not uniformly distributed. Numerous accounts of blow-down and wind-throw disturbances can be found in the literature, but almost all of them are qualitative descriptions (e.g., Richards, 1996; Whitmore, 1989; citations in Clark, 1990; Nelson et al., 1994). Bruenig (1989) used aerial photographs taken in different years to demonstrate the importance of windstorms in creating wind-throws in forested wetlands in Borneo. Bruenig found that the rates of single-tree gaps varied among forest types from 0.2–3.0% of the area per year, but that the rate of disturbance in one forest type appeared to be about 1% for single tree-fall gaps and for openings created by wind-throws. Bruenig suggested that forests with taller trees and trees with a high height/diameter ratio are more susceptible to lightning and wind-throw. Forested wetlands appear to be particularly susceptible to blowdown and wind-throw disturbances because the trees have very shallow roots. Clearly, there have been too few quantitative studies to characterize the importance of blow-downs and wind-throws in tropical forests, even though their impacts on forest structure and

225

species composition are long-lived (Hubbell and Foster, 1986). Most catastrophic wind disturbances are associated with tropical cyclones, the general term for hurricanes and typhoons, which are the most destructive types of windstorms in the tropics and subtropics; their damaging effects can extend far into the temperate zone (Encyclopædia Britannica, Inc., 1992; Boose et al., 1994). They are intense storms (as large as 150–250 km across) with maximum wind speeds that are, at least, 32.7 m s−1 . Heavy rain is also usually associated with tropical cyclones. The number of tropical cyclones per year, world-wide, varies between approximately 30 and 100, and they occur primarily near Southeast Asia, the Caribbean, and adjacent waters (Vega and Binkley, 1993), and in the southwest Pacific and Australian waters (Fig. 8.1). In addition to disturbing natural ecosystems, tropical cyclones also cause enormous economic destruction and human suffering (Encyclopædia Britannica, Inc., 1992). Background canopy disturbance Overview Canopy disturbance is now viewed as integral to an understanding of tropical forest ecology (Hartshorn, 1978; Whitmore, 1975), and much of the data cited in this chapter were obtained as part of efforts to characterize what we have defined as the “background” pattern of disturbance in tropical forests, disturbances that are usually caused by branch-falls, standing dead trees, and the fall of one to several adjacent trees. The average rate of gap formation in humid lowland tropical forests has been reported to be 1 ha−1 yr−1 with a range of 0.7–2.6, opening ~1–2% of the forest per year with a range of 0.5–3.6% yr−1 (Denslow, 1987; Swaine et al., 1987; Whitmore, 1989; Hartshorn, 1990; Jans et al., 1993; Yavitt et al., 1995; Lugo and Scatena, 1996). Average gap size, shortly after gap formation, ranges from 54 to 120 m2 (see review in Jans et al., 1993). Small gaps are much more common than large gaps (Lawton and Putz, 1988; Uhl et al., 1988a; Chandrashekara and Ramakrishnan, 1994). In lowland moist and wet forests, the largest gaps range considerably in size. Brokaw (1982a,b) measured gap sizes of 232 m2 and 342 m2 in an old forest on Barro Colorado Island (BCI), Panama. Yavitt et al. (1995) measured a gap size of 604 m2 in a young forest there. Sanford et al. (1986) measured gaps as large as 781 m2 at La Selva, Costa Rica. Gap turnover is a function of

226

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

Fig. 8.1. Major tracks (arrows) and frequency (shaded areas) of hurricanes and typhoons. Source: Encyclopædia Britannica, Inc. (1992).

the size of gaps and their rate of occurrence and can be thought of as the number of years it would take to cover 100% of the forest in gaps (assuming there is no gap overlap). Estimated gap turnover rates range from 60 to 145 years in humid lowland forests (see review in Jans et al., 1993), and rates of gap formation can vary considerably from year to year (Mart´ınez-Ramos et al., 1988; Lawton and Putz, 1988; Dickinson et al., 1999). The variation is even greater when rare and large disturbance events are included in the estimates of turnover rates (Brokaw, 1985b; Leighton and Wirawan, 1986; Whigham et al., 1990). Variation among forests Similarities among forests in gap size, frequency, and turnover have been emphasized (Lawton and Putz, 1988; Hartshorn, 1990; Kapos et al., 1990; Jans et al., 1993; Yavitt et al., 1995). The similarities, however, are likely owing to a bias towards studying humid lowland forests in the neo-tropics (P. Hall, pers. commun., 1997) not obviously affected by major periodic disturbance (see summary data above). There are fewer data on background canopy disturbance rates for forests subject to major periodic disturbances, dry tropical forests, montane forests, wetland forests, and paleotropical forests, but differences among forest types are to be expected. Several examples follow. In a wet forest in Puerto Rico periodically disturbed by periodic catastrophic wind events, mean gap size (76 m2 ) and gap formation rate (0.8 gaps ha−1 yr−1 ) are below the midrange values for humid forests not impacted by periodic

major disturbance (Scatena and Lugo, 1995). The gap formation rate in a semideciduous forest in the Mexican Yucatan (1300 mm annual rainfall) periodically hit by major disturbance is exceedingly low at 0.2 gaps ha−1 , resulting in 0.07% of the forest being disturbed in an average year (Dickinson et al., 1999). The average gap size in the Yucatan forest (180 m2 ; Dickinson et al., 1999), a montane forest site (>135 m2 ; Lawton and Putz, 1988), and in a West African forest with a long dry season (244 m2 ) studied by Jans et al. (1993). Jans et al. also reported a correspondingly long gap turnover time (244 years) for that site. Compared with humid lowland neo-tropical forests, gap turnover times also were long (154–375 yr) in two equatorial Southeast Asian forests (Jengka Forest Reserve, Peninsular Malaysia: Poore, 1968; Samarinda, East Kalimantan, Borneo: Riswan et al., 1985). Long gap turnover times can be partly explained by small gap size in the Jengka (Whitmore, 1975). East Kalimantan is periodically affected by severe droughts, which may explain long turnover times there (Leighton and Wirawan, 1986). In the following section and in Table 8.2, we describe what we expect to be the main sources of variation among sites in background canopy disturbance (also see Brokaw, 1985b). These include: wind regime, forest

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

227

Table 8.2 Predictions about variation in background canopy disturbance regimes #

Prediction

1

Dry, high-altitude forests, and extremely nutrient-poor forests should have the lowest average gap size owing to small crown sizes.

2

Forests dominated by tall trees with thin crowns will have linear gaps, with less effect on understory light levels.

3

Forests subject to natural and anthropogenic major disturbance (fires, blow-downs, hurricanes, agricultural clearing, logging, landslides, flooding events, and river meanders) will have smaller gaps and a lower rate of gap formation than forests that do not experience such events or that have not experienced such an event in >200 years.

4

Shallow rooting of trees on fertile, extremely nutrient-poor, or waterlogged soils will increase the rate of tree-fall and prevalence of uprooting (and soil disturbance), while seasonally dry forests and forests on moderately infertile and well-drained soils will exhibit low rates of tree-fall and more stem-snapping.

5

Forests on slopes will have a higher rate of tree-fall than forests on flat terrain only if soils and parent material are unstable and rooting is shallow.

6

Forests with high abundance of woody vines should exhibit larger gap sizes and more frequent gaps.

7

High rates of standing mortality will result in a smaller average gap size and lower overall rates of gap formation.

8

Except where continuous strong winds have large effects on structure, forests with higher frequencies of strong, gusty winds (e.g., where rainstorms are more frequent) should have higher rates of gap formation than forests where winds are less gusty.

9

Dry tropical forests should have smaller gaps and lower rates of gap formation than humid forests, owing to smaller crown sizes, a more stable tree architecture, higher rates of standing mortality, periodic fire (if applicable), and a shorter period in which most gap causal factors operate.

structure (as affected by site conditions, biogeography and species composition, and prior catastrophic disturbances), tree anchorage, standing mortality, and vines and epiphytes. The effects of wind are often mediated by the other factors. We also compare dry and wet tropical forests, which appear to differ in several of the above factors, and we predict that they will prove to vary considerably in background canopy disturbance. Wind regime Forests subject to more frequent strong and gusty winds should have more canopy disturbances than forests where winds are calm (Grace, 1977; Richards, 1996; see also Table 8.2, Prediction 8). In terms of damage to trees, high mean wind speeds, strong gusts, and abrupt changes in wind direction are probably most likely to cause damage (Gloyne, 1968; Grace, 1977). Within a given forest, sites that are most exposed to gusty winds have higher tree-fall rates (see review in Brokaw, 1985b). Strong gusts (from 780 to over 100 km hr−1 ) have been reported to occur on multiple occasions in a given year in several sites (Bultot and Griffiths, 1972; Lawton and Putz, 1988; Uhl et al., 1988a). Whitmore (1975) suggested that variation among forests in the frequency of convectional storms and squall lines would explain variation among forests in disturbance. Squall lines along the Malaysian Peninsula are reported to be the

most important cause of large gaps (Whitmore, 1975). The generally calmer winds of equatorial Southeast Asia have been hypothetically linked to high rates of standing mortality and low overall gap formation rates in forests there (P. Hall, pers. commun., 1997). In forests with distinct dry seasons, a peak in gap formation occurs during the middle of the wet season [see review in Brokaw (1985b), and also Matelson et al. (1995)], although not all wet season gaps can be ascribed to wind (see below). Ninety percent of the gaps sampled at La Selva in Costa Rica were formed during the 6 wettest months (Hartshorn, 1989), and in a cloud forest in Monteverde (Costa Rica) 70% of the gaps formed in one year were caused during one severe, but common type of storm (see Lawton and Putz, 1988). Dry tropical forests are also likely to have more wind-induced damage during the wet season, as most of the strong wind gusts occur during the wet season (Bultot and Griffiths, 1972). In addition to causing individual tree-fall gaps, wind damage can enlarge existing gaps by toppling trees at the gap edge (Bruenig, 1989; Whitmore, 1984), or cause clustering of gaps owing to increased turbulence around existing gaps where the canopy is uneven (Poorter et al., 1994). Wind is often the coup de grace to a tree made vulnerable by such things as disease, poor rooting conditions owing to waterlogged soils, and rain-loading

228

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

(Gloyne, 1968; Whitmore, 1975; Brokaw, 1985b; Richards, 1996). Though rain-loading of the canopy can cause large, live trees to fall in calm, rainy weather (Strong, 1977), standing mortality cannot generally be directly attributable to wind, although extreme winds may lead to eventual standing mortality (Shaw, 1983; Whigham et al., 1991). Where gap formation by standing mortality is minimal, we expect that the largest proportion of gaps would be caused by uprooting, snapping, and live limb-fall owing to wind. Trees fall in the direction of the prevailing winds in a subalpine temperate forest, strong evidence of the importance of wind in those environments (Wooldridge et al., 1995). Wind speeds vary in several ways, most importantly perhaps with geographic setting and the frequency of storms (Whitmore, 1975). Average coastal wind speeds are generally higher than inland wind speeds at the same altitude (Grace, 1977). Windstorms may often be more common somewhat inland from the coast owing to orographic rain storms (Dale, 1959; Whitmore, 1975). Average wind speed increases with altitude (Grace, 1977), although canopy gaps do not seem to become more frequent at higher altitudes (Lawton and Putz, 1988; Matelson et al., 1995). Trees whose physiognomy is shaped to a large degree by strong and continuous winds, such as trees in exposed cloud forests, are not as prone to wind damage as trees that are not so conditioned (Lawton and Putz, 1988; Brokaw, 1985b; Richards, 1996). This is likely in large part owing to the increased turbulence associated with rough rather than smooth canopy topographies (Gloyne, 1968; Grace, 1977). Topography significantly affects wind speed and direction above canopies (Grace, 1977). When wind speeds are high, forests on the lee side of ridges can experience severe turbulence (Grace, 1977). Wind is responsible for the formation of disturbances over a wide range of scales, from single tree-falls (e.g., Denslow, 1987) to blow-downs (Dunn et al., 1983; Uhl et al., 1988a; Clark, 1990; Nelson et al., 1994), and swaths of hurricane-disturbed forest (Everham and Brokaw, 1996; Richards, 1996). In forests that are not typically affected by major wind disturbance, the incidence of multi-tree wind-throws may be the most important factor in disturbance regimes and resulting patterns of species abundance differentiating one forest from another (Denslow, 1987). Forest structure Effects of site conditions: Variation among forests in

above-ground structure (shoot characteristics, including crown size, average tree height, and allometry) should be reflected in the size distributions of tree-fall gaps, their effect on light conditions at ground level, and on the frequency of gap creation. A positive relationship between tree size (measured as trunk diameter) and gap size [i.e., area of the hole in the canopy that extends to ground level; see Brokaw (1982a) and Popma et al. (1988)] has been documented (Brokaw, 1982a; Lawton and Putz, 1988; Clark and Clark, 1996). Small gap sizes in montane and drier forests are a reflection of this (see p. 226 above). It is not clear whether bole height, bole thickness, or crown size most influences gap size. However, the findings that the height at which the bole gave way has little effect on gap size (Lawton and Putz, 1988) and that gaps formed by uprooting trees are no larger than gaps formed by snapped trees (Jans et al., 1993) indicate that crown size is most important (Richards, 1996; see also Table 8.2, Prediction 1). All else being equal, large crowns are on tall thick boles (King, 1991) and are poised to do maximum damage to surrounding trees when they fall. Tree architecture varies with wind regime, rainfall, soils, topography, altitude and latitude [see review of hypotheses regarding physiological limits on tree size by Stevens and Perkins (1992)]. Rainfall patterns are primary determinants of above-ground forest structure (Beard, 1955; Holdridge et al., 1971; Ellenberg, 1979; Lieberman et al., 1996). Forests with the highest densities of emergents that have large crowns should occur in lowland tropical wet and moist forests on well-drained soils (Holdridge et al., 1971; Grubb, 1989), especially on lower slopes or in areas with little topographic relief (Ashton and Hall, 1992). Forest height in a wet Puerto Rican forest is highest on better drained areas and lowest where soils are often waterlogged (Lugo et al., 1995). Kira (1978) found that the tallest forests at a number of Southeast Asian sites occur where rainfall was around 2000 mm yr−1 and was evenly distributed throughout the year. Tree height drops both in wetter sites and in sites with decreasing amounts and increasing seasonality of rainfall. In concert, the height to diameter ratio decreases in drier and more seasonal forests, that is, a tree of 1 m dbh becomes shorter (Kira, 1978). Thus, crown size may not decrease as quickly as tree height with decreasing rainfall. High stand turnover rates (as at La Selva) should lead to low densities of very large trees (Lieberman et al., 1985; Clark, 1996). Stand and basal area turnover appear to increase generally in

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

productive sites (Phillips et al., 1994), consequently, such sites may have more, but smaller, gaps. Forests on extremely nutrient-poor soils should also have small tree-fall gaps as the trees are typically of low stature, although the effects of low nutrient status are often confounded with periodic moisture deficits (Whitmore, 1975; Grubb, 1989). Kapos et al. (1990) found larger gaps, and larger trees, on a relatively nutrient-rich site than on a nutrient-poor site. Periodic drought in eastern Borneo killed the largest and tallest trees, leading to a broken emergent stratum (Leighton and Wirawan, 1986; Richards, 1996; see below). Large trees on ridges are particularly vulnerable to drought-related mortality, since the soils dry quickly (Leighton and Wirawan, 1986; Ashton and Hall, 1992; Richards, 1996). This effect may have contributed to lower rates of gap formation and lower gap sizes on ridges compared with slopes in West African seasonal forests (Poorter et al., 1994). Rooting was impeded on hill crests (Poorter et al., 1994), perhaps compounding an effect of periodic drought. Tree height and crown size decrease with increasing altitude (Holdridge et al., 1971; Whitmore, 1975; Ellenberg, 1979; Lieberman et al., 1996). Consistently stronger winds at higher altitudes (Grace, 1977) lead to shorter trees with thicker boles, primarily due to the mechanical effects of continuous winds on tree growth (Lawton and Dryer, 1980; Telewski, 1995). Also, there appears to be a temperature effect on forest height along altitude gradients (Pendry and Proctor, 1996). A hypothesis that there should exist a pattern of increasing crown width from the temperate to the tropical zone was proposed by Terborgh (1985), but King (1991) did not find such a pattern. Forests dominated by tall trees with small crowns produce small gaps that are often linear in shape, providing little opportunity for regeneration of lightdemanding species (Putz and Appanah, 1987; see also Table 8.2, Prediction 2). The distribution of forests in which tall trees with small crowns predominate is unclear, but examples include certain zones in Sarawak peat swamps (Whitmore, 1975) and Pasoh Forest Reserve in Peninsular Malaysia (Putz and Appanah, 1987). Both of these sites are nutrient-poor, and trees also experience periodic moisture stress (Whitmore, 1975; Putz and Appanah, 1987). Canopy trees in the tallest forests may generally show a reduction in crown expansion as compared with canopy trees in somewhat shorter forests (King, 1996).

229

Biogeographic and species-composition effects: Biogeographic effects on forest structure may occur; they should not be confused with site effects, and the effects of past disturbance on species composition. Variation in species composition within and among biogeographic regions may affect disturbance regimes when different species have different stem architecture (Beard, 1945a; Wadsworth and Englerth, 1959; Jans et al., 1993), wood properties (Putz et al., 1983), rooting patterns (Everham and Brokaw, 1996), and modes of death (Brokaw, 1985b). The high density of very tall, emergent trees in some dipterocarpdominated forests may be related to unique aspects of the reproductive biology and ecology of this Southeast Asian family (Ashton, 1988). Forests in Borneo often have a towering emergent canopy of dipterocarps, and on certain sites 80–100% of all individuals in the upper canopy are dipterocarps. Forests east of Wallace’s Line (where dipterocarps are less species-rich), in Africa and in the Neotropics typically have a sparse cover of emergent trees above the main canopy (Ashton, 1988). Higher densities of large emergents may lead to larger gaps when emergents fall. High rates of standing mortality in certain equatorial forests in Southeast Asia (Table 8.3) may also have a biogeographic component (Hall, 1991). Effects of prior catastrophic disturbances: While most studies on background canopy disturbance have been done in forests not subjected to major disturbance, major disturbances have large and long-term effects on canopy structure (Johns, 1986; Foster, 1988; Saldarriaga et al., 1988) and subsequent background canopy gap disturbance (Spies and Franklin, 1989; Lorimer, 1989). Any disturbance that results in high mortality of large canopy trees would be expected to lead to a period of lower rates of formation of large tree-fall gaps and smaller gap sizes, as the large dead canopy trees are replaced by individuals from smaller size classes (Hartshorn, 1978; Brokaw, 1982b; Denslow and Hartshorn, 1994; Dahir and Lorimer, 1996; see also Table 8.2, Prediction 3). Such disturbances (see Johns, 1986) include blowdowns (Whitmore, 1975), fire (Leighton and Wirawan, 1986), landslides (Guariguata, 1990), agricultural clearing (Saldarriaga et al., 1988), logging, flooding events (Mori and Becker, 1991; Gullison et al., 1996), severe droughts (Leighton and Wirawan, 1986; Woods, 1989) and river meanders (Foster, 1990). As the frequency of large-scale disturbances increases, tree-fall gaps may

230

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

Table 8.3 Standing mortality of tropical forest trees Region

Climate 1

Standing mortality 2 (%)

Soil fertility 3

Source

Bako, Sarawak, NW Borneo

4167 [0]

84

low

Hall (1991)

Mersing, Sarawak, NW Borneo

3905 [0]

68

high

Hall (1991)

Lambir, Sarawak, NW Borneo

2874 [0]

65

moderate

Hall (1991)

Lower Montane Venezuela

1650 [2]

64

−

Carey et al. (1994) 4

Lowland Venezuela

2725 [1–2]

60

−

Carey et al. (1994) 5

Pasoh, peninsular Malaysia

1900 [1]

45

low

Putz and Appanah (1987)

Paracou, French Guiana

3000 [3]

44

low

Durrieu de Madron (1994)

La Selva, Costa Rica

4000 [0]

40

high

Lieberman and Lieberman (1987)

Amazonas, Brazil

2186 [4]

26

low

Rankin de Merona et al. (1990)

Barro Colorado Island, Panama

2656 [3]

14

high

Putz and Milton (1982)

Amazonas, Venezuela

3500 [0]

¾ 10

low

Uhl (1982b)

1

Annual rainfall in mm. The number of consecutive months with precipitation below 100 mm is shown in brackets. All stems that died after excluding stems for which the cause of mortality was unknown. Consequently, estimates in the Table differ somewhat from the original citation. 3 Soil fertility, often estimated by the current authors. 4 Data are averaged over 9 plots in the State of Merida. 5 Data are averaged over 8 plots in the States of Merida, Delta Amacuro, and Bol´ıvar. 2

occur less and less frequently, and eventually become inconsequential to tree population dynamics (Lorimer, 1989). In the Luquillo Experimental Forest in Puerto Rico, where hurricane return times are about 60 years, periodic hurricanes open more of the canopy than do background canopy gaps, in all topographic positions apart from the often waterlogged riparian valleys (Lugo and Scatena, 1996). Extreme cases of a reduction in the importance of background canopy gaps may occur in the “hurricane scrubs” of North Queensland, Australia (Webb, 1958) and in forests in the Caribbean that are damaged frequently and severely (Beard, 1945b). Even minor hurricanes would be expected to reduce tree-fall rates in the years after the hurricane event, as wind disturbance would fell trees that had rot in their trunk or were poorly rooted in the substrate (Putz and Sharitz, 1991). Major canopy disturbances also tend to reduce canopy height owing to the felling of the tallest trees during severe wind events (Wadsworth and Englerth, 1959; Foster, 1988). In forests subjected to frequent major canopy damage, tree heights should remain lower as a result of bole thickening and reduced height growth in response to increased lateral illumination (see Holbrook and Putz, 1989). These effects on tree size should tend to reduce gap size when trees fall. Recovery following complete canopy removal may

provide a maximum bound on the time to recovery of forest canopy structure. The gap disturbance regime (frequency and size) of seasonally dry forests receiving around 2600 mm of rainfall on Barro Colorado Island, Panama, following abandonment of agriculture, was similar to nearby primary forest after 70–80 years (Yavitt et al., 1995). In upland, nutrient-poor, wetforest sites in the upper Amazon basin, stems 40– 60 cm dbh were prevalent in a stand 60 years old, but after clearing it took 190 years for these forests to regain the previous basal area and biomass (Saldarriaga et al., 1988). After approximately 200 years, forests which formed on the trailing edge of river meanders were still undergoing structural changes in species composition towards larger emergents, which would continue to impact background canopy disturbance (Foster, 1990). Recovery following catastrophic wind disturbance, measured in a variety of ways, may take from decades to centuries, depending in part on severity of damage (Everham and Brokaw, 1996). Catastrophic wind events in Puerto Rico recur at a frequency less than the time to forest recovery (Lugo and Scatena, 1996). Tree anchorage Deeply rooted species (Touliatos and Roth, 1971) and forests in which most individuals are deeply rooted

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

(Brokaw, 1985b) or otherwise firmly anchored by root grafting (Basnet et al., 1992) or root penetration into soil parent material on shallow soils (Wadsworth and Englerth, 1959; Basnet et al. 1992) suffer less damage in general and less uprooting in particular from strong winds (Everham and Brokaw, 1996; see also Table 8.2, Prediction 4). Deep rooting is impeded in waterlogged soils (Hartshorn, 1978) or where impermeable soil horizons occur (Richards, 1996). Rooting depth may be superficial in soils where deeper layers are exceedingly nutrient-poor or highly acid (Richards, 1996). Relatively shallow rooting also appears to occur in soils with high fertility in wet forests (Brokaw, 1985b). On the other hand, deep rooting appears to occur on soils of moderately low fertility, with somewhat open nutrient cycles, and where seasonal drought confers a premium on deep rooting (Richards, 1996). In support of the hypothesized relationship between rooting patterns and gap formation, Kapos et al. (1990) found lower rates of tree-fall on a more infertile soil. Hartshorn (1978) found higher rates of tree-fall on waterlogged soils where deep rooting was impeded (see also Scatena and Lugo, 1995; Lugo et al., 1995). Firm anchorage may result in a lower frequency of tree-fall and a higher proportion of bole-snapping relative to uprooting (Richards, 1996). Uprooting, as opposed to bole snapping, creates disturbance in the soil and litter layer, which is important in vegetation response. Whether a tree was uprooted or snapped may have little effect on gap size, however (Lawton and Putz, 1988; Jans et al., 1993), because uprooting or snapping may often depend more on wood characteristics than on tree size (Putz et al., 1983). In contrast, in Nouragues, French Guiana, relatively small uprooted trees tended to create gaps that were larger than would be expected from their size (van der Meer and Bongers, 1996a). The increase in gap size was owing to slightly higher numbers of fallen trees in gaps caused by uprooted trees on shallow soils where, compared with the parts of the plot with deeper soils, there was poor root-system development and anchorage. Everham and Brokaw (1996) in a review of uprooting during catastrophic wind events, found that the highest rates of uprooting occurred on the wettest sites. Wadsworth and Englerth (1959) reported high rates of uprooting on deep soils and high rates of stem breakage on shallow soils. Uhl et al. (1988a) reported a blow-down in which 80% of trees were uprooted on nutrient-poor soils where tree roots were shallow.

231

However, most stems that die outside of these blowdowns do so after snapping (Uhl, 1982b). Of the forests listed in Table 8.4, few had percentages of uprooting above 50%, after excluding standing mortality and other modes of tree damage and gap formation that do not involve uprooting or trunk breakage. The highest rate was at La Selva, a wet site with relatively nutrient-rich soils. In most forests (Table 8.4), the majority of tree-fall and gap-making events cause little or no soil disturbance. Although there are no strong patterns evident in the Table, several dry and moist forest sites (e.g., Amazonas, Pasoh, Tai and Zagne) had low to moderate rates of uprooting. As expected if relative soil fertility affects uprooting, increasing soil fertility from Tai to Zagne and Para coincides with increasing rates of uprooting. Jans et al. (1993) explained the higher percentage of uprooting at Para (compared with nearby sites Zagne and Tai) to impeded rooting owing to waterlogging of soils and higher gravel content at Para. In contrast, the high rate of stem snapping at Tai (Vooren, 1986; Jans et al., 1993) was attributed to good rooting conditions (Jans et al., 1993). The relatively large size of gaps caused by uprooting in parts of the sample plot with shallow soils appears to account for the increase in the proportion of uprooting among gap-makers over what might be expected from the proportion of uprooting among fallen trees as a whole (compare the 1996a and 1996b data of van der Meer and Bongers in Table 8.4). We have no explanation as to why the leeward cloud forest at Monteverde had a relatively low percentage of uprooting (Matelson et al., 1995) compared with all other sites and with Monteverde windward cloud forest (Lawton and Putz, 1988). In considering the data in Table 8.4, it must be borne in mind that high rates of standing mortality should generally lead to low overall rates of gap formation (e.g., the Sarawak and Pasoh sites in Tables 8.3 and 8.4). Topographic position influences anchorage and treefall rates, but its effects appear to depend on the instability of the soil and underlying parent material, and on waterlogging or other factors that lead to shallow rooting (Table 8.2, Prediction 5). There are few data on the effects of slopes on canopy gap formation. Canopy gaps may often be more frequent on slopes than on flat terrain (Oldeman, 1978; Berner, 1992; Lugo et al., 1995), but not at all sites (Kapos et al., 1990; Hubbell and Foster, 1986; Scatena and Lugo, 1995). In a Costa Rican montane oak–bamboo forest, an increase in gap-formation probably involved

232

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

Table 8.4 Relative percentage of gap-makers or fallen trees that were either uprooted or snapped across the main stem Site

Sample 1 Climate 2

Soil fertility 3

La Selva, Costa Rica

GM

4000 [0]

high

90

−

Hartshorn (1980)

Lower Montane Venezuela

FT

1650 [1–2]

−

69

31

Carey et al. (1994) 4

Uprooted (%)

Snapped (%)

Source

Bako, Sarawak, Borneo

FT

4167 [0]

low

56

44

Hall (1991)

Lambir, Sarawak, Borneo

FT

2874 [0]

moderate

54

46

Hall (1991)

Lowland, Venezuela

FT

2725 [1–1.5]

−

52

48

Carey et al. (1994) 5

Monteverde,6 Costa Rica

GM 7

>2500 [−]

−

48

52

Lawton and Putz (1988)

Mersing, Sarawak, Borneo

FT

3905 [0]

high

47

53

Hall (1991)

Nouragues, French Guiana

GM

3000 [3]

low

45

55

van der Meer and Bongers (1996a)

Para, Ivory Coast

GIST

2100 8 [4]

low 9

43

57

Jans et al. (1993)

Zagne, Ivory Coast

GIST

1650 8

low 9

40

60

Jans et al. (1993)

Pasoh, Malaysia

FT

1900 [1]

low

40

60

Putz and Appanah (1987)

[4]

Amazonas, Brazil

FT

2186 [4]

low

36

64

Rankin de Merona et al. (1990)

Nouragues, French Guiana

FT

3000 [3]

low

34

66

van der Meer and Bongers (1996b)

Monteverde,6 Costa Rica

FT

>2500 [−]

−

26

74

Matelson et al. (1995)

Barro Colorado Island, Panama

FT 7

2656 [3]

high

25

75

Putz et al. (1983)

Tai, Ivory Coast

FT 7

1875 8 [4]

low 9

23

77

Vooren (1986)

Barro Colorado Island, Panama

FT

2656 [3]

high

22

78

Putz and Milton (1982)

Noh Bec, Quintana Roo, Mexico

GM 7

1500 [6]

moderate

18

82

Dickinson et al. (1999)

Tai, Ivory Coast

GIST

1875 8 [4]

low 9

15

85

Jans et al. (1993)

1

Data sources differed in whether the sample included gap-makers (GM), gap-makers from single tree-fall gaps (GIST), or fallen trees (FT) – whether they caused gaps or not. 2 Annual rainfall in mm. The number of consecutive months with < 100 mm of precipitation is shown in brackets. 3 Often estimated by the current authors. 4 Data are averaged over 9 plots in the State of Merida. 5 Data are averaged over 8 plots in the States of Merida, Delta Amacuro, and Bol´ıvar. 6 Leeward (Lawton and Putz, 1988) and windward (Matelson et al., 1995) cloud forest. Rainfall does not include cloud deposition. 7 Trees that fell over at ground level, often owing to a rotten base, are included in the “snapped” category because they cause minor soil disturbance. 8 Estimated by M. Dickinson. 9 F. Bongers, pers. commun., 1997.

higher growth and mortality rates on slopes, downslope leaning of boles, and larger crown volumes on the down-slope side of trees (Berner, 1992). Asymmetric crowns may increase the probability of tree-fall even on flat sites (Young and Hubbell, 1991; Richards, 1996). Soil instability on slopes amplifies the effects of other conditions that would predispose a tree to being toppled (Denslow, 1987), but soil types and their parent materials are differentially resistant to slippage (Guariguata, 1990) and differentially suitable for anchorage. For instance, in palm forests in the Luquillo Experimental Forest in Puerto Rico, tree falls are more frequent on steep slopes with shallow clayey

and often waterlogged soils (Lugo and Scatena, 1996; Lugo et al., 1995). In contrast, Wadsworth and Englerth (1959), also in Puerto Rico, noted that uprooting during hurricane winds was prevented in shallow soils on slopes by root penetration into cracks in stable parent rock, as compared with deeper soils in valleys. As may be the case with the palm forests noted above, waterlogging appears to exert a stronger effect than slope in the Bisley catchments of the Luquillo Experimental Forest where background canopy gap formation rate is highest on waterlogged soils in riparian valleys, but lower on slopes and in upland valleys (Scatena and Lugo, 1995).

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

Standing mortality Standing dead trees tend to form smaller gaps (Brokaw, 1985b; Putz and Appanah, 1987; Hall, 1991; Krasny and Whitmore, 1992; Jans et al., 1993; Midgley et al., 1995) than tree-falls, and high rates of standing mortality should translate into lower overall rates of gap formation because many trees that die standing will not form discernable canopy gaps (Table 8.2, Prediction 7; see also Putz and Appanah, 1987). The depressive effect of high rates of standing mortality on rates of canopy gap formation and on gap sizes should be particularly acute where the trees that die standing are larger than the trees that are uprooted or snapped, as is the case in certain lower montane and lowland sites in Venezuela (Carey et al., 1994) and at Mersing, Sarawak (Hall, 1991), but not in La Selva, Costa Rica (Lieberman et al., 1985). The depressive effect of standing mortality of large trees on gap size is exemplified by a study of three West African sites, among which gap-size distribution and rate of formation did not differ, owing to a preponderance of gaps formed by branch-fall from standing dead emergents in the site with the largest trees (Vooren, 1986; Jans et al., 1993). If rates of uprooting and snapping had been equal, gap sizes should have been larger at the site with the larger trees (Jans et al., 1993). The proportion of canopy gaps caused by dead or dying trees varies considerably: 72% in a temperate forest in New York (U.S.A.) (Krasny and Whitmore, 1992); 70% in a South African forest (Midgley et al., 1995); at least 50% in a West African forest (Jans et al., 1993); 27% in a semi-deciduous forest in the Yucatan (Dickinson et al., 1999); and at least 2% in a cloud forest at Monteverde, Costa Rica (Lawton and Putz, 1988). Standing mortality often leads to gaps that may best be described as gradual gaps (Krasny and Whitmore, 1992), formed by slowly dying trees that drop limbs one by one or by uprooting or snapping after the tree has died. Accordingly, some of the percentages above are probably underestimates of the role of standing mortality in gap creation. Although there are few sites for which data are available, the highest rates of standing mortality appear to be in equatorial Southeast Asia (Table 8.3). Comparably high rates of standing mortality have also been reported in two temperate forests (Krasny and Whitmore, 1992; Midgley et al., 1995). Data suggest that gap formation rates are also lower in equatorial Southeast Asia (see p. 226 above; see also Putz and

233

Milton, 1982). Rates of standing mortality should be highest where wind gusts are not frequent and strong trees are not architecturally prone to structural failure, trees are deeply rooted, and rare and severe droughts occur. From Table 8.3, it appears that soil fertility has little effect, although the site with the highest rate of standing mortality has low soil fertility (Hall, 1991). Average annual rainfall appears to have no effect, although Hall (1991) attributed high rates of standing mortality to periodic moisture deficits in freely draining soils. Such moisture deficits can occur quickly under certain soil conditions (Richards, 1996). The locations of sites (Table 8.3) examined by Hall (1991) did not appear seriously affected by the 1982– 1983 drought that caused such severe mortality in east Bornean forests. Lack of frequent strong gusty winds may also help explain high rates of standing mortality in equatorial Southeast Asia (P. Hall, pers. commun., 1997), while high rates of tree snapping in the Venezuelan Amazon (Uhl, 1982b) may account for the lowest proportion of standing mortality in Table 8.3. Wind appears to be an important cause of tree-fall in Barro Colorado Island and La Selva (p. 227; see also Brokaw, 1985b), forests with low rates of standing mortality. Different rates of standing mortality among forests may also be related to differences among species in propensity to die standing (Seth et al., 1960; Brokaw, 1985b; Jans et al., 1993; Condit et al., 1995). In a French Guianan forest dominated by species with pyramidal crowns, standing death was more common, while in forests with predominately umbrella-shaped crowns, wind-throw was more common (Ri´era, 1995). Senescence (Swaine et al., 1987), lightning strikes (Putz and Appanah, 1987; Lawton and Putz, 1988; Bruenig, 1989; Smith et al., 1994; Magnusson et al., 1996), defoliation by herbivores (Whitmore, 1975), fungal pathogens (Whitmore, 1975; Swaine et al., 1987) and termites (Putz and Appanah, 1987) also cause standing mortality. Vines and epiphytes High abundance of woody vines may increase gap frequency, while high frequencies of vine interconnections among adjacent trees may increase gap size (Putz, 1984; see also Table 8.2, Prediction 6). Woody vines appear to increase the likelihood that a tree will fall or die due to increased mechanical stress on the shoot and to shading of the tree’s foliage (Putz, 1984). In a somewhat different way, epiphytes may increase rates of gap formation by adding weight to the

234

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

canopy of trees in wet montane forests, especially in the wet season (Strong, 1977; Matelson et al., 1995). Vines may increase gap size by pulling down trees that are either linked with the gap-making tree or are hooked by woody vine tangles on the falling tree (Putz, 1984). There is ample evidence for this effect from vine-cutting experiments preceding logging operations (Putz, 1985). In contrast, woody vines may also have a dampening effect on tree-fall frequency and gap size by stabilizing trees through their connections with neighbors (Putz, 1984). Vine abundance is highest in areas subject to disturbance (Webb, 1958; Whitmore, 1974; Hegarty and Caball´e, 1991) and may be least in forests on nutrient-poor soils (Grubb, 1989). Dry to wet forest gradient It has been suggested that tropical dry forests should be less dynamic than tropical wet forests (Hartshorn, 1978; S.H. Bullock, pers. commun., 1996). But there have been few studies in which gap size and disturbance frequency have been described for a forest receiving less than 2000 mm of rainfall. Jans et al. (1993) and Dickinson et al. (1999) provide evidence suggesting that rates of gap formation and gap sizes are lower in forests with long dry seasons (see above, p. 226). The pattern of decreasing gap disturbance rates along the wet to dry forest gradient is obviously not well established, but several hypotheses have been advanced as to why drier forests should have lower rates of gap disturbance than wetter forests (see Table 8.2, Prediction 9): (1) Trees in drier forests have smaller crowns which produce smaller canopy gaps (Holdridge et al., 1971). However, as forest height decreases, it takes less of a gap to create an equivalent light gap, as gap aperture is the key variable (Canham et al., 1990; Lawton, 1990). Also, light limitation in the understory of the shortest forests may not be the most serious problem for tree regeneration (Lieberman and Li, 1992) so that light gaps lose their importance. Below-ground gaps may become effectively smaller for a given gap aperture, because of higher root/shoot ratios in drier forests (Cuevas, 1995). (2) Trees become shorter for a given trunk diameter as annual rainfall decreases and seasonality increases (Kira, 1978). Short trees with thick boles should be less subject to snapping (Putz et al., 1983), and shorter boles present a shorter moment arm that should lead to decreases in uprooting. Accordingly, rates of gap formation should be lower.

(3) Trees more often die standing in drier forests, leading to smaller tree-fall gaps (S.H. Bullock, pers. commun.). High rates of standing mortality would be expected on drier sites because of better anchorage, relative low height/diameter ratios, shorter periods during which gap-causing agents operate, and relatively low rain-loading of the tree canopies. Although standing mortality has been found to be important in forests that receive 2000 mm rainfall per year or less (Vooren, 1986; Putz and Appanah, 1987; Jans et al., 1993), it is also a feature of wet and moist forests, and there are too few data from drier forests to assess this hypothesis (Table 8.3). Standing mortality is associated with droughts in dry and moist tropical forests (Whigham et al., 1990; Swaine, 1992; Condit et al., 1995). Evidence from the 1982– 1983 El Ni˜no Southern Oscillation event support the notion that forests experiencing a yearly dry season may be less affected than wetter forests (Richards, 1996). Condit et al. (1995) reported 3% mortality (versus 2% during a non-drought period) on Barro Colorado Island, Panama, a forest where there are three months with less than 100 mm of rainfall. In contrast, mortality rates were substantial in forests in which rainfall does not typically drop below 100 mm in any given month. Woods (1989) reported 12–28% mortality in previously logged wet forest in Sabah, Malaysian Borneo. Similarly, Leighton and Wirawan (1986) reported 37% and 71% mortality of large trees (¾ 60 cm dbh) on two ridge plots, and 39–60% mortality of large trees (¾ 50 cm dbh) on mostly dry ridges at another site, in East Kalimantan on the island of Borneo. On wetter alluvial soils, mortality was approximately half as great as on ridges (Leighton and Wirawan, 1986; see also Ashton et al., 1995). Severe droughts of the magnitude of the 1982–1983 event recur in east Borneo at intervals of fifty to several hundred years (Leighton and Wirawan, 1986). (4) Rates of tree mortality are lower in drier forests, therefore rates of gap formation are lower. In contrast to this expectation, Lugo and Scatena (1996) found no relationship between relative mortality (% yr−1 ) and rainfall in a large sample of sites (see also Swaine, 1992). As disturbance-regime data are usually area-based, we used data from Phillips and Gentry (1994) to test for a relationship between rainfall (range 1500–4746 mm yr−1 ) and the number of stems that died per hectare per

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

year. As with relative mortality, no relationship was found (F (2,28) = 0.35; R2 = 0.02; P = 0.71). (5) Drier forests may burn periodically (Uhl et al., 1988b; Swaine, 1992; Snook, 1993), leading to long periods without appreciable formation of treefall gaps as forest structure recovers (Dahir and Lorimer, 1996). (6) Rates of gap formation are lower in drier tropical forests owing to a shortened wet season, while gap-causing agents (such as gusty wind, rainfall, and lightning strikes) are operative over longer periods in forests that receive high rainfall and are aseasonal. (7) Less rain-loading of tree canopies occurs in dry forests because of lower epiphyte biomass (Holdridge et al., 1971). Lower rates of gap formation should result.

Catastrophic wind disturbances Blow-downs and wind-throws in which a number of trees are damaged or killed are larger disturbances than those created by the death of single trees (Table 8.1). Art (1993) has defined blow-downs as “an extensive toppling of trees by wind within a relatively small area, greatly altering the small-scale climate within the ecosystems”. Blow-downs and wind-throws appear to be fairly common in tropical and subtropical forests, but clearly there have been too few quantitative studies to characterize their importance in tropical forests, even though their impacts last for a long time and clearly influence forest dynamics (Hubbell and Foster, 1986). Hurricanes and typhoons are the most destructive types of tropical windstorms. They occur in all regions of the world, and their impacts can be devastating (e.g., Yih et al., 1991). Many areas with vast expanses of tropical forest (e.g., South America, Africa, large areas in Asia and Australasia), however, are not impacted by hurricanes and typhoons (Fig. 8.1). In recent years, there have been a large number of articles in which the impacts of hurricanes and typhoons are considered and several authors have written summary articles (e.g., Brokaw and Walker, 1991; Tanner et al., 1991; Smith et al. 1994; Everham and Brokaw, 1996). While it is possible to predict an average return time for hurricanes within tropical and subtropical areas (Everham and Brokaw, 1996), there are few examples where it has been shown that the structure and dynamics of the forest are strongly influenced

235

by periodic hurricane events (Roth, 1992; Zimmerman et al., 1994) as much as or more than by background canopy gaps. It has been predicted, however, that forest structure and floristics would be influenced by repeated hurricane or typhoon damage (e.g., Odum, 1970; Fraver et al., 1998). Coastal mangroves may represent one of the few types of tropical and subtropical forests in which periodic wind disturbance has a dominant influence on the physical structure of the canopy (Roth, 1992). Predictions about the distribution and intensity of hurricanes and typhoons may mean little when one is considering a particular area of tropical forest, because the damage effects are greatest near the center of the storm, and there are few examples to demonstrate that specific forests have been heavily damaged by hurricanes at a frequency equaling the regional return frequency. Three examples demonstrate this point. Hurricane Gilbert heavily damaged a forest in the northeast Yucatan Peninsula in 1988 (Whigham et al., 1991; Harmon et al., 1994; Whigham et al., 1998). The northeast Yucatan was struck by 50–60 hurricanes between 1886 and 1968 (Alaka, 1976) which would represent a return time of approximately 1.6 years for the region. Our studies have been ongoing since 1984. Hurricane Gilbert has been the only storm during that 12-year period to do any significant damage to the forest. Long-term residents could not remember any storm that damaged the local forests as much as Hurricane Gilbert did. Records of hurricanes passing through the Sian-Ka’an Biosphere Reserve, southcentral Yucatan Peninsula, show the same pattern. The area encompassed by the reserve has been struck by 11 hurricanes between 1893 and 1982 and each of them followed a different path (L´opez Ornat, 1983), indicating that few areas would have been heavily damaged more than once or twice during the period of approximately 90 years, even though the return interval, based on long-term records for the area, would be about nine years. The Luquillo Experimental Forest in Puerto Rico is one of the few areas where long-term climatological and vegetation data are available to evaluate the occurrence and impact of hurricanes. The northeastern portion of Puerto Rico was struck by approximately 45 hurricanes between 1886 and 1968 (Alaka, 1976) for a return frequency of 1.8 years. Heavy to moderate damage to the Luquillo Forest occurred three times between 1928 and 1932 (a 1.3-year return interval) but the next damaging storms did not occur until 24 years (1932–1956)

236

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

and 33 years (1956–1989) had passed (Zimmerman et al., 1994). Since Hurricane Hugo in September 1989, four hurricanes have passed near the Luquillo Forest, but none has caused any significant damage (J. Zimmerman, pers. commun., 1996). The extent of hurricane and typhoon impacts can range from very small to very large, and the impacts in any particular location will be influenced by a number of factors, both biotic (e.g., stem size, stem condition, species, presence of pathogens, structural complexity of the stand) and abiotic (e.g., storm intensity, topography, soil characteristics, disturbance history) (Everham and Brokaw, 1996; Imbert et al., 1996). Topography, for example, has been shown to be important in the Luquillo Forest in Puerto Rico where the landscape is rugged compared to the more uniform landscape of New England (Lugo, 1995). The factors which mediate the effects of wind as a cause of background canopy disturbance may often be obscured in intense hurricanes and typhoons (Everham and Brokaw, 1996).

COMMUNITY AND ECOSYSTEM RESPONSES TO CANOPY DISTURBANCE

Background canopy gaps, blow-downs, and catastrophic wind events in tropical and sub-tropical forests create a continuous range of conditions that set the stage for community and ecosystem response (Lugo and Scatena, 1996; Vandermeer et al., 1996). We first describe community-level responses by focusing on both changes in species composition and pathways of initial response to mainly catastrophic wind disturbance. We then compare responses between background canopy disturbances and catastrophic wind disturbances. We close this section by discussing changes in ecosystem processes, associated primarily with catastrophic wind disturbance events that cause large changes in the physical structure of forests. Community-level responses A framework for understanding responses The interplay between characteristics of the disturbed site (see p. 228 above), species availability, and post-disturbance species performance determine changes in species composition and relative abundance immediately following the disturbance, and set the stage for long-term successional changes (Pickett et al.,

1987; Pickett and McDonnell, 1989; Everham and Brokaw, 1996; Vandermeer et al., 1996). Species availability (Pickett and McDonnell, 1989) after a wind disturbance includes propagules (seeds for most species) that are dispersed into a site at the time of the disturbance event, occur in a seed bank (generally in the soil in tropical forests; Garwood, 1989), or are dispersed to the site after the event from reproductive individuals that either occur outside the area of disturbance or were present in the disturbed area and recovered from damage incurred during the disturbance (Noble and Slatyer, 1980; Everham and Brokaw, 1996). Species availability also includes intact or damaged adults and juveniles that were present at the time of the disturbance. Sprouting of persistent meristems from roots and stems of intact or damaged plants is an important component of recovery (Everham and Brokaw, 1996). Species performance (i.e., establishment and growth of individuals that originated from seed or persisted through the disturbance event) is determined by the autecology of the species (e.g., germination, growth, and assimilation patterns), environmental conditions (e.g., light and soil moisture) prior to and following the disturbance, and species interactions (e.g., disease, herbivory, and competition; Pickett and McDonnell, 1989). Throughout the literature on responses to catastrophic wind disturbances and background canopy gaps, species have been grouped by relative shade tolerance (Brokaw, 1985b; Clark and Clark, 1987; Denslow, 1987; Swaine and Whitmore, 1988; Brokaw and Scheiner, 1989; Everham and Brokaw, 1996). The grouping of plants into shade intolerant and shadetolerant species necessarily involves breaking up a continuum of conditions and responses, as well as lumping species with different life-history attributes. Thus there is room for much refinement (Clark and Clark, 1992; Grubb, 1996) in categorizing species responses. We believe that the most useful approach for comparing species responses to wind disturbances is to differentiate species that have relatively abundant seedlings and saplings below a closed canopy (shadetolerant) from those that may persist for relatively short periods below a closed canopy, but grow very little, if at all, beyond their seed reserves (shade-intolerant: Clark and Clark, 1987; Swaine and Whitmore, 1988). Pioneer species (Whitmore, 1989; Clark and Clark, 1992; Kennedy and Swaine, 1992) are a subset of the shade-intolerant group of species, and are short-lived, grow quickly in high light, and have light wood.

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

237

Table 8.5 Predictions about absolute change in species composition and the prevalence of different pathways of response to catastrophic wind disturbance in tropical forests #

Prediction

Species composition 1

Absolute change in patterns of species relative abundance (species composition) should be greatest where catastrophic wind events have the longest return interval

2

Large changes in species composition should be associated with catastrophic wind events that damage forests established on old fields or plantations

3

Little change in species composition should occur in forests frequently hit by catastrophic winds, because of dominance by species that resist and are resilient to damage (i.e., survive well and sprout readily)

Response pathways 4

Regrowth from damaged stems predominates after catastrophic wind events in the tropics

5

The importance of regrowth decreases with increasing mortality rates, damage, and uprooting

6

The importance of regrowth should be greatest in wet sites within humid tropical forests, in high-altitude tropical forests, and in dry tropical forests

7

The importance of release of understory trees after catastrophic wind disturbance is currently underestimated, primarily owing to a lack of long-term studies of regrowth

8

Recruitment of short-lived shade-intolerant species with dormant seeds should be greatest where mortality and damage is greatest and uprooting more common

9

Recruitment of short-lived shade-intolerant species should be minimal in forests with a long return time for catastrophic events

10

Recruitment of shade-intolerant species with non-dormant seeds should be minimal in catastrophic wind disturbances owing to reductions in populations of seed dispersers and nonexistent or inconsequential post-event fruiting

11

Repression by vines of tree release and recruitment should occur where damage is severe and frequent

12

Repression by vines should affect species composition of recovery owing to differential abilities among tree species to shed and avoid vines

Shade-intolerant species are most likely to regenerate from seeds that germinate at the time of gap formation (Lieberman and Lieberman, 1987; Kennedy and Swaine, 1992), while the shade-tolerant species are more likely to colonize a gap from established seedlings or saplings (Brokaw, 1985b; Brokaw and Scheiner, 1989; Connell, 1989; Brown and Whitmore, 1992) or from sprouts of damaged seedlings or saplings (Putz and Brokaw, 1989). This may often be primarily a function of mortality rates below a closed canopy (see Lieberman et al., 1990). The response of shadeintolerant species to wind disturbances is further influenced by whether the species have or do not have a dormant seed bank (e.g., Garwood, 1989). Species that do not maintain a dormant seed bank can only respond to a wind disturbance if seeds are recruited to the site from individuals at the site that produced seed shortly before or after the disturbance, or from reproductive individuals not at the disturbance site but close enough for seeds to be dispersed by wind or animals.

Responses to catastrophic wind disturbance Major pathways of response to catastrophic wind disturbance: Responses to catastrophic wind disturbances have been discussed in terms of recovery pathways (Everham and Brokaw, 1996), including: (1) regrowth from sprouting of damaged stems; (2) release of established seedlings and saplings of primarily shade tolerant species that were present in the understory at the time of disturbance; (3) recruitment of seedlings of primarily shade-intolerant species from dormant seeds in the soil seed bank, or from seed dispersal to the site just before or following the disturbance; and (4) repression of vegetation by fast-growing species of vines, herbs, and shrubs. Table 8.5 lists 12 generalized predictions about community responses to catastrophic wind disturbance. In this section we discuss the basis for each of the predictions. We expect exceptions to the generalizations, because they are based on a review of the literature which includes findings from catastrophic wind events of widely different intensities and from forests that differ in disturbance history, soils, and

238

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

type (e.g., wet and dry forests, lowland and montane, temperate and tropical). Change in species composition: Absolute change in relative abundance patterns (species composition) in response to a catastrophic wind event should depend, in large part, on the characteristics of the disturbed site and on the species composition of the forest before the event, both of which are related to the severity, frequency, and recency of past events (Richards, 1996; see also Table 8.5, Predictions 1–3). The greatest change in species composition has been predicted to occur in forests that are infrequently impacted by catastrophic wind disturbances. This response is mostly owing to the increased importance of shade-intolerant species which often have low abundance prior to the disturbance and have increased opportunities for recruitment following the disturbance (Richards, 1996). Wind damage in forests with a low frequency of disturbance should be greater because canopy trees are taller and the topography of the forest canopy is more irregular and thus more prone to wind damage (Foster and Boose, 1992; Poorter et al., 1994; Everham and Brokaw, 1996; Richards, 1996). An exception to this prediction occurred in Nicaragua where storms are infrequent, hurricane damage was severe, and recruitment was limited (Boucher et al., 1994; Vandermeer et al. 1995, 1996). The absolute change in species composition would not be expected to be great in forests that suffer frequent wind disturbance, where species are better adapted to resist and respond to damage (Frangi and Lugo, 1991; Richards, 1996). Large changes in species composition have been shown to occur in the temperate zone when a canopy of relatively shade-intolerant species is heavily disturbed and replaced by shade-tolerant species that were present in the understory (Spurr, 1956; Webb, chapter 7, this volume). This general process has been termed accelerated succession (Spurr, 1956; Abrams and Scott, 1989). In contrast, catastrophic wind events in the tropics are often characterized by little change in species composition (Everham and Brokaw, 1996). Differences between tropical and temperate forests may be due, in part, to the fact that many of the temperate forests in which studies have been done developed on sites that had been previously disturbed by logging or had been cleared for agriculture and then abandoned (Spurr, 1956; Foster, 1988). The same type of response may be anticipated in the tropics where plantations (Lugo, 1992; Parrotta, 1995; Fu

et al., 1996) and secondary forests (Lugo, 1992) often have a diverse understory differing in composition from the canopy trees and where plantations often are more severely damaged during catastrophic events than natural primary and secondary forests (see Everham and Brokaw, 1996; Fu et al., 1996). Plantations or natural forests dominated by fast-growing, soft-wooded species would be highly susceptible to wind damage and resulting mortality (Putz et al., 1983; Zimmerman et al., 1994; Fu et al., 1996) and should show low levels of regrowth (Zimmerman et al., 1994). Relatively little change in species composition would be expected where species are resistant and resilient to damage. The importance of resistant and resilient species should increase in forests that are frequently damaged by wind (Brokaw and Walker, 1991; Frangi and Lugo, 1991; Walker, 1991; Bellingham et al., 1994; Zimmerman et al., 1994; Matelson et al., 1995; Everham and Brokaw, 1996; Scatena and Lugo, 1995). Hard-wooded, shade-tolerant species have been found to survive damage well and to sprout more readily than soft-wooded species (Putz et al., 1983; Zimmerman et al., 1994). Also, palms appear to resist wind damage and readily recover from defoliation (Frangi and Lugo, 1991). Forests that experience frequent windstorms, but that are not severely damaged owing to the dominance of resistant and resilient species, should be dominated by shade-tolerant species with seedling and sapling pools (Richards, 1996). Limited opportunities for recruitment in gaps would result from both reduced damage to established trees and a pulsed pattern of disturbance with many years between events, rendering inviable a life-history involving short life-span and recruitment from seed in gaps (Noble and Slatyer, 1980). Also, as we argue below (pp. 239–241), there are several general barriers to recruitment in tropical cyclonic-storm disturbances that would appear to make recruitment a less than ideal regeneration strategy. Regrowth by sprouting: In their review, Everham and Brokaw (1996) suggested that regrowth from surviving stems predominates following catastrophic wind disturbance in the tropics because most events cause low to moderate damage and mortality (Table 8.5, Predictions 4–6). Sprouting occurs after mechanical damage to individuals that were damaged by wind, by falling debris, or by the fall of large individuals. Several variables appear to influence the importance of recovery by regrowth. Everham and Brokaw (1996) predicted that regrowth would become less important as direct

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

or indirect mortality from wind disturbance increased. Also, the importance of recovery by regrowth may decrease as the proportion of uprooted trees increases; uprooted stems are, in some cases, less likely to sprout than snapped trees (Putz et al., 1983; Bellingham et al., 1994; Everham and Brokaw, 1996). On moister sites, sprouting appears to be a common tree response, but increases in density due to sprouting may be negated by increased mortality of uprooted trees, which appears to be common on wet sites (Everham and Brokaw, 1996). The importance of stem sprouting as a major pathway of recovery following wind disturbance also appears to be greater in drier forests and in forests at higher elevations (Ewel, 1977; Ewel, 1980; Murphy and Lugo, 1986), although this tendency has not been evaluated in terms of forest response to hurricane damage. Release: Our predictions regarding release as an important recovery mechanism are based on only a few studies of this phenomenon in tropical forests (Table 8.5, Prediction 7). There are at least four reasons why the importance of release after catastrophic wind events is likely to be underestimated. First, for seedlings and saplings, the difference between release and regrowth is trivial; to be in either response category, the plants had to be present before the disturbance event. Damaged seedlings and saplings would fall into the regrowth category, while individuals that were not damaged would arbitrarily fall into the release category. Second, we expect that short-term studies of regrowth seriously overestimate the eventual importance of sprouts, because of the development of disease associated with wounds and a concomitant decrease in wind-firmness (Roth and Hepting, 1943; Shigo, 1984; Putz and Sharitz, 1991). Large trees often cannot recover effectively after severe damage (Oldeman, 1978), and may be very likely to die, even long after the event that damaged them (Shaw, 1983; Walker, 1995). Third, catastrophic wind disturbance may initiate changes in canopy dominance, beginning with an increase in the importance of short-lived pioneers which, after several decades, are replaced by shade-intolerant species that persist until the next disturbance event (Weaver, 1986). The fourth reason why release may be underestimated comes from evidence that catastrophic events appear to lead to enhanced growth of suppressed seedlings and saplings. Defoliation and structural damage to canopy trees results in an increase in the amount of light in the understory, at least for a couple

239

of years (Fern´andez and Fetcher, 1991; Bellingham et al., 1996). In Nicaragua, although the majority of stems were sprouts, released seedlings and saplings appeared to account for a large proportion of the stems encountered in post-hurricane inventories (Yih et al., 1991; Boucher et al., 1994; Vandermeer et al., 1995). In Puerto Rico, falling litter and debris from Hurricane Hugo killed 60% of the seedling pool of one species, but the remaining seedlings of that species responded with a strong increase in growth (You and Petty, 1991). An increase in growth in smaller trees following hurricane disturbance was observed in the Yucatan following Hurricane Gilbert (Whigham et al., 1991). The abundance of the long-lived Shorea parvifolia in the storm forest of Kelantan disturbed in the late 1800’s may be an example of an important episode of release (Wyatt-Smith, 1954). Release has also been found to be important in temperate forests recovering from catastrophic wind disturbance (e.g., Spurr, 1956; Foster, 1988; Platt and Schwartz, 1990; Merrens and Peart, 1992; Webb, chapter 7, this volume). Even if regrowth of damaged stems is more important than release among the set of individuals that successfully captures space in the canopy after a given disturbance event, a generalized release response in past disturbance events will still have played a central role in that prevalence of regrowth (Foster, 1988; Connell, 1989; Clark and Clark, 1992; You and Petty, 1991). Recruitment: Recruitment rarely appears to be the dominant recovery pathway following catastrophic wind disturbances (Everham and Brokaw, 1996; Table 8.5, Predictions 8–10). Recruitment, nonetheless, is generally a component of recovery (Weaver, 1986; Frangi and Lugo, 1991; Walker, 1991; Bellingham et al., 1994). Recruitment should be greatest following wind disturbances that cause extensive damage and mortality, particularly in forests that are infrequently impacted by major storm events (Everham and Brokaw, 1996; Richards, 1996). Recruitment is, however, likely to vary spatially within the area impacted by wind disturbances. The greatest damage is more likely to occur near the center of the storm than at the periphery (Lugo et al., 1983; Richards, 1996), and recruitment would thus be more important near the storm center. Recruitment would also be expected to be more important in topographically exposed sites that are more heavily damaged (Bellingham, 1991; Frangi and Lugo, 1991; Foster and Boose, 1992); particularly

240

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

at higher-elevation sites where damage is more severe and less spatially variable (Beard, 1945a,b; Brokaw and Grear, 1991). Several examples in which high levels of recruitment have occurred seem to counter the general perception that regrowth is the primary recovery mechanism following wind disturbance. Whitmore’s (1974) 21-year study of the role of catastrophic wind disturbance in tropical forests contrasted sites on Kolombangara, in the Solomon Islands, that were subject to severe damage by past hurricane events with those that had been more sheltered from hurricane damage. The canopies of the more hurricane-prone areas were dominated by long-lived shade-intolerant species while the sheltered sites had a high representation of nonpioneer species. The shade-intolerant species suffered heavy damage during disturbance events, yet recruited well afterwards from seed. Snook (1993) also reported successful recruitment by long-lived shade-intolerant species after hurricanes in the Yucatan Peninsula of Mexico. Lack of recruitment by shade-intolerant species following catastrophic wind disturbances may be related to the presence of few large gaps, lack of exposed soil, the presence of a thick litter layer, and lack of propagules (Everham and Brokaw, 1996; Walker, 1999). Catastrophic wind events create a range of gap sizes dominated by small gaps, although the size distribution shifts to a larger average gap size as storm intensity increases (Everham and Brokaw, 1996). Except for patches that are severely damaged, the canopy may not remain open long enough for pioneers to avoid suppression, as regrowth following defoliation and minor branch damage can be rapid (Bellingham et al., 1994). Perhaps more important than a lack of large gaps is the high deposition of litter and lack of disturbance to the substrate and understory necessary for germination and establishment of pioneer seedlings (Weaver, 1986; Everham and Brokaw, 1996). “Litter gaps”, where the litter layer is removed and mineral soil is exposed, are thought to be indispensable for high rates of germination and establishment of small-seeded species in general and pioneers in particular (Putz, 1983; Putz and Appanah, 1987; Raich and Christensen, 1989; Kennedy and Swaine, 1992; Molofsky and Augspurger, 1992; Grubb, 1996). Litter gaps are often associated with uprooted trees, nurse logs, and steep slopes (Grubb, 1996). Uprooting mixes the soil, which may enhance germination rates (Putz, 1983; Putz and

Appanah, 1987). Experimental litter removals after a hurricane were particularly beneficial to pioneer species (Guzm´an-Grajales and Walker, 1991). The importance of litter gaps is supported by the finding that fires that follow hurricanes, by removing litter, lead to high levels of establishment of pioneer species, particularly if fire intensity is high enough to kill vegetation that would otherwise sprout profusely (Oliver and Larson, 1990; Snook, 1993; Everham and Brokaw, 1996). Even if conditions are right for germination, growth and survival of shade-intolerant seedlings, dormant or newly dispersed seeds may not be available (Schupp and Fuentes, 1995; Everham and Brokaw, 1996). Lack of a response to major disturbance by short-lived shadeintolerant species has been attributed to the absence of both a seed bank and reproductive individuals owing to a long disturbance return time or the spatial pattern of disturbance (Noble and Slatyer, 1980; Uhl et al., 1988a; Brand and Parker, 1995; Peterson and Carson, 1996). One possible example of lack of response by short-lived shade-intolerant species owing to lack of propagules were the severely hurricane-damaged Nicaraguan forests described by Boucher et al. (1994) and Vandermeer et al. (1995). The hurricane return time for the area is long, perhaps too long for adults and seed banks to persist between events. Because major wind disturbances should significantly reduce background rates of canopy gap formation (Dahir and Lorimer, 1996), there may be few opportunities for regeneration between events. Unique problems of seed input may be associated with catastrophic winds. When defoliation is extensive, seed fall just after the event can be minimal (Lindo, 1968), although defoliation has been reported to trigger a large flowering response, at least in some species, and lead to an increase in seed fall compared to normal levels (Everham and Brokaw, 1996; Lugo and Scatena, 1996). This later seed rain may often be irrelevant, however, because of a premium on early establishment (Brokaw, 1985a; Brown and Whitmore, 1992; Kennedy and Swaine, 1992) and because the seed bank is likely to be the predominant source of viable seeds in background canopy gaps (Garwood, 1989) where defoliation does not occur. In large gaps created by catastrophic winds, seed rain may be further constrained among wind-dispersed pioneer species by the distance to potential seed sources. A large proportion of species have animal-dispersed seeds (Swaine and Whitmore, 1988; Levey et al.,

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

1994), and the dispersers often avoid large gaps in the canopy (Schupp et al., 1989). Seed and fruit dispersers may leave hurricane-damaged areas because of a lack of seeds and fruit (Ackerman et al., 1991; Yih et al., 1991). One might expect that, where hurricane damage is most severe, flowering and fruiting would be most attenuated. This would particularly affect those species that do not have dormant seeds, a seeding and sapling bank, or a strong sprouting response, but might have little effect on pioneers. If this is the case, it could partially explain the lack of an overwhelming recruitment response in some severely damaged forests (Boucher et al., 1994; Vandermeer et al., 1995, 1996). Repression: Long-term repression of recovery as a result of the proliferation of herbs and vines (Everham and Brokaw, 1996) appears to be infrequent in the tropics (Table 8.5, Predictions 11, 12). In high-altitude dwarf forests in Puerto Rico, recovery was dominated by ferns and grasses (Weaver, 1986; Walker et al., 1996), although the cause of a lack of tree recovery was not clear. In gaps in Samoan lowland forest created by catastrophic winds, dense growth of grasses and ferns suppressed tree regeneration (Wood, 1970). These would seem to be examples of the arrestedsuccession effect, where shrub and herb communities delay invasion by trees in the temperate zone (Putz and Canham, 1992). The climbing habit of vines, however, creates a different situation. Repression by vines has been documented in several severely and frequently disturbed forests. Frequent hurricanes in wet forests of northeastern Australia can lead to the formation of “cyclone scrub”, a short-statured forest with emergent climber towers and abundant vines throughout the canopy (Webb, 1958). These forests occur where winds from frequent storms are locally intensified by topography (Webb, 1958). A similar situation occurs in Nigeria as a result of frequent tornados (Jones, 1955a,b). Recruitment and release can be suppressed by vines in large blow-downs (Lindo, 1968; Wood, 1970) and after intensive logging (Putz, 1991) where forest structure is conducive to vine proliferation (Hegarty and Caball´e, 1991; Putz, 1985). Soil disturbance owing to logging promotes high levels of vine establishment (Putz, 1985, 1991). Proliferation of vines appears to delay forest structural development, but, except for frequently disturbed forests (Webb, 1958), the effect is temporary (Wyatt-Smith, 1954; Webb, 1958; Whitmore, 1974; Whitmore, 1975). Putz (1980) listed several characteristics that enable trees

241

to avoid and rid themselves of vines, these include: fast height and diameter growth, regular self-pruning of branches, large compound leaves, and vine-removing symbionts. The ability to avoid and shed vines provides a mechanism by which vine repression may alter the species composition of trees following severe damage (Table 8.5). In support of this general expectation, vine tangles in big gaps (¾ 1 ha) in Costa Rica exclude “all but the very fast growing pioneers” (Hartshorn, 1980). Also, palms are “climber shedders/vine tangle surmounters par excellence” owing to their methods of leaf production and shedding (Putz, 1980). It is apparent that not all severely damaged forests are prone to repression by vines (Walker et al., 1996). One possible example of this is the “hurricane forests” of the Caribbean island of St. Vincent, where damage is frequent and uprooting predominant, but where vine proliferation is apparently absent (Beard, 1945b). If blow-down is not prevalent, severe winds and major branch loss may rid trees of vines. In the forests damaged by Hurricane Gilbert in northeastern Yucatan (Whigham et al., 1991), almost all vines were eliminated from canopy trees, but they survived the disturbance and within a month most began to branch profusely. For the first year after the disturbance it appeared that vines would play an important role in recovery, but within five years almost all of the vine branches that were marked in 1988 had died, and most of the trees were still free of vines even though some had heavy vine loads prior to the hurricane (D. Whigham and E. Cabrera, pers. observ.). This could potentially explain the lack of vine repression in some relatively wind-resistant forests. Contrasts between responses to catastrophic wind and other canopy disturbances It should be apparent that much of the literature upon which an understanding of response to catastrophic wind disturbance is based comes from studies of background canopy disturbance, and response to background canopy gaps is often a good model for understanding response to catastrophic wind events (Ackerman et al., 1991; Everham and Brokaw, 1996). However, Ackerman et al. (1991) suggested that response to background canopy disturbance is an imperfect analog for response to both the least and the most severe damage by tropical cyclonic storms, while moderate damage should most closely approximate background canopy disturbance. Comparisons are made difficult, in part, because recovery in background

242

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

canopy gaps is typically studied only where gaps extend into the understory (Brokaw, 1982a; Popma et al., 1988; Lieberman et al., 1989). In contrast, in studies of catastrophic wind disturbance, gaps (or heavily disturbed patches) are not generally distinguished from areas where only upper canopy damage occurs (Frangi and Lugo, 1991; Walker, 1991). We are not aware of any studies which explicitly compare regeneration in sites dominated by background canopy disturbance with that in forests damaged by catastrophic events [(the closest would be that of Whitmore (1974)]. Forests whose disturbance regimes are dominated by catastrophic wind events should differ from forests dominated by background canopy disturbance in the frequency of large gaps, in the lack of seed sources for various reasons, and in poor recruitment even where seeds are present. Foster and Boose (1992) and Everham and Brokaw (1996) conceived of a gradient of disturbance from small gaps, created by standing dead trees or branch falls, to larger gaps formed by the uprooting or breakage of one or many trees, and, at the extreme, to large gaps with indistinct edges created by damaged and defoliated trees during a catastrophic wind event. Less severe catastrophic wind events may cause minimal upper-canopy damage, and there may only be a few small gaps that extend to the understory (Everham and Brokaw, 1996). When compared with background canopy disturbance, catastrophic wind events can create equally small gaps as well as much larger gaps (Foster and Boose, 1992; Everham and Brokaw, 1996). For both types of disturbance, small gaps are much more frequent than large gaps, although, despite their low frequency, large gaps may cover a comparatively large area (see Lawton and Putz, 1988; Brokaw, 1985b; Foster and Boose, 1992). Timing is obviously different, with background canopy disturbances occurring continuously whereas catastrophic events occur in a pulsed and infrequent manner (Oldeman, 1989). The general assumption is that release, sprouting, and ingrowth of gap-edge trees dominate in small gaps, while recruitment becomes more prevalent in large gaps (Dunn et al., 1983; Brokaw, 1985a,b; Denslow, 1987; Raich and Christensen, 1989; Whitmore, 1989; Everham and Brokaw, 1996) in both disturbance types. Similarly, predictions and data from forests dominated by background canopy gaps (Hartshorn, 1978; Denslow, 1980; Putz and Appanah, 1987; Uhl et al., 1988a; Brokaw, 1985b; Ashton, 1981; Jans et al., 1993) and from forests damaged by catastrophic wind

events (Weaver, 1986; Runkle, 1990; Walker, 1991; Bellingham et al., 1994; Richards, 1996) suggest that it is the frequency of large gaps and severely damaged patches that should determine the relative abundance of shade-intolerant species in a forest. Where background canopy disturbance dominates in the tropics, shadetolerant species comprise by far the largest group (Clark and Clark, 1987; Whitmore, 1989; Welden et al., 1991; Lieberman et al., 1995). Few authors have found (Whitmore, 1974) or predict (Richards, 1996) a dominance by shade-intolerant species in forests damaged frequently by catastrophic wind disturbances. This may be owing to the finding that most catastrophic wind events cause only minor to moderate damage (Everham and Brokaw, 1996). Even where gap sizes are large, relatively poor recruitment responses have been shown to occur (Uhl et al., 1988a; Boucher et al., 1994; Vandermeer et al., 1995) owing to lack of propagules and post-dispersal phenomena. In a large blow-down in the Venezuelan Amazon in which uprooting predominated, Uhl et al. (1988a) ascribed a lack of pioneer response to a lack of seed sources, which, in turn, may be due to poor regeneration of these species in the prevailing small gaps (see also Peterson and Carson, 1996). Putz and Appanah (1987) have suggested, however, that a lack of propagules should not be expected to be more strongly associated with forests damaged by catastrophic wind events (Putz and Appanah, 1987). Long intervals between events in forests whose disturbance regimes are dominated by catastrophic winds would tend to reduce the relative abundance of both adults and seed pools of short-lived pioneers (Noble and Slatyer, 1980). In cyclonic storm disturbance, lack of a recruitment response may be due to propagule limitation related to defoliation and branch damage of canopy trees (Lindo, 1968) and to the lack of seed dispersers in heavily damaged areas (Ackerman et al., 1991; Yih et al., 1991). Even where propagules are present, recruitment may be reduced in catastrophic wind disturbances by a lack of disturbed and litter-free soil (Guzm´anGrajales and Walker, 1991; see review in Everham and Brokaw, 1996) and by rapid vegetative recovery that would compete with new seedlings (Everham and Brokaw, 1996). Similarly, both lack of litter-free soil in forests where uprooting is uncommon (e.g., Putz and Appanah, 1987) and competition from established individuals (Kennedy and Swaine, 1992) have been related to reduced recruitment in background canopy

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

disturbances. Lack of disturbed soil in catastrophic wind disturbances, even where mortality and structural damage rates are high, may not be as generally important as previously suggested (Everham and Brokaw, 1996): the proportion of uprooting among fallen trees ranges from 15% to 88% in tropical forests hit by catastrophic winds (Everham and Brokaw, 1996) while similar rates of uprooting (15%–90%) are reported for tree-falls in forests dominated by a background canopy disturbance regime (Table 8.3). In this section, we have provided a series of predictions (Table 8.5) regarding changes in species composition and recovery pathways following canopy wind disturbance. No clear patterns emerge, because there have not been enough long-term studies of tropical forests to elicit broad geographic patterns or determine whether or not predictions, such as those that we provide, adequately describe recovery processes following background and catastrophic wind disturbances. A general model that might be useful in designing future investigations to test these and other predictions comes from succession research on prairie pothole wetlands in the United States (van der Valk, 1981). Prairie pothole wetlands occur over a broad range of hydrogeomorphic settings in which water quality (fresh or saline) and disturbances, both natural (wet and dry cycles) and anthropogenic (agriculture, fire), play important roles in determining the status of vegetation at any moment in time. Typically, vegetation undergoes dramatic changes in response to short- and long-term changes in water levels. Van der Valk based his model on “a Gleasonian approach” to understanding vegetation change at the ecosystem level, based on the ecological characteristics of individual species. He suggested that species or groups of species with similar life-history traits respond similarly to disturbance, whether of natural or anthropogenic origin. In applying this conceptual approach to tropical forests, the first step would be to characterize the disturbance regime of the forest. Once the disturbance regime was adequately characterized, the responses of individual species (or groups of species such as shade-tolerant and shadeintolerant) to the range of disturbances would be evaluated. In prairie pothole wetlands, for example, the interactions between water levels and germination characteristics of seeds of species in the soil seed bank were useful in predicting vegetation responses. In tropical forests, the soil seed bank appears to be less important, but species responses over the range of sizes of wind disturbances (e.g., ingrowth in small

243

canopy gaps, and sprouting in larger openings) appears to be most important. Ongoing long-term research in the Luquillo Experimental Forest in Puerto Rico (Zimmerman et al., 1996) should provide the types of data that will be required to test predictions such as those that we have provided in Table 8.5. Ecosystem-level responses Ecosystem-level processes (e.g., rates of decomposition, rates of nutrient uptake and release, primary production) should be influenced by wind-generated disturbances, especially in larger disturbances where changes in the physical structure of ecosystems are greater. Many changes in ecosystem-level processes are mediated through alterations of microclimate, changes in site water balance, and additions of nutrients and carbon associated with the destruction of biomass (Frangi and Lugo, 1991; Lodge and McDowell, 1991; Sanford et al., 1991; Lugo and Scatena, 1996; Scatena et al., 1996). Microclimate modifications primarily involve increases and changes in the quantity and quality of light (Raich, 1989; Brown, 1996) and associated variables such as soil and air temperatures and relative humidity (Brown, 1993). Reductions in leaf biomass, and destruction of branches and boles increase amounts of photosynthetically active radiation (Bellingham et al., 1996), decrease fine-root biomass in the disturbance site (Parrotta and Lodge, 1991; Silver et al., 1996), and increase soil moisture owing to decreased evapotranspiration (Vitousek and Denslow, 1986). The degree of change in microclimatic variables following canopy disturbance is related to disturbance size (Chazdon and Fetcher, 1984; Canham et al., 1990; Bellingham et al., 1996), but there is often considerable spatial variation within the disturbance site (Brown, 1993). The microclimate changes as vegetation and the ecosystem recover from disturbance. Bellingham et al. (1996) found that photosynthetically active radiation increased with the severity of hurricane damage (i.e., amount of defoliation) but that light levels had returned to almost pre-hurricane levels within 33 months. Fern´andez and Fetcher (1991) found similar results in Puerto Rico 14 months after Hurricane Hugo. Increased light levels in single tree-fall gaps also decrease with time, but the causes of the changes would be different from those measured in hurricane-damaged forests because of differences in ecosystem recovery responses

244

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

(e.g., ingrowth versus sprouting in the understory: Bellingham et al., 1996). There have been few studies of root biomass, root turnover, and root growth in gaps. Silver et al. (1996) found that root biomass in the Luquillo Experimental Forest (Puerto Rico) was only beginning to recover six years after Hurricane Hugo; though most of the soil nutrient pools had returned to pre-hurricane levels. Wind-generated disturbances of all sizes significantly increase the necromass of fallen leaves, branches, trunks and upturned roots (Frangi and Lugo, 1991; Harmon et al., 1994; Smith et al., 1994). Not only does the amount of dead biomass increase in hurricaneimpacted forests (Harmon et al., 1994), but the standing stocks of nutrients associated with the dead biomass increase tremendously because of the relatively high nutrient content of the leaves and wood brought down by the disturbance (Tanner et al., 1991). Frangi and Lugo (1991) found that the amounts of nitrogen, phosphorus, potassium, calcium, and magnesium in aboveground necromass had increased by 19%, 18%, 17%, 23%, and 16% respectively, 7 months after Hurricane Hugo. Whigham et al. (1991) measured an increase of approximately 50% in coarse woody debris in a dry tropical forest following Hurricane Gilbert. Standing stocks of calcium, potassium, magnesium, and nitrogen increased between 22% and 36% and phosphorus and magnesium increased by more than 55%. The amount of leaf material deposited by Hurricane Gilbert was 29–98% higher than the total annual amounts of leaf litter-fall measured for the four years prior to the hurricane (Whigham et al., 1991). The concentrations and total amounts of nutrients in leaf litter-fall associated with Hurricane Gilbert were in many instances 100–300% higher than they had been in the four previous non-hurricane years (Table 8.6). Release of nutrients through leaching and decomposition of leaves and wood would potentially increase nutrient availability, as well as losses of nutrients to groundwater and surface water, and to the atmosphere by gas emission. There have been, however, few studies in which nutrient cycling has been examined at the ecosystem level in tropical and subtropical forests following disturbances similar to those caused by wind damage (Parker, 1985; Vitousek and Denslow, 1986; Uhl et al., 1988a; Lodge and McDowell, 1991; Walker, 1999). Hartshorn (1978), Whitmore (1978) and Orians (1982) each predicted that soil properties would change following canopy disturbance, and that tree uprooting in single tree-fall gaps would also change

the characteristics of exposed soils (Schaetzl et al., 1989). Vitousek and Denslow (1986) found higher soil moisture in tree-fall gaps, but no significant increase in soil nutrients between gap and non-gap habitats, even though soils associated with uprooted trees had lower nitrogen, phosphorus and carbon, and higher rates of nitrogen mineralization. Parker (1985) found similar results in disturbed and undisturbed areas in the same forest type as that studied by Vitousek and Denslow. Uhl et al. (1988a) found no evidence to support the hypothesis “that treefall gaps might represent zones of high nutrient leakage” in humid tropical forests in Venezuela. There is some evidence for increased nutrient availability in larger wind-disturbed areas (Lodge and McDowell, 1991) but not as much as occurs in response to forest clear-cutting (Steudler et al., 1991). In addition, many of the changes in nutrient cycling in response to large-scale disturbances such as hurricanes appear to have little long-term impact on ecosystem function. Table 8.6 Biomass and nutrients (totals and concentrations) in leaf litter-fall for Hurricane Gilbert compared to the range of values measured in the 4 years prior to the hurricane 1 Leaf litter

Percent. increase 2

Biomass

29–98

Phosphorus Total

133–250

Concentratiion

60–166

Potassium Total

92–189

Concentration

143–382

Calcium Total

43–133

Concentration

8–17

Magnesium Total

33–100

Concentration

0–(−6)

Manganese

1

Total

50–200

Concentration

2–11

Data compiled from Whigham et al. (1991). Percent increase = 100 (x1 − x0 )/x0 ; where x0 is the mass or concentration before the hurricane, and x1 during it. 2

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS

As indicated earlier, many of the ecosystem responses related to nutrient cycling can be ascribed to large inputs of coarse woody debris and the release of nutrients through leaching and decomposition (Frangi and Lugo, 1991; Harmon et al., 1994; Smith et al., 1994). In this context, the Luquillo Experimental Forest in Puerto Rico has been the most intensively studied site (Zimmerman et al., 1996). Sanford et al. (1991) used simulation modeling to predict that an initial decrease in nitrogen mineralization would be followed by higher rates following hurricane damage. Their predictions with respect to nitrogen dynamics were verified by Lodge et al. (1991), who measured elevated soil ammonium concentrations and higher rates of nitrification 17 and 7 months, respectively, after Hurricane Hugo, and increased mineralization was later suggested by increased nitrate levels in stream water. Sanford et al. (1991) also predicted that a period of phosphorus immobilization after the hurricane would be followed in about two years by an increase in phosphorus availability. Soil phosphorus availability was not measured at the Luquillo site following Hurricane Hugo, but Lodge and McDowell (1991) used results from a long-term fertilization study at Luquillo to suggest that responses to increased amounts of phosphorus occur, but only after several years of fertilizer addition. This conclusion is supported by long-term studies of phosphorus addition to a tropical dry forest in the Yucatan (Whigham and Lynch, 1998). There was no detectable response in tree growth, leaf litter-fall biomass and nutrients to four years (1984–1988) of phosphorus fertilization of the forest (Whigham et al., 1998). One year after Hurricane Gilbert (1988), phosphorus concentrations in leaf litterfall were higher in each of four years after the hurricane than they had been during four pre-hurricane years. It is unclear, however, whether the increased nutrient levels in leaf litter-fall were due to increased phosphorus availability following four years of fertilization and/or to responses to Hurricane Gilbert. Reduced fine-root biomass and added labile carbon in disturbed sites (Parrotta and Lodge, 1991) seem to influence the rates of emission of gases containing nitrogen and carbon. Steudler et al. (1991) consistently found lower rates of loss of carbon dioxide, decreased rates of emission of methane, and higher rates of nitrous oxide production in hurricanedisturbed areas than in undisturbed sites. All these trends are consistent with the hypothesis that lower oxygen levels in the soil alter microbially-mediated

245

processes such as denitrification, methanogenesis, and soil respiration. We know of no data in which disturbance sites have been monitored until there is no evidence of the past disturbance. The studies of Bruenig (1989) and Whitmore (1989) suggest, however, that tropical and subtropical forests are very dynamic and that few areas remain in equilibrium for long periods of time. Zimmerman et al. (1996) provided a conceptual view of temporal patterns of recovery following catastrophic disturbance. Several authors have noted that it is possible to identify large-scale disturbance sites decades after the event occurred (e.g., Richards, 1996). Even though there have been few long-term studies of ecosystem-level processes in single treefall gaps or in areas impacted by blow-downs or hurricanes, the responses that have been measured are consistent enough to suggest that the impacts last for periods from a few months up to three years, and that ecosystem recovery occurs rapidly (Whigham and Lynch, 1998). We believe that largescale and long-term impacts to wind damaged forests, particularly those damaged by hurricanes, would only occur following massive destruction of biomass by fires. Harmon et al. (1994) examined several sites in the Yucatan Peninsula, and found that sites that had been impacted by the hurricane and by fire had the highest amounts of coarse woody debris, and they estimated that it would take between 30 and 150 years to return to pre-hurricane levels.

CONCLUSIONS

There is ample evidence that wind influences tropical forests, and that the loss of biomass that results directly or indirectly from wind occurs over a range of scales. In this chapter we have chosen to divide the continuum of wind disturbances into background disturbances that result in the death of one or a few trees, and catastrophic events that open larger areas of forest (e.g., wind-throws and damage caused by hurricanes). The division that we have made is, of course, arbitrary, and was done primarily to separate wind disturbances into those that received the most scientific attention (background disturbances) from those that have not been adequately studied (catastrophic disturbances). One obvious conclusion is that there is still little known about the distribution of and long-term responses to catastrophic wind disturbances, which appear to be

246

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

widespread. Except for a series of recent studies of hurricanes (see articles in Biotropica, 23, 1991 and 28, 1996), most of the information that has been compiled about recovery from catastrophic events has been anecdotal or primarily qualitative. Recovery from catastrophic wind events can, however, be quite rapid even after extensive damage (Whigham and Lynch, 1998). With the advent of a more complete coverage of tropical areas using satellites and of improved techniques for evaluating satellite images, it should now be possible to evaluate more adequately the distribution and frequency of occurrence of catastrophic wind disturbances in tropical forests and monitor recovery from them. We focussed our discussion on elements of wind disturbance regimes that have important effects on species, community and ecosystem responses. The following general conclusions are offered: (1) Gap sizes should be largest where there is an abundance of large trees and a low rate of standing mortality, where vines are abundant, and where wind knocks down gap-edge trees. (2) Forests of small stature (dry forests, high-altitude forests, and nutrient-poor forests) should have the smallest gaps, while forests on slightly seasonal sites with adequate soil nutrient levels should have the largest gaps, because the trees are large. (3) Rates of gap formation should be highest where tree architecture is least vulnerable to wind, there is poor anchorage, and standing mortality rates are low, and in otherwise calm regions where gusty winds are most frequent. (4) Soil disturbance should be greatest where uprooting rates are highest. This should occur where anchorage is poor owing to shallow rooting and unstable soils and parent material, and where overall rates of gap formation are highest. (5) Wind is an important cause of background canopy disturbance, and appears to be the primary cause in some, although not all, forests not frequently affected by catastrophic wind and other major disturbances. (6) High rates of background canopy disturbance, as with major disturbance, should lead to a reduction in the susceptibility of the forest to damage, because when a large or vulnerable tree is felled, it is replaced by a younger individual that is less likely to be prone to damage. (7) The long-term impacts of catastrophic wind disturbances are poorly understood, especially

ecosystem-level responses, but recent longer-term studies in the Caribbean are showing a wide range of responses at species, community, and ecosystem scales. (8) Wind disturbances may be less important in parts of the Paleotropics, but many more studies are needed to verify this conclusion. (9) There is a need for long-term studies of species, community, and ecosystem recovery from wind disturbances. Most studies, to date, have focussed on the effects of wind disturbances more than on the recovery. (10) The effects of catastrophic winds, because of their long-sustained winds with speeds that are hard to predict, often cannot be predicted from what is known of the factors that influence background canopy-gap disturbance regimes.

ACKNOWLEDGMENTS

The authors would like to thank Lawrence Walker, David Foster, and Sarah Webb for reviewing the manuscript and providing many useful comments and suggestions. We would also like to note the assistance of Pamela Hall and Neil Gale. Funding for some of the research reported by Dickinson and Whigham was provided by the Smithsonian Institution Scholarly Studies Program, U.S. Man and the Biosphere Program, and the National Science Foundation. N. Brokaw’s participation in writing this chapter was supported by grant BSR8811902 from the National Science Foundation to the Institute for Tropical Ecosystem Studies, University of Puerto Rico, and to the International Institute of Tropical Forestry, as part of the Long-Term Ecological Research Program in the Luquillo Experimental Forest. Additional support for Brokaw came from the U.S. Forest Service (Department of Agriculture) and the University of Puerto Rico.

REFERENCES Abrams, M.D. and Scott, M.L., 1989. Disturbance-mediated accelerated succession in two Michigan forest types. For. Sci., 35: 42–49. Ackerman, J.D., Walker, L.R., Scatena, F.N. and Wunderle Jr., J.M., 1991. Ecological effects of hurricanes. Bull. Ecol. Soc. Am., 72: 178–180. Alaka, M.A., 1976. Climatology of Atlantic tropical storms and hurricanes. In: W. Schwerdtfeger (Editor), Climates of Central

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS and South America, World Survey of Climatology. Elsevier, Amsterdam, pp. 479–509. Art, H.W. (Editor), 1993. The Dictionary of Ecology and Environmental Sciences. Henry Holt, New York, 632 pp. Ashton, P.M.S., Gunatilleke, C.V.S. and Gunatilleke, I.A.U.N., 1995. Seedling survival and growth of four Shorea species in a Sri Lankan rainforest. J. Trop. Ecol., 11: 263–279. Ashton, P.S., 1981. The need for information regarding tree age and growth in tropical trees. In: F.H. Bormann and G. Berlyn (Editors), Bulletin of the Yale University School of Forestry and Environmental Studies. Yale University School of Forestry and Environmental Studies, New Haven, pp. 306. Ashton, P.S., 1988. Dipterocarp biology as a window to the understanding of tropical forest tructure. Annu. Rev. Ecol. Syst., 19: 347–370. Ashton, P.W. and Hall, P., 1992. Comparisons of structure among mixed dipterocarp forests of north-western Borneo. J. Ecol., 80: 459–481. Basnet, K., Likens, G.E., Scatena, F.N. and Lugo, A.E., 1992. Hurricane Hugo: damage to a tropical rain forest in Puerto Rico. J. Trop. Ecol., 8: 47–55. Beard, J.S., 1945a. The mora forests of Trinidad, British West Indies. J. Ecol., 33: 173–192. Beard, J.S., 1945b. The progress of plant succession on the Soufri`ere of St. Vincent. J. Ecol., 33: 1–9. Beard, J.S., 1955. The classification of tropical American vegetationtypes. Ecology, 36: 89–100. Bellingham, P.J., 1991. Landforms influence patterns of hurricane damage: evidence from Jamaican montane forests. Biotropica, 23: 427–433. Bellingham, P.J., Tanner, E.V.J. and Healey, J.R., 1994. Sprouting of trees in Jamaican montane forests after a hurricane. J. Ecol., 82: 747–758. Bellingham, P.J., Tanner, E.V.J., Rich, P.M. and Goodland, T.C.R., 1996. Changes in light below the canopy of a Jamaican montane rainforest after a hurricane. J. Trop. Ecol., 12: 699–722. Berner, P.O.B., 1992. Effects of Slope on the Dynamics of a Tropical Montane Oak–Bamboo Forest in Costa Rica. Ph.D. Dissertation, University of Florida, Gainesville, Florida. Boose, E.R., Foster, D.R. and Fluet, M., 1994. Hurricanes and landscape-level disturbance in tropical and temperate forests. Ecol. Monogr., 64: 369–400. Boucher, D.H., Vandermeer, J.H., Mallona, M.A., Zamora, N. and Perfecto, I., 1994. Resistance and resilience in a directly regenerating rainforest: Nicaraguan trees of the Vochysiaceae after Hurricane Joan. For. Ecol. Manage., 68: 127–136. Brand, T. and Parker, V.T., 1995. Scale and general laws of vegetation dynamics. Oikos, 73: 375–380. Brokaw, N.V.L., 1982a. The definition of treefall gap and its effect on measures of forest dynamics. Biotropica, 14: 158–160. Brokaw, N.V.L., 1982b. Treefalls: frequency, timing, and consequences. In: E.G. Leigh Jr., A.S. Rand and D.M. Windsor (Editors), The Ecology of a Tropical Forest: Seasonal Rhythms and Long-term Changes. Smithsonian Institution Press, Washington, D.C., pp. 101–108. Brokaw, N.V.L., 1985a. Gap-phase regeneration in a tropical forest. Ecology, 66: 682–687. Brokaw, N.V.L., 1985b. Treefalls, regrowth, and community structure in tropical forests. In: S.T.A. Pickett and P.S. White (Editors), The

247

Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York, pp. 53–69. Brokaw, N.V.L. and Grear, J.S., 1991. Forest structure before and after Hurricane Hugo at three elevations in the Luquillo mountains, Puerto Rico. Biotropica, 23: 386–392. Brokaw, N.V.L. and Scheiner, S.M., 1989. Species composition in gaps and structure of a tropical forest. Ecology, 70: 538–541. Brokaw, N.V.L. and Walker, L.R., 1991. Summary of the effects of Caribbean hurricanes on vegetation. Biotropica, 23: 442–446. Brown, N., 1993. The implications of climate and gap microclimate for seedlings growth conditions in a Bornean lowland rain forest. J. Trop. Ecol., 9: 153–168. Brown, N., 1996. A gradient of seedling growth from the centre of a tropical rain forest canopy gap. For. Ecol. Manage., 82: 239–244. Brown, N.D. and Whitmore, T.C., 1992. Do dipterocarp seedlings really partition tropical rain forest gaps? Philos. Trans. R. Soc. London Ser. B, 335: 369–378. Bruenig, E., 1989. Forested wetlands in Borneo. In: A.E. Lugo, M. Brinson and S. Brown (Editors), Forested Wetlands. Elsevier, Amsterdam, pp. 293–344. Bultot, F. and Griffiths, J.F., 1972. The equatorial wet zone. In: J.F. Griffiths (Editor), Climates of Africa. Elsevier, Amsterdam, pp. 259–311. Canham, C.D. and Marks, P.L., 1985. The response of woody plants to disturbance: patterns of establishment and growth. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York, pp. 197–216. Canham, C.D., Denslow, J.S., Platt, W.J., Runkle, J.R., Spies, T.A. and White, P.S., 1990. Light regimes beneath closed canopies and tree-fall gaps in temperate and tropical forests. Can. J. For. Res., 20: 620–631. Carey, E.V., Brown, S., Gillespie, A.J.R. and Lugo, A.E., 1994. Tree mortality in mature lowland tropical moist and tropical lower montane moist forests in Venezuela. Biotropica, 26: 255–265. Chandrashekara, U.M. and Ramakrishnan, P.S., 1994. Vegetation and gap dynamics of a tropical wet evergreen forerst in Western Ghats of Kerala, India. J. Trop. Ecol., 10: 337–354. Chazdon, R.L. and Fetcher, N., 1984. Photosynthetic light environments in a lowland tropical rain forest in Costa Rica. J. Ecol., 72: 553–564. Clark, D.A. and Clark, D.B., 1987. An´alisis de la regeneraci´on de a´ rboles del dosel en bosque muy h´umedo tropical: aspectos te´oricos y pr´acticos. Rev. Biol. Trop., 35: 41–54. Clark, D.A. and Clark, D.B., 1992. Life history diversity of canopy and emergent trees in a neotropical rain forest. Ecol. Monogr., 62: 315–344. Clark, D.B., 1990. The role of disturbance in the regeneration of neotropical moist forests. In: K.S. Bawa and M. Hadley (Editors), Reproductive Ecology of Tropical Forest Plants. The Parthenon Publishing Group, Park Ridge, New Jersey, pp. 291–315. Clark, D.B., 1996. Abolishing virginity. J. Trop. Ecol., 12: 735–739. Clark, D.B. and Clark, D.A., 1996. Abundance, growth and mortality of very large trees in neotropical lowland rain forest. For. Ecol. Manage., 80: 235–244. Condit, R., Hubbell, S.P. and Foster, R.B., 1995. Mortality rates of 205 Neotropical tree and shrub species and the impact of a severe drought. Ecol. Monogr., 65: 419–439.

248

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

Connell, J.H., 1989. Some processes affecting the species composition in forest gaps. Ecology, 70: 560–562. Cuevas, E., 1995. Biology of the belowground system of tropical dry forests. In: S.H. Bullock, H.A. Mooney and E. Median (Editors), Seasonally Dry Tropical Forests. Cambridge University Press, Cambridge, pp. 362–383. Dahir, S.E. and Lorimer, C.G., 1996. Variation in canopy gap formation among developmental stages of northern hardwood forests. Can. J. For. Res., 26: 1875–1892. Dale, W.L., 1959. The rainfall of Malaya I. J. Trop. Geogr., 13: 23–37. Denslow, J.S., 1980. Gap partitioning among tropical rainforest trees. Biotropica, 12: 47–55, supplement. Denslow, J.S., 1987. Tropical rainforest gaps and tree species diversity. Annu. Rev. Ecol. Syst., 18: 431–451. Denslow, J.S. and Hartshorn, G.S., 1994. Tree-fall gap environments and forest dynamic processes. In: L.A. McDate, K.S. Brown, H.A. Hespenheide and G.S. Hartshorn (Editors), La Selva: Ecology and Natural History of a Neotropical Rain Forest. University of Chicago Press, Chicago, pp. 120–127. Dickinson, M.D., Whigham, D.F. and Hermann, S.M., 1999. Tree regeneration in natural treefall and felling disturbance in a selectively lugged semideciduous tropical forest in southeastern Mexico. For. Ecol. Manage., in press. Dunn, C.P., Guntenspergen, G.R. and Dorney, J.R., 1983. Catastrophic wind disturbance in an old-growth hemlock–hardwood forest, Wisconsin. Can. J. Bot., 61: 211–217. Durrieu de Madron, L., 1994. Mortalit´e des arbres en forˆet primaire de Guyane Fran¸caise. Bois For. Trop., 239: 43–55. Ellenberg, H., 1979. Man’s influence on tropical mountain ecosystems in South America. J. Ecol., 67: 401–416. Encyclopædia Britannica, Inc., 1992. The New Encyclopaedia Britannica, Vol. 9. The University of Chicago, Chicago, pp. 58– 64. Everham III, E.M. and Brokaw, N.V.L., 1996. Forest damage and recovery from catastrophic wind. Bot. Rev., 62: 113–185. Ewel, J., 1977. Differences between wet and dry successional tropical ecosystems. Geo-Eco-Trop, 1: 103–117. Ewel, J., 1980. Tropical succession: manifold roots to maturity. Biotropica, 12: 2–7, supplement. Fern´andez, D.S. and Fetcher, N., 1991. Changes in light availability following Hurricane Hugo in a subtropical montane forest in Puerto Rico. Biotropica, 23: 393–399. Foster, D.R., 1988. Species and stand response to catastrophic wind in central New England. J. Ecol., 76: 135–151. Foster, D.R. and Boose, E.R., 1992. Patterns of forest damage resulting from catastrophic wind in central New England, U.S.A. J. Ecol., 80: 79–98. Foster, R.B., 1990. Long-term change in the successional forest community of the R´ıo Manu floodplain. In: A.H. Gentry (Editor), Four Neotropical Forests. Yale University Press, New Haven, Connecticut, pp. 565–572. Frangi, J.L. and Lugo, A.E., 1991. Hurricane damage to a flood plain forest in the Luquillo Mountains of Puerto Rico. Biotropica, 23: 324–335. Fraver, S., Brokaw, N.V.L. and Smith, A.P., 1998. Delimiting the gap phase in the growth cycle of a Panamanian forest. J. Trop. Ecol., 14: 673–681. Fu, S., Pedraza, C.R. and Lugo, A.E., 1996. A twelve-year comparison

of stand changes in a mahogany plantation and a paired natural forest of similar age. Biotropica, 28: 515–524. Garwood, N.C., 1989. Tropical soil seed banks: A review. In: M.A. Leck, V.T. Parker and R.L. Simpson (Editors), Ecology of Soil Seed Banks. Academic Press, New York, pp. 149–210. Glitzenstein, J.S. and Harcombe, P.A., 1988. Effects of the December 1983 tornado on forest vegetation of the Big Thicket, southeast Texas, U.S.A. For. Ecol. Manage., 25: 269–290. Gloyne, R.W., 1968. The structure of wind and its relevance to forestry. For. Supplement, 20: 7–19. Grace, J., 1977. Plant Response to Wind. Academic Press, London. Grime, J.P., 1979. Plant Strategies and Vegetation Processes. John Wiley, New York. Grubb, P.J., 1989. The role of mineral nutrients in the tropics: a plant ecologist’s view. In: J. Proctor (Editor), Mineral Nutrients in Tropical Forest and Savanna Ecosystems. Blackwell Scientific Publications, Oxford, pp. 417–439. Grubb, P.J., 1996. Rainforest dynamics: the need for new paradigms. In: D.S. Edwards (Editor), Tropical Rainforest Research – Current Issues. Kluwer Academic Publishers, Dordrecht, pp. 215–233. Guariguata, M.R., 1990. Landslide disturbance and forest regeneration in the upper Luquillo Mountains of Puerto Rico. J. Ecol., 78: 814–832. Gullison, R.E., Panfil, S.N., Strouse, J.J. and Hubbell, S.P., 1996. Ecology and management of mahogany (Swietenia macrophylla King) in the Chimanes Forest, Beni, Bolivia. Bot. J. Linn. Soc., 122: 9–34. Guzm´an-Grajales, S. and Walker, L.R., 1991. Differential seedling responses to litter after Hurricane Hugo in the Luquillo Experimental Forest, Puerto Rico. Biotropica, 23: 407–413. Hall, P., 1991. Structure, Stand Dynamics and Species Compositional Change in Three Mixed Dipterocarp Forests of Northwest Borneo. Ph.D. Dissertation, Boston University, Boston, Massachusetts, 701 pp. Harmon, M.E., Whigham, D.F., Sexton, J. and Olmsted, I., 1994. Decomposition and mass of woody debris in the dry tropical forests of northeastern Yucatan Peninsula, Mexico. Biotropica, 27: 305–316. Hartshorn, G.S., 1978. Tree falls and tropical forest dynamics. In: P.B. Tomlinson and M.H. Zimmerman (Editors), Tropical Trees as Living Systems. Cambridge University Press, Cambridge, pp. 617– 638. Hartshorn, G.S., 1980. Neotropical forest dynamics. Biotropica, 12: 23–30, supplement. Hartshorn, G.S., 1989. Gap-phase dynamics and tropical tree species richness. In: L.B. Holm-Nielsen, I.C. Nielsen and H. Balslev (Editors), Tropical Forests: Botanical Dynamics, Speciation, and Diversity. Academic Press, London, pp. 65–73. Hartshorn, G.S., 1990. An overview of neotropical forest dynamics. In: A.H. Gentry (Editor), Four Neotropical Rainforests. Yale University Press, New Haven, pp. 585–599. Hegarty, E.E. and Caball´e, G., 1991. Distribution and abundance of vines in forest communities. In: H. Mooney and F.E. Putz (Editors), Biology of Vines. Cambridge University Press, Cambridge, pp. 313–335. Holbrook, N.M. and Putz, F.E., 1989. Influence of neighbors on tree form: effects of lateral shade and prevention of sway on the allometry of Liquidambar styraciflua (Sweet Gum). Am. J. Bot., 76: 1740–1749.

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS Holdridge, L.R., Grenke, W.C., Hatheway, W.H., Liang, T. and Tosi, J.A., 1971. Forest Environments in Tropical Life Zones. A Pilot Study. Pergamon Press, New York, 747 pp. Hubbell, S.P. and Foster, R.B., 1986. Canopy gaps and the dynamics of a neotropical forest. In: M.J. Crawley (Editor), Plant Ecology. Blackwell Scientific Publications, Oxford, pp. 77–96. Imbert, D., Labbe, P. and Rousteau, A., 1996. Hurricane damage and forest structure in Guadeloupe, French West Indies. J. Trop. Ecol., 12: 663–680. Jans, L., Poorter, L., van Rompaey, R.S.A.R. and Bongers, F., 1993. Gaps and forest zones in tropical moist forest in Ivory Coast. Biotropica, 25: 258–269. Johns, R.J., 1986. The instability of the tropical ecosystems in New Guinea. Blumea, 31: 341–371. Jones, E.W., 1955a. Ecological studies on the rain forest of southern Nigeria IV. J. Ecol., 43: 564–594. Jones, E.W., 1955b. Ecological studies on the rain forest of southern Nigeria IV (contd.). J. Ecol., 44: 83–117. Kapos, V., Pallant, L., Bien, A. and Freskos, S., 1990. Gap frequencies in lowland rain forest sites on contrasting soils in Amazonian Ecuador. Biotropica, 22: 218–225. Kellman, M. and Tackaberry, R., 1993. Disturbance and tree species coexistence in tropical riparian forest fragments. Global Ecol. Biogeogr. Lett., 3: 1–9. Kennedy, D.N. and Swaine, M.D., 1992. Germination and growth of colonizing species in artificial gaps of different sizes in dipterocarp rain forest. Philos. Trans. R. Soc. London Ser. B, 335: 357–366. King, D.A., 1991. Tree size. Natl. Geogr. Res. Explor., 7: 342–351. King, D.A., 1996. Allometry and life history of tropical trees. J. Trop. Ecol., 12: 25–44. Kira, T., 1978. Community architecture and organic matter dynamics in tropical lowland rain forests of Southeast Asia with special reference to Pasoh Forest, West Malaysia. In: P.B. Tomlinson and M.H. Zimmerman (Editors), Tropical Trees as Living Systems. Cambridge University Press, Cambridge, pp. 561–590. Krasny, M.E. and Whitmore, M.C., 1992. Gradual and sudden forest canopy gaps in Alllegheny northern hardwood forests. Can. J. For. Res., 22: 139–143. Lauer, W., 1983. Climate and weather. In: H. Lieth and M.J.A. Werger (Editors), Tropical Rain Forest Ecosystems. Elsevier, Amsterdam, pp. 7–54. Lawton, R.O., 1990. Canopy gaps and light penetration into a windexposed tropical lower montane rain forest. Can. J. For. Res., 20: 659–667. Lawton, R.O. and Dryer, V., 1980. The vegetation of the Monteverde Cloud Forest Reserve. Brenesia, 18: 101–116. Lawton, R.O. and Putz, F.E., 1988. Natural disturbance and gap-phase regeneration in a wind-exposed tropical cloud forest. Ecology, 69: 764–777. Leighton, M. and Wirawan, N., 1986. Catastrophic drought and fire in Borneo tropical rain forest associated with the 1982–1983 El Nino Southern Oscillation event. In: G.T. Prance (Editor), Tropical Rain Forests and the World Atmosphere. Westview Press, Boulder, Colorado, pp. 75–102. Levey, D.J., Moermond, T.C. and Denslow, J.S., 1994. Frugivory: an overview. In: L.A. McDade, K.S. Brown, H.A. Hespenheide and G.S. Hartshorn (Editors), La Selva: Ecology and Natural History of a Neotropical Rainforest. University of Chicago Press, Chicago, pp. 282–294.

249

Lieberman, D. and Li, M., 1992. Seedling recruitment patterns in a tropical dry forest in Ghana. J. Vegetation Sci., 3: 375–382. Lieberman, D. and Lieberman, M., 1987. Forest tree growth and dynamics at La Selva, Costa Rica (1969–1982). J. Trop. Ecol., 3: 347–358. Lieberman, D., Lieberman, M., Peralta, R. and Hartshorn, G.S., 1985. Mortality patterns and stand turnover rates in a wet tropical forest in Costa Rica. J. Ecol., 73: 915–924. Lieberman, D., Hartshorn, G.S., Lieberman, M. and Peralta, R., 1990. Forest dynamics at La Selva Biological Station, Costa Rica, 1969– 1985. In: A.H. Gentry (Editor), Four Neotropical Forests. Yale University Press, New Haven, pp. 509–521. Lieberman, D., Lieberman, M., Peralta, R. and Hartshorn, G.S., 1996. Tropical forest structure and composition on a large-scale altitudinal gradient in Costa Rica. J. Ecol., 84: 137–152. Lieberman, M., Lieberman, D. and Peralta, R., 1989. Forests are not just swiss cheese: canopy sterogeometry of non-gaps in tropical forests. Ecology, 70: 550–552. Lieberman, M., Lieberman, D., Peralta, R. and Hartshorn, G.S., 1995. Canopy closure and the distribution of tropical forest tree species in La Selva, Costa Rica. J. Trop. Ecol., 11: 161–178. Lindo, L.S., 1968. The effect of hurricanes on the forests of British Honduras. Commonw. For. Rev., 47: 248. Lodge, D.J. and McDowell, W.H., 1991. Summary of ecosystem-level effects of Caribbean hurricanes. Biotropica, 23: 373–378. Lodge, D.J., Scatena, F.N., Asbury, C.E. and S´anchez, M.J., 1991. Fine litterfall and related nutrient inputs resulting from Hurricane Hugo in subtropical wet and lower montane rain forests in Puerto Rico. Biotropica, 23: 336–342. L´opez Ornat, A., 1983. Localizaci´on y medio f´ısico. In: Centro de Investigaciones de Quintana Roo, Sian Ka’an. Estudios Preliminares de una zona en Quintana Roo propresta como Reserva de la Bi´osfera. Centro de Investigaciones de Quintana Roo, Puerto Moreloa, Quintana Roo, Mexico, pp. 19–49. Lorimer, C.G., 1989. Relative effects of small and large disturbances on temperate hardwood forest structure. Ecology, 70: 565–567. Lugo, A.E., 1992. Comparison of tropical tree plantations with secondary forests of similar age. Ecol. Monogr., 62: 1–41. Lugo, A.E., 1995. Reconstructing hurricane passages over forests: a tool for understanding multiple-scale responses to disturbance. Trends Ecol. Evol., 10: 98–99. Lugo, A.E. and Scatena, F.N., 1996. Background catastrophic tree mortality in tropical moist, wet, and rain forests. Biotropica, 28: 505–589. Lugo, A.E., Applefield, M., Pool, D.J. and McDonald, R.B., 1983. The impact of Hurricane David on the forests of Dominica. Can. J. For. Res., 13: 201–211. Lugo, A.E., Bokkestijn, A. and Scatena, F.N., 1995. Structure, succession, and soil chemistry of palm forests in the Luquillo Experimental Forest. In: A.E. Lugo and C. Lowe (Editors), Tropical Forests: Management and Ecology. Springer-Verlag, New York, pp. 142–177. Magnusson, W.E., Lima, A.P. and de Lima, O., 1996. Group lightning mortality of trees in a Neotropical forest. J. Trop. Ecol., 12: 899–904. Mart´ınez-Ramos, M., Alvarez-Buylla, E.R., Sarukh´an, J.K. and Pi˜nero, D., 1988. Treefall age determination and gap dynamics in a tropical forest. J. Ecol., 76: 700–716. Matelson, T.J., Nadkarni, N.M. and Solano, R., 1995. Tree damage

250

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

and annual mortality in a Montane forest in Monteverde, Costa Rica. Biotropica, 27: 441–447. Merrens, E.J. and Peart, D.R., 1992. Effects of hurricane damage on individual growth and stand structure in a hardwood forest in New Hampshire, U.S.A. J. Ecol., 80: 787–795. Midgley, J.J., Cameron, M.C. and Bond, W.J., 1995. Gap characteristics and replacement patterns in the Knysna Forest, South Africa. J. Vegetation Sci., 6: 29–36. Molofsky, J. and Augspurger, C.K., 1992. The effect of leaf litter on early seedling establishment in a tropical forest. Ecology, 73: 68–77. Mori, S.A. and Becker, P., 1991. Flooding affects survival of Lecythidaceae in Terra Firme Forest near Manaus, Brazil. Biotropica, 23: 87–90. Murphy, P.G. and Lugo, A.E., 1986. Ecology of tropical dry forest. Annu. Rev. Ecol. Syst., 17: 67–88. Nelson, B.W., Kapos, V., Adams, J.B., Oliveira, W.J., Braun, O.P.G. and do Amaral, I.L., 1994. Forest disturbance by large blowdowns in the Brazilian Amazon. Ecology, 75: 853–858. Noble, R.R. and Slatyer, R.O., 1980. The use of vital attributes to predict successional changes in plant communities subject to recurrent disturbances. Vegetatio, 43: 5–21. Odum, H.T., 1970. Rain forest structure and mineral cycling homeostatis. In: H.T. Odum and R.F. Pigeon (Editors), A Tropical Rain Forest. National Technical Information Service, Springfield, pp. H3–H52. Oldeman, R.A.A., 1978. Architecture and energy exchange of dicotyledonous trees in the forest. In: P.B. Tomlinson and M.H. Zimmerman (Editors), Tropical Trees as Living Systems. Cambridge University Press, Cambridge, pp. 535–560. Oldeman, R.A.A., 1989. Dynamics in tropical rain forests. In: L.B. Holm-Nielsen, I.C. Nielsen and H. Balselev (Editors), Tropical Forests: Botanical Dynamics, Speciation, and Diversity. Academic Press, New York, pp. 3–21. Oliver, C.D. and Larson, B.C., 1990. Forest Stand Dynamics. McGraw-Hill, Inc., New York, 544 pp. Orians, G.H., 1982. The influence of tree-falls in tropical forests in tree species diversity. Trop. Ecol., 23: 255–279. Parker, G., 1985. The Effect of Disturbance on Water and Solute Budgets of Hill-Slope Tropical Rainforests in Northeastern Costa Rica. Ph.D. Dissertation, University of Georgia, Athens, 161 pp. Parrotta, J.A., 1995. Influence of overstory composition on understory colonization by native species in plantations on a degraded tropical site. J. Vegetation Sci., 6: 627–636. Parrotta, J.A. and Lodge, D.J., 1991. Fine root dynamics in a subtropical wet forest following hurricane disturbance in Puerto Rico. Biotropica, 23: 343–347. Pendry, C.A. and Proctor, J., 1996. The causes of altitudinal zonation of rain forests on Bukit Belalong, Brunei. J. Ecol., 84: 407–418. Peterson, C.J. and Carson, W.P., 1996. Generalizing forest regeneration models: the dependence of propagule availability on disturbance history and stand size. Can. J. For. Res., 26: 45–52. Phillips, O.L. and Gentry, A.H., 1994. Increasing turnover through time in tropical forests. Science, 263: 954–958. Phillips, O.L., Hall, P., Gentry, A.H., Sawyer, S.A. and Vasquez, R., 1994. Dynamics and species richness of tropical rain forests. Proc. Natl. Acad. Sci. USA, 91: 2805–2809. Pickett, S.T.A. and McDonnell, M.J., 1989. Changing perspectives

in community dynamics: a theory of successional forces. Trends Ecol. Evol., 4: 241–245. Pickett, S.T.A. and White, P.S. (Editors), 1985. Patch dynamics: a synthesis. In: The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York, pp. 371–384. Pickett, S.T.A., Collins, S.L. and Armesto, J.J., 1987. Models, mechanisms, and pathways of succession. Bot. Rev., 53: 335–371. Pickett, S.T.A., Kolasa, J., Armesto, J.J. and Collins, S.L., 1989. The ecological concept of disturbance and its expression at various hierarchical levels. Oikos, 54: 129–136. Platt, W.J. and Schwartz, M.W., 1990. Temperate hardwood forests. In: R.L. Myers and J.J. Ewel (Editors), Ecosystems of Florida. University of Central Florida Press, Orlando, pp. 194–229. Poore, M.E.D., 1968. Studies in Malaysian rain forest 1. The forest on triassic sediments in Jengka Forest Reserve. J. Ecol., 56: 143–196. Poorter, L., Jans, L., Bongers, F. and van Rompaey, R.S.A.R., 1994. Spatial distribution of gaps along three catenas in the moist forest of Tai National Park, Ivory Coast. J. Trop. Ecol., 10: 385–398. Popma, J., Bongers, F., Mart´ınez-Ramos, M. and Veneklass, E., 1988. Pioneer species distribution in treefall gaps in Neotropical rain forest; a gap definition and its consequences. J. Trop. Ecol., 4: 77–88. Putz, F.E., 1980. Lianas vs. trees. Biotropica, 12: 224–225. Putz, F.E., 1983. Treefall pits and mounds, buried seeds, and the importance of soil disturbance to pioneer trees on Barro Colorado Island, Panama. Ecology, 64: 1069–1074. Putz, F.E., 1984. The natural history of lianas on Barro Colorado Island, Panama. Ecology, 65: 1713–1724. Putz, F.E., 1985. Woody vines and forest management in Malaysia. Commonw. For. Rev., 64: 359–365. Putz, F.E., 1991. Silvicultural effects of lianas. In: F.E. Putz and H.A. Mooney (Editors), The Biology of Vines. Cambridge University Press, Cambridge, pp. 493–501. Putz, F.E. and Appanah, S., 1987. Buried seeds, newly dispersed seeds, and the dynamics of a lowland forest in Malaysia. Biotropica, 19: 326–333. Putz, F.E. and Brokaw, N.V.L., 1989. Sprouting of broken trees on Barro Colorado Island, Panama. Ecology, 70: 508–512. Putz, F.E. and Canham, C.D., 1992. Mechanisms of arrested succession in shrublands: root and shoot competition between shrubs and tree seedlings. For. Ecol. Manage., 49: 267–275. Putz, F.E. and Milton, K., 1982. Tree mortality rates on Barro Colorado Island. In: E.G. Leigh, A.S. Rand and D.M. Windsor (Editors), Ecology of a Tropical Forest: Seasonal Rhythms and Long Term Changes. Smithsonian Institution Press, Washington, DC, pp. 95–100. Putz, F.E. and Sharitz, R.R., 1991. Hurricane damage to old-growth forest in Congaree Swamp National Monument, South Caroliana, U.S.A. Can. J. For. Res., 21: 1765–1770. Putz, F.E., Coley, P.D., Lu, K., Montalvo, A. and Aiello, A., 1983. Uprooting and snapping of trees: structural determinants and ecological consequences. Can. J. For. Res., 13: 1011–1020. Raich, J.W., 1989. Seasonal and spatial variation in the light environment in a tropical Dipterocarp forest and gaps. Biotropica, 21: 299–302. Raich, J.W. and Christensen, N.L., 1989. Malaysian dipterocarp forest: Tree seedling and sapling species composition and smallscale disturbance patterns. Natl. Geogr. Res., 5: 348–363. Rankin de Merona, J.M., Hutchings, H.R.W. and Lovejoy, T.E.,

BACKGROUND CANOPY GAP AND CATASTROPHIC WIND DISTURBANCES IN TROPICAL FORESTS 1990. Tree mortality and recruitment over a five-year period in undisturbed upland rainforest of the central Amazon. In: A.H. Gentry (Editor), Four Neotropical Forests. Yale University Press, New Haven, pp. 573–584. Richards, P.W., 1996. The Tropical Rain Forest – An Ecological Study. Cambridge University Press, Cambridge, 575 pp. Ri´era, B., 1995. Rˆole des perturbations actuelles et pass´ees dans la dynamique et la mosa¨ıque foresti`ere. Rev. Ecol. Terre Vie, 50: 209–222. Riswan, S., Kenworthy, J.B. and Kartawinata, K., 1985. The estimation of temporal processes in tropical rain forest: a study of primary mixed dipterocarp forest in Indonesia. J. Trop. Ecol., 1: 171–182. Roth, E.R. and Hepting, G.H., 1943. Origin and development of oak stump sprouts as affecting their likelihood to decay. J. For., 41: 27–36. Roth, L.C., 1992. Hurricanes and mangrove regeneration: Effects of Hurricane Joan, October 1988, on the vegetation of Isla del Venado, Bluefields, Nicaragua. Biotropica, 24: 375–384. Runkle, J.R., 1990. Gap dynamics in an Ohio Acer–Fagus forest and speculations on the geography of disturbance. Can. J. For. Res., 20: 632–641. Runkle, J.R. and Yetter, T.C., 1987. Treefalls revisited: gap dynamics in the southern Appalachians. Ecology, 68: 417–424. Rykiel Jr., E.J., 1985. Towards a definition of ecological disturbance. Aust. J. Ecol., 10: 361–365. Saldarriaga, J.G., West, D.C., Tharp, M.L. and Uhl, C., 1988. Longterm chronosequence of forest succession in the upper R´ıo Negro of Colombia and Venezuela. J. Ecol., 76: 938–958. Sanford Jr., R.L., Braker, H.E. and Hartshorn, G.S., 1986. Canopy gaps in a primary neotropical lowland forest. J. Trop. Ecol., 2: 277–282. Sanford Jr., R.L., Parton, W.J., Ojima, D.S. and Lodge, D.J., 1991. Hurricane effects on soil organic matter dynamics and forest production in the Luquillo Experimental Forrest, Puerto Rico: results of simulation modeling. Biotropica, 23: 364–372. Scatena, F.N. and Lugo, A.E., 1995. Geomorphology, disturbance, and the soil and vegetation of two subtropical wet steepland watersheds of Puerto Rico. Geomorphology, 13: 199–213. Scatena, F.N., Moya, S., Estrada, C. and Chinea, J.D., 1996. The first five years in the reorganization of aboveground biomass and nutrient use following Hurricane Hugo in the Bisley Experimental Watersheds, Luquillo Experimental Forest, Puerto Rico. Biotropica, 28: 424–440. Schaetzl, R.J., Burns, S.F., Johnson, D.L. and Small, T.W., 1989. Tree uprooting: review of impacts on forest ecology. Vegetatio, 79: 165–176. Schupp, E.W. and Fuentes, M., 1995. Spatial patterns of seed dispersal ´ and the unification of plant population ecology. Ecoscience, 2: 267–275. Schupp, E.W., Howe, H.F., Augspurger, C.K. and Levey, D.J., 1989. Arrival and survival in tropical treefall gaps. Ecology, 70: 562–564. Schwerdtfeger, W., 1976. Climates of Central and South America. Elsevier, Amsterdam, 532 pp. Seth, K.K., Waheed Khan, M.A. and Yadav, J.S.P., 1960. Sal mortality in Bihar. Indian For., 86: 645–679. Shaw, W.B., 1983. Tropical cyclones: determinants of pattern and

251

structure in New Zealand’s indigenous forests. Pac. Sci., 37: 405–414. Shigo, A.L., 1984. Compartmentalization: a conceptual framework for understanding how trees grow and defend themselves. Annu. Rev. Phytopathol., 22: 189–214. Silver, W.L., Scatena, F.N., Johnson, A.H., Siccama, T.G. and Watt, F., 1996. At what temporal scales does disturbance affect belowground nutrient pools? Biotropica, 28: 441–457. Smith III, T.J., Robblee, M.B., Wanless, H.R. and Doyle, T.W., 1994. Mangroves, hurricanes, and lightning strikes. Bioscience, 44: 256–262. Snook, L.K., 1993. Stand Dynamics of Mahogany (Swietenia macrophylla King) and Associated Species after Fire and Hurricane in the Tropical Forests of the Yucatan Peninsula, Mexico. Ph.D. Dissertation. Yale University, New Haven, Connecticut, 254 pp. Sousa, W.P., 1984. The role of disturbance in natural communities. Annu. Rev. Ecol. Syst., 15: 353–391. Spies, T.A. and Franklin, J.F., 1989. Gap characterisitcs and vegetation response in coniferous forests of the Pacific Northwest. Ecology, 70: 543–545. Spurr, S.H., 1956. Natural restocking of forests following the 1938 hurricane in central New England. Ecology, 37: 443–451. Steudler, P.A., Melillo, J.M., Bowden, R.D., Castro, M.S. and Lugo, A.E., 1991. The effects of natural and human disturbances on soil nitrogen dynamics and trace gas fluxes in a Puerto Rican wet forest. Biotropica, 23: 356–363. Stevens, G.C. and Perkins, A.L., 1992. The branching habits and life history of woody plants. Am. Nat., 139: 267–275. Strong Jr., D.R., 1977. Epiphyte loads, tree falls, and perennial forest disruption: a mechanism for maintaining higher tree species richness in the tropics without animals. J. Biogeogr., 4: 215–218. Swaine, M.D., 1992. Characterisitcs of dry forest in West Africa and the influence of fire. J. Vegetation Sci., 3: 365–374. Swaine, M.D. and Whitmore, T.C., 1988. On the definition of ecological species groups in tropical rain forests. Vegetatio, 75: 81–86. Swaine, M.D., Lieberman, D. and Putz, F.E., 1987. The dynamics of tree populations in tropical forest: a review. J. Trop. Ecol., 3: 359–366. Tanner, E.V.J., Kapos, V. and Healey, J.R., 1991. Hurricane effects on forest ecosystems in the Caribbean. Biotropica, 23: 513–521. Telewski, F.W., 1995. Wind-induced physiological and developmental responses in trees. In: M.P. Coutts and J. Grace (Editors), Wind and Trees. Cambridge University Press, Cambridge, pp. 237–263. Terborgh, J., 1985. The vertical component of plant species diversity in temperate and tropical forests. Am. Nat., 126: 760–776. Touliatos, P. and Roth, E., 1971. Hurricanes and trees: ten lessons from Camille. J. For., May: 285–288. Uhl, C., 1982a. Recovery following disturbances of different intensities in the Amazon rain forest of Venezuela. Intercienca, 7: 19–24. Uhl, C., 1982b. Tree dynamics in a species rich tierra firme forest in Amazonia, Venezuela. Acta Cient. Venez., 33: 72–77. Uhl, C., Clark, K., Dezzeo, N. and Maquirino, P., 1988a. Vegetation dynamics in Amazonian treefall gaps. Ecology, 69: 751–763. Uhl, C., Kauffman, J.B. and Cummings, D.L., 1988b. Fire in the Venezuelan Amazon 2: Environmental conditions necessary for

252

Dennis F. WHIGHAM, Matthew B. DICKINSON and Nicholas V.L. BROKAW

forest fires in the evergreen rainforest of Venezuela. Oikos, 53: 176–184. van Andel, J. and van den Bergh, J.P., 1987. Disturbance of grasslands: outline of the theme. In: J. van Andel (Editor), Disturbance in Grasslands. Dr W. Junk Publishers, Dordrecht, pp. 3–13. van der Meer, P.J. and Bongers, F., 1996a. Formation and closure of canopy gaps in the rain forest at Nouragues, French Guiana. Vegetatio, 126: 167–179. van der Meer, P.J. and Bongers, F., 1996b. Patterns of tree-fall and branch-fall in a tropical rain forest in French Guiana. J. Ecol., 84: 19–29. van der Valk, A.G., 1981. Succession in wetlands: A Gleasonian approach. Ecology, 62: 688–696. Vandermeer, J.H., Mallona, M.A., Boucher, D.H., Yih, K. and Perfecto, I., 1995. Three years of ingrowth following catastrophic hurricane damage on the Caribbean coast of Nicaragua: evidence in support of the direct regeneration hypothesis. J. Trop. Ecol., 11: 465–471. Vandermeer, J.H., Boucher, D.H., Perfecto, I. and Granzow de la Cerda, I., 1996. A theory of disturbance and species diversity: evidence from Nicaragua after Hurricane Joan. Biotropica, 28: 600–613. Vega, A.J. and Binkley, M.S., 1993. Tropical cyclone formation in the North Atlantic basin, 1960–1989. Climate Res., 3: 221–232. Vitousek, P.M. and Denslow, J.S., 1986. Nitrogen and phosphorus availability in treefall gaps of a lowland tropical rain forest. J. Ecol., 74: 1167–1178. Vooren, A.P., 1986. Nature and origin of tree and branch fall in the Tai Forest (Ivory Coast). Neth. J. Agric. Sci., 34: 112–115. Wadsworth, F.H. and Englerth, G.H., 1959. Effects of the 1956 hurricane on forests in Puerto Rico. Caribb. For., 20: 38–51. Walker, L.R., 1991. Tree damage and recovery from Hurricane Hugo in Luquillo Experimental Forest, Puerto Rico. Biotropica, 23: 379–385. Walker, L.R., 1995. Timing of post-hurricane tree mortality in Puerto Rico. J. Trop. Ecol., 11: 315–320. Walker, L.R., 1999. Seedling and sapling dynamics in treefall pits in Puerto Rico. Biotropica, in press. Walker, L.R., Lodge, D.J., Brokaw, N.V.L. and Waide, R.B., 1991. An introduction to hurricanes in the Caribbean. Biotropica, 23: 313–316. Walker, L.R., Zimmerman, J.K., Lodge, D.J. and Guzm´anGrajales, S., 1996. An altitudinal comparison of growth and species composition in hurricane-damaged forests in Puerto Rico. J. Ecol., 84: 877–889. Weaver, P.L., 1986. Hurricane damage and recovery in montane forest of the Luquillo Mountains of Puerto Rico. Caribb. J. Sci., 22: 53–70. Webb, L.J., 1958. Cyclones as an ecological factor in tropical lowland rainforest, north Queensland. Aust. J. Ecol., 6: 220–228. Welden, C.W., Hewett, S.W., Hubbell, S.P. and Foster, R.B., 1991. Sapling survival, growth, and recruitment: relationship to canopy height in a neotropical forest. Ecology, 72: 35–50. Whigham, D.F. and Lynch, J.F., 1998. Responses of plants and birds to hurricane disturbance in a dry tropical forest in Quintana Roo, Mexico. In: F. Dallmeier and J.A. Comiskey (Editors), Forest Biodiversity in North, Central and South America, and the Caribbean. Research and Monitoring. Parthenon Publishing Group, Pearl River, pp. 165–186.

Whigham, D.F., Zugasty Towle, P., Cabrera Cano, E., O’Neill, J. and Ley, E., 1990. The effect of annual variation in precipitation on growth and litter production in a tropical dry forest in the Yucatan of Mexico. Trop. Ecol., 31: 23–34. Whigham, D.F., Olmsted, I., Cabrera Cano, E. and Harmon, M.E., 1991. The impact of Hurricane Gilbert on trees, litterfall, and woody debris in a dry tropical forest in the northeastern Yucatan Peninsula. Biotropica, 23: 434–441. Whigham, D.F., Lynch, J.F. and Dickinson, M.B., 1998. Dynamics and ecology of natural and managed forests in Quintana Roo, Mexico. In: R. Primack, D. Bray, H. Galletti and I. Ponciano (Editors), Conservation and Community Development in the Maya Forest of Belize, Guatemala, and Mexico. Island Press, pp. 267–281. White, P.S. and Pickett, S.T.A., 1985. Natural disturbance and patch dynamics: an introduction. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York, pp. 3–13. Whitmore, T.C., 1974. Change with Time and the Role of Cyclones in Tropical rain forest on Kolombangara, Solomon Islands. University of Oxford, Oxford, 78 pp. Whitmore, T.C., 1975. Tropical Rain Forests of the Far East. Clarendon Press, Oxford, 282 pp. Whitmore, T.C., 1978. Gaps in the forest canopy. In: P.B. Tomlinson and M.H. Zimmerman (Editors), Tropical Trees as Living Systems. Cambridge University Press, Cambridge, pp. 639–655. Whitmore, T.C., 1984. Tropical Rain Forests of the Far East. Clarendon Press, Oxford, 352 pp. Whitmore, T.C., 1989. Changes over twenty-one years in the Kolombangara rain forests. J. Ecol., 77: 469–483. Wood, T.W.W., 1970. Wind damage in the forest of Western Samoa. Malay. For., 33: 92–99. Woods, P., 1989. Effects of logging, drought, and fire on structure and composition of tropical forests in Sabah, Malaysia. Biotropica, 21: 290–298. Wooldridge, G., Musselman, R. and Massman, W., 1995. Wind throw and airflow in a subalpine forest. In: M.P. Coutts and J. Grace (Editors), Wind and Trees. Cambridge University Press, Cambridge, pp. 358–375. Wyatt-Smith, J., 1954. Storm forest in Kelantan. Malay. For., 17: 5–11. Yavitt, J.B., Battles, J.J., Lang, G.E. and Knight, D.H., 1995. The canopy gap regime in a secondary Neotropical forest in Panama. J. Trop. Ecol., 11: 391–402. Yih, K., Boucher, D.H., Vandermeer, J.H. and Zamora, N., 1991. Recovery of the rainforest of southeastern Nicaragua after destruction by Hurricane Joan. Biotropica, 223: 106–113. You, C. and Petty, W.H., 1991. Effects of Hurricane Hugo on Manilkara bidentata, a primary tree species in the Luquillo Experimental Forest of Puerto Rico. Biotropica, 23: 400–406. Young, T.P. and Hubbell, S.P., 1991. Crown asymmetry, treefalls, and repeat disturbance of broad-leaved forest gaps. Ecology, 72: 1466–1471. Zimmerman, J.K., Everham III, E.M., Waide, R.B., Lodge, D.J., Taylor, C.M. and Brokaw, N.V.L., 1994. Responses of tree species to hurricane winds in subtropical wet forest in Puerto Rico: implications for tropical tree life histories. J. Ecol., 82: 911–922. Zimmerman, J.K., Willig, M.R., Walker, L.R. and Silver, W.L., 1996. Introduction: disturbance and Caribbean ecosystems. Biotropica, 28: 414–423.

Chapter 9

FOREST HERBIVORY: INSECTS T.D. SCHOWALTER and M.D. LOWMAN

INTRODUCTION

Herbivory is the process of feeding on any plant parts, including foliage, stems, roots, flowers, fruits, or seeds. It is important in that it reduces the density of plants or plant materials, opens the canopy, transfers mass and nutrients to the forest floor, stimulates plant growth under certain conditions, and affects habitat and resource conditions for other organisms. Many studies have addressed the effects of disturbance, including ecosystem-management practices, on herbivorous insect species that are economically important (e.g., Schowalter et al., 1986; Mattson and Haack, 1987; Paine and Baker, 1993) and the effects of herbivore outbreaks on plant growth and mortality (reviewed by Schowalter et al., 1986). However, measurements of the magnitude of herbivory in different ecosystems and under different environmental conditions have used various non-comparable techniques, which hinders interpretation (reviewed by Lowman, 1995). Few studies have assessed the relationship between herbivory and disturbances or environmental changes, or the effects of herbivory on other ecosystem processes. Although loss of plant material through herbivory generally is negligible, or at least inconspicuous, periodic outbreaks of herbivorous insects can denude or kill plants of selected species over many square kilometers. This capacity to alter ecosystem structure dramatically has led to the widely-held view that herbivorous insects are biotic agents of disturbance (e.g., Veblen et al., 1994; D’Antonio et al., Chapter 17, this volume). However, insect herbivores (unlike abiotic disturbances) are an integral component of the ecosystem and respond (as do other organisms) to change in environmental conditions. Schowalter (1985) and Willig and McGinley (Chapter 27, this volume) have argued that integral ecosystem effects (e.g., stimulated

growth of non-host plants) should not be considered disturbance. We recognize that both perspectives contribute to understanding herbivory and its relationship to disturbance. Our challenge is to represent herbivores both as integral components of communities that adapt and respond to environmental change, and as agents of change that affect ecosystem structure and function in ways similar to abiotic disturbances. At issue is the threshold at which herbivory shifts from a normal trophic process to a disturbance. Normal levels of herbivory should not be considered disturbance, but, at some undefined levels of intensity, scale, and frequency, outbreaks have effects on ecosystem structure and function similar to those of fire, storm, drought, or flood. A key aspect of herbivory is its selectivity with respect to plant species affected. Such specificity ensures a diversity of indirect effects on other components of ecosystems, in the same way that various species show different tolerances to other disturbances or environmental changes. Outbreaks typically are triggered by changing environmental conditions, especially disturbances (or their suppression), that alter the physiological condition or abundance of host plants or predators. However, outbreak populations can spread to neighboring patches, altering community composition, canopy coverage, and biogeochemical cycling processes. These herbivore effects may function as a regulatory mechanism to stabilize ecosystem processes following disturbances or during environmental changes (Schowalter, 1985). This chapter addresses relationships between disturbance and herbivory in forest ecosystems. We review measurements of herbivory, attributes of herbivore outbreaks as agents of disturbance, disturbance-related factors that trigger growth of herbivore populations, and the effects of herbivory on forest structure and

253

254

ecosystem processes, including promotion of future disturbances. Comparison of data from temperate and tropical forests suggests general relationships between disturbance and herbivory. We note that our discussion focuses on herbivore relationships to discrete disturbances such as fire, storm, drought, or flood, rather than to longer-term environmental changes such as atmospheric pollutants or climate change, which are beyond the scope of this volume. Nevertheless, where appropriate, we consider these long-term environmental changes as they affect herbivore outbreaks. We also do not address exotic herbivores separately. Outbreaks of native or exotic herbivores depend on suitable environmental conditions (i.e., abundant and suitable hosts, and inadequate regulation by predators) and can have similar impact on ecosystem structure and function. Of course, exotic herbivores show a greater capacity to decimate their hosts, at least until their hostspecific predators are introduced.

TYPES AND PATTERNS OF HERBIVORY

Types of herbivory Plants are complex structures. They include nutritious parts (cytoplasm and fluids) surrounded by cell walls composed of lignin and cellulose, which insects cannot digest directly. Exploitation of plant resources is inhibited further by various biochemical defenses that are toxic or interfere with digestion. Herbivorous insects have evolved morphological, physiological, behavioral, symbiotic, and other adaptations to counter these barriers. However, these adaptations typically restrict herbivorous insects to one host (monophagy) or a few species of hosts (polyphagy) with similar defensive characteristics. Herbivores affect ecosystem structure and function to varying degrees depending on the degree of specificity and the dominance status of hosts. Insects can be classified into different feeding guilds, or functional groups, based on their mode of exploiting plants for food. Specific groups of plantfeeders include chewers (consumers of foliage, stems, flowers, pollen, seeds, and roots); miners and borers (which eat wood, bark, or one or more of the tissue layers between the intact upper and lower epidermis of leaves); gall-formers (which reside and feed within the plant and induce the production of abnormal growth reactions by the plant tissues); sap-suckers (which have

T.D. SCHOWALTER and M.D. LOWMAN

specialized mouthparts and survive by tapping into plant fluids); and seed-eaters and fruit-eaters (which consume the reproductive parts of plants) (Romoser and Stoffolano Jr, 1994). Only seed eaters, seedlingeaters, and some tree-killing bark beetles are true plant predators; most herbivores are more correctly classified as parasites because they do not kill their hosts but feed on the living plant (Price, 1980). These different modes of consumption affect plants in different ways. For example, folivores (species that feed on foliage) directly reduce the area of photosynthetic tissue, whereas sap-sucking insects affect the flow of fluids and nutrients throughout the plant. Folivory is the best-studied aspect of herbivory. In fact, the term herbivory often is used even when folivory alone is measured, because loss of foliage is the most obvious and most easily measured aspect of herbivory. Folivory represents the direct consumption of photosynthetically active material. Consequently, the loss of leaf area by a tree can be used as a relative term to indicate the severity of an herbivore outbreak. In contrast, other herbivores such as sap-suckers or rootborers cause less conspicuous damage to trees which is more difficult to measure, and also may have longerterm impacts (e.g., by disease transmission). Patterns of herbivory All plant species support characteristic associated insect herbivores, although some host a greater diversity of herbivores and suffer higher levels of herbivory than do others. Some plants can sustain high levels of herbivory and survive, whereas other species suffer mortality at significantly lower levels. The consequences of herbivory vary significantly, not just among plant species, but also in conjunction with different spatial and temporal factors (Huntly, 1991). For example, the combination of drought and herbivory or the effects of forest fragmentation on herbivore activity can significantly affect the ability of the host plant to respond. The relative timing of herbivory and the intervals between attacks also have important effects on the overall impact on the forest ecosystem. Measurement of herbivory is usually expressed in temporal units (e.g., daily or annual rates), and ranges from negligible to several times the standing-crop biomass of foliage (Table 9.1), depending on forest type, environmental condition, and regrowth capability of the plant (Lowman, 1995). Lowman (1992)

FOREST HERBIVORY: INSECTS

255

Table 9.1 Herbivory measured in temperate and tropical forests (including understory) 1 Location

Technique 2

Forest type

Level of grazing

Source

tropical rainforest

7.5% (new leaves)

1

Stanton (1975)

tropical rainforest

30% (old leaves)

1

Stanton (1975)

Panama

tropical rainforest

13%

1

Wint (1983)

Barro Colorado Island (Panama)

tropical rainforest

8% (6% insect; 1–2% vertebrates)

2,5

Leigh and Smythe (1978)

15%

2,5

Leigh and Windsor (1982)

Tropical Costa Rica

Puerto Rico

understory only

21% (but up to 190%)

3

Coley (1983)

tropical rainforest

7.8%

2

Odum and Ruiz Reyes (1970)

5.5–16.1%

4

Benedict (1976)

2–6%

4

Schowalter (1994)

2–13%

4

Schowalter and Ganio (1999)

Venezuela

understory only

0.1–2.2%

4

Golley (1977)

New Guinea

tropical rainforest

9–12%

1

Wint (1983)

Australia

sclerophyll

15–300%

4

Lowman and Heatwole (1992)

subtropical rainforest

14.6%

4

Lowman (1984)

tropical rainforest

8–12%

4

Lowman et al. (1993)

temperate deciduous

7–10%

1

Bray (1964)

2–10%

4

Reichle et al. (1973)

1–5%

4

Schowalter et al. (1981)

0.28) indicate that fire may have both positive and negative effects on productivity. The effects of fire on productivity were relatively larger than the effects of grazing, and showed a significant pattern along the precipitation gradient. The dispersion of the points around the equality line of Fig. 11.7 is larger than the dispersion of the points around the grazed vs. ungrazed line of Fig. 11.5. Thus, both positive and negative effects of fire on productivity appear to be more intense than the effects of grazing. This is clearly shown by Fig. 11.8: fire might increase Fig. 11.6. Effects of grazing on ANPP (above-ground net primary productivity by 300%, or reduce it to less than 20% production) (calculated as (ANPPgrazed − ANPPungrazed )/ANPPungrazed ) of control treatments. The proportional effect of fire of grassland and savanna sites along a gradient of mean annual precipitation. The dashed line indicates the null effects of grazing on productivity was positively associated with mean annual precipitation (r 2 = 0.30, P < 0.00001, d.f. = 62): on ANPP. the positive cases were on the more humid side of the

300

M. OESTERHELD, J. LORETI, M. SEMMARTIN AND J.M. PARUELO

gradient and the negative cases on the drier side of the gradient, with a transition between 600–700 mm. Climatic fluctuations

Fig. 11.7. Relationship between ANPP (above-ground net primary production) of burned and unburned plots of grassland and savanna sites comprising a wide range of primary productivity. The full line corresponds to the best-fit line, and the dashed line represents the equality line, where ANPP of burned plots is equal to the ANPP of unburned plots. Solid circles and triangles correspond to sites receiving less and more than 600 mm of mean annual precipitation respectively.

We compiled a data set of the normalized difference vegetation index (a surrogate for above-ground net primary productivity) for 13 grassland sites in North and South America. Site selection was based on the following criteria: a broad gradient of precipitation had to be encompassed, annual precipitation data had to be available, and natural grassland had to be the dominant vegetation type within the scanning unit or pixel. We used the Pathfinder Land Program data set (National Oceanic and Atmospheric Administration/National Aeronautic and Space Administration; James and Kalluri, 1994), which covers the 1981–1992 period and is based on maximum composites for tenday periods. Spatial resolution is 8 km. Data for normalized difference vegetation index were transformed into productivity by means of the equation provided by Paruelo et al. (1997). To make this analysis comparable to our previous analyses, we plotted maximum and minimum productivity for the 12-year period as a function of average productivity (Fig. 11.9). The figure shows that variations in productivity were, in relative terms, much greater on the drier side of the gradient. Both regression lines significantly differed from the equality line. The relationship between maximum and average productivity was ln ANPPMAX = 0.84 + 0.90 × ln ANPPAVG  2  r = 0.98, P < 0.00001, d.f . = 12 .

(11.8)

The intercept was significantly larger than 0 and the slope significantly lower than 1. The relationship between minimum and average productivity was ln ANPPMIN = −2.11 + 1.31 × ln ANPPAVG   2 r = 0.96, P < 0.00001, d.f . = 12 .

(11.9)

It had the opposite pattern: the intercept was significantly lower than 0 and the slope significantly greater than 1. Thus, relative fluctuations in productivity from year to year tend to be smaller as mean annual rainfall Fig. 11.8. Effects of burning on ANPP (above-ground net primary increases (Fig. 11.10). Extreme productivity values production) (calculated as (ANPPburned − ANPPunburned )/ANPPunburned ) were 80–90% greater or smaller than the mean in dry of grassland and savanna sites along a gradient of mean annual sites, and only 20% in humid sites. precipitation. The dashed line indicates the null effects of burning We investigated to what extent this interannual on ANPP. variability in productivity was related to precipitation

GRAZING, FIRE, AND CLIMATE EFFECTS ON GRASSLANDS AND SAVANNAS

301

fluctuations. We analyzed the relationship between the relative variations in productivity shown in Fig. 11.10 and the relative variation of the precipitation of the year in which the maximum or the minimum productivity was recorded [(annual precipitation for maximum or minimum ANPP – mean precipitation)/mean precipitation]. This analysis showed that 50% of the interannual fluctuations in productivity were accounted for by the year’s relative deviation in precipitation with respect to the mean (r 2 = 0.50, P < 0.0001). Discussion of disturbance effects The relatively mild effects of grazing and their uniformity throughout the gradient of rainfall reflect the importance of compensatory mechanisms in all sorts of systems and conditions. Compensatory growth, the increase in production per unit of remaining biomass after grazing (McNaughton, 1983b), is responsible not only for the positive effects of grazing on productivity but also for the relatively low magnitude of the negative effects. Since percent consumption increases along the gradient, our results indicate that compensatory growth also increases as precipitation increases. Without this increasing compensatory growth, the effects of grazing should have been all negative and directly related to consumption: more negative from the dry to the wet end of the gradient. Thus, the lack of strong effects on productivity and the common pattern along the gradient are in fact the result of strong feedback processes, which, through drastic changes in species composition, canopy structure, canopy photosynthesis, within-plant resource allocation, nutrient cycling, and water economy, among others, buffer potential changes in the functional, ecosystem-level variable, productivity (McNaughton, 1979, 1983b; Detling, 1987). It is particularly interesting that grazing effects on species composition are strongly influenced by the position of a system along the precipitation gradient (Milchunas and Lauenroth, 1993); grazing has minor effects on species composition of dry grasslands and large effects in more humid grasslands and savannas. Thus, the more or less similar relative effects of grazing on productivity along the gradient are maintained despite strong structural changes. A challenging, and promising, aspect of our analysis is the difference between fire and grazing effects along the gradient. It has been stated many times that fire effects on productivity vary from predominantly negative

Fig. 11.9. Fluctuation of ANPP in grasslands along a gradient of productivity. Solid squares correspond to maximum ANPP (aboveground net primary production) and open circles correspond to minimum ANPP in a series of 12 years between 1981 and 1992.

Fig. 11.10. Relative variation of ANPP (above-ground net primary production) of grasslands along a gradient of mean annual precipitation. The relative variation was calculated as (ANPPMAX − ANPPAVG)/ANPPAVG for the maximum data and as (ANPPMIN − ANPPAVG)/ANPPAVG for the minimum data. The dashed line is the equality line that indicates the null effects of climate fluctuations on ANPP.

in arid sites to positive in humid sites (Daubenmire, 1968; Vogl, 1974; Anderson, 1982; Bragg, 1995). However, we do not know of a quantitative test of

302

M. OESTERHELD, J. LORETI, M. SEMMARTIN AND J.M. PARUELO

that statement such as the one we have presented. The observation that fire generally does not have negative effects on productivity in humid sites, as grazing does, and does not have positive effects in drier sites, as grazing does, poses interesting questions regarding the different mechanisms through which these two agents of disturbance affect productivity. Interannual variability in productivity has been an important aspect of ecosystem studies because it affects one’s ability to predict productivity. Lauenroth and Sala (1992) have shown that 40% of the variation in productivity of a shortgrass site was explained by annual precipitation. Thus, productivity would fluctuate following the wide relative fluctuations in precipitation that characterize those semiarid environments (Le Hou´erou et al., 1988; Lauenroth and Burke, 1995). Our analysis based on patterns of the normalized difference vegetation index suggests that this pattern of variation closely matches the pattern of the coefficient of variation of rainfall along the gradient. For North American grasslands, the relative variability of the integral of normalized difference vegetation index (our estimator of above-ground net primary productivity) decreased exponentially with an increase in mean annual precipitation (Paruelo and Lauenroth, 1998). Our analysis fails to consider the interaction among disturbances, a potentially important aspect of disturbance phenomena, which has received particular attention recently (Collins, 1987; Hobbs et al., 1991; Briggs and Knapp, 1995; Noy-Meir, 1995). We do not have enough data to study these interactive effects at the large scale we have selected for our chapter. In any case, the patterns we have shown are strong enough, despite any potential interactive effect that we have not accounted for.

A CONCEPTUAL MODEL

Our analyses allowed us to build a conceptual model of the relative effects of three types of disturbance in grasslands and savannas (Fig. 11.11). The central element of this model is that both the regimes and the ranking of importance of the effects of these three types of disturbance depend on the position of a particular system on the gradient of mean annual precipitation. The effect of a disturbance on the productivity of a system located at any point along the gradient will be a function of the disturbance regime at that point, and the

Fig. 11.11. A conceptual model of the relative effects of grazing, fire, and climate on ANPP (above-ground net primary production) along a gradient of precipitation. The lines enclosing the response areas for each disturbance represent approximate boundaries of variation of ANPP according to Figs. 11.6, 11.8 and 11.10. Grazing effects are bounded by thick lines, fire effects are bounded by dashed lines and enclose a dotted area, climate effects are bounded by thin lines and enclose an area with stripes.

way that system responds to disturbances with changes in productivity. Between 200 and 450 mm, herbivory rates as a proportion of consumption and fire frequency are very low. In contrast, the systems are exposed to wide relative climatic fluctuations from year to year that can drastically change water economy, largely on the input side. Grazing in these systems may either increase or decrease productivity, but its effects are relatively mild compared to the interannual variation in productivity. One does not know what the effect of an eventual fire will be, but an extrapolation of our curve suggests that it may be severe. Productivity of these systems will fluctuate greatly around the mean, mainly as a consequence of interannual climatic fluctuations, reaching values ranging from less than one third of the mean to twice the mean. Between 450–700 mm of precipitation, grazing intensity increases, but still is a low proportion of productivity. This, together with slow decomposition rates limited by low soil water and high lignin content of the litter, sets the stage for the occurrence of fire with increasing frequency. Fire depends on fuel accumulation over more than one year. Grazing, as

GRAZING, FIRE, AND CLIMATE EFFECTS ON GRASSLANDS AND SAVANNAS

in any segment of the gradient, may have both positive and negative effects on productivity, negative effects being more frequent than positive. Fire in these systems, however, usually decreases productivity. The relative effects of fire and grazing in this segment of the gradient are of the same magnitude. Year-to-year variations in precipitation are much lower than in the driest end of the gradient, and so is the variation in productivity. It is therefore in this intermediate portion of the gradient that the three agents of disturbance have effects of similar relative magnitude. The presence or absence of grazing, the occurrence of fire, or the occurrence of an unusual year may be equally important in changing productivity. Above 700 mm of rainfall, grazing intensity in systems dominated by large ungulates may be very high, with a more patchy distribution in native systems than in livestock production systems. Fire frequency is much higher, favored by high annual production and the occurrence of a dormant season. Interannual fluctuations in weather are minimal in relative terms. Grazing has minor relative effects on productivity due to compensatory growth, but fire increases productivity up to five times the mean. Climatic fluctuations, in contrast, can only change productivity by less than 25%. Thus, productivity of these systems will fundamentally depend on fire occurrence in the first place, and grazing in the second. Grazing may have a larger effect in these systems through the regulation of the fire regime than by herbage removal per se. Ecologists have known for decades that mean annual productivity of grasslands and savannas is linearly related to mean annual precipitation (Walter, 1939 cited by Rutherford, 1980; McNaughton, 1985; Sala et al., 1988a; McNaughton et al., 1993). Our results provide a quantitative measure of the variation that may be observed around that mean as a consequence of disturbance agents. These results have several implications. First, livestock and wildlife managers can use them when making decisions about setting longterm levels of herbivore populations: the food base will fluctuate in different ways for different systems, and that variation may cascade to affect animal populations and the human societies based upon their production of economic goods. Second, these results can be used to rank the potential importance of each of these agents in driving productivity fluctuations for particular ecosystems along the precipitation gradient. Instead of the qualitative suggestions from fire ecologists, grazing ecologists, climatologists, and grassland ecologists

303

in general stressing the importance of one factor or considering them all equally important, a more balanced, broader view of their relative effects is now available, together with a more precise reference to the heterogeneity of responses across the biome. Finally, the results suggest that grasslands and savannas of intermediate mean precipitation are the most stable in terms of fluctuations in productivity, whereas the two extreme ends of the gradient are more prone to change. However, by controlling fire regime, grazing may reduce fluctuations in productivity at the humid end of the gradient, providing more stability to those systems. This may actually be a byproduct of human utilization of grasslands and savannas with livestock at much higher densities than wildlife (Oesterheld et al., 1992). ACKNOWLEDGEMENTS

We thank D. Frank for providing published and unpublished data, A. Breltr´an and E. Jobb´agy for providing climatic data, and R. Fern´andez Alduncin for sending us the literature we could not locate in Argentina. L. Walker, K. Bradbury and an anonymous reviewer significantly improved an earlier version of this manuscript. Our work is supported by CONICET, University of Buenos Aires, and Fundaci´on Antorchas. REFERENCES Anderson, R.C., 1982. An evolutionary model summarizing the roles of fire, climate and grazing animals in the origin and maintenance of grasslands: An end paper. In: J.R. Estes, R.J. Tyrl and J.N. Brunken (Editors), Grasses and Grasslands: Systematics and Ecology. University of Oklahoma Press, Norman, Oklahoma, pp. 297–308. Archer, S., 1989. Have southern Texas savannas been converted to woodlands in recent history? Am. Nat., 134: 545–61. Archer, S., 1995. Tree-grass dynamics in a Prosopis-thornscrub savanna parkland: Reconstructing the past and predicting the future. Ecoscience, 2: 83–99. Archer, S., Scifres, C., Bassham, C.R. and Maggio, R., 1988. Autogenic succession in a subtropical savanna: conversion of grassland to thorn woodland. Ecol. Monogr., 58: 111–127. Belsky, A.J., 1990. Tree grass ratios in East African savannas. A comparison of existing models. J. Biogeogr., 17: 483–490. Bidwell, T.G., Engle, D.M. and Claypool, P.L., 1990. Effects of spring headfires and backfires on tallgrass prairie. J. Range Manage., 43: 209–212. Biondini, M.E. and Manske, L., 1996. Grazing frequency and ecosystem processes in a northern mixed prairie, USA. Ecol. Appl., 6: 239–256.

304

M. OESTERHELD, J. LORETI, M. SEMMARTIN AND J.M. PARUELO

Bragg, T.B., 1982. Seasonal variations in fuel and fuel consumption by fires in a bluestem prairie. Ecology, 63: 7–11. Bragg, T.B., 1995. The physical environment of great plains grasslands. In: A. Joern and K.H. Keeler (Editors), The Changing Prairie. North American Grasslands. Oxford University Press, New York, pp. 49–81. Breman, H. and de Wit, C., 1983. Rangeland productivity and exploitation in the Sahel. Science, 221: 1341–1347. Briggs, J.M. and Knapp, A.K., 1995. Interannual variability in primary production in tallgrass prairie: Climate, soil moisture, topographic position, and fire as determinants of aboveground biomass. Am. J. Bot., 82: 1024–1030. Briggs, J.M., Seastedt, T.R. and Gibson, D.J., 1989. Comparative analysis of temporal and spatial variability in above-ground production in a deciduous forest and prairie. Holarct. Ecol., 12: 130–136. Burke, I.C., Yonker, C.M., Parton, W.J., Cole, C.V., Flach, K. and Schimel, D.S., 1989. Texture, climate, and cultivation effects on soil organic content in US grassland soils. Soil Sci. Soc. Am. J., 53: 800–805. Center, D.M., Vaughn, C.E. and Jones, M.B., 1989. Effects of management on plant production and nutrient cycling on 2 annual grassland sites. Hilgardia, 57: 1–40. Chaneton, E.J., Lemcoff, J.H. and Lavado, R.S., 1996. Nitrogen and phosphorus cycling in grazed and ungrazed plots in a temperate subhumid grassland in Argentina. J. Appl. Ecol., 33: 291–302. Coe, M., Cumming, D. and Philipson, J., 1976. Biomass and production of large African herbivores in relation to rainfall and primary production. Oecologia, 22: 341–354. Collins, S.L., 1987. Interactions of disturbances in tallgrass prairie: a field experiment. Ecology, 68: 1243–1250. Collins, S.L., 1990. Introduction: fire as a natural disturbance in tallgrass prairie. In: S. Collins and L. Wallace (Editors), Fire in North American Tallgrass Prairies. University of Oklahoma Press, Norman, Oklahoma, pp. 3–7. Cook, G.D., 1994. The fate of nutrients during fires in a tropical savanna. Aust. J. Ecol., 19: 359–365. Corona, M.E.P., Ciudad, A.G., Criado, B.G. and Dealdana, B.R.V., 1995. Patterns of aboveground herbage production and nutritional quality structure on semiarid grasslands. Commun. Soil Sci. Plant Anal., 26: 1323–1341. Coughenour, M.B., 1991. Biomass and nitrogen responses to grazing of upland steppe on Yellowstone northern winter range. J. Appl. Ecol., 28: 71–83. Coupland, R., 1992. Overview of the grasslands of North America. In: R.T. Coupland (Editor), Natural Grasslands. Introduction and Western hemisphere. Ecosystems of the World 8A. Elsevier, Amsterdam, pp. 147–150. Couteaux, M.M., Bottner, P. and Berg, B., 1995. Litter decomposition, climate and litter quality. Trends Ecol. Evol., 10: 63–66. Daubenmire, R., 1968. The ecology of fire in grasslands. Adv. Ecol. Res., 5: 209–266. Day, T.D. and Detling, J.K., 1990. Grassland patch dynamics and herbivore grazing preference following urine deposition. Ecology, 71: 180–188. Detling, J.K., 1987. Grass response to herbivory. In: J.L. Capinera (Editor), Integrated Pest Management on Rangeland: a Short Grass Prairie Perspective. Westview Press, Boulder, Colorado, pp. 56–68.

Detling, J.K., 1988. Grasslands and savannas. Regulation of energy flow and nutrient cycling by herbivores. In: L.R. Pomeroy and J.J. Alberts (Editors), Concepts of Ecosystem Ecology. A Comparative View. Springer-Verlag, New York, pp. 131–148. Dodd, J.L., 1994. Desertification and degradation in Sub-Saharan Africa –The role of livestock. Bioscience, 44: 28–34. East, R., 1984. Rainfall, soil nutrients status and biomass of large African savanna mammals. Afr. J. Ecol., 22: 245–270. Epstein, H.W., Lauenroth, W.K., Burke, I.C. and Coffin, D.P., 1997. Productivity patterns of C3 and C4 functional types in the U.S. great plains. Ecology, 78: 722–731. Facelli, J.M., 1988. Response to grazing after nine years of cattle exclusion in a Flooding Pampa grassland, Argentina. Vegetatio, 78: 21–26. Frank, D.A. and McNaughton, S.J., 1992. The ecology of plants, large mammalian herbivores, and drought in Yellowstone National Park. Ecology, 73: 2043–2058. Frank, D.A. and McNaughton, S.J., 1993. Evidence for the promotion of aboveground grassland production by native large herbivores in Yellowstone National Park. Oecologia, 96: 157–161. Frank, D.A., Inouye, R., Huntly, N., Minshall, G.W. and Anderson, J.E., 1994. The biogeochemistry of a North-temperate grassland with native ungulates: nitrogen dynamics in Yellowstone National Park. Biogeochemistry, 26: 163–188. Frost, P. and Robertson, F., 1987. The ecological effects of fire in savannas. In: B. Walker (Editor), Determinants of tropical savannas. ICSU Press, Miami, Florida, pp. 93–140. Gosz, J.R., 1993. Ecotone hierarchies. Ecol. Appl., 3: 369–376. Hart, S.C., Firestone, M.K., Paul, E.A. and Smith, J.L., 1993. Flow and fate of soil nitrogen in an annual grassland and a young mixed-conifer forest. Soil Biol. Biochem., 25: 431–442. Hobbs, N.T., Schimel, D.S., Owensby, C.E. and Ojima, D.S., 1991. Fire and grazing in the tallgrass prairie. Contingent effects on nitrogen budgets. Ecology, 72: 1374–82. Hobbs, N.T., Baker, D.L., Bear, G.D. and Bowden, D.C., 1996. Ungulate grazing in sagebrush grassland: mechanisms of resource competition. Ecol. Appl., 6: 200–217. Hulbert, L.C., 1988. Causes of fire effects in tallgrass prairie. Ecology, 69: 46–58. Hunt, H.W., Ingham, E.R., Coleman, D.C., Elliot, E.T. and Reid, C.P., 1988. Nitrogen limitation of production and decomposition in prairie, mountain meadow, and pine forest. Ecology, 69: 1009–1016. Huntley, B., 1982. Southern African savannas. In: B. Huntley and B. Walker (Editors), Ecology of Tropical Savannas. Springer-Verlag, Berlin, pp. 101–119. Jackson, L.E., Strauss, R.B., Firestone, M.K. and Bartolome, J.W., 1990. Influence of tree canopies on grassland productivity and nitrogen dynamics in deciduous oak savanna. Agric. Ecosyst. Environ., 32: 89–106. James, M.E. and Kalluri, S.N.V., 1994. The Pathfinder AVHRR land data set: an improved coarse resolution data set for terrestrial monitoring. Int. J. Remote Sensing, 15: 3347–3363. Jaramillo, V.J. and Detling, J.K., 1992. Small-scale heterogeneity in a semi-arid North American grassland. 2. Cattle grazing of simulated urine patches. J. Appl. Ecol., 29: 9–13. Jobb´agy, E., Paruelo, J. and Le´on, R.J.C., 1995. Estimaci´on del r´egimen de precipitaci´on a partir de la distancia a la cordillera en el noroeste de la Patagonia. Ecol. Austral, 5: 47–54.

GRAZING, FIRE, AND CLIMATE EFFECTS ON GRASSLANDS AND SAVANNAS Joffre, R., 1990. Plant and soil nitrogen dynamics in Mediterranean grasslands. A comparison of annual and perennial grasses. Oecologia, 85: 142–150. Kucera, C.L., 1981. Grasslands and fire. In: H.A. Mooney, T.M. Bonnicksen, N.L. Christensen, J.E. Lofan and W.A. Reiners (Editors), Fire Regimes and Ecosystem Properties. U.S. Forest Service General Technical Reports WO-26, pp. 90–111. Lauenroth, W.K., 1979. Grassland primary production: North American grasslands in perspective. In: N. French (Editor), Perspectives in Grassland Ecology. Springer-Verlag, New York, pp. 3–24. Lauenroth, W.K. and Burke, I.C., 1995. Great Plains, climate variability. Encycl. Environ. Biol., 2: 237–249. Lauenroth, W.K. and Sala, O.E., 1992. Long-term forage production of North American shortgrass steppe. Ecol. Appl., 2: 397–403. Le Hou´erou, H.N., Bingham, R.L. and Skerbek, W., 1988. Relationship between the variability of primary production and the variability of annual precipitation in world arid lands. J. Arid Environ., 15: 1–18. Le´on, R.J.C. and Facelli, J., 1981. Descripci´on de una coenoclina en el SW del Chubut. Rev. Fac. Agron., 2: 163–171. Mamolos, A.P., Elisseou, G.K. and Veresoglou, D.S., 1995. Depth of root activity of coexisting grassland species in relation to N and P additions, measured using nonradioactive tracers. J. Ecol., 83: 643–652. McNaughton, S.J., 1979. Grazing as an optimization process: grass– ungulate relationships in the Serengeti. Am. Nat., 113: 691–703. McNaughton, S.J., 1983a. Serengeti grassland ecology: the role of composite environmental factors and contingency in community organization. Ecol. Monogr., 53: 291–320. McNaughton, S.J., 1983b. Compensatory plant growth as a response to herbivory. Oikos, 40: 329–336. McNaughton, S.J., 1984. Grazing lawns: animals in herds, plant form, and coevolution. Am. Nat., 124: 863–886. McNaughton, S.J., 1985. Ecology of a grazing ecosystem: The Serengeti. Ecol. Monogr., 55: 259–294. McNaughton, S.J., 1993. Grasses and grazers, science and management. Ecol. Appl., 3: 17–20. McNaughton, S.J., Coughenour, M.B. and Wallace, L., 1982. Interactive processes in grassland ecosystems. In: J. Estes, R. Tyrl and J. Brunken (Editors), Grasses and Grasslands. Systematics and Ecology. University of Oklahoma Press, Norman, Oklahoma, pp. 167–194. McNaughton, S.J., Ruess, R.W. and Seagle, S.W., 1988. Large mammals and process dynamics in African ecosystems. Bioscience, 38: 794–800. McNaughton, S.J., Oesterheld, M., Frank, D.A. and Williams, K.J., 1989. Ecosystem-level patterns of primary productivity and herbivory in terrestrial habitats. Nature, 341: 142–144. McNaughton, S.J., Oesterheld, M., Frank, D.A. and Williams, K.J., 1991. Primary and secondary production in terrestrial ecosystems. In: J. Cole, G. Lovett and S. Findlay (Editors), Comparative Analyses of Ecosystems. Patterns, Mechanisms, and Theories. Springer-Verlag, New York, pp. 120–139. McNaughton, S.J., Sala, O.E. and Oesterheld, M., 1993. Comparative ecology of African and South American arid to subhumid ecosystems. In: P. Goldblatt (Editor), Biological Relationships Between Africa and South America. Yale University Press, New Haven, Connecticut, pp. 548–567.

305

Medina, E. and Silva, J.F., 1990. Savannas of northern South America. A steady state regulated by water–fire interactions on a background of low nutrient availability. J. Biogeogr., 17: 403–414. Meentemeyer, V., 1978. Macroclimate and lignin control of litter decomposition rates. Ecology, 59: 465–472. Milchunas, D.G. and Lauenroth, W.K., 1993. Quantitative effects of grazing on vegetation and soils over a global range of environments. Ecol. Monogr., 63: 327–366. Milchunas, D.G., Varnamkhasti, A.S., Lauenroth, W.K. and Goetz, H., 1995. Forage quality in relation to long-term grazing history, current-year defoliation, and water resource. Oecologia, 101: 366–374. Noy-Meir, I., 1995. Interactive effects of fire and grazing on structure and diversity of Mediterranean grasslands. J. Vegetation Sci., 6: 701–710. Odum, E.P., 1969. The strategy of ecosystem development. Science, 164: 262–270. Oesterheld, M., Sala, O.E. and McNaughton, S.J., 1992. Effect of animal husbandry on herbivore-carrying capacity at a regional scale. Nature, 356: 234–236. Owensby, C.E., Paulsen, G.M. and McKendrick, J.D., 1970. Effect of burning and clipping on big bluestem reserve carbohydrates. J. Range Manage., 23: 358–362. Owensby, C.E., Coyne, P.I. and Auen, L.M., 1993. Nitrogen and phosphorus dynamics of a tallgrass prairie ecosystem exposed to elevated carbon dioxide. Plant Cell Environ., 16: 843–850. Painter, E. and Belsky, A.J., 1993. Application of herbivore optimization theory to rangelands of the western United States. Ecol. Appl., 3: 2–9. Pandey, C.B. and Singh, J.S., 1992. Rainfall and grazing effects on net primary productivity in a tropical savanna, India. Ecology, 73: 2007–2021. Papanastasis, V.P., 1980. Effects of season and frequency of burning on a phryganic rangeland in Greece. J. Range Manage., 33: 251–255. Parton, W., Schimel, D.S., Cole, C. and Ojima, D.S., 1987. Analysis of factors controlling soil organic matter levels in Great Plains grasslands. Soil Sci. Soc. Am. J., 51: 1173–1179. Paruelo, J.M. and Lauenroth, W.K., 1996. Relative abundance of plant functional types in grasslands and shrublands of North America. Ecol. Appl., 6: 1212–1224. Paruelo, J.M. and Lauenroth, W.K., 1998. Interannual variability of NDVI and its relationship to climate for North American shrublands and grasslands. J. Biogeogr., 25: 721–733. Paruelo, J.M., Epstein, H.E., Lauenroth, W.K. and Burke, I.C., 1997. ANPP estimates from NDVI for the Central Grassland Region of the US. Ecology, 78: 953–958. Paruelo, J.M., Jobb´agy, E., Sala, O.E., Lauenroth, W.K. and Burke, I.C., 1998. Functional and structural convergence of temperate grassland and shrubland ecosystems. Ecol. Appl., 8: 194–206. P´erez Corona, M.E., Garc´ıa Ciudad, A., Garc´ıa Criado, B. and V´azquez de Aldana, B.R., 1995. Patterns of aboveground herbage production and nutritional quality structure on semiarid grasslands. Commun. Soil Sci. Plant Anal., 26: 1323–1341. Ripley, E., 1992. Water flow. In: R.T. Coupland (Editor), Natural grasslands. Introduction and Western Hemisphere. Ecosystems of the World 8A. Elsevier, Amsterdam, pp. 55–74. Risser, P., 1990. Landscape processes and the vegetation of the North American Grassland. In: S. Collins and L. Wallace (Editors), Fire

306

M. OESTERHELD, J. LORETI, M. SEMMARTIN AND J.M. PARUELO

in North American Tallgrass Prairies. University of Oklahoma Press, Norman, Oklahoma, pp. 133–146. Risser, P.G., 1985. Grasslands. In: B.F. Chabot and H.A. Mooney (Editors), Physiological Ecology of North American Plant Communities. Chapman and Hall, New York, pp. 232–256. Risser, P.G. and Parton, W.J., 1982. Ecosystem analysis of the tallgrass prairie nitrogen cycle. Ecology, 163: 1342–1351. Risser, P.G., Birney, E.C., Blocker, H.D., May, S.W., Parton, W.J. and Wiens, J.A., 1981. The True Prairie Ecosystem. Hutchinson Ross, Stroudsburg, Pennsylvania, 557 pp. Rutherford, M.C., 1980. Annual plant production–precipitation relations in arid and semiarid regions. S. Afr. J. Sci., 76: 53–56. Sala, O.E., Oesterheld, M., Le´on, R.J.C. and Soriano, A., 1986. Grazing effects upon plant community structure in subhumid grasslands of Argentina. Vegetatio, 67: 27–32. Sala, O.E., Parton, W.J., Joyce, L.A. and Lauenroth, W.K., 1988a. Primary production of the central grassland region of the United States. Ecology, 69: 40–45. Sala, O.E., Biondini, M.E. and Lauenroth, W.K., 1988b. Bias in estimates of primary production: an analytical solution. Ecol. Model., 44: 43–55. Sala, O.E., Golluscio, R.A., Lauenroth, W.K., Milchunas, D.G. and Burke, I.C., 1991. Coupling of water and nutrient capture strategies in the Patagonian steppe. Bull. Ecol. Soc. Am., 72: 238. Schimel, D.S., Stillwell, M.A. and Woodmansee, R.G., 1985. Biogeochemistry of C, N, and P in a soil catena of the shortgrass steppe. Ecology, 66: 276–282. Schimel, D.S., Parton, W.J., Kittel, T.G.F., Ojima, D.S. and Cole, C.V., 1990. Grassland biogeochemistry – Links to atmospheric processes. Climatic Change, 17: 13–26. Schimel, D.S., Kittel, T.G.F., Knapp, A.K., Seastedt, T.R., Parton, W.J. and Brown, V.B., 1991. Physiological interactions along resource gradients in a tallgrass prairie. Ecology, 72: 672–685. Schulze, E.-D., Mooney, H.A., Sala, O.E., Jobb´agy, E., Buchmann, N., Bauer, G., Canadell, J., Jackson, R.B., Loreti, J., Oesterheld, M. and Ehleringer, J.R., 1996. Rooting depth, water availability, and vegetation cover along an aridity gradient in Patagonia. Oecologia, 108: 503–511. Seagle, S.W. and McNaughton, S.J., 1992. Spatial variation in forage nutrient concentrations and the distribution of Serengeti grazing ungulates. Landscape Ecol., 7: 229–241. Seastedt, T.R., 1995. Soil systems and nutrient cycles of the north American prairie. In: A. Joern and K.H. Keeler (Editors), The Changing Prairie. North American Grasslands. Oxford University Press, New York, pp. 157–174. Seastedt, T.R., James, S.W. and Todd, C.D., 1988. Interactions among soil invertebrates, microbes and plant growth in the tallgrass prairie. Agric. Ecosyst. Environ., 24: 219–228. Semmartin, M. and Oesterheld, M., 1996. Efectos del pastoreo sobre la productividad primaria: dise˜no espacial, competencia y

disponibilidad de recursos. Congreso Argentino de Producci´on Animal, Vol 20. Termas de R´ıo Hondo, Santiago del Estero, Argentina, p. 168. Sinclair, A., 1979. The Serengeti environment. In: A. Sinclair and M. Norton-Griffiths (Editors), Serengeti. Dynamics of an Ecosystem. University of Chicago Press, Chicago, pp. 31–45. Singh, R.S., 1993. Effect of winter fire on primary productivity and nutrient concentration of a dry tropical savanna. Vegetatio, 106: 63–71. Smallwood, K.S. and Schonewald, C., 1996. Scaling population density and spatial pattern for terrestrial, mammalian carnivores. Oecologia, 105: 329–335. Soriano, A., 1983. Deserts and semi-deserts of Patagonia. In: N.E. West (Editor), Temperate Deserts and Semi-Deserts. Ecosystems of the World 5. Elsevier, Amsterdam, pp. 423–459. Soriano, A., 1992. Rio de la Plata grasslands. In: R.T. Coupland (Editor), Natural Grasslands. Introduction and Western Hemisphere. Ecosystems of the World 8A. Elsevier, Amsterdam, pp. 367–407. Steuter, A.A., 1987. C3 /C4 production shift on seasonal burns – northern mixed prairie. J. Range Manage., 40: 27–31. Trlica Jr., M.J. and Schuster, J.L., 1969. Effects of fire on grasses of the Texas High Plains. J. Range Manage., 22: 329–333. Trollope, W., 1984. Fire in savanna. In: P. Booysen and N. Tainton (Editors), Ecological Effects of Fire in South African Ecosystems. Springer-Verlag, Berlin, pp. 149–176. Turner, C.L., Seastedt, T.R. and Dyer, M.I., 1993. Maximization of aboveground grassland production – The role of defoliation frequency, intensity, and history. Ecol. Appl., 3: 175–186. Vogl, R.J., 1974. Effects of fire on grasslands. In: T.T. Kozlowski and C.E. Ahlgren (Editors), Fire and Ecosystems. Academic Press, New York, pp. 139–134. Walker, B.H. and Noy-Meir, I., 1982. Aspects of the stability and resilience of Savanna ecosystems. In: B. Huntley and B. Walker (Editors), Ecology of Tropical Savannas. Springer-Verlag, Berlin, pp. 556–590. Walter, H., 1939. Grassland, Savanne und Busch der arideren Teile Afrikas in ihrer o¨ kologischen Bedingtheit. Jahrb. Wiss. Bot., 87: 750–860. Walter, H., 1977. Zonas de Vegetaci´on y Clima. Omega, Barcelona, 245 pp. Wedin, D.A., 1995. Species, nitrogen, and grassland dynamics: The constraints of stuff. In: C. Jones and J.H. Lawton (Editors), Linking Species and Ecosystems. Chapman and Hall, pp. 253–262. Whicker, A.D. and Detling, J.K., 1988. Ecological consequences of prairie dog disturbances. Bioscience, 38: 778–785. Wink, R.L. and Wright, H.A., 1973. Effects of fire on an ash juniper community. J. Range Manage., 26: 326–329. Zaady, E., Groffman, P.M. and Shachak, M., 1996. Litter as a regulator of N and C dynamics in macrophytic patches in Negev desert soils. Soil Biol. Biochem., 28: 39–46.

Chapter 12

DISTURBANCE IN DESERTS James A. MacMAHON

INTRODUCTION

In this chapter I discuss disturbances in deserts of the world. On the surface this seems a straightforward task which could be completed by listing natural disturbances and documenting their effects on organisms and then presenting a self-conscious treatment of anthropogenic disturbances. A more complete approach might include consideration of recovery from disturbance, and would give special attention to the effects of disturbance on animals as well as their role as agents of disturbance, rather than discussing only the responses of plants. Indeed, this is the general approach that I take. However, a further understanding of disturbance in deserts requires elaboration of some characteristics of deserts that make the term “disturbance” somewhat difficult to define rigidly, and that suggest additional topics for this chapter. The ambiguity of the word disturbance has frequently been mentioned in the literature (e.g., Grime, 1979). A seminal definition was proposed by White and Pickett (1985) when they suggested that “A disturbance is any relatively discrete event in time that disrupts ecosystem, community, or population structure and changes resources, substrate availability, or physical environment.” White and Pickett observe that disturbance is relative to dimensions in space (organism size) and time (organism life-span), and that a common consequence of disturbance is an increase in the patchiness of an area. More recently Huston (1994) elaborated on this conceptual approach when he defined disturbance as “. . . any process or condition external to the natural physiology of living organisms that results in the sudden mortality of biomass in a community on a time scale significantly shorter (e.g., several orders of magnitude faster) than that of the accumulation of

the biomass.” His definition emphasizes the time-scale consideration but not that of size. Superficially, these definitions suggest that discussion of disturbance in deserts should be limited to consideration of the sudden death of individuals, or of parts of individuals such as leaves of plants, and the subsequent development of increased patchiness. Generally, consideration of the effects of contemporary climate change or other long-interval events would not be included as a disturbance using the perspective developed above. However, some desert plants, at least as clones, may live for 10 000 years (Vasek, 1980); thus, climate change occurring over several decades, the rate that seems to be occurring now, may be a “disturbance” because the environment is changing several orders of magnitude faster than the accumulation of biomass, as indicated by life-span, of some desert plants. Because of this I consider recent, rapid climate change and buildup of carbon dioxide as disturbances and treat them separately. Other workers have adopted a similar perspective, at least for plants (Bazzaz, 1996; Smith et al., 1997). A second problem in defining disturbance is that associated with separating disturbance factors that are inherent to the system and upon which the system depends, as opposed to those that are “foreign” to it (Vogl, 1980). Vogl (1980) discussed a variety of “inherent” disturbance factors such as rain and floods, wind and storms, fire, snow and frost, erosion, and perturbations caused by animals or man. I will discuss most of these categories for deserts, but generally treat these factors as disturbances of a foreign nature because they occur with unpredictable frequency and amplitude. At the outset, one must constrain one’s view of deserts. The literature contains many definitions of deserts. Meigs (1953) developed a widely used

307

308

James A. MacMAHON

Table 12.1 Arid areas of the world 1,2 Area (total) Africa (30 312) North America (21 322) South America (17 818) Asia (43 770) Australia (7618) Europe (500) World (130 737) 1 2 3

Arid 3

Semi-arid

Total

Arid + semiarid 2

Arid 2

12 819

2951

15 770

52

42.3

1125

1935

3060

14

5.3

1358

1268

2626

14

7.6

10 235

4817

15 052

34

23.4

3250

1375

4625

61

42.7

80

20

32

22.1

0.100 28 889

0.300 12 651

0.400 41 540

Data are reworked from Le Hou´erou (1992a). Data are millions of km2 except the last two columns which are arid or arid + semiarid as a percent of the total continental or world area. Here, “arid” includes extremely arid and hyper-arid lands.

approach for UNESCO that includes rainfall and temperature. This system recognizes hyper-arid, arid, semiarid, and several more mesic subdivisions. Arid and hyper-arid areas receive less than 200 mm (~8 inches) of annual precipitation and semi-arid areas receive between 200 and 600 mm (up to 24 inches). Le Hou´erou (1992a) defined arid lands by the ratio between precipitation and potential evapotranspiration. Using his system, Africa has the greatest absolute area of arid land and is closely tied with Australia as to the proportion of total land area that is arid (Table 12.1). The continent with the smallest proportion of arid land is Europe. The largest single area of desert, the Sahara of North Africa, covers nearly 9×106 km2 , an area approximately the size of the United States. In this chapter, I will discuss disturbances in hyper-arid and arid areas and drier semi-arid areas (transitions), calling them collectively deserts. I will not include semi-arid grasslands [but see Oesterheld et al. (Chapter 11, this volume) and Ghersa and Le´on (Chapter 20, this volume) for coverage of grasslands]. Deserts range from areas devoid of any conspicuous vegetation to areas moderately well vegetated with shrubs and sub-trees, a scattering of grasses, and a variety of annuals and succulents. Deserts represent a highly variable group of ecosystems that occur in areas ranging from temperate to tropical zones around the world. I specifically will not refer to those cold, highlatitude areas termed polar deserts. Generally, deserts are caused by one of four phenomena. First is the rain-shadow effect whereby moisture is lost from air as it moves inland over mountains. Moisture-laden air condenses to form rain

as it moves up over mountains. As the air crosses the mountains and descends it becomes drier and creates arid conditions. The Mojave Desert of North America is predominantly a rain-shadow desert, as are some deserts of Central Asia. Another source of condensation of water out of the air, and thus an increased drying power of the air, is that caused by air crossing cold ocean currents. Examples of this occur in Africa forming the Namib Desert, along the coast of Peru forming the Atacama Desert, and the peninsula of Baja California forming a portion of the Sonoran Desert. The available moisture in these coastal deserts is often in the form of fog rather than rain. The third possibility is that some land lies in the interior of a continent where moisture-laden air is not common because of the distance from sources of moisture. Some deserts of central Australia and of China are caused by the continental nature of their climates. Finally, there is the effect of high-pressure zones that occur at about 30º North and 30º South latitude that are caused by Hadley Cells. These convection entities are driven by solar energy and the spinning of the earth on its axis. Air lifted at the equator produces rain. As the air moves both north and south of the equator it descends, dries, and creates high-pressure areas that preclude inward movement of moist air. The Sahara and Kalahari, in northern and southern Africa respectively, are deserts of this type. Disruption of any of these climate patterns can be a major disturbance to deserts, altering both their inhabitants and their geomorphology. For introductions to the specific deserts of the world (Fig. 12.1), their causes, and biological and geological characteristics, the reader is referred to the following

DISTURBANCE IN DESERTS

309

Fig. 12.1. Hot deserts and their transitions. Seasons of precipitation are indicated. This figure includes most of the arid areas referred to in this chapter except some “cold” deserts of Asia and the United States. Reprinted from Evenari (1985) with permission.

publications: Petrov (1976); West (1983); Evenari et al. (1985, 1986); Walter and Breckle (1986); Allan and Warren (1993); Arritt (1993); Lovegrove (1993); and MacMahon (1999).

NATURAL DISTURBANCE IN DESERTS

Deserts are ecosystems occurring in extreme environments. Because desert organisms are constantly exposed to extreme and often unpredictable values of environmental factors, especially temperature, annual precipitation, and insolation, they are frequently at the limits of their tolerance ranges, and slight changes in the values of these factors can move the organism into zones of their tolerance curves where they cannot survive – that is, mortality occurs or reproduction is not possible. Additionally, because of the open, sparse vegetation, geomorphic processes are changing the landscape at an appreciable rate, often too fast for the vegetation to adjust. Vegetation itself controls some geomorphic processes such as deposition and deflation of soils, and some geomorphic instability

may foster diversity (McAuliffe, 1994). Some weather variables act as agents of disturbance, and at the same time they foster other types of disturbance. Examples include: fires, which are particularly likely following periods of above-average rainfall; floods that occur frequently in response to high-intensity, short-duration rainfall and cause erosion; drought, a phenomenon that occurs periodically in deserts and may destabilize soils; outbreaks of animal pests; extreme swings of temperature due in part to the relative lack of insulating cloud cover; mass movement caused by a variety of geomorphic processes; and, finally, global climate change, the longer-term climatic changes that can change a particular site from desert status to that of some other community type or to that of a more extreme desert. Given this wide range of types and frequency of disturbances and the long time scale required for desert communities to recover, perhaps 1000 years for some species (McAuliffe, 1988, 1994), it is reasonably argued that desert systems are seldom, if ever, in equilibrium (Sullivan, 1996). This perspective is the basis for a series of models of deserts as “pulse–

310

reserve” systems (Noy-Meir, 1973). In a sense, desert systems respond to pulses of favorable rainfall and nutrient availability, and then persist using the reserves developed during the last pulse. The argument that deserts are not equilibrium systems has changed the way that we view their potential management (Westoby et al., 1989; Laycock, 1991). If all of the short-term possibilities for disturbances in deserts are combined with the longer-term prospects of global changes that can increase the variability of climatic factors, then one has to ask whether or not the concept of disturbance, other than those extreme disturbances caused by human beings, is of the same importance in deserts as in more mesic systems. Deserts are so often “disturbed” that disturbance may be one of the defining characteristics of the system rather than a periodic anomaly as in some other systems reviewed in this book. Additionally, the slow recovery of deserts following disturbance, because of the low and variable rainfall, may make them more vulnerable to further disturbance. An example will demonstrate this natural variation. Goldberg and Turner (1986) reported on vegetation changes over a period of 72 years (1906–1978) in a Sonoran Desert site on Tumamoc Hill, near Tucson, Arizona. While virtually no directional change in vegetation was noted, significant fluctuations in cover and density coincided with sequences of very wet or very dry years. Establishment of new plants was episodic rather than an annual event. After listing all of the potential changes that can occur in a desert, I have to indicate that there can also be significant persistence of system components despite disturbance. Some plant species and even some individual plants can persist for long periods of time. For example, clones of creosotebush (Larrea tridentata) survive for nearly 10 000 years in the same spot (Vasek, 1980), and at least 15 woody plant species in the Sonoran Desert are known to survive over 100 years (Bowers et al., 1995). These longevity data highlight one of the apparent contradictions of desert communities. Despite the rigors of the desert environment, individual plants persist for long periods, and communities may persist for thousands of years (Axelrod, 1979); yet if one looks at the year-to-year variation, for example, the ephemeral nature of the species mix and abundance of annuals, there may be dramatic changes in the composition of individual small plots over very short time scales. All of these factors make it necessary to

James A. MacMAHON

approach the topic of disturbance more carefully than one might in ecosystems where weather and climatic factors may neither vary as much nor reach the same extremes as in deserts. Temperature Desert plants and animals are exposed to extremely high temperatures, and to wide variation in temperatures in a 24-hour period. Prolonged exposure to freezing temperatures is uncommon. As might be expected, desert organisms seem well adapted to extremely high temperatures either by avoidance, a common method used by both plants and animals, or endurance by shutting down photosynthesis, having microphyllous leaves and shedding these during periods of prolonged intense heat, or any one of dozens of other adaptive syndromes. High temperature, per se, probably kills many seedlings but many fewer established plants (Smith et al., 1997). Some succulent species may heat up to 65ºC, for short periods, with no ill effects (Nobel, 1988). The secondary effects of high temperature, as a synergist with low water availability to increase drying power of air or decrease soil moisture content, are often more likely to kill plants and animals than the direct effects. In contrast, low temperatures are known to cause death in a variety of established desert species. One of the best-studied is the saguaro cactus (Carnegiea gigantea; Steenbergh and Lowe, 1977), which succumbs to freezing temperatures. Death from freezing requires both very low temperatures and lengthy periods of exposure. Bowers (1981) analyzed temperature data for Tucson, Arizona. She observed that catastrophic freezes, those that kill desert plants, occurred four times between 1946 and 1979, and that they were more common in the past 100 years than in the previous century. Catastrophic freezes are characterized by minimum temperatures of at least −8.4 to −5.6ºC and durations of at least 15 to 20 hours. Freezing damage to plants during catastrophic freezes has extended hundreds of kilometers south of Arizona to the southern border of the Mexican state of Sonora, where in 1937 temperatures reached −8.9 to −6.7ºC, damaging genera such as Ficus, Pithecellobium, and Randia. Since 1900, some of the genera experiencing damage in Arizona include: Ambrosia, Celtis, Encelia, Jatropha, Olneya, Sapium, a variety of cacti, but especially saguaro, and a variety of subtrees and shrubs (Bowers, 1981). These

DISTURBANCE IN DESERTS

genera include many of the dominants on Sonoran Desert sites. The nurse-plant phenomenon, where young plants cluster under the canopies of other species, may relate to freezing avoidance. Many plants that cluster beneath other plants in their seedling stages may gain the advantage of a warmer microenvironment during cold periods in the cover of the nurse plant than they would if exposed in the open (Nobel, 1988). It is difficult to prove that this is the major advantage to the plant, since the environment under a nurse plant also provides shade, organic matter, the availability of mycorrhizal fungi, protection from herbivory (e.g., McAuliffe 1984a,b), and more favorable water holding capacity because of higher soil organic matter, as well as a variety of nutrients. The nurse-plant phenomenon has been studied extensively (e.g., Nobel, 1988; McAuliffe, 1984a; Yeaton and Manzanares, 1986; Georgiadis, 1989; Cody, 1993). Finally, temperature as an agent of disturbance has been implicated as a controlling factor in the distribution of a life form. Von Willert et al. (1992) extensively reviewed the biology of succulent plants, especially species in the Namib Desert. They suggested that a major limiting factor for this life-history strategy is disturbance in the form of freezing temperatures – that is, those below −4º or −5ºC during the growing season. They suggested that this relationship explains the paucity of succulents in the deserts of Central Asia, the Great Basin of North America, and the Patagonian desert of Argentina. Werger (1983) made similar suggestions earlier. These assertions remain to be rigorously tested. Water Water is probably the environmental factor that most often “drives” a variety of ecosystem processes in deserts (Noy-Meir, 1973, 1974; Crawford and Gosz, 1986). Water and wind share several characteristics. Either can act as agents of erosion or deposition; thus, excesses of either can be damaging, while moderate amounts of both are necessary for the functioning of desert ecosystems. Perhaps the most important effect of water is that it causes the patterning of desert vegetation in time and space (Allen, 1991). Water erosion may be more important in semi-arid regions while wind may dominate as the erosive force in arid areas (West, 1988). In the Patagonian deserts and semi-

311

deserts, 28 590 km2 of land have been altered by water erosion (Soriano, 1983). Several aspects of water in deserts, such as availability, predictability, and flooding, will be considered independently. Lack of water can be a significant detriment to desert plants although, even after long periods of drought, many systems seem to have the capacity to “spring back” to relatively normal community composition (Goldberg and Turner, 1986). Certainly a few species may be lost during periods of drought, but the majority of characteristic dominants seem either able to survive or to have sufficient seed reserves in the soil to re-establish after drought. This is not unexpected, given the periodic nature of drought in almost all deserts. Plant clumps (resource islands) may increase survivorship of individual plants during drought (Reynolds et al., 1999). Interestingly, clumpage does not matter. There is a great degree of unpredictability in the availability of water in desert ecosystems. Worldwide, there is a correlation between the year-to-year variation in rainfall and the amount of rainfall. As rainfall decreases, its coefficient of variation (and thus unpredictability) increases dramatically (Fig. 12.2). Other calculations for predictability have been suggested (Weis and Schwartz, 1988). This unpredictability, to a large extent, drives desert systems, even influencing primary production (Le Hou´erou et al., 1988). Nonetheless, some animals (Noy-Meir, 1974) and plants have adapted to the periodic paucity of water. For many plants, an unusual series of back-toback, above-average rainfall years are needed before establishment of a new generation can occur. Single good years of precipitation may cause germination, but these seedlings subsequently die if there is not a second year of above-average rainfall to allow establishment of the root systems. Such episodes, on average, occur with a frequency of about once in forty years (MacMahon and Wagner, 1985) in some areas. This differs markedly from the situation in grasslands and forests, where establishment is nearly an annual phenomenon. Observations of the age structure of populations of barrel cacti (Ferocactus) in the northern Mojave Desert confirm this periodicity (Ehleringer and House, 1984). This episodic establishment has also been noted for a variety of other North American plants, for example ironwood, Olneya tesota (B´urquez and Quintana, 1994) and the agave, Agave deserti (Jordan and Nobel, 1979). During the first half of

312

James A. MacMAHON

Fig. 12.2. Relationship between coefficient of variation (standard deviation/mean annual value) and the mean annual precipitation (mm) for some North American hot-desert sites. Localities are: 1, Yuma, Arizona; 2, Inyo, California; 3, Blythe, California; 4, Mojave, California; 5, Bakersfield, California; 6, Las Vegas, Nevada; 7, Wellton, Arizona; 8, Barstow, California; 9, Lucerne Valley, California; 10, Gila Bend, Arizona; 11, Guaymas, Sonora; 12, Saltillo, Coahuila; 13, La Paz, Baja California Sur; 14, Chihuahua, Chihuahua; 15, Organ Pipe Cactus National Monument, Arizona; 16, Ajo, Arizona; 17, Tucson, Arizona. Reprinted from Evenari (1985) with permission.

this century creosotebush (Larrea) and paloverde (Cercidium) declined with virtually no recruitment of new individuals. Mortality of some species during this period coincided with a prolonged drought from 1936 to 1964. New plants established in an episodic manner following unusually heavy precipitation during certain seasons (Turner, 1990). The re-greening of the Sahel following a long period of drought is another example, in part, of this process (Tucker et al., 1991). Much of the effect of periodic droughts on vegetation and periods of plant establishment can now be observed by the use of satellite imagery (e.g., Nicholson et al., 1990; Peters et al., 1993). This should allow us to detail and quantify this phenomenon at a landscape scale in the future. The specific season of rainfall often determines the characteristics of rains and even the response of perennials to rainfall (Ehleringer et al., 1991). In the North American deserts, winter rains are of long duration, great areal extent and low intensity.

In contrast, summer rains are characterized by short duration, low areal extent, and high intensity. Many other areas of the world including Asia, Australia, and Africa (Walter and Breckle, 1986), experience these same differences in seasonal rainfall characteristics. Summer rains rapidly saturate the ground surface, causing it to become water-repellent. This leads to surface flow of water, which moves particulate matter, including litter and soil, to stream channels, alters the surface characteristics of the soil in ways that influence plants and animals, and ultimately causes a feedback to additional runoff events (Fig. 12.3). One interesting feedback is that digging (bioturbation) by animals – for instance, isopods (Hemilepistus reaumuri) and porcupines (Hystrix indica) – in the Negev significantly increases erosion (Yair, 1995). Whitford and Kay (1999) review bioturbation in desert soils. A result of surface flow is that ephemeral stream channels often have a richer vegetation, (1) because the water seeps to deeper levels under such storm conditions, and (2) there

DISTURBANCE IN DESERTS

313

Fig. 12.3. A flow diagram of a conceptual model of the interactions among ecosystem components and rainfall as they create a feedback to desert soil properties in the Negev Desert. From Yair and Shachak (1982) with permission.

is an accumulation of organic matter, which further aids plant growth. Some detailed experiments in the Mojave Desert, in which surface flow was diverted, raise questions about the magnitude of the effects on the soil surface (Schlesinger and Jones, 1984; Schlesinger et al., 1989). These studies suggest that there is little effect on soil properties as a result of artificially diverting surface flow away from alluvial piedmonts. Although plants grow better where the water has not been diverted, this is interpreted as a result of a differential distribution of biomass of shrubs in natural areas, as compared to the experimental plots. The lack of an effect may be caused by the low energy of winter rainfall in the Mojave Desert. Runoff and sediment yield are often the same on vegetated and denuded plots if the precipitation is of low energy (Bolin and Ward, 1987). Studies of runoff from grasslands and shrublands in the Chihuahuan desert show more N loss in bare areas, less in shrublands, and least in deserts (Schlesinger et al., 1999). Phosphorus loss was small in all habitats. The overland movement of water from high-intensity storms often leads to the formation of ephemeral ponds or lakes in low-lying areas. Such water bodies (playas) are a common feature of desert landscapes where

soils are not porous enough for percolation to prevent accumulation of water on the surface. Playas are generally less than 100 km2 in area, and, although they may look similar, the 50 000 or so that occur worldwide are often geomorphically quite distinct (Neal, 1969, 1975). Playas have been studied in a number of places, and many authors have found significant adaptations of animals and plants to playa environments. Changes in the seasonality of rainfall, such as might occur under various scenarios of climate change, could prevent the formation of playas and thus contribute to the loss of a unique flora and fauna. In some areas, high rainfall amounts may simply cause floods, to which the organisms are not adapted. In such cases, some plants and animals may be disturbed, and in some cases killed or uprooted, but this would generally be an unusual, transient situation. Flooding on most desert surfaces is usually not a phenomenon of any appreciable geomorphic consequence, since the vast bulk of the water is diverted into stream channels, which periodically cut down into the substrate as they accumulate water. Thus, in normal desert systems, if the water can percolate, it adds to the soil water reserves. If rainfall is at high intensity and water flows across the surface, it may remove

314

organic matter from the immediate vicinity of plants; the organic matter is deposited elsewhere, and plants on scoured surfaces are already somewhat tolerant of low-nutrient systems. Deflation by surface water and wind movement through interplant spaces often causes the development of phytogenic mounds around the bases of plants, especially around clumps of prostrate species (Goudie and Wilkinson, 1977). In contrast, clumps of plants may trap fine materials moved by water or wind, aiding in the formation of lenses of relatively rich soils referred to as “islands of fertility” (Schlesinger et al., 1996). Interestingly, creosotebushes with conical crowns trap less material than those with hemispherical crowns (De Soyza et al., 1997). Such capture processes may actually direct succession in some habitats (Vasek and Lund, 1980). Wind and water erosion in deserts is generally greater on non-vegetated surfaces than on those that are vegetated (Fig. 12.4) (Schimpf and MacMahon, 1981). Vegetation decreases both wind and soil erosion and, to a point, affects the severity of drought (Fig. 12.5). Additionally, the seasonal pattern of rainfall and its energy influences the shapes of the curves in Fig. 12.4.

Fig. 12.4. A conceptual model of the dependence of erosion by wind (dashed lines) and water (solid lines) on rainfall and vegetation cover. Curve A is relative sediment yield with vegetation cover and B is without. Curve C is relative wind erosion with vegetation cover and D is without. Note that the response axis has no scale, but shows the relative magnitude of the effect based on existing data. From World Meteorological Organization, Geneva, Switzerland (1983) with permission.

James A. MacMAHON

Grime (1979) rightly pointed out that the most difficult environments for plants to adapt to are those with extreme and unpredictable values of environmental variables. Deserts clearly are environments with these characteristics, especially for water availability. Wind Wind can have a variety of effects on deserts that are similar to those of water, while others need to be considered separately. Wind effects are not independent of rainfall or temperature, and they are altered by the degree of vegetation cover (Fig. 12.6). A minor effect of wind is direct physical damage in the uprooting of plants. This is an unusual phenomenon, normally occurring when speeds approach 100 km hr−1 , but it does occur at lower velocities in tall plants such as saguaros, especially if they have been girdled at their bases by the feeding of jackrabbits (Lepus californicus). Additionally, wind may increase the drying power of the air, desiccating plants and animals to critical levels. The aridity of Patagonian deserts and semideserts is due to wind increasing the drying of surfaces rather than just to the low rainfall directly (Soriano, 1983). Finally, perhaps the most important role of wind is as a carrier for particulate matter, both organic and inorganic. In some cases, the movement of materials is a positive influence and in others it can be quite negative, especially when the material carried is deposited in ways or in quantities that disrupt the functioning of plants or animals. The importance of wind is quite variable from place to place. In North American deserts the effects of wind are generally unimportant. In North Africa, winds are strong and violent, and may blow on at least 50 days in the spring with average speeds of 20–28 km hr−1 ; they can commonly attain speeds of up to 60 km hr−1 , with maxima of over 100 km hr−1 (Grenot, 1974). Dust is regularly removed from deserts and transported elsewhere. Saharan dust is often deposited in southern Europe (Mattsson and Nihlen, 1996). The red rain that occurs in the British Isles is also an example of Saharan dust being deposited in Europe. Similarly, hazes that appear over the Arctic during the summer probably have a source in the Central Asian deserts (Allan and Warren, 1993). It has even been postulated that Saharan dust is a source of nutrients for Amazonian rainforests, where the airborne input of materials may

DISTURBANCE IN DESERTS

315

Fig. 12.5. Interrelationships among drought severity, plant cover (%) and soil erosion. On barren ground the effects of water and wind erosion and consequently drought severity are increased. From World Meteorological Organization, Geneva, Switzerland (1983) with permission.

Fig. 12.6. A conceptual model of the general relations between annual rainfall and soil erosion on vegetated surfaces for various amounts of rainfall; and the increased erosion caused by wind when vegetation is removed. From World Meteorological Organization, Geneva, Switzerland (1983) with permission.

316

reach 26.9 kg−1 yr−1 for phosphate phosphorus, and 12.6 kg−1 yr−1 for potassium (Reichholf, 1986). The effect of wind is especially noticeable in areas of sanddune development – first, because the sand dunes are formed as a result of winds, and second, because their subsequent movement is wind-driven. In the Takla Makan Desert of northwestern China, an area covering 320 000 km2 , dunes are moving and expanding the extent of that desert. Some of these dunes reach over 300 m in height. There are two interesting relationships in respect to the effect of wind on vegetation. In areas with virtually no vegetation and extremely low rainfall, movement by wind is not an obvious factor. Presumably, surfaces at these sites have become stabilized, perhaps by the formation of desert pavements, after being exposed to winds for very long times. In contrast, areas with some rainfall seem to generate the greatest amount of particulate matter for movement. Part of this is organic matter. In the Namib Desert where little vegetation occurs on sand dunes, the transport of organic matter by wind and its deposition on dunes provides the organic matter that is the basis for a food chain including detritivorous beetles, especially tenebrionids such as Onomacris spp. (Lovegrove, 1993). Were the wind patterns to change and the source of organic matter to be cut off, it is likely that there would be significant negative effects on animals in this dune system, including not only the beetles but also the lizards that feed on them. Clearly, sand dunes represent a specialized desert landform with properties very different from those of less mobile desert substrates. Detailed discussion of sand dunes is beyond the scope of this particular chapter. The reader is referred to Bowers (1982) and Danin (1996) for biological effects of dune environments. For the present purpose, unusual movements of dunes mediated by human activity, for instance by destabilizing existing dunes, are a disturbance. The normal movement of dunes has to be considered as an integral part of the disturbance regime of the dune system, but a foreign disturbance when dune sands cover adjacent non-dune vegetation (Goudie, 1978, 1983). Studies in the Mojave Desert of California determined that wind and the consequent deposition of dust on plants increased leaf temperature and lowered maximum rates of photosynthesis for three species (Atriplex canescens, Hymenoclea salsola, and Larrea tridentata; Sharifi et al., 1997). Interestingly,

James A. MacMAHON

at Rock Valley, Nevada, another Mojave Desert site, dust did not appear to alter vegetation composition (Rundel and Gibson, 1996). In some areas of the Mojave Desert the use of off-road vehicles seems to be responsible, in part, for dust storms (Webb and Wilshire, 1983). Plumes from such events may cover more than 1700 km2 and can be seen from space, as was the case in the western Mojave Desert in California on 1 January 1973 (Nakata et al., 1976). Dust storms may cause death to plants. In the Nile Delta, military operations between 1940 and 1943, in which tanks and other military vehicles broke up the undisturbed hard desert surfaces, increased the frequency of dust storms (Smith, 1984). A potential consequence of increases in winds and their related geomorphic effects is that they may influence a variety of organisms; for instance, many animals with waxy protective layers which prevent desiccation, may have these abraded, causing death through the destruction of the water barrier. Wind-transported dust can cover roads and pipelines, powerlines can be destroyed, and dwellings can be buried. These disturbances may have major economic consequences (Allan and Warren, 1993). Like many other “disturbances” discussed in this chapter, wind is a natural phenomenon in deserts. The geomorphic results of wind are often either erosive or depositional events, occurring over long periods of time. Numerous organisms are well adapted to loose substrates deposited by wind and their subsequent movement. Wind causes geomorphic disturbance only when it occurs in an area not generally exposed to wind, or when an area that had become stabilized by the establishment of natural ecosystems has been destabilized through human influences or by natural forces. Fire Fire is an unusual phenomenon in undisturbed deserts. If one looks at the aridity gradient from transitional desert grasslands to extreme deserts essentially without vegetation, the frequency of fire decreases dramatically to zero. This is no surprise; it is due to an increasingly meager and dispersed fuel load along the gradient. In altered habitats there are significant changes caused by different disturbances. Deserts, in response to unusually high annual rainfall at least two years in succession and/or invasion by alien plants, become more susceptible to fire. In contrast, when transition grasslands are exposed to drought or high grazing

DISTURBANCE IN DESERTS

intensity, fire frequency tends to decrease. These effects are significant, because many of the grassland components in the transitions to deserts are highly adapted to fire, and the compositional integrity of these ecosystems may depend on fire, whereas many desert species cannot survive fires. Thus, the grasslands are invaded by desert species and the desert either rebounds (O’Leary and Minnich, 1981) or is dramatically altered (McLaughlin and Bowers, 1982). In the presence of alien annuals that perpetuate a more frequent fire cycle (Rundel and Gibson, 1996), some desert shrub species will simply be displaced. The recovery of desert systems from fire may depend, in part, on the effects of the vegetation before the fire on the soils. For example, in some areas that had supported creosotebush, annuals did not colonize after a fire, whereas interspaces were colonized (Adams et al., 1970). The reason for this phenomenon is not clear. For particular species, fire may be especially damaging. A single fire in the Sonoran Desert may destroy up to 68% of all of the individuals of saguaro (Carnegiea gigantea; Rogers, 1985). Since saguaros require 30 years to reach reproductive maturity (Steenbergh and Lowe, 1977), fires with a return frequency of less than 30 years could essentially remove saguaro from the communities. There may be numerous species that respond similarly but that have not been as well studied. In contrast, creosotebush (Larrea tridentata) may burn to the ground but sprout back in a short time (O’Leary and Minnich, 1981). This sprouting ability is undoubtedly an adaptation to being covered by loose moving soil, but is also a preadaptation to fire. A second example, with additional ramifications for the effect of fire, involves ironwood (Olneya tesota), generally the tallest and oldest (up to 1200 yr) subtree in the Sonoran Desert. More species of perennials occur under ironwood canopies than in the open. Of 65 species in an area, 52 occur under ironwood, 31 of them only in this situation (B´urquez and Quintana, 1994). If an alien species, buffel grass (Cenchrus ciliaris = Pennisetum ciliare) is established there, it builds up fuel and increases fire probability, and the consequent fires kill ironwood and its associated perennials. The post-fire environment is dominated by buffel grass, and perennials cannot reestablish, the composition of the communities thus being altered (B´urquez and Quintana, 1994). Some of the interesting general characteristics of desert fires have been highlighted for the Sonoran Desert. In a period of 29 years, 1611 fires consumed

317

41 000 ha in an area of 391 000 ha of upland Sonoran Desert (Schmid and Rogers, 1988). Fires caused by lightning were fewer in number, but burned twice as great an area as those set by humans, and covered a greater area in desert than in non-desert vegetation (Rogers, 1986). When two consecutive years of aboveaverage winter precipitation occurred, the density of native annuals increased significantly. As annuals died and created fuel, the desert, as expected, became more susceptible to burning, and fire frequency increased more than after single wet years (Rogers and Vint, 1987). It has been estimated that, in natural systems in the Sonoran Desert, the return time between fires is about 295 years (Rogers, 1986). This return time would allow many species to re-establish, since average-sized fires would require 276 years to burn every hectare of land, all other things being equal. If the fire return time is shortened by climate change or the invasion by alien annuals, it could be devastating to native communities. Detailed data such as these are less available for natural systems elsewhere. In all areas of the world, however, an increase of alien annual species in deserts generally leads to an increase in fire frequency (Brown and Gubb, 1986). This has been especially well documented in the Great Basin of North America, where the invader Bromus tectorum has increased fire frequency in the Great Basin Desert dominated by sagebrush, and in many areas has virtually eliminated big sagebrush (Artemisia tridentata) (Smith and Nowak, 1990). Invasion of Bromus rubens has had a similar effect in the Mojave Desert (Rundel and Gibson, 1996). It is interesting to note that alien species are not invaders only on sites that have been disturbed by human forces (Knapp, 1996). Many sites that are completely undisturbed have been invaded by these species, leading to the consequent increase in the susceptibility of the system to fire and perhaps its longterm alteration (Brandt and Rickard, 1994). In Australia, fire has demonstrably influenced aridland vegetation in the pre- and post-Quaternary (Kemp, 1981). Fires have been used by aboriginals for the past 40 000 years, and may have substantially influenced the flora because of differential effects on fire-susceptible and fire-tolerant species (Williams and Calaby, 1985; Walter and Breckle, 1986). Burned areas are also general features of the contemporary Australian desert environment. In the 1974–75 fire season, 12% of the continent, including much desert, was burned (Cunningham, 1981). In contrast, some places in South Africa, for example the

318

Karoo, are generally thought to be little affected by fire as a significant factor, despite the fact that limited local fires are caused by lightning (Huntley, 1984). As mentioned above, the influence of fire in areas transitional to grasslands is generally to reduce the shrub component, since many desert shrub species are susceptible to fire. Occupation of transitional sites by humans has decreased fire frequency, favoring shrub invasion. The relative importance of fire in increasing shrub-dominated lands at the expense of grasslands has been debated for decades. The actual cause of these invasions is likely quite complex. In addition to fire, climate change and grazing have been proposed as the agents of change. Whatever the cause, shrub incursions into grasslands have been demonstrated photographically (Hastings and Turner, 1965; Humphrey, 1974, 1987). Desert animals are often relatively little affected by fire for two reasons. First, aboveground species may be large enough and swift enough to outrun the fire, while other species seek refuge below ground (Polis and Yamashita, 1991), and are insulated by a few centimeters of soil from the effects of the fire. This even includes relatively slow-moving species like desert tortoises (Gopherus agassizii) (pers. observ.). Animals Most of this chapter emphasizes the responses of plants to disturbance regimes. Obviously, as the plant community is altered, so are the associated animal communities. While this is not the place for a detailed analysis of animal communities, I want to mention three kinds of interactions between plants and animals in the context of disturbance. First, animals can positively influence plants in ways that may help the plant avoid the effects of disturbance; second, animals may act as disturbing agents; and, third, animals themselves may respond to disturbances directly or to alterations in plants caused by disturbances. I will briefly discuss their positive and negative effects on plants and will not discuss direct effects of disturbance on animals here, but include them under the various agents of disturbance where appropriate. Aside from the obvious influence of animals on processes such as pollination, movement of mycorrhizal spores, and dispersal of seeds (Chambers and MacMahon, 1994), their digging activities may increase the rates of water infiltration around the bases of plants by as much as 21% (Laundre, 1993). Digging

James A. MacMAHON

also aerates soils that would otherwise be compacted, and may move substantial amounts of organic matter from above the ground to below it, in essence enriching below-ground soils (Lightfoot and Whitford, 1990). Decreases of some animal populations may act as a disturbance factor for some co-evolved plants, or even for animal species that merely occur in the same community. There are a variety of ways in which animals negatively influence plants. Obviously, they consume plants or plant parts. However, to qualify as a disturbance, this would have to be in significant proportions that are not usually seen. Some cases of consumption that reach the point of being a disturbance include outbreaks of grasshoppers (Locustana and Schistocerca species) and lepidopterous larvae in several African deserts (Werger, 1986) where many plant species are decimated. Fortunately for humans, grasshopper control measures have reduced the effects of these periodic outbreaks. In the past, outbreaks have been very serious and have been recorded for thousands of years. According to Popov et al. (1984, p. 150), Pliny recorded an outbreak in 125 BC that caused death to 800 000 people in Cyrenaica and 300 000 in Tunisia. There are also cases where single species of plants have essentially been removed through animal feeding. Crawford (1991) observed larvae of chrysomelid beetles completely defoliating the evening primrose (Oenothera) plants on New Mexican dunefields. Tarbush (Flourensia cernua) endures up to 30% defoliation by larval Lepidoptera (especially Bucculatrix flourensiae) as well as by a chrysomelid beetle (Zygogramma tortuosa) (Schowalter, 1996). The longterm effects of these defoliations are not known, but simulation of browsing in other systems suggests that there may be significant effects (Bilbrough and Richards, 1993). Exclosure studies suggest the possible magnitude of both positive and negative consumer effects on plants. In a 50-year study of exclusion of lagomorphs (Lepus californicus and Sylvilagus audubonii) from creosotebush-dominated communities in New Mexico, there was a 30-fold increase in the basal area of spike dropseed (Sporobolus contractus), and significant increases of honey mesquite (Prosopis glandulosa), tarbush, and mariola (Parthenium incanum), in the exclosure plots (Gibbens et al., 1993). Clearly, under conditions of high animal population densities, longterm influences on shrub and grass populations may be significant.

DISTURBANCE IN DESERTS

The numbers of seeds that animals collect may influence certain highly-favored plant species (Crist and MacMahon, 1992). In a transition zone between the Chihuahuan Desert and desert grassland in southeastern Arizona, exclusion of three species of granivorous kangaroo rats (Dipodomys spp.) caused increases in density of tall perennial and annual grasses, and in populations of rodents characteristic of arid grasslands (Brown and Heske, 1990). The continuation of these studies has suggested that exclusion of birds and rodents increased density of winter and summer annuals, but that the winter annuals were more sensitive (Guo et al., 1995). Finally, these general animal effects may be more complicated than they appear. Individual plants respond differently to animal damage, and the animals respond differently to individual plants of the same species. Individual creosotebushes varied from being seldom and lightly browsed to having 90% of their branches clipped in a single month. In an experiment, jackrabbits browsed more heavily on plants with a history of being browsed than on those which had been only lightly browsed, and less often on artificially “browsed” (clipped) shrubs than on controls, the latter result suggesting induced resistance to browsing and the former suggesting low constitutive resistance (Ernest, 1994). Most of the examples of the negative effects of plant consumption by animals presented above have related to warm deserts; however, there are similar effects, especially through girdling caused by rodent populations, on cold-desert shrubs such as Artemisia tridentata (Parmenter et al., 1987). There are some indirect effects of animal feeding activities. For some plants, such as saguaro, significant amounts of tissue may be removed from individuals by browsing rodents (Neotoma), lagomorphs, and bighorn sheep (Ovis canadensis). In most cases the amount of material removed does not directly kill the cactus, even though it may appear dramatic. However, the individuals become more susceptible to death from freezing, from being blown over, and from fungal infections (Steenbergh and Lowe, 1977). Natural repair Natural ecosystem repair is generally referred to as “succession” (Clements, 1916). There have been varying viewpoints about whether or not succession occurs in deserts. The controversy is often stimulated by the observation that, following a disturbance,

319

there does not seem to be a set of plant species replacements leading to a climax, but rather the species forming the “climax vegetation” seem also to form the pioneering vegetation; there may also be several simultaneously co-occurring groups of species early in succession, which include colonizers as well as species characteristic of mature stands. In an experimental study of disturbance in the Negev Desert, Evenari and Gutterman (1976) found that 16 of 20 species typical of mature communities germinated during the first year after the disturbance. No “pioneer” flora was obvious. Recently, workers have warned against using the term “succession” because it obfuscates the actual dynamic processes involved in vegetation regeneration (Webb et al., 1987; McAuliffe, 1988). It should be noted that, although McAuliffe (1988) suggested not using the word “succession”, he used the term in a paper in 1991 indicating that the process may occur in deserts, but is not typical, and requires more careful interpretation in arid areas than in more mesic areas where the concept was developed (Goldberg and Turner, 1986; McAuliffe, 1991). Some factors causing deserts to appear to undergo atypical succession include timing of establishment, the biology of individual species, and the particular agent of disturbance, among others. Following a disturbance in an arid area, there may be a long period before establishment occurs, mostly because of the episodic nature of the conditions necessary for establishment (MacMahon, 1981). (See the section on Water, pp. 311–314.) When plants do establish, some species require the presence of others to provide cover for their seedlings, the nurseplant phenomenon discussed above. Under certain circumstances, the very same species may be able to establish in the open. This compounds the problem of defining a typical successional sequence. One observation (McAuliffe, 1988) suggests that, if only presence/absence data are used there are few differences between early and later successional stands in desert areas, paralleling the Evenari and Gutterman (1976) results. For some sites, especially transitional sites, there may be some species that can be predicted to occur early in the process of recovery – pioneerlike species (McLendon and Redente, 1990). For any disturbed site, the actual species list for the community depends on the type and age of disturbance, and whether it was natural or anthropogenic in origin (Webb et al., 1987). Additionally, the type and degree of disturbance affects the trajectory of succession,

320

James A. MacMAHON

Fig. 12.7. Digraph model showing principal transitions among states in a species rich Sonoran Desert community. This model depicts a realistic set of relations among desert species, in which every state can reach every other state, eventually. This potential outcome, in part, helps to explain the mosaic-like patches of species in deserts and suggests that a single-trajectory for a successional pathway is unlikely. See text for discussion. Reprinted from McAuliffe (1988) with permission.

but not the long-term outcome – that is, a variety of trajectories seem to converge on the same final vegetation (McLendon and Redente, 1990). A further complication is that alien species, now abundant in deserts around the world, can slow the establishment of native species by pre-empting a site (Allen and Knight, 1984). In the Canyonlands National Park, Utah, Kleiner (1983) found that arid grasslands subject to grazing may return to a state mimicking their pre-disturbance condition in what appears to be a directional way when grazing is prevented. This return (recovery) included the development of cryptogamic crusts. In other areas where grazing has been studied, the season of grazing had more effect on the trajectory of recovery than the grazing intensity (Whisenant and Wagstaff, 1991). As mentioned previously, the use of nurse-plants by some species, conditioning the early successional environment, is typical in many deserts. For example, in the Karoo, mound-building members of the Mesembryanthemaceae establish early, and the mounds are later invaded by woody species. Such sequences of dependent species look like successional seres. This process is also affected by burrowing of animals (Yeaton and Esler, 1990). One problem with applying the normal concept of succession to deserts is that many of the plants are

extremely long-lived, even persisting for thousands of years. Given this span of time, some plants may persist while the substrates on which they occur are dramatically altered, as are other species in the community. McAuliffe (1991) observed that alluvial terraces of different age supported different plant communities, illustrating the importance of substrates in determining the trajectory of succession. Later, he pointed out that, in contrast to a common assumption, desert soils may indeed be rather well developed and very old, even though they may not appear so if one only examines their surface properties (McAuliffe, 1994). He also asserted that invasion by undesirable species such as the non-native annuals mentioned above is likely, in part, to be a function of the geomorphic surfaces on which the native community occurs. I believe that the process of succession or natural repair does occur in deserts, but (1) the establishment phase may be episodic, (2) the pioneer species may (MacMahon, 1981; Zedler, 1981) or may not (Evenari and Gutterman, 1976; McAuliffe, 1988) be the same as the later successional species, and (3) the geomorphic surfaces may direct the successional outcome as can the type and intensity of the disturbance. Deserts tend to return to their former states, ceteris paribus, even though it may take a long time. For example, there is very slow recovery of vegetation following

DISTURBANCE IN DESERTS

construction of pipelines and roads (Vasek et al., 1975a,b); McAuliffe (1988) estimated that 1000 years may be required for regeneration of the creosotebush community. Carpenter et al. (1986) found that recovery of “old-fields” in the Mojave Desert took 65–130 years, the time required varying with elevation. McAuliffe (1988) presented a system of the states of spaces on a desert landscape and the transitions among these states, which suggests that these approximate a Markov chain. Markov chain models are based on transitions among the states of a system and the characteristic probabilities of these transitions, which depend only on the current state of the system. McAuliffe started with a simple model of transitions among three states: areas occupied by two plant species (Ambrosia and Larrea) and open space. Later he created a digraph model (Fig. 12.7) of a more complex Arizona Upland desert community. In this instance, as was the case for the simple three-state model, any state can be reached from any other state. The time to reach a particular state depends on the probabilities of the state transitions along the whole state-transition path. This model accurately mimicked data collected by McAuliffe in the Sonoran Desert, and was consistent with the relation between a species and its nurse plant, and other biological relationships. The observation that any state can, with some probability, change to any other state suggests at least one reason that a classic single-trajectory view of succession would not be appropriate in deserts. McAuliffe’s approach provides a possible explanation of the apparent variation in the patterns of recovery of disturbed sites in deserts, which has caused so much confusion over the applicability of the concept of succession to desert communities.

321

not likely to directly influence desert plants, which are already adapted to extremes of temperature, nor the animals associated with them. However, associated with average temperature changes are changes in the extremes of temperature and in world-wide circulation patterns and consequently the patterns of rainfall (Woodward, 1987). Schneider and Root (1996) have warned that there may be a great variety of such “surprises” that are far more important than average temperature changes. One might anticipate that, if both the high and low temperatures become more extreme, they may exceed the tolerance limits of some species, particularly if the duration of freezing temperatures increases by as little as 10–20 hours. One may also anticipate that changes in the patterns of world-wide circulation will change the relative proportions of summer and winter rainfall over deserts. This could cause changes in community composition that are related to a complex set of interactions (Cook and Irwin, 1992), which differ according to whether drought is present or absent (Fig. 12.8) (Post,

Global change Global change involves a variety of world-wide dramatic alterations in the earth’s atmosphere and in ecosystems (Solomon and Shugart, 1993). The four causes of change most often mentioned are: an increase in human population; deforestation; an increase in ambient carbon dioxide levels; and an increase in the earth’s mean annual temperature. I will discuss human population influences later (pp. 323–325). Here, I confine my comments to climate change and direct effects of an increase in carbon dioxide on deserts, ignoring deforestation because deserts are usually devoid of typical trees. A small increase in average annual temperature is

Fig. 12.8. Feedbacks in terrestrial ecosystem responses to climate change induced by carbon dioxide. Arrows with plus signs (+) indicate processes that have positive effects or increase the rates of other processes. Arrows with minus signs (−) indicate processes that have the opposite effects. GPP represents gross primary production. From Post (1993) with permission.

322

James A. MacMAHON

Fig. 12.9. Hypothesized response of ecosystems to enhanced carbon dioxide in relation to prevailing nutrient and water availability. Those ecosystems that have been studied in the field are marked with heavy outlines; those where aspects of the system have been studied under controlled environmental conditions are shown using broken outlines. The remaining ecosystems are largely unstudied. Note that deserts are generally predicted to be responsive to changes in carbon dioxide, and that different deserts differ in their responses. From Mooney et al. (1991) with permission.

1993). It is predicted for both Australia and the Great Basin of North America that winter precipitation will decrease and summer precipitation increase, conditions to which the vegetation is not adapted. Also, summer precipitation is likely to increase the proportion of grasses and, in turn, the frequency and intensity of fires. Often overlooked are the direct effects of increase in carbon dioxide concentration. These include effects on photosynthetic rates, stomatal conductance, decomposition rates, herbivore consumption, and competition (Bazzaz, 1996). It is postulated that such direct effects occurred in Chihuahuan Desert ecosystems during the last glaciation (Cole and Monger, 1994). These studies indicate a shift on alluvial fans in New Mexico from

domination by C4 grasses to domination by C3 shrubs between 7000 and 9000 years ago, correlated with a rapid increase in carbon dioxide concentration. There is no reason to believe that contemporary systems might not respond in a similar manner, changing the composition of vegetation and a variety of autogenic ecosystem processes including fire frequency (Betancourt, 1996). The assertion about the role of carbon dioxide in shrubland expansion is, however, not without controversy (Archer et al., 1995). Likely direct responses of vegetation to carbon dioxide concentration and to climate change will vary from biome to biome and even within biomes, as indicated for deserts on alluvial or other surfaces (Fig. 12.9). It has been argued by Schlesinger et al.

DISTURBANCE IN DESERTS

323

(1990) that the boundaries of desert vegetation might be the best places to look for vegetation shifts produced by climate change because of the sensitivity of arid systems to carbon dioxide concentration. To date, these boundary changes are not obvious, although this does not mean that they are not occurring (Schlesinger and Gramenopoulos, 1996). The need for careful, long-term monitoring of changes of vegetation composition in arid transitional areas is clearly indicated.

HUMANS AS AGENTS OF DISTURBANCE

Deserts are home to about 13% (Allan and Warren, 1993) of the world’s human population. Thus, despite their rigorous environments, they are extensively used and inhabited by human beings. Natural disturbances to desert systems can obviously affect many humans and, conversely, human populations can act as agents of disturbance for deserts. In many cases it is difficult to separate natural disturbances from those induced by humans. Historical records of the use of deserts over thousands of years are often reasonably good – for instance, the occupation of deserts by early Egyptian cultures (Zahran and Willis, 1992); but the consequences of those uses are generally not well documented (Werger, 1983). When scientists began studying the nature of deserts, they often viewed landscapes that were already altered. For example, parts of the well-studied Sonoran Desert of Mexico have been occupied by the hunter–gatherer Seri Indians for at least 2000 years with effects that cannot be quantified (Felger and Moser, 1985). The human use of deserts and of transitional areas, especially those with grasslands, has caused the processes of desertification and desertization (Mainguet, 1994). World-wide, more “desert-like” areas are being created at an ever increasing rate. While estimates vary, it is thought that nearly 6×106 ha are subjected to desertification each year (Kassas, 1995). The implication is that human mismanagement, in combination with drought conditions in some areas, is turning grasslands into shrublands that have the physiognomy of a desert. There are even cases where the actions of humans, through the removal of plants by domestic livestock, are thought to have changed the albedo of the earth’s surface, leading to an altered rainfall pattern and, consequently, to desertification (Charney, 1975, 1977). It is suggested that the removal of vegetation increases surface reflectivity, in turn changing the patterns of

Fig. 12.10. Daily precipitation differences that are calculated to occur at various latitudes in the Sahel if vegetation is removed and there are albedo changes of either 14 or 35%. Other percentage changes in albedo would lead to slightly different values; however, these two changes bracket the most likely scenarios. From Charney (1975) with permission.

vertical air movement and consequently amounts of rainfall (Fig. 12.10). I will generally not consider areas that have clearly become deserts through human activities, although in some areas of the world what has caused desert conditions may not be obvious. It is important to add a caveat to my description of the desertification/mismanagement cycle. Natural forces can cause “desertification”. Changes in climatic patterns can certainly tip the balance between shrublands and grasslands, and have done so in the past in many areas including the Sahara Desert (Le Hou´erou, 1992b), Australia (Smith, 1982), the Negev Desert (Evenari et al., 1971), the Chihuahuan Desert (Dick-Peddie, 1993), and the Karakum (Petrov, 1976). While there does not seem to be any world-wide pattern of change in the total annual rainfall in arid lands, many areas have experienced prolonged declines in precipitation – for example, northern Chile (Burgos et al., 1991),

324

North Africa (Le Hou´erou, 1993), and the southwestern United States and adjacent Mexico (Humphrey, 1987). The apparent magnitude of desertification can be appreciated by the example of the expansion of the Sahara Desert into the semi-arid Sahel. In the 1960s and 1970s, the Sahara expanded southward by a distance of about 350 km. It is hypothesized that this expansion was caused, in part, by human activities (livestock production, fuel-gathering, population growth), but was significantly driven by a drought that was severe from 1968 to 1973, and continued at some level to the late 1980s and early 1990s. Rains returned in the 1980s and some vegetation was restored (Tucker et al., 1991). The return of vegetation caused many to suggest that anthropogenic influences were, in fact, minor and that this waxing and waning of the Sahara was a regular desert phenomenon. Regardless of the final resolution of this debate (Hutchinson, 1996; Rietkerk et al., 1996), a large area extending beyond the Sahel (Ellis and Galvin, 1994), with its dependent inhabitants, was changed from productive grasslands to a lowerproduction desert-like landscape, presumably by a combination of natural and anthropogenic disturbances. Introductions of animals that are managed for human benefit, such as cattle, goats, and sheep, have dramatic direct effects on vegetation, soils, run-off, infiltration rates, etc. For example, most of Afghanistan and Iran are so altered by domestic animals that the original vegetation is hard to discern (Breckle, 1983). Introduced species that become feral may also have influence. For example, non-native honeybees (Apis mellifera) remove at least 90% of the pollen of saguaro cacti (Schmidt and Buchmann, 1986). This dramatic use of pollen by an introduced species may have significant effects on normal pollinators of saguaros such as bats, native bees, doves, and moths. The presence of feral burros (Equus asinus) in Death Valley National Park (U.S.A.) has decreased concentrations of native species such as Ambrosia dumosa, Oryzopsis hymenoides, and Sphaeralcea ambigua and some grasses (Loope et al., 1988). These reductions are caused by browsing, grazing, and trampling the native plants. The effect of introduction of rabbits into Australian arid lands is well documented. In addition to altering soil by their burrowing activities, rabbits have displaced native animals, and consumption by rabbits has decreased establishment of some plant species, for instance some acacias (Williams and Calaby, 1985). The introduction of plants such as tamarisk (Tamarix spp.) can have significant negative influences on

James A. MacMAHON

animals and plants. In Death Valley, the presence of tamarisk, a desert riparian species, has lowered the water table, threatening native marsh plants as well as the desert pupfish Cyprinodon nevadensis (Loope et al., 1988). In the Namib Desert of southwestern Africa, introduced species such as Nicotiana glauca (desert tobacco) may be crowding out some native riparian species (Loope et al., 1988). Introduced annuals can alter fire frequency; however, their direct competitive effects are also of concern in some areas. For example, the introduced grass Pennisetum ciliare seems to be displacing the brittle bush (Encelia farinosa) in the Sonoran Desert, at least on some longterm observation plots (Burgess et al., 1991). Such competitive displacement of species is a common result of the introduction of aliens (Williamson, 1996). Tourism effects are highly variable in desert areas; but visitors, anxious to experience deserts and their stark beauty, often trample native vegetation and alter the course of waterways because of rilling effects produced by “path-making”; their footprints and vehicle tracks can disturb microphytic soil crusts, decreasing nitrogenase activity by 30–100% (Belnap, 1996), and have other damaging effects on these somewhat fragile systems (Cunningham, 1981). Offroad vehicles negatively affect flora and fauna of desert areas throughout the world (Seely and Hamilton, 1978); they cause instability of some geological substrates, while compacting others (Webb and Wilshire, 1983). Compacted soils prevent natural system repair, and may take 80–140 years to regain their pre-disturbance physical properties. Interestingly, this process is fastest in colder areas, probably because of the soil-loosening effects of freezing and thawing (Webb et al., 1986). Surprisingly, some deserts recover from human perturbations as extreme as atomic bomb testing in relatively short periods (Yool, 1998). Human repair of disturbed systems It is not possible to cover fully the attempts of humans to repair the results of disturbances in desert ecosystems. Suffice it to say that the unusual characteristics of the biotic and abiotic environment of deserts cause humans to adopt a set of restoration, reclamation, or rehabilitation techniques different from those they would use in more mesic areas. Throughout the arid world, workers have been attempting to divorce themselves from the traditional agricultural approaches to restoration, and to make new sets of

DISTURBANCE IN DESERTS

plans tailored specifically to desert conditions. These include: dealing with the dispersion pattern of plants during the restoration process to mimic the clumps that provide the nurse-plant effect and accumulate organic matter; attention to the establishment phase of recovery so as to obviate the effects of low and unpredictable rainfall; and other actions tailored to these environments (Majer, 1989; Aronson et al., 1993; MacMahon, 1997). The need to develop these specific techniques has been well summarized for plants (Allen, 1994), soils and their microbial populations (Kieft, 1991), and animals (Majer, 1989). As a final note, because desert plants are so physiologically resilient, once they are established they may actually be used in semi-arid areas to reclaim sites after severe disturbance. Examples of such rehabilitation efforts include the use of cacti (Opuntia spp.), Atriplex species, and other plants in the Mediterranean Basin to control the erosional effects of both water and wind, thus halting the negative effects of opening the vegetation to erosion (Le Hou´erou, 1996a,b). While not all of these past efforts have been successful, there are many desert species that will likely be found to be useful in the future.

CONCLUSION

I have discussed natural disturbance processes in desert systems and, in a much more superficial way, alluded to anthropogenic disturbances. One might ask the question, “Why is disturbance in deserts important?” In natural systems, disturbance is a characteristic of the desert that organisms seem to withstand through survival or regeneration. Anthropogenic disturbances often take deserts beyond their capacity to maintain themselves by introducing new agents of stress or creating values of environmental factors that exceed the limits of organisms. The loss of desert species and the degradation of desert communities is a loss for humankind that is important in the same ways that the more publicized loss of rainforests is important. Many plants and animals in deserts represent untapped genetic resources, act as valuable sources of food, fiber and forage, are known to contain many medicinal products (Goodin and Northington, 1985), and provide sites upon which agriculture can be developed if scientific knowledge is carefully implemented to maintain sustainability of those land surfaces (Hoekstra and Shachak, 1999).

325

With a burgeoning human population, all species occurring in deserts, as well as the desert ecosystems themselves, are necessary to provide for the welfare of humans. At the same time, some deserts must be set aside in their natural state, with their natural disturbance regimes, as benchmarks against which overall changes in the global environment and man’s influence on natural systems can be measured. Economic significance of deserts does not encompass the aesthetic attributes of natural desert systems. Many people, including myself, find a solace and spirituality in deserts which is unmatched elsewhere and cries out for protection from anthropogenic disturbances.

REFERENCES Adams, S., Strain, B.R. and Adams, M.S., 1970. Water-repellent soils, fire, and annual plant cover in a desert scrub community of southeastern California. Ecology, 51: 696–700. Allan, T. and Warren, A., 1993. Deserts. The Encroaching Wilderness. Oxford University Press, New York, 176 pp. Allen, E.B., 1991. Temporal and spatial organization of desert plant communities. In: J. Skujins (Editor), Semiarid Lands and Deserts. Marcel Dekker, New York, pp. 193–208. Allen, E.B., 1994. Restoration ecology: Limits and possibilities in arid and semiarid lands. In: B.A. Roundy, E.D. McArthur, J.S. Haley and J.S. Mann (Editors), Proceedings: Wildland Shrub and Arid Land Restoration Symposium, General Technical Report INT-GTR-315. USDA, Forest Service, Intermountain Research Station, Ogden, Utah, pp. 1–9. Allen, E.B. and Knight, D.H., 1984. The effects of introduced annuals on secondary succession in sagebrush-grassland, Wyoming. Southwest. Nat., 29: 407–421. Archer, S., Schimel, D.S. and Holland, E.A., 1995. Mechanisms of shrubland expansion: land use, climate, or CO2 . Climatic Change, 28: 91–99. Aronson, J., Floret, C., Le Floc’h, E., Ovalle, C. and Pontanier, R., 1993. Restoration and rehabilitation of degraded ecosystems in arid and semi-arid lands. I. A view from the south. Restoration Ecol., 1: 8–17. Arritt, S., 1993. The Living Earth Book of Deserts. Reader’s Digest Association, Inc., Pleasantville, New York, 224 pp. Axelrod, D.I., 1979. Age and origin of Sonoran Desert vegetation. Occas. Pap. Calif. Acad. Sci., 132. Bazzaz, F.A., 1996. Plants in Changing Environments: Linking Physiological, Population, and Community Ecology. Cambridge University Press, Cambridge, 320 pp. Belnap, J., 1996. Soil surface disturbances in cold deserts: effects on nitrogenase activity in cyanobacterial-lichen soil crusts. Biol. Fertil. Soils, 23: 362–367. Betancourt, J.L., 1996. Long- and short-term climate influences on southwestern shrublands. In: J.R. Barrow, E.D. McArthur, R.E. Sosebee and R.J. Tausch (Editors), Proceedings: Shrubland

326 Ecosystem Dynamics in a Changing Environment. USDA, Forest Service, Intermountain Research Station, Ogden, Utah, pp. 5–9. Bilbrough, C.J. and Richards, J.H., 1993. Growth of sagebrush and bitterbrush following simulated winter browsing: mechanisms of tolerance. Ecology, 74: 481–492. Bolin, S.B. and Ward, T.J., 1987. An analysis of runoff and sediment yield from natural rainfall plots in the Chihuahuan Desert. In: E.F. Aldon, C.E. Gonzales Vicente and W.H. Moir (Editors), Strategies for Classification and Management of Native Vegetation for Food Production in Arid Zones: Symposium Proceedings. General Technical Report RM-150. USDA, Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado, pp. 196–200. Bowers, J.E., 1981. Catastrophic freezes in the Sonoran Desert. Desert Plants, 2: 232–236. Bowers, J.E., 1982. The plant ecology of inland dunes in western North America. J. Arid Environ., 5: 199–220. Bowers, J.E., Webb, R.H. and Rondeau, R.J., 1995. Longevity, recruitment and mortality of desert plants in Grand Canyon, Arizona, USA. J. Vegetation Sci., 6: 551–564. Brandt, C.A. and Rickard, W.H., 1994. Alien taxa in the North American shrub-steppe four decades after cessation of livestock grazing and cultivation agriculture. Biol. Conservation, 68: 95–105. Breckle, S.W., 1983. Temperate deserts and semi-deserts of Afghanistan and Iran. In: N.E. West (Editor), Temperate Deserts and Semi-deserts. Ecosystems of the World 5. Elsevier, Amsterdam, pp. 271–319. Brown, C.J. and Gubb, A.A., 1986. Invasive alien organisms in the Namib Desert, Upper Karoo and the arid and semi-arid savannas of western southern Africa. In: I.A.W. Macdonald, E.J. Kruger and A.A. Ferrar (Editors), The Ecology and Management of Biological Invasions in Southern Africa. Oxford University Press, Cape Town, South Africa, pp. 93–108. Brown, J.H. and Heske, E.J., 1990. Control of a desert–grassland transition by a keystone rodent guild. Science, 250: 1705–1707. Burgess, T.L., Bowers, J.E. and Turner, R.M., 1991. Exotic plants at the Desert Laboratory, Tucson, Arizona. Madro˜no, 38: 96–114. Burgos, J.J., Fuenzalida, P.H. and Molion, L.C.B., 1991. Climate change predictions for South America. Climatic Change, 18: 223–239. B´urquez, A. and Quintana, M.A., 1994. Islands of diversity: Ironwood ecology and the richness of perennials in a Sonoran Desert biological reserve. In: G.P. Nabhan and J.L. Carr (Editors), Ironwood: An Ecological and Cultural Keystone of the Sonoran Desert. Occasional Papers in Conservation Biology, 1. Conservation International, Washington, D.C., pp. 9–27. Carpenter, D.E., Barbour, M.G. and Bahre, C.J., 1986. Old field succession in Mojave Desert scrub. Madro˜no, 33: 111–122. Chambers, J.C. and MacMahon, J.A., 1994. A day in the life of a seed: Movements and fates of seeds and their implications for natural and managed systems. Annu. Rev. Ecol. Syst., 25: 263–292. Charney, J.G., 1975. Dynamics of deserts and drought in the Sahel. Q. J. R. Meteorol. Soc., 101: 193–202. Charney, J.G., 1977. A biogeophysical feedback mechanism in arid lands. In: Y. Mundlack and S.F. Singer (Editors), Arid Zone Development. Ballinger, Cambridge, Massachusetts, pp. 181– 188. Clements, F.E., 1916. Plant Succession: An Analysis of the

James A. MacMAHON Development of Vegetation. Carnegie Institution of Washington Publication Number 242, 512 pp. Cody, M.L., 1993. Do cholla cacti (Opuntia spp., subgenus Cylindropuntia) use or need nurse plants in the Mojave Desert? J. Arid Environ., 24: 139–154. Cole, D.R. and Monger, H.C., 1994. Influence of atmospheric CO2 on the decline of C4 plants during the last deglaciation. Nature, 368: 533–536. Cook, J.G. and Irwin, L.L., 1992. Climate–vegetation relationships between the Great Plains and Great Basin. Am. Midl. Nat., 127: 316–326. Crawford, C.S., 1991. Animal adaptations and ecological processes in desert dunefields. J. Arid Environ., 21: 245–260. Crawford, C.S. and Gosz, J.R., 1986. Dynamics of desert resources and ecosystem processes. In: N. Polunin (Editor), Ecosystem Theory and Application. John Wiley, New York, pp. 63–88. Crist, T.O. and MacMahon, J.A., 1992. Harvester ant foraging and shrub-steppe seeds: interactions of seed resources and seed use. Ecology, 73: 1768–1779. Cunningham, G.M., 1981. Desert vegetation. In: R.H. Groves (Editor), Australian Vegetation. Cambridge University Press, Cambridge, pp. 377–389. Danin, A., 1996. Plants of Desert Dunes. Adaptations of Desert Organisms. Springer, Berlin, 177 pp. de Soyza, A.G., Whitford, W.G., Martinez-Meza, E. and Van Zee, J.W., 1997. Variation in creosotebush (Larrea tridentata) canopy morphology in relation to habitat, soil fertility and associated annual plant communities. Am. Midl. Nat., 137: 13–26. Dick-Peddie, W.A., 1993. New Mexico Vegetation, Past, Present, and Future. University of New Mexico Press, Albuquerque, 244 pp. Ehleringer, J.R. and House, D., 1984. Orientation and slope preference in barrel cactus (Ferocactus acanthodes) at its northern distribution limit. Great Basin Nat., 44: 133–139. Ehleringer, J.R., Phillips, S.L., Schuster, W.S.F. and Sandquist, D.R., 1991. Differential utilization of summer rains by desert plants. Oecologia, 88: 430–434. Ellis, J. and Galvin, K.A., 1994. Climate patterns and land-use practices in the dry zones of Africa. BioScience, 44: 340–349. Ernest, K.A., 1994. Resistance of creosotebush to mammalian herbivory: temporal consistency and browsing-induced changes. Ecology, 75: 1684–1692. Evenari, M., 1985. The desert environment. In: M. Evenari, I. Noy-Meir and D.W. Goodall (Editors), Hot Deserts and Arid Shrublands. Ecosystems of the World 12A. Elsevier, Amsterdam, pp. 1–22. Evenari, M. and Gutterman, Y., 1976. Observations on the secondary succession of three plant communities in the Negev Desert, Israel. I. Artemisietum herbae albae. In: R. Jacques (Editor), Etudes de Biologie Vegetale – Hommage au Prof. P. Chouard. CNRS, Paris, pp. 57–86. Evenari, M., Shanan, L. and Tadmor, N., 1971. The Negev. The Challenge of a Desert. Harvard University Press, Cambridge, Massachusetts, 345 pp. Evenari, M., Noy-Meir, I. and Goodall, D.W. (Editors), 1985. Hot Deserts and Arid Shrublands. Ecosystems of the World 12A. Elsevier, Amsterdam, 365 pp. Evenari, M., Noy-Meir, I. and Goodall, D.W. (Editors), 1986. Hot Deserts and Arid Shrublands. Ecosystems of the World 12B. Elsevier, Amsterdam, 451 pp.

DISTURBANCE IN DESERTS Felger, R.S. and Moser, M.B., 1985. People of the Desert and the Sea: Ethnobotany of the Seri Indians. University of Arizona Press, Tucson, 435 pp. Georgiadis, N.J., 1989. Microhabitat variation in an African savanna: Effects of woody cover and herbivores in Kenya. J. Trop. Ecol., 5: 93–108. Gibbens, R.P., Havstad, K.M., Billheimer, D.D. and Herbel, C.H., 1993. Creosotebush vegetation after 50 years of lagomorph exclusion. Oecologia, 94: 210–217. Goldberg, D.E. and Turner, R.M., 1986. Vegetation change and plant demography in permanent plots in the Sonoran Desert. Ecology, 67: 695–712. Goodin, J.R. and Northington, D.K., 1985. Plant Resources of Arid and Semiarid Lands. A Global Perspective. Academic Press, Orlando, Florida, 338 pp. Goudie, A. and Wilkinson, J., 1977. The Warm Desert Environment. Cambridge University Press, Cambridge, 88 pp. Goudie, A.S., 1978. Dust storms and their geomorphological implications. J. Arid Environ., 1: 291–310. Goudie, A.S., 1983. Dust storms in space and time. Prog. Phys. Geogr., 7: 502–530. Grenot, C.J., 1974. Physical and vegetational aspects of the Sahara Desert. In: G.W. Brown Jr. (Editor), Desert Biology, Volume II. Academic Press, New York, pp. 103–164. Grime, J.P., 1979. Plant Strategies and Vegetation Processes. Wiley, New York, 222 pp. Guo, Q., Thompson, D.B., Valone, T.J. and Brown, J.H., 1995. The effects of vertebrate granivores and folivores on plant community structure in the Chihuahuan Desert. Oikos, 73: 251–259. Hastings, J.R. and Turner, R.M., 1965. The Changing Mile: An Ecological Study of Vegetation Change with Time in the Lower Mile of an Arid and Semi-arid Region. University of Arizona Press, Tucson, 317 pp. Hoekstra, T.W. and Shachak, M. (Editors), 1999. Arid Lands Management: Toward Ecological Sustainability. University of Illinois Press, Urbana and Chicago, 279 pp. Humphrey, R.R., 1974. Fire in the deserts and desert grassland of North America. In: T.T. Kozlowski and C.E. Ahlgren (Editors), Fire and Ecosystems. Academic Press, Inc., New York, pp. 365–400. Humphrey, R.R., 1987. 90 Years and 535 Miles. Vegetation Changes Along the Mexican Border. University of New Mexico Press, Albuquerque, 448 pp. Huntley, B.J., 1984. Characteristics of South African biomes. In: P.V. Booysen and N.M. Tainton (Editors), Ecological Effects of Fire in South African Ecosystems. Springer-Verlag, Berlin, pp. 1– 17. Huston, M., 1994. Biological Diversity. The Coexistence of Species on Changing Landscapes. Cambridge University Press, Cambridge, England, 681 pp. Hutchinson, C.F., 1996. The Sahelian desertification debate: a view from the American south-west. J. Arid Environ., 33: 519–524. Jordan, P.W. and Nobel, P.S., 1979. Infrequent establishment of seedlings of Agave deserti (Agavaceae) in the northwestern Sonoran Desert. Am. J. Bot., 66: 1079–1084. Kassas, M., 1995. Desertification: a general review. J. Arid Environ., 30: 115–128. Kemp, E.M., 1981. Pre-Quaternary fire in Australia. In: A.M. Gill, R.H. Groves and I.R. Noble (Editors), Fire and the Australian Biota. Australian Academy of Sciences, Canberra, pp. 3–21.

327 Kieft, T.L., 1991. Soil microbiology in reclamation of arid and semiarid lands. In: J. Skujins (Editor), Semiarid Lands and Deserts. Marcel Dekker, Inc., New York, pp. 209–256. Kleiner, E.F., 1983. Successional trends in an ungrazed, arid grassland over a decade. J. Range Manage., 36: 114–118. Knapp, P.A., 1996. Cheatgrass (Bromus tectorum L) dominance in the Great Basin Desert. Global Environ. Change, 6: 37–52. Laundre, J.W., 1993. Effects of small mammal burrows on water infiltration in a cool desert environment. Oecologia, 94: 43–48. Laycock, W.A., 1991. Stable states and thresholds of range condition on North American rangelands: A viewpoint. J. Range Manage., 44: 427–433. Le Hou´erou, H.N., 1992a. An overview of vegetation and land degradation in world arid lands. In: H.E. Dregne (Editor), Degradation and Restoration of Arid Lands. Texas Tech University, Lubbock, Texas, pp. 127–163. Le Hou´erou, H.N., 1992b. Outline of the biological history of the Sahara. J. Arid Environ., 22: 3–30. Le Hou´erou, H.N., 1993. Changements climatiques et desertisation. Secheresse, 4: 95–111. Le Hou´erou, H.N., 1996a. The role of cacti (Opuntia spp.) in erosion control, land reclamation, rehabilitation and agricultural development in the Mediterranean Basin. J. Arid Environ., 33: 135–159. Le Hou´erou, H.N., 1996b. Drought-tolerant and water-efficient trees and shrubs for rehabilitation of tropical and subtropical arid lands of Africa and Asia. Land Husbandry, 1: 43–64. Le Hou´erou, H.N., Bingham, R.L. and Skerbek, W., 1988. Relationship between the variability of primary production and the variability of annual precipitation in world arid lands. J. Arid Environ., 15: 1–18. Lightfoot, D.C. and Whitford, W.G., 1990. Phytophagous insects enhance nitrogen flux in a desert creosotebush community. Oecologia, 82: 18–25. Loope, L.L., Sanchez, P.G., Tarr, P.W., Loope, W.L. and Anderson, R.L., 1988. Biological invasions of arid land nature reserves. Biol. Conservation, 44: 95–118. Lovegrove, B., 1993. The Living Deserts of Southern Africa. Fernwood Press, Cape Town, South Africa, 224 pp. MacMahon, J.A., 1981. Successional processes: Comparisons among biomes with special reference to probable roles of and influences on animals. In: D.C. West, H.H. Shugart and D.B. Botkin (Editors), Forest Succession: Concept and Application. Springer-Verlag, New York, pp. 277–304. MacMahon, J.A., 1997. Ecological restoration. In: G. Meffe and C. Carroll (Editors), Principles of Conservation Biology, 2nd Edition. Sinauer Associates, Sunderland, Massachusetts, pp. 479–511. MacMahon, J.A., 1999. Warm deserts. In: M.G. Barbour and W.D. Billings (Editors), North American Terrestrial Vegetation, 2nd Edition. Cambridge University Press, New York, in press. MacMahon, J.A. and Wagner, F.H., 1985. The Mojave, Sonoran and Chihuahuan deserts of North America. In: M. Evenari, I. Noy-Meir and D.W. Goodall (Editors), Hot Deserts and Arid Shrublands. Ecosystems of the World 12A. Elsevier, Amsterdam: 105–202. Mainguet, M., 1994. Desertification: Natural Background and Human Mismanagement, 2nd Edition. Springer-Verlag, Berlin, 314 pp. Majer, J.D., 1989. Animals in Primary Succession: The Role of Fauna

328 in Reclaimed Lands. Cambridge University Press, Cambridge, 547 pp. Mattsson, J.O. and Nihlen, T., 1996. The transport of Saharan dust to southern Europe: a scenario. J. Arid Environ., 32: 111–119. McAuliffe, J.R., 1984a. Sahuaro–nurse tree associations in the Sonoran Desert: competitive effects of sahuaros. Oecologia, 64: 319–321. McAuliffe, J.R., 1984b. Prey refugia and the distributions of two Sonoran Desert cacti. Oecologia, 65: 82–85. McAuliffe, J.R., 1988. Markovian dynamics of simple and complex desert plant communities. Am. Nat., 131: 459–490. McAuliffe, J.R., 1991. Demographic shifts and plant succession along a late Holocene soil chronosequence in the Sonoran Desert of Baja California. J. Arid Environ., 20: 165–178. McAuliffe, J.R., 1994. Landscape evolution, soil formation, and ecological patterns and processes in Sonoran Desert bajadas. Ecol. Monogr., 64: 111–148. McLaughlin, S.P. and Bowers, J.E., 1982. Effects of wildfire on a Sonoran Desert plant community. Ecology, 63: 246–248. McLendon, T. and Redente, E.F., 1990. Succession patterns following soil disturbance in a sagebrush steppe community. Oecologia, 85: 293–300. Meigs, P., 1953. World distribution of arid and semi-arid homoclimates. In: Review of Research on Arid Zone Hydrology, Arid Zone Programme, 1. UNESCO, Paris, pp. 203–209. Mooney, H.A., Drake, B.G., Luxmoore, R.J., Oechel, W.C. and Pitelka, L.F., 1991. Predicting ecosystem responses to elevated CO2 concentrations. BioScience, 41: 96–104. Nakata, J.K., Wilshire, H.G. and Barnes, G.G., 1976. Origin of Mojave Desert dust plumes photographed from space. Geology, 4: 644–648. Neal, J.T., 1969. Playa variation. In: W.G. McGinnies and B.J. Goldman (Editors), Arid Lands in Perspective. University of Arizona Press, Tucson, pp. 15–44. Neal, J.T., 1975. Playas and Dried Lakes. Occurrence and Development. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania, 411 pp. Nicholson, S.E., Davenport, M.L. and Malo, A.R., 1990. A comparison of the vegetation response to rainfall in the Sahel and East Africa, using Normalized Difference Vegetation Index from NOAA AVHRR. Climatic Change, 17: 209–241. Nobel, P.S., 1988. Environmental Biology of Agaves and Cacti. Cambridge University Press, Cambridge, 270 pp. Noy-Meir, I., 1973. Desert ecosystems: environment and producers. Annu. Rev. Ecol. Syst., 4: 25–51. Noy-Meir, I., 1974. Desert ecosystems: higher trophic levels. Annu. Rev. Ecol. Syst., 5: 195–214. O’Leary, J.F. and Minnich, R.A., 1981. Postfire recovery of creosote bush scrub vegetation in the western Colorado desert. Madro˜no, 28: 61–66. Parmenter, R.R., Mesch, M.R. and MacMahon, J.A., 1987. Shrub litter production in a sagebrush-steppe ecosystem: rodent population cycles as a regulating factor. J. Range Manage., 40: 50–54. Peters, A.J., Reed, B.C., Eve, M.D. and Havstad, K.M., 1993. Satellite assessment of drought impact on native plant communities of southeastern New Mexico, U.S.A. J. Arid Environ., 24: 305–319. Petrov, M.P., 1976. Deserts of the World. John Wiley, New York, 447 pp.

James A. MacMAHON Polis, G.A. and Yamashita, T., 1991. The ecology and importance of predaceous arthropods in desert communities. In: G.A. Polis (Editor), The Ecology of Desert Communities. University of Arizona Press, Tucson, pp. 180–222. Popov, G.B., Wood, T.G. and Haggis, M.J., 1984. Insect pests of the Sahara. In: J.L. Cloudsley-Thompson (Editor), Sahara Desert. Pergamon Press, Oxford, pp. 145–174. Post, W.M., 1993. Uncertainties in the terrestrial carbon cycle. In: A.M. Solomon and H.H. Shugart (Editors), Vegetation Dynamics and Global Change. Chapman and Hall, New York, pp. 116–132. Reichholf, J.H., 1986. Is Saharan dust a major source of nutrients for the Amazonian rain forest? Stud. Neotrop. Fauna Environ., 21: 251–255. Reynolds, J.F., Virginia, R.A., Kemp, P.R., de Soyza, A.G. and Tremmel, D.C., 1999. Impact of drought on desert shrubs: Effects of seasonality and degree of resource island development. Ecol. Monogr., 69: 69–106. Rietkerk, M., Ketner, P., Stroosnijder, L. and Prins, H.H.T., 1996. Sahelian rangeland development: a catastrophe? J. Range Manage., 49: 512–519. Rogers, G.F., 1985. Mortality of burned Cereus giganteus. Ecology, 66: 630–632. Rogers, G.F., 1986. Comparison of fire occurrence in desert and nondesert vegetation in Tonto National Forest, Arizona. Madro˜no, 33: 278–283. Rogers, G.F. and Vint, M.K., 1987. Winter precipitation and fire in the Sonoran Desert. J. Arid Environ., 13: 47–52. Rundel, P.W. and Gibson, A.C., 1996. Ecological Communities and Processes in a Mojave Desert Ecosystem: Rock Valley, Nevada. Cambridge University Press, Cambridge, 369 pp. Schimpf, D.J. and MacMahon, J.A., 1981. Water as a factor in the biology of North American desert plants. In: D.D. Evans and J.L. Thames (Editors), Water in Desert Ecosystems, 12. Dowden, Hutchinson and Ross, Stroudsburg, Pennsylvania, pp. 114–171. Schlesinger, W.H. and Gramenopoulos, N., 1996. Archival photographs show no climate-induced changes in woody vegetation in the Sudan, 1943–1994. Global Change Biol., 2: 137–141. Schlesinger, W.H. and Jones, C.S., 1984. The comparative importance of overland runoff and mean annual rainfall to shrub communities of the Mojave Desert. Bot. Gaz., 145: 116–124. Schlesinger, W.H., Fonteyn, P.J. and Reiners, W.A., 1989. Effects of overland flow on plant water relations, erosion, and soil water percolation on a Mojave Desert landscape. Soil Sci. Am. J., 53: 1567–1571. Schlesinger, W.H., Reynolds, J.F., Cunningham, G.L., Huenneke, L.F., Jarrell, W.M., Virginia, R.A. and Whitford, W.G., 1990. Biological feedbacks in global desertification. Science, 247: 1043–1048. Schlesinger, W.H., Raikes, J.A., Hartley, A.E. and Cross, A.F., 1996. On the spatial pattern of soil nutrients in desert ecosystems. Ecology, 77: 364–374. Schlesinger, W.H., Abrahams, A.D., Parsons, A.J. and Wainwright, J., 1999. Nutrient losses in runoff from grassland and shrubland habitats in Southern New Mexico: I. rainfall simulation experiments. Biochemistry, 45: 21–34. Schmid, M.K. and Rogers, G.F., 1988. Trends in fire occurrence in the Arizona Upland subdivision of the Sonoran Desert, 1955 to 1983. Southwest. Nat., 33: 437–444. Schmidt, J.O. and Buchmann, S.L., 1986. Floral biology of the

DISTURBANCE IN DESERTS saguaro Cereus giganteus, I. Pollen harvest by Apis mellifera. Oecologia, 69: 491–498. Schneider, S.H. and Root, T.L., 1996. Ecological implications of climate change will include surprises. Biodiversity and Conservation, 5: 1109–1119. Schowalter, T.D., 1996. Arthropod associates and herbivory on tarbush in southern New Mexico. Southwest. Nat., 41: 140–144. Seely, M.K. and Hamilton III, W.J., 1978. Durability of vehicle tracks on three Namib Desert substrates. S. Afr. J. Wildl. Res., 8: 107–111. Sharifi, M.R., Gibson, A.C. and Rundel, P.W., 1997. Surface dust impacts on gas exchange in Mojave Desert shrubs. J. Appl. Ecol., 34: 837–846. Smith, G., 1984. Climate. In: J.L. Cloudsley-Thompson (Editor), Sahara Desert. Pergamon Press, Oxford, pp. 17–39. Smith, J.M.B., 1982. A History of Australasian Vegetation. McGrawHill Book Company, Sydney, 168 pp. Smith, S.D. and Nowak, R.S., 1990. Ecophysiology of plants in the intermountain lowlands. In: C.B. Osmond, L.F. Pitelka and G.M. Hidy (Editors), Plant Biology of the Basin and Range. SpringerVerlag, Berlin, pp. 179–241. Smith, S.D., Monson, R.K. and Anderson, J.E., 1997. Physiological Ecology of North American Desert Plants. Springer-Verlag, Berlin. 286 pp. Solomon, A.M. and Shugart, H.H., 1993. Vegetation Dynamics and Global Change. Chapman and Hall, New York, 338 pp. Soriano, A., 1983. Deserts and semi-deserts of Patagonia. In: N.E. West (Editor), Temperate Deserts and Semi-deserts. Ecosystems of the World 5. Elsevier, Amsterdam, pp. 423–460. Steenbergh, W.F. and Lowe, C.H., 1977. Ecology of the Saguaro. II. Reproduction, Germination, Establishment, Growth and Survival of the Young Plant. National Park Service, Scientific Monograph Series, No. 8, U.S. Government Printing Office, Washington, D.C., 242 pp. Sullivan, S., 1996. Towards a non-equilibrium ecology: perspectives from an arid land. J. Biogeogr., 23: 1–5. Tucker, C.J., Dregne, H.E. and Newcomb, W.W., 1991. Expansion and contraction of the Sahara Desert from 1980 to 1990. Science, 253: 299–301. Turner, R.M., 1990. Long-term vegetation change at a fully protected Sonoran Desert site. Ecology, 71: 464–477. Vasek, F.C., 1980. Creosote bush: Long-lived clones in the Mojave Desert. Am. J. Bot., 67: 246–255. Vasek, F.C. and Lund, L.J., 1980. Soil characteristics associated with a primary plant succession on a Mojave Desert dry lake. Ecology, 61: 1013–1018. Vasek, F.C., Johnson, H.B. and Brum, G.D., 1975a. Effects of power transmission lines on vegetation of the Mojave Desert. Madro˜no, 23: 114–130. Vasek, F.C., Johnson, H.B. and Eslinger, D.H., 1975b. Effects of pipeline construction on creosote bush scrub vegetation of the Mojave Desert. Madro˜no, 23: 1–13. Vogl, R.J., 1980. The ecological factors that produce perturbationdependent ecosystems. In: J. Cairns Jr. (Editor), The Recovery Process in Damaged Ecosystems. Ann Arbor Science Publishers, Ann Arbor, Michigan, pp. 63–94. von Willert, D.J., Eller, B.M., Werger, M.J.A., Brinckmann, E. and Ihlenfeldt, H.-D., 1992. Life Strategies of Succulents in Deserts:

329 With Special Reference to the Namib Desert. Cambridge University Press, Cambridge, 340 pp. Walter, H. and Breckle, S.W., 1986. Tropical and Subtropical Zonobiomes. Springer-Verlag, Berlin, 465 pp. Webb, R.H. and Wilshire, H.G. (Editors), 1983. Environmental Effects of Off-Road Vehicles. Springer-Verlag, New York, 534 pp. Webb, R.H., Steiger, J.W. and Wilshire, H.G., 1986. Recovery of compacted soils in Mojave Desert ghost towns. Soil Sci. Soc. Am. J., 50: 1341–1344. Webb, R.H., Steiger, J.W. and Turner, R.M., 1987. Dynamics of Mojave Desert shrub assemblages in the Panamint Mountains, California. Ecology, 68: 478–490. Weis, I.M. and Schwartz, S.S., 1988. The calculation and interpretation of climatic predictabilities. J. Biogeogr., 15: 419–429. Werger, M.J.A., 1983. Vegetation geographical patterns as a key to the past, with emphasis on the dry vegetation types of South Africa. Bothalia, 14: 405–410. Werger, M.J.A., 1986. The Karoo and southern Kalahari. In: M. Evenari, I. Noy-Meir and D.W. Goodall (Editors), Hot Deserts and Arid Shrublands. Ecosystems of the World 12B. Elsevier, Amsterdam, pp. 283–359. West, N.E. (Editor), 1983. Temperate Deserts and Semi-Deserts. Ecosystems of the World 5. Elsevier, Amsterdam, 522 pp. West, N.E., 1988. Spatial pattern–functional interactions in shrubdominated plant communities. In: C.M. McKell (Editor), The Biology and Utilization of Shrubs, 5. Academic Press, Inc., New York, pp. 283–305. Westoby, M., Walker, B. and Noy-Meir, I., 1989. Opportunistic management for rangelands not at equilibrium. J. Range Manage., 42: 266–274. Whisenant, S.G. and Wagstaff, F.J., 1991. Successional trajectories of a grazed salt desert shrubland. Vegetatio, 94: 133–140. White, P.S. and Pickett, S.T.A., 1985. Natural disturbance and patch dynamics: An introduction. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, Florida, pp. 3–13. Whitford, W.G. and Kay, F.R., 1999. Bioperturbation by mammals in deserts: a review. J. Arid Environ., 41: 203–230. Williams, O.B. and Calaby, J.H., 1985. The hot deserts of Australia. In: M. Evenari, I. Noy-Meir and D.W. Goodall (Editors), Hot Deserts and Arid Shrublands. Ecosystems of the World 12A. Elsevier, Amsterdam, pp. 269–312. Williamson, M., 1996. Biological Invasions. Chapman and Hall, London, 244 pp. Woodward, F.I., 1987. Climate and Plant Distribution. Cambridge University Press, Cambridge, 174 pp. World Meteorological Organization, 1983. Meteorological Aspects of Certain Processes Affecting Soil Degradation – Especially Erosion. Tech Note Number 178, World Meteorological Organization Number 591:149, Geneva, Switzerland. Yair, A., 1995. Short and long term effects of bioturbation on soil erosion, water resources and soil development in an arid environment. Geomorphology, 13: 87–99. Yair, A. and Shachak, M., 1982. A case-study of energy, water and soil flow-chains in an arid ecosystem. Oecologia, 54: 389–397. Yeaton, R.I. and Esler, K.J., 1990. The dynamics of a succulent karoo vegetation. Vegetatio, 88: 103–113. Yeaton, R.I. and Manzanares, A.R., 1986. Organization of

330 vegetation mosiacs in the Acacia schaffneri–Opuntia streptacantha association, southern Chihuahuan Desert, Mexico. J. Ecol., 74: 211–217. Yool, S.R., 1998. Multi-scale analysis of disturbance regimes in the northern Chihuahuan Desert. J. Arid Environ., 40: 467–483. Zahran, M.A. and Willis, A.J., 1992. The Vegetation of Egypt. Chapman and Hall, London, 424 pp.

James A. MacMAHON Zedler, P.H., 1981. Vegetation change in chaparral and desert communities in San Diego County, California. In: D.C. West, H.H. Shugart and D.B. Botkin (Editors), Forest Succession. Springer-Verlag, New York, pp. 406–430.

Chapter 13

DISTURBANCE REGIMES IN NORTH AMERICAN WETLANDS Karen L. McKEE and Andrew H. BALDWIN

INTRODUCTION

The term “wetland” refers to land that is either periodically saturated with or permanently covered by water, exhibits unique soil conditions, and contains hydrophytes (plants adapted to periodic flooding). Ecosystems or communities that can be considered to be associated with wetlands occur in all climatic regions, can exist under widely variable hydrogeologic conditions, and contain very different species assemblages (Whigham et al., 1993). Their diversity and transitional position in the landscape have led to difficulties in classification of wetland types, confusion regarding which areas are in fact wetlands, and problems in determining the areal extent of wetlands. Wetland classification systems are usually based on some combination of vegetation, hydrology, and waterquality features [see, for instance, Cowardin et al. (1979) for wetlands in the United States], but no one system adequately classifies all wetland types worldwide. Wetlands may be tidal or non-tidal, coastal or inland, freshwater or saline. They may be dominated by trees, shrubs, herbaceous macrophytes, or mosses. Wetlands may be further categorized according to soil type (e.g., mineral or peat) or water quality (e.g., acidic or basic, nutrient-rich or nutrient-deficient). The variability among wetlands of different regions and their importance to humans throughout history have resulted in many different terms describing wetlands, including the familiar marshes, swamps, mires, fens, and bogs. Other terms (e.g., mangroves, pocosins, salinas, savannas, and playas) may be applied to specific wetland types in certain regions (Chapman, 1977; Larsen, 1982; Gore, 1983; Whigham et al., 1993). For the purposes of this chapter, we have provided a simplified classification scheme that will define some common types for the reader unfamiliar

with wetlands and to guide the more knowledgeable reader (Table 13.1). Because wetlands develop in the zone between dry land and open water, they possess qualities transitional between terrestrial and aquatic systems. Wetlands may play a biogeochemical role as source, sink, or transformer of chemicals due to their intermediate position between terrestrial (source) and aquatic (sink) systems. Wetlands also generally exhibit a high primary production compared to upland or aquatic systems. The hydrology of wetlands, which varies from intermittently to permanently flooded, is a major determinant of ecosystem structure and function (Mitsch and Gosselink, 1993). Hydrology determines their unique physico-chemical attributes; controls the movement of water, sediment, nutrients, and toxins through them, and influences the occurrence and distribution of plants and animals within them. Hydrology controls abiotic conditions such as soil moisture, oxygen, and nutrient availability, and influences biotic processes such as dispersal of diaspores. Coastal wetlands are additionally affected by ocean tides, which generate fluctuations in water level and/or salinity. These factors in turn affect the distribution and relative abundance of plant and animal species, and ecosystem functions such as productivity, energy flow, and nutrient cycling. Wetland organisms are varied in terms of size, form, complexity, and mobility. They may be planktonic (floating), natant (swimming), benthic (living in or on the soil substrate), rooted, or epiphytic (attached to macrophytes). Plants and animals range in size from microscopic to massive. The microbiotic community includes bacteria, protozoans, algae, and fungi. Creeks and channels and the water column overlying the wetland surface contain phytoplankton (dinoflagellates, diatoms, and blue-green algae), which are important primary producers, and zooplankton, which include

331

332

Karen L. McKEE and Andrew H. BALDWIN

Table 13.1 Common wetland types covered in this chapter and some of their distinguishing characteristics Wetland type

Synonymous terms

Marsh

Dominant vegetation

Hydrology

herbaceous (grasses, rushes, sedges, forbs)

Water chemistry 1 /soil type

generally non-acidic, mineral or organic soil tidal or non-tidal

>18‰ salinity 1

Brackish marsh

same

0.5–18‰ salinity

Freshwater marsh

same

50 yr). Similarly, the scale of open-pit workings has reached massive proportions, as demonstrated by the Kennecott Copper Mine in Bingham (Utah, U.S.A) with an open pit 3 km wide and 800 m deep (Williamson et al., 1982). Mining wastes: direct and indirect disturbances In many of the forms of surface and deep mining (Table 14.3) it is the production and disposal of waste which can cause the most extensive and long-lasting disturbance to land. For example, the legacy from a declining industry using deep mining for coal in Britain was that, in 1972, the National Coal Board owned 2000 spoil heaps containing 2 109 t of mining waste

on an area of 15 000 ha, with about an equal number of spoil heaps being associated with other coal-mining companies and operations (Thomson and Rodin, 1972; Glover, 1978). Much of this colliery spoil had become extremely acidic over the years, with the oxidation of pyrite releasing sulphuric acid into the surface layers (Palmer, 1978); thus, even though many of the heaps had been abandoned for many years, there had been little natural colonization by vegetation. The relationships between the primary and secondary phases of mining and waste production are shown for non-ferrous metal production in Fig. 14.1. The disposal of rock and overburden, the construction of impoundments (dams) for the fine tailings (1000 km2 (Hutchinson and Whitby, 1974). A more recent air-pollution concern may have considerable global impact in the future. It is the emission of methane, an important greenhouse gas, from the activities of underground and surface mining themselves. These emissions are geographically spread around the world, and have been estimated at 45 1012 g yr−1 (Williams and Mitchell, 1994).

370

John A. COOKE

Table 14.4 Severity of the main factors which influence plant establishment on some wastes and substrates of mining origin 1 Waste type

Adverse texture/stucture 2

Low nutrients

Extreme pH

Toxicity

High salinity

Coal spoil

2

2

0–2

0–1

0–1

Oil-shale

1–2

2

1–2

0

0–1

Iron-ore mining

1

1

0

0

0–1

Heavy metal wastes

2

2

2

1–2

0–1

Gold wastes

2

2

2

0–2

0–2

Bauxite (red mud)

2

2

2

2

2

Acid rocks

2

1

1

0

0

Calcareous rocks

2

2

1

0

0

Sand and gravel

1

1

0

0

0

China clay wastes

2

2

1

0

0

Strip-mining (coal)

1–2

1–2

0–2

0

0–1

1

1–2

0

0

0–1

Coastal sands 1

Adapted from Bradshaw and Chadwick (1980). Key to importance of limiting factors (difference in severity due to variation in materials and situations): 0, negligible; 1, moderate; 2, severe. 2

ECOSYSTEMS OF MINED LAND

Succession and reclamation Mining is a temporary land-use. It is not sustainable at any one place because the mineral deposit is finite and eventually exhausted. The land surface is changed by mining activity, and the disturbance may persist for a long time. Reclamation is the process whereby the land surface is returned to some form of beneficial use. The terms restoration, rehabilitation, and replacement represent the goals of the reclamation process to achieve the pre-mined state or some other new land use (Bradshaw, 1990b). Restoration, although used more generally in the earlier literature (e.g., Johnson and Bradshaw, 1979), now is sometimes used exclusively to refer to restoration of the original (pre-mining) ecosystem with all its structural and functional aspects. Rehabilitation is the term used for the progression towards the reinstatement of the original ecosystem, and replacement is the creation of an alternative ecosystem (Bradshaw, 1990b). Recovery of mined land occurs when the land is largely left to natural processes after disturbance. In practice, the terms recovery, restoration, rehabilitation, and replacement may all be described as resetting the ecological clock (Cairns, 1991). Despite the lack of unifying theories of succession

(Miles, 1987), it is important to understand the basic processes concerning ecosystem development in the contexts both of abandoned derelict mine sites and of the practice of ecological restoration of mined land. These processes include the mechanisms of plant and animal colonization, the establishment of species populations, the development of ecosystem structure and function, and the limits and time-scales involved. Ecosystems of abandoned mined land In natural succession where there is usually no soil, resource availability and resource demand during colonization and early succession are particularly important (Vitousek and Walker, 1987). Compared to normal soils, mining substrates, which may have been derived from deep in the earth, or are wastes produced from the processing of the minerals or rock surfaces left after extraction, may present extreme challenges to colonization by plants and the formation of any kind of self-sustaining ecosystem (Table 14.4). Mining substrates show considerable variation in their physical and chemical nature, as indicated in Table 14.4. Severe values (Category 2) of the factors mentioned are likely to inhibit natural colonization by most plant species for many years although often a few species (which may be particularly tolerant or have tolerant ecotypes or populations) may form a sparse vegetation cover.

MINING

Physical texture may be very coarse, as in some quarry wastes and coal spoils, or very fine, as in milled tailings. Fine texture without organic matter leads to high bulk densities, extreme compaction, low water infiltration rates, and surface waterlogging. Nearly all mine substrates have very low levels of macronutrients (especially nitrogen, phosphorus, and potassium). Low pH is a problem in wastes containing iron pyrite which on weathering will generate sulphuric acid and (if there is no capacity in the waste to neutralize acids) causes pH values of 50 yr). Also, the ecosystems which do develop are often different from the typical climax vegetation of the region because of differences in the soils of mined land. Unless specific restoration goals are integrated into the mine planning, it is unlikely that the pre-mined ecosystem will develop naturally. Where local topsoil can be replaced quickly, as in the examples of progressive reclamation during strip or dredge mining, and where ecological principles, developed through research, guide the restoration planning, then success may be achieved. In many mining situations, more modest goals may have to be accepted, simple ecosystems being established and maintained for aesthetic and safety reasons. However, it is important to try to maintain future options for longer-term ecological sustainability.

ACKNOWLEDGEMENTS

The author would like to thank: Dehn von Ahlefeldt (University of Natal), Alan Baker (University of Sheffield) and Ronwyn Stander (Richards Bay Minerals) for help with the photographs; and the South African Foundation for Research Development, and the University of Natal Research Fund for financial support.

REFERENCES Alvarez, H., Ludwig, J.A. and Harper, K.T., 1974. Factors influencing plant colonization of mine dumps at Park City, Utah. Am. Midl. Nat., 92: 1–11. Antonovics, J., Bradshaw, A.D. and Turner, R.G., 1971. Heavy metal tolerance in plants. Adv. Ecol. Res., 7: 1–85. Australian Environment Protection Agency, 1995. Rehabilitation and Revegetation. Best Practice Environmental Management in Mining. Commonwealth of Australia, Barton, 36 pp. Baker, A.J.M. and Proctor, J., 1990. The influence of cadmium, copper, lead, and zinc on the distribution and evolution of metallophytes in the British Isles. Plant Syst. Evol., 173: 91–108. Baker, S.R., Gardener, J.H. and Ward, S.C., 1995. Bauxite mining environmental management and rehabilitation practices in Western Australia. World’s Best Practice in Mining and

John A. COOKE Processing, Conference, Sydney. Australasian Institute of Mining and Metallurgy, pp. 43–53. Barbour, A.K., 1994. Mining non-ferrous metals. In: R.E. Hester and R.M. Harrison (Editors), Mining and its Environmental Impact. Issues in Environmental Science and Technology. Royal Society of Chemistry, Letchworth, England, pp. 1–15. Bradshaw, A.D., 1983. The Reconstruction of Ecosystems: Presidential address to the British Ecological Society, December 1982. J. Appl. Ecol., 20: 1–17. Bradshaw, A.D., 1990a. Restoration: an acid test for ecology. In: W.R. Jordan, M.E. Gilpin and J.D. Aber (Editors), Restoration Ecology: A Synthetic Approach to Ecological Research. Cambridge University Press, Cambridge, pp. 23–29. Bradshaw, A.D., 1990b. The reclamation of derelict land and the ecology of ecosystems. In: W.R. Jordan, M.E. Gilpin and J.D. Aber (Editors), Restoration Ecology: A Synthetic Approach to Ecological Research. Cambridge University Press, Cambridge, pp. 53–74. Bradshaw, A.D. and Chadwick, M.J., 1980. The Restoration of Land: The Ecology and Reclamation of Derelict and Degraded Land. Blackwell, Oxford, 317 pp. Bradshaw, A.D., Marrs, R.H. and Roberts, R.D., 1982. Succession. In: B.N.K. Davis (Editor), Ecology of Quarries: the Importance of Natural Vegetation. Institute of Terrestrial Ecology, Abbots Ripton, pp. 47–52. Brenner, F.J., Werner, M. and Pike, J., 1984. Ecosystem development and natural succession in surface coal mine reclamation. Miner. Environ., 6: 10–22. Brooks, R.R. and Malaisse, F., 1985. The Heavy Metal-Tolerant Flora of Southcentral Africa. A.A. Balkema, Rotterdam, 199 pp. Buchanan, D.J. and Brenkley, D., 1994. Green Coal Mining. In: R.E. Hester and R.M. Harrison (Editors), Mining and its Environmental Impact. Issues in Environmental Science and Technology. Royal Society of Chemistry, Letchworth, England, pp. 71–95. Buckley, G.P., 1989. Biological Habitat Construction. Belhaven Press, London, pp. 363. Cairns, J.J., 1991. The status of the theoretical and applied science of restoration ecology. Environ. Prof., 13: 186–194. Camp, P.D., 1990. Rehabilitation after dune mining at Richards Bay Minerals. S. Afr. Min. World, October: 34–37. Camp, P.D. and Weisser, P.J., 1991. Dune rehabilitation, flora and plant succession after mining at Richards Bay, South Africa. In: D.A. Everard and G.P. Von Maltitz (Editors), Dune Forest Dynamics in Relation to Land-Use Practices. Foundation for Research Development, Pretoria, pp. 106–123. Chaney, W.R., Pope, P.E. and Byres, W.R., 1995. Tree survival and growth on land reclaimed in accord with public law 95–87. J. Environ. Qual., 24: 630–634. Cooke, J.A. and Morrey, D.R., 1981. Heavy metals and fluoride in soils and plants associated with metalliferous mine wastes in the Northern Pennines. In: P.J. Say and B.A. Witton (Editors), Heavy Metals in Northern England: Environmental and Biological Aspects. University of Durham, Durham, pp. 153–164. Daily, G.C., 1995. Restoring value to the world’s degraded lands. Science, 269: 350–354. Daniels, W.L. and Zipper, C.E., 1988. Improving coal surface mine reclamation in the Central Appalachian region. In: J. Cairns Jr (Editor), Rehabilitating Damaged Ecosystems. CRC Press, Boca Raton, pp. 139–162.

MINING Davis, B.N.K., 1979. Chalk and limestone quarries as wildlife habitats. Miner. Environ., 1: 48–56. Davis, B.N.K., 1982. Ecology of Quarries: the Importance of Natural Vegetation. Institute of Terrestrial Ecology, Abbots Ripton, 77 pp. Davis, B.N.K., Lakhani, K.H., Brown, M.C. and Park, D.G., 1985. Early seral communities in a limestone quarry: an experimental study of treatment effects on cover and richness of vegetation. J. Appl. Ecol., 22: 473–490. Duckham, F., 1969. Introduction. In: R.L. Galloway (Editor) A History of Coal Mining in Great Britain. David and Charles, Newton Abbot, pp. 273, reprint of 1882 edition. Ellis, D.V., 1988. Case histories of coastal and marine mines. In: W. Salomons and U. Forstner (Editors), Chemistry and Biology of Solid Waste: Dredged Material and Mine Tailings. SpringerVerlag, Berlin, pp. 73–100. Ernst, W.H.O., 1988. Response of plants to mine tailings and dredged materials. In: W. Salomons and U. Forstner (Editors), Chemistry and Biology of Solid Waste: Dredged Material and Mine Tailings. Springer-Verlag, Berlin, pp. 54–69. Ferguson, K.D. and Erickson, P.M., 1988. Pre-mine prediction of acid-mine drainage. In: W. Salomons and U. Forstner (Editors), Environmental Management of Solid Waste: Dredged Material and Mine Tailings. Springer-Verlag, Berlin, pp. 24–43. Foord, S.H., van Aarde, R.J. and Ferreira, S.M., 1994. Seed dispersal by vervet monkeys in rehabilitating coastal dune forests at Richards Bay. S. Afr. J. Wildl. Res., 24: 56–59. Galloway, R.L., 1882. A History of Coal Mining in Great Britain. David and Charles, Newton Abbott, 273 pp., reprinted 1969. Gibson, D.J., Johnson, F.L. and Risser, P.G., 1985. Revegetation of unreclaimed coal strip-mines in Oklahoma. II. Plant Communities. Reclam. Revegetation Res., 4: 31–47. Glover, H.G., 1978. The disposal of coal mine spoil in the United Kingdom. In: G.T. Goodman and M.J. Chadwick (Editors), Environmental Management of Mineral Wastes. Nato Advanced Study Institutes Series. Series E: Applied Science. Sijthoff and Noordhoff, Alphen aan den Rijn, pp. 35–69. Gorsira, B. and Risenhoover, K.L., 1994. An evaluation of woodland restoration on strip-mined lands in east Texas. Environ. Manage., 18: 787–793. Gunn, J., Bailey, D. and Gagen, P., 1992. Landform Replication as a Technique for the Reclamation of Limestone Quarries. Department of the Environment, HMSO, London, 38 pp. Guo, H., Wu, D. and Zhu, H., 1989. Land restoration in China. J. Appl. Ecol., 26: 787–792. Hodgson, J.G., 1989. Selecting and Managing Plant Materials Used in Habitat Reconstruction. In: G.P. Buckley (Editor), Biological Habitat Construction. Belhaven Press, London, pp. 45–67. Hogan, G.D., Courtin, G.M. and Rauser, W.E., 1977a. The effects of soil factors on the distribution of Agrostis gigantea on a mine waste site. Can. J. Bot., 55: 1038–1042. Hogan, G.D., Courtin, G.M. and Rauser, W.E., 1977b. Copper tolerance in clones of Agrostis gigantea from a mine waste site. Can. J. Bot., 55: 1043–1050. Holl, K.D. and Cairns Jr., J., 1994. Vegetational community development on reclaimed coal surface mines in Virginia. Bull. Torrey Bot. Club, 121: 327–337. Humphries, R.N., 1980. The development of wildlife interest in limestone quarries. Reclam. Rev., 3: 197–207. Hutchinson, T.C. and Whitby, L.M., 1974. Heavy metal pollution

383 in the Sudbury mining and smelting region of Canada. 1. Soil and vegetation contamination by nickel, copper, and other metals. Environ. Conservation, 1: 123–132. Jasper, D.A., Robson, A.D. and Abbott, L.K., 1987. The effect of surface mining on the infectivity of vesicular-arbuscular mycorrhizal fungi. Aust. J. Bot., 35: 641–652. Johnson, F.L., Gibson, D.J. and Risser, P.G., 1982. Revegetation of unreclaimed coal strip-mines in Oklahoma. 1. Vegetation structure and soil properties. J. Appl. Ecol., 19: 453–463. Johnson, M.S. and Bradshaw, A.D., 1979. Ecological principles for the restoration of disturbed and degraded land. Adv. Appl. Biol., 4: 141–200. Johnson, M.S., Putwain, P.D. and Holliday, R.J., 1978. Wildlife conservation value of derelict metalliferous mine workings in Wales. Biol. Conservation, 14: 131–148. Johnson, M.S., Cooke, J.A. and Stevenson, J.K., 1994. Revegetation of metalliferous wastes and land after metal mining. In: R.E. Hester and R.M. Harrison (Editors), Mining and its Environmental Impact. Issues in Environmental Science and Technology. Royal Society of Chemistry, Letchworth, England, pp. 31–48. Klemow, K.M. and Raynal, D.J., 1981. Population ecology of Melilotus alba in a limestone quarry. J. Ecol., 69: 33–44. Land Use Consultants, 1992. Amenity Reclamation of Mineral Workings: Main Report. Land Use Consultants, Department of the Environment, HMSO, London, 242 pp. Leopold, D.J. and Wali, M.K., 1992. The rehabilitation of forest ecosystems in the eastern United States and Canada. In: M.K. Wali (Editor), Ecosystem Rehabilitation. Volume 2. SPB Academic Publishing, The Hague, pp. 187–231. Lubke, R.A., Moll, J.B. and Avis, A.M., 1993. Rehabilitation Ecology. In: C.E. Services (Editors), Environmental Impact Assessment, Eastern Shores of Lake St. Lucia (Kingsa/Trojan Lease Area). CSIR, Pretoria, pp. 251–302. Luken, J.O., 1990. Directing Ecological Succession. Chapman and Hall, London, 251 pp. McNeilly, T., 1990. Evolutionary lessons from degraded ecosystems. In: W.R. Jordan, M.E. Gilpin and J.D. Aber (Editors), Restoration Ecology: A Synthetic Approach to Ecological Research. Cambridge University Press, Cambridge, pp. 271–286. Mentis, M.T. and Ellery, W.N., 1994. Post-mining rehabilitation of dunes on the north-east coast of South Africa. S. Afr. J. Sci., 90: 69–74. Miles, J., 1987. Vegetation succession: past and present perceptions. In: A.J. Gray, M.J. Crawley and P.J. Edwards (Editors), Colonisation, Succession and Stability. Blackwell, Oxford, pp. 1– 29. Mining Annual Review, 1985. Mining Annual Review. Mining Journal Ltd, London, 556 pp. Mining Annual Review, 1995. Mining Annual Review. Mining Journal Ltd, London, 248 pp. Morrey, D.R., Baker, A.J.M. and Cooke, J.A., 1988. Floristic variation in plant communities on metalliferous mining residues in the northern and southern Pennines, England. Environ. Geochem. Health, 10: 11–20. Nichols, O., Wykes, B.J. and Majer, J.D., 1989. The return of vertebrate and invertebrate fauna to bauxite mined areas in southwestern Australia. In: J.D. Majer (Editor), Animals in Primary Succession: the Role of Fauna in Reclaimed Lands. Cambridge University Press, Cambridge, pp. 397–422.

384 Palmer, M.E., 1978. Acidity and nutrient availability in colliery spoil. In: G.T. Goodman and M.J. Chadwick (Editors), Environmental Management of Mineral Wastes. Nato Advanced Study Institutes Series. Series E: Applied Science. Sijthoff and Noordhoff, Alphen aan den Rijn, pp. 85–126. Pierzynski, G.M., Schnoor, J.L., Banks, M.K., Tracy, J.C., Licht, L.A. and Erickson, L.E., 1994. Vegetative remediation at Superfund sites. In: R.E. Hester and R.M. Harrison (Editors), Mining and its Environmental Impact. Issues in Environmental Science and Technology. Royal Society of Chemistry, Letchworth, England, pp. 49–69. Ratcliffe, D.A., 1974. Ecological effects of mineral exploitation in the United Kingdom and their significance to nature conservation. Proc. R. Soc. London Ser. A., 339: 355–372. Richardson, J.A., Davis, B.N.K. and Evans, M.E., 1980. Disused quarries. In: T.C. Dunn (Editor), The Magnesian Limestone of Durham County. The Gilpin Press, Houghton le Spring, pp. 61– 68. Riley, C.V., 1957. Reclamation of coal strip-mined lands with reference to wildlife plantings. J. Wildl. Manage., 21: 402–413. Ritcey, G.M., 1989. Tailings Management: Problems and Solutions in the Mining Industry. Vol. 6. Process Metallurgy. Elsevier Amsterdam, 970 pp. Safaya, N.M. and Wali, M.K., 1992. Applicability of U.S. environmental laws in the developing countries: an analysis of ecological and regulatory concepts. In: M.K. Wali (Editor), Ecosystem Rehabilitation, Vol. 1. SPB Academic Publishing, The Hague, pp. 143–155. Schuster, W.S. and Hutnik, R.J., 1987. Community development on 35-year-old planted minespoil banks in Pennsylvania. Reclam. Revegetation Res., 6: 109–120. Sengupta, M., 1993. Environmental Impacts of Mining: Monitoring, Restoration and Control. Lewis, Boca Raton, 494 pp. Simon, E., 1978. Heavy metals in soils, vegetation development and heavy metal tolerance in populations from metalliferous areas. N. Phytol., 81: 175–188. Smith, R.A.H. and Bradshaw, A.D., 1979. The use of metal tolerant plant populations for the reclamation of metalliferous wastes. J. Appl. Ecol., 16: 595–612. South African Mining, 1996. South African Mining; Coal, Gold, and Base Minerals. February. Thomson Publishing, Randburg, p 53. Thatcher, F.M., 1979. A Study of the Vegetation Established on the Slimes Dams of the Witwatersrand. PhD, University of the Witwatersrand, Johannesburg, 518 pp. Thirgood, J.V., 1978. Approaches to land reclamation in Britain and North America. In: G.T. Goodman and M.J. Chadwick (Editors), Environmental Management of Mineral Wastes. Nato Advanced Study Institutes Series. Series E: Applied Science. Sijthoff and Noordhoff, Alphen aan den Rijn, pp. 1–18. Thomson, G.M. and Rodin, S., 1972. Colliery Spoil Tips – after Aberfan. The Institution Civil Engineers, London, 60 pp.

John A. COOKE Tracey, W.H. and Glossop, B.L., 1980. Assessment of topsoil handling techniques for rehabilitation of sites mined for bauxite within the Jarrah forest of Western Australia. J. Appl. Ecol., 17: 195–201. United Kingdom Department of the Environment, 1994. The Reclamation and Management of Metalliferous Mining Sites. Department of Environment, HMSO, London, 168 pp. United States National Research Council, 1981. Surface Mining: Soil, Coal, and Society. National Research Council, National Academy Press, Washington D.C., 233 pp. Usher, M.B., 1979. Natural communities of plants and animals in disused quarries. J. Environ. Manage., 8: 223–236. Usher, M.B. and Jefferson, R.G., 1990. The concepts of colonization and succession: their role in nature reserve management. In: S.H. Hillier, D.W.H. Walton and D.A. Wells (Editors), Calcareous Grasslands – Ecology and Management. Bluntishham Books, Huntingdon, pp. 149–153. van Aarde, R.J., Ferreira, S.M., Kritzinger, J.J., van Dyk, P.J., Vogt, M. and Wassenaar, T.D., 1996. An evaluation of habitat rehabilitation on coastal dune forests in northern KwaZulu–Natal, South Africa. Restoration Ecol., 4: 334–345. Vitousek, P.M. and Walker, L.R., 1987. Colonization, succession and resource availability: ecosystem level interactions. In: A.J. Gray, M.J. Crawley and P.J. Edwards (Editors), Colonization, Succession and Stability. Blackwell, Oxford, pp. 207–224. Wagner, W.L., Martin, W.C. and Aldon, E.F., 1978. Natural succession on strip mined lands in northwestern New Mexico. Reclam. Rev., 1: 67–73. Wells, J.D., van Meurs, L.H. and Rabie, M.A., 1992. Terrestrial minerals. In: R.F. Fruggle and M.A. Rabie (Editors), Environmental Management in South Africa. Juta, Cape Town, pp. 337–379. Wild, H., 1978. The vegetation of heavy metal and other toxic soils. In: M.J.A. Werger (Editor), Biogeography and Ecology of Southern Africa. W. Junk, The Hague, pp. 1303–1332. Willems, J.H., 1990. Calcareous grasslands in continental Europe. In: S.H. Hillier, D.W.H. Walton and D.A. Wells (Editors), Calcareous Grasslands – Ecology and Management. Bluntishham Books, Huntingdon, pp. 3–10. Williams, A. and Mitchell, C., 1994. Methane emissions from coal mining. In: R.E. Hester and R.M. Harrison (Editors), Mining and its Environmental Impact. Issues in Environmental Science and Technology. Royal Society of Chemistry, Letchworth, England, pp. 97–109. Williams, S.T., McNeilly, T. and Wellington, E.M.H., 1977. The decomposition of vegetation growing on metal mine waste. Soil Biol. Biochem., 9: 271–275. Williamson, N.A., Johnson, M.S. and Bradshaw, A.D., 1982. Mine Wastes Reclamation. The Establishment of Vegetation on Metal Mine Wastes. Mining Journal Books, London, 103 pp.

Chapter 15

DISTURBANCE ASSOCIATED WITH MILITARY EXERCISES Stephen DEMARAIS, David J. TAZIK, Patrick J. GUERTIN and Eric E. JORGENSEN

INTRODUCTION

Disturbances to ecosystems caused by military activities must be distinguished according to whether they occur during times of war or peace. In war, environmental planning is not a predominant concern. During peacetime, disturbances may accompany military training activities (Lanier-Graham, 1993), often with an opportunity to document, minimize, and mitigate their negative impacts on the environment. Wartime disturbances often are catastrophic and at large spatial and temporal scales. Disturbance impacts may cross regional and national boundaries, particularly for such conflicts as World Wars I and II. During wartime battles, with victory the primary concern, there is little concern for adverse effects on the environment; environmental planning and management play little part. Impacts are not studied in advance, and documentation may not be undertaken at all until political arenas are stabilized. Defeat of a nation or group of peoples can be facilitated by eliminating their environmental resources. General Sherman of the Union Army practiced the “scorched earth” approach as he made his way to capture Atlanta, Georgia, during the Civil War in the United States during the 1860s, burning all resources of potential use by his enemy. During World War II, large areas of reclaimed lands in the Benelux countries were flooded with sea water to impede the German invasion (Barrow, 1991). During the Vietnam War, up to 40% of the land area of Vietnam, and 44% of all forests, were sprayed with defoliants (Gradwohl and Greenberg, 1988). This defoliation, combined with effects from bombs and heavy equipment, destroyed forest vegetation over an estimated 22×106 ha in South and North Vietnam (Barrow, 1991). Recovery of these

defoliated forests has been slow and soil erosion prevalent on steep slopes (Freedman, 1989). Direct impacts of military munitions can be significant. In Vietnam, an estimated 25×106 bomb and shell craters displaced about 3×109 m3 of soil (Gradwohl and Greenberg, 1988). A 227 kg high explosive bomb can form a crater 14 m in diameter and 9 m deep, and the crater may still be plainly visible 25 years later. Roughly 3.5×106 such bombs were dropped in Vietnam during 1968 and 1969 (Stanford Biology Study Group, 1971). Turning to military activities in peacetime, the management of military lands involves facilitating use of those lands for training. Inevitably, use leads to disturbance. The focus of this chapter is on disturbance caused by military training activities, and management scenarios that reduce or mitigate such disturbance. Military training, while not as dramatic a disturbance as the direct effects of war, can have a widespread influence on the land. Most of our information comes from lands administered by the United States Department of Defense (DoD). We will discuss the impacts of army training in four countries (Australia, Canada, Germany, and the United States). It is the policy of the United States Army to maintain training lands in a condition which closely mimics the natural conditions under which actual warfare would be conducted, as well as for wildlife habitat and other natural values (Hinchman et al., 1990; Goodman, 1996). In particular, it is the goal of the Department of Defense to reduce or avoid long-term impacts on natural resources caused by military training (Prose, 1985). Military land-use is frequently intensive, especially where maneuvers are conducted with tracked vehicles (Shaw and Diersing, 1990). For instance, at Pinon Canyon Maneuver Site, Fort Carson, Colorado, after

385

386

Stephen DEMARAIS, David J. TAZIK, Patrick J. GUERTIN and Eric E. JORGENSEN

two years (six training rotations), a line-transect study in three out of five training areas, with observations at one-meter intervals, showed that 40% of the surface area had been impacted by tracked vehicles (Shaw and Diersing, 1990). Under such training regimes, military land-managers are responsible for maintaining and rehabilitating training lands (Diersing and Severinghaus, 1984). Importantly, many habitats and species are protected from private development by the presence of military installations (Owens, 1990; Creswell, 1994; Goodman, 1996). Also, base longevity and maintenance of realistic habitat conditions for training require proactive resource management (Hinchman et al., 1990; O’Neil et al., 1990; Pearson et al., 1990). Therefore, it is good policy for the military to manage its lands in ways consistent with sound stewardship. Disturbances resulting from military activities are multi-faceted, affecting ecosystems at several points; soil structure is damaged through compaction causing erosion, and vegetation is damaged or destroyed causing modification and loss of wildlife habitat. Such damage to the soil and the vegetation compromise the realism of the training exercise, creating an incentive to train in alternative undisturbed locations. This expansion of the training grounds creates further disturbance (Hinchman et al., 1990; Trumbull et al., 1994). The management approach on lands of the United States Department of Defense has increasingly incorporated the concept of ecosystem management. This concept allows the Department to consider the full array of natural resources on lands under their stewardship (Goodman, 1996). For some time, it was a priority to identify “indicator species” which could be used to indicate early stages of habitat disturbance (e.g., Diersing and Severinghaus, 1985). More recently, documentation and monitoring of long-term change has received emphasis. The approach has thus focused on development of repeatable methodologies, appropriate in many ecosystems (Tazik et al., 1992b). Description of military lands Management of military lands is a unique challenge: testing of and training in the use of advanced weapon systems that are of longer range and more devastating requires a more extensive land area. Not surprisingly, the Department of Defense is the third largest steward of land in the government of the United States, with management authority over 10×106 ha

(Goodman, 1996). Nearly half of this land (4.8×106 ha) is controlled by the United States Army (Shaw and Diersing, 1990); consequently, the Army has the largest impact and the most control over the disturbance of military lands. Worldwide, armed forces control between 7.5×105 and 1.5×106 ha of the earth’s land area (Thomas, 1995). Military lands exist in almost every ecosystem covered in this volume. Diversity is the predominant factor in their choice. Thus, many of the disturbance effects and disturbed ecosystems discussed in this volume find expression on military lands. For example, lands for training in ground combat within the United States managed by the Army, Marine, and Army National Guard occur in lands dominated by boreal forest (13.2%), chaparral–oak woodlands (1.3%), eastern deciduous forest (10.4%), grasslands (5.0%), mesquite grasslands (0.3%), montane woodland brush (10.7%), northern desert (12.4%), northern hardwood–conifer forest (4.5%), oak savanna (1.3%), Pacific rainforest (0.9%), pinyon–juniper–oak woodland (2.5%), southeast evergreen forest (11.8%), southern desert scrub (25.5%), and tropical vegetation (0.2%) (Smith, 1986; Evinger, 1995). This chapter reviews the mechanisms and the impacts associated with military activities. Emphasis is placed on military training activities because these are the impacts most frequently studied and documented. To understand the impacts of military disturbance better, we describe the principal mechanism of disturbance, the training exercise. Military impacts on soil, hydrology, plants, and animals and the larger concepts of community and ecosystem are reviewed. Management efforts to minimize and mitigate environmental impacts are described and examples are presented.

TRAINING DISTURBANCE REGIMES

The importance of both spatial and temporal factors and their interaction in characterizing land-disturbance regimes has long been recognized (Moloney and Levin, 1996). The temporal components include both frequency and return interval. Spatial components include distribution, and area or size. Additionally, factors of magnitude, such as intensity and severity, along with the occurrence of other disturbances, are important to consider in the relationships between the temporal and spatial factors (Pickett and White, 1985). Disturbance regimes derived from military activities are likely

DISTURBANCE ASSOCIATED WITH MILITARY EXERCISES

to include several agents of disturbance including intense vehicle traffic, explosive munitions, and spills of petroleum, oil, and lubricants (Conrad et al., 1994). These disturbances are rarely independent of each other and often occur as part of a single military event (Department of the Army, 1988b, 1991). Consequently, defining military disturbance would include several regimes. Although not standard military terminology, five regimes that can be recognized are mechanizedmaneuver training, infantry-maneuver training, livefire training, command and support, and combat engineering. The mechanized maneuver regime consists of unit exercises designed to simulate actual mechanized combat situations (Department of the Army, 1984). These exercises usually occur in open terrain with slopes less than 20% (Krzysik, 1994). Maneuvering with tracked vehicles is the major agent of impact on training lands (Conrad et al., 1994). The primary environmental disturbances caused during mechanized maneuvers are disturbance to the soil and vegetation from vehicle tracks. Depending on environmental conditions, tracked-vehicle traffic can result in soil compaction, comminution of surface particles, and upheaval, crushing and/or uprooting of vegetation (Krzysik, 1994; Thurow et al., 1995; Wilson, 1988). Constructed defenses often include excavations deep enough to accommodate an armored vehicle. Anti-tank ditches are wide, deep trenches designed to prohibit vehicle crossing. Excavations are usually filled in at the conclusion of an exercise (Department of the Army, 1988c,d). Infantry-maneuver training. Infantry units are trained to fight in dispersed formations (Department of the Army, 1986) where a platoon of roughly 30 to 40 soldiers would have a frontage of 100 to 150 m in offensive actions and 200 m in defensive actions. Common to these training exercises are cutting vegetation for camouflage and digging fox-holes, which usually must be refilled. Live-fire training. Live-fire weapon systems include rifles, tank guns, anti-tank missile systems, and selfpropelled and towed howitzers, multiple-launch rocket systems, and mortars. Munitions vary with the system, but include high explosives, chemical obscurants, ball ammunition, illumination, and kinetic penetrating antiarmor projectiles (Department of the Army, 1993). The large quantities of these munitions expended over an extended period of time can lead to potential environmental contamination (Getz et al., 1996). Impact of

387

munitions can also cause soil displacement, damage to vegetation, fires, incidental killing of wildlife, and changes in wildlife behavior (Severinghaus et al., 1980; Gese et al., 1989; Dinkins et al., 1992; Tazik et al., 1992a). The result of repetitive use of a limited number of firing points can best be described as similar to the damage to soils and vegetation occurring on a construction site. The command and support regime is a conglomeration of units which include headquarters, supply, maintenance, and other non-combat units in static or semi-static bivouac positions. Environmental damage from bivouac positions can vary with the type and size of unit, but usually consists of compacted soils, loss of lower vegetation layers, damage to trees (Trumbull et al., 1994), and excavation for fortification of the positions. The engineering regime represents the activities of combat engineering units, both in conjunction with combat units and separately (Department of the Army, 1988a). Major activities include obstacle development and destruction, gap-crossing, and construction of emplacements which involve earth moving. Engineers often use explosives for demolitions and related work (Department of the Army, 1989, 1993). Temporal and spatial considerations The spatial characteristics of training activities vary among the disturbance regimes. For example, a heavily mechanized infantry battalion including 100 tracked vehicles requires up to 24 800 ha to conduct a “move to contact” exercise and 13 800 ha for a “defensive operation”. A light infantry company made up of 108– 120 soldiers requires 7000 ha for a “move to contact” exercise and 1600 ha for a “defensive operation”. Additionally, topography, vegetation, and geographic shape influence the types of training activities. Wooded and hilly areas are used for bivouac and other static activities. Flat or rolling grasslands and similar open terrain is often used for mechanized-maneuver activities. Given these considerations, the location of activities within an area is often predetermined by topography and vegetation conditions. Temporal features of military training are controlled by a complex mixture of variables, including available land, topography, and vegetation. Impacts from training in any particular area depend greatly on the types of units stationed there. Most United States training areas large enough to support mechanized-maneuver

388

Stephen DEMARAIS, David J. TAZIK, Patrick J. GUERTIN and Eric E. JORGENSEN

training are located west of the Mississippi River, and have the potential of being impacted by the most environmentally damaging of military training activities (Conrad et al., 1994). The land base available for training activities has a strong influence on intensity and frequency of use; installations with a high concentration of military units per unit area of suitable land, as in some European countries, often experience high intensity of use.

EFFECTS OF TRAINING DISTURBANCE REGIMES

Immediate effects versus long-term cumulative effects Disturbance effects from military activities may occur in the short term, but may often persist for decades (Wilshire and Nakata, 1976; Webb and Wilshire, 1980; Lathrop, 1983; Prose, 1985) especially in ecosystems with low productivity. Cumulative effects are common in military training because of the insular nature of training facilities (surrounded by land not subject to military impact). Maneuvers (e.g., vehicle and troop movement, camping) occur repeatedly on the same sites (Trumbull et al., 1994). For instance, relatively minor short-term trampling events can be repeated over many years, causing a cumulative detrimental impact to soils and plants (Trumbull et al., 1994). Such long-term use of sites without rest may limit the potential for recovery and its rate, and ultimately impact small mammal and avian populations dependent upon specific vegetation and soil conditions. For instance, Severinghaus et al. (1980) determined that burrowing mammals were very sensitive to soil changes induced by maneuvers. Disturbance effects are so variable in their dependence on season, ecosystem, and substrate that generalizations concerning the short-term or cumulative nature of impacts are difficult. For instance, in Canadian prairie the season of impact was found to be a more important determinant of disturbance than the number of vehicle passages (Wilson, 1988). Certainly, some types of disturbance will impact plants, animals, and their habitat so as to change community composition and modify the physical characteristics of the habitat. Of course, other species will take their place unless the disturbance is severe. The desirability or acceptability of these changes needs to be assessed on a case-bycase, site-by-site basis.

Soil effects Disturbance of soil structure is a common consequence of military training (Goran et al., 1983; Diersing and Severinghaus, 1984; O’Neil et al., 1990; Pearson et al., 1990). Erosion potential is increased on disturbed soils (Diersing and Severinghaus, 1984; O’Neil et al., 1990; Pearson et al., 1990; Trumbull et al., 1994) and infiltration rate is decreased (Trumbull et al., 1994). Compaction is a well-studied consequence of soil disturbance (Becher, 1985). Compaction varies with the moisture content at the time of impact, parent material, vegetation type, and the characteristics of the vehicle (United States Bureau of Land Management, 1980; Adams et al., 1982; Becher, 1985; Thurow et al., 1993). Timing of training exercises relative to ecosystem moisture regimes therefore may be expected to influence observed effects (Wilson, 1988; Thurow et al., 1993). Compaction effects are more pronounced and longer-lasting when vehicles pass over wet soil (Thurow et al., 1993). Compaction can reduce aeration and inhibit root growth, nutrient uptake, and seedling emergence (Chancellor, 1977). Thus, plant-community effects and recovery time are related to the extent of compaction (Webb and Wilshire, 1980; United States Bureau of Land Management, 1980; Prose, 1985; Thurow et al., 1993). Potential for wind erosion generally increases on soils subject to vehicle passages, especially when they are dry. Wind was cited as a factor at such dissimilar locations as Fort Bliss in the Chihuahuan Desert of New Mexico (Marston, 1986; Gillespie, 1987) and at Fort Lewis in the Pacific Border Province of Washington (Pearson et al., 1990). After the Gulf War, wind erosion increased where moving vehicles had exposed fine substrate materials after breaking through the “desert pavement” (a layer of pebbles and cryptogamic crust left behind after fine materials have blown away) (El-Baz, 1992). Wind erosion causes dune formation and dust storms (Krzysik, 1985; McDonald, 1995). Heavy metals and other contaminants have been deposited into soils because of training exercises (Peters and Miller, 1993; Hinsenveld, 1995; Freese and Riesbeck, 1995) and because of activities collateral to military missions such as oil-well fires (Sadiq et al., 1992) and construction (Hagarty et al., 1993). These contaminants have the potential to enter food webs.

DISTURBANCE ASSOCIATED WITH MILITARY EXERCISES

Hydrology effects Water infiltration tends to be lowered in compacted sites, runoff thereby being increased and erosion accelerated (Hillel, 1980; Prose, 1985). However, prairie soils at Fort Hood, Texas, were compacted only when vehicles passed over wet soils (Thurow et al., 1993). Decreased infiltration rates appeared to promote dominance by early-successional plants, even after some soil physical properties (e.g., bulk density) had recovered to their pre-disturbance condition (Thurow et al., 1993). Removal of vegetation and damage to soil causes increases in runoff rates and the amount of transported sediment by lowering field capacity (e.g., Riggins et al., 1989). This change increases the frequency and severity of flooding, including flash floods, and causes siltation. Somewhat paradoxically, imprints from the treads of tracked vehicles store water, thereby lowering erosion potential for a time (Riggins et al., 1989). Effects on plants Effects on plants include direct crushing and killing; which ultimately causes reduced stem density (Trumbull et al., 1994). Root systems are directly impacted by vehicles, as a result of alteration of bulk density and the rate of water infiltration (Trumbull et al., 1994). These effects can produce site characteristics inconsistent with the autecological requirements of the species, and may reduce individual vigor. The relative cover of dominant plant species remained altered in the Mojave Desert 36 years after a military maneuver (Lathrop, 1983). In the same ecosystem, training activities have been found to disturb vegetation by reducing the density and cover of creosotebush (Larrea tridentata) and other plants (Krzysik, 1985). Cool-season grasses and warm-season forbs replaced warm-season grasses in locations subject to trackedvehicle impact over two years at Fort Carson, Colorado, apparently because of competitive interactions associated with spring rain events (Diersing and Severinghaus, 1984; Shaw and Diersing, 1990). Also, woody plants were damaged and killed, allowing increased cover of undesirable and disturbance-related snakeweed (Gutierrezia sarothrae) and kochia (Kochia scoparia) (Diersing and Severinghaus, 1984). Pitting (forming small basins or pits in the soil to catch and hold water and store moisture for plant use) promoted

389

the establishment of cool-season grasses and forbs to reduce erosion potential on sites subject to heavy vehicle impact (Berlinger and Cammack, 1990). Wind erosion can transport seeds, expose roots, and decrease soil fertility (Marston, 1986). Ultimately, these effects would result in establishment of weedy invaders and habitat perturbation [permanent modification of a habitat from one condition to another: White and Pickett (1985)]. Effects on animals Physical modification of habitat resulting in changed levels of available resources is the primary disturbance affecting vertebrate populations on military installations. Modifications include clearing of woodland and understory, the mixing or removal of soil and detritus, modification or removal of food resources, and general degradation of habitat (Severinghaus et al., 1980; Severinghaus and Severinghaus, 1982). Some animal species are adversely affected, some benefit, and others are not impacted (e.g., O’Neil et al., 1990; Pearson et al., 1990). This habitat modification can lead to species replacement (Diersing and Severinghaus, 1984), for example, by an increase in abundance of early-sere species at the expense of uncommon climax species and endemics. Severinghaus et al. (1980) found that small mammals, particularly those associated with the soil surface and sub-surface, were adversely affected by maneuver activities through clearing and compacting of the soil, vegetation disturbance, and resultant erosion. Moderate habitat modification can increase the abundance of white-footed mice (Peromyscus leucopus), possibly in response to invasion by weedy, early-successional forbs (Diersing and Severinghaus, 1984). In the Mojave Desert, loss of shrub cover was related to lower relative abundance of the little pocket mouse (Perognathus longimembris) and southern grasshopper mouse (Onychomys torridus) (Krzysik, 1985). Birds are generally adversely affected by activities related to maneuvers. The frequency with which 17 out of 19 bird species were observed was lower on shortterm and long-term training areas compared to a control area at Fort Knox, Kentucky (Severinghaus et al., 1980). Bird species richness in woodlands decreased at Fort Carson, Colorado, following disturbance (Tazik, 1991). Habitat modification from woodland to open woodland or forest-edge communities initially was beneficial or neutral for some bird species (edge specialists,

390

Stephen DEMARAIS, David J. TAZIK, Patrick J. GUERTIN and Eric E. JORGENSEN

seed-eaters) whereas continued habitat modification had an adverse effect on all species (Severinghaus et al., 1980). At the National Training Center in the Mojave Desert, California, disturbance (removal of shrub cover) had an adverse effect on sparrows (Amphispiza belli, A. bilineata, Spizella breweri, S. passerina and Zonotrichia leucophrys), meadowlarks (Sturnella neglecta), and thrashers (Toxostoma lecontei) (Krzysik, 1985). Bird biomass decreases as training activities increase, except in mature forested habitats where large trees tend to be resistant to maneuver-induced damage (Severinghaus and Severinghaus, 1982). Prairie species seem to be more resilient to disturbance impacts than woodland species (Tazik, 1991), perhaps because they are adapted to a climax habitat with less vertical structure and complexity. Changes in the physical structure of the habitat may have an impact on animal populations. In anaerobic marsh sediments at Fort Richardson, Alaska, mortality of waterfowl (Anas sp. and Cygnus sp.) was linked to ingestion of white phosphorus, which resists environmental breakdown (Racine et al., 1992). Blowing dust can be detrimental to desert fauna (Marston, 1986; El-Baz, 1992), but long-term impacts on animal populations and communities are not known. Community effects Species replacement (loss of one species while gaining another with no apparent change in net diversity) is an important issue on military lands (Severinghaus and Severinghaus, 1982; Diersing and Severinghaus, 1984), particularly in regard to climax species. Replacement of these frequently rare or sensitive species by common mid- and early-successional species reduces landscape diversity (Scott et al., 1996). For example, earlysuccessional annuals replaced perennial grasses at Fort Hood, Texas (Johnson, 1982; Thurow et al., 1993). Exotics invaded sites subject to vehicle impact during spring in Manitoba (Wilson, 1988). Generally, vehicle impact decreases cover and allows invasion by annual grasses and forbs. Disturbance of climax-type vegetation frequently increases biodiversity (e.g., Lathrop, 1983). These changes can last for many years (Lathrop, 1983). However, continued periodic disturbance ultimately will degrade habitats and their ability to support biodiversity. For instance, at Fort Leonard Wood, Missouri, plant species richness decreased on sites subject to periodic encampments for 20–40 years (Trumbull et al.,

1994). These observations are consistent with Connell’s (1978) intermediate-disturbance hypothesis. Ecosystem effects Military bases frequently are virtual islands of relatively natural habitat surrounded by properties with a variety of contrasting land uses (Tennesen, 1993; Creswell, 1994; Goodman, 1996). The potential for disturbance to these regional ecosystems is of four types: (1) impacts on endemic, particularly threatened and endangered, biota (Creswell, 1994); (2) habitat fragmentation allowing exotic or weedy species to become established (Wilson, 1988); (3) impacts which move beyond the boundaries of the base (McDonald, 1995); and (4) cases where the protection provided by the base results in it being the last and best example of the native ecosystem and its habitats in an otherwise highly developed area (e.g., the Mediterranean ecosystem of southern California). Toxic contaminants may disturb surrounding ecosystems through escape from the confines of a local military base. Substantial effort has been expended in research on methods for immobilizing and cleaning contaminants (Hagarty et al., 1993; Peters and Miller, 1993). Heavy metals and other contaminants deposited into water and soils purposely in the form of spent munitions or accidentally by spills, holding ponds, or fire (Deneke et al., 1975; Sweazy et al., 1977; Machin and Ehresmann, 1985) have the potential to enter food webs (Sweazy et al., 1977; Peters and Miller, 1993; Genskow, 1994; Hinsenveld, 1995; Richter and Franke, 1995; Freese and Riesbeck, 1995). This impact may occur through the contaminants being taken up by plants, or through transport in ground and surfacewater (Peters and Miller, 1993; Freese and Riesbeck, 1995).

MANAGEMENT EFFORTS

Regulatory context The United States military has become increasingly challenged by a dramatic expansion in environmental compliance requirements (Conrad et al., 1994). During 1980–1990, the code of national environmental regulations doubled to nearly 10 000 pages (Butts, 1990). The Department of Defense manages over 800 installations covering more than 10×106 ha, each of which must

DISTURBANCE ASSOCIATED WITH MILITARY EXERCISES

comply with national, state, and local regulations. The policy of the Department of Defense requires any activities that it controls outside the United States to follow generally accepted environmental standards adopted for similar situations in the United States (Department of Defense, 1991). A variety of United States environmental laws currently exist affecting the nature, frequency, and extent of ground disturbance caused by military operations [see review by Conrad et al. (1994), and also Eckert and Carroll, Chapter 30, this volume]. The Endangered Species Act which requires that adverse impacts on threatened or endangered species and their critical habitat should be avoided, and that programs for their conservation should be put in place, has had the greatest impact of any legislation on military training in the United States. The National Environmental Policy Act (NEPA) requires the Department of Defense to consider the environmental consequences of their actions and to document these considerations. Assessments like those for the National Environmental Policy Act are also required on overseas bases. The Clean Water Act requires control of soil erosion that results in sediment deposition in surface waters. Section 404 of the Act requires permits for dredge and fill operations and the delineation of wetlands. The Resource Conservation and Recovery Act imposes comprehensive regulations on hazardous wastes. The Sikes Act requires that installations should be managed so as to ensure sustained multiple use of natural resources. The Migratory Bird Treaty Act authorizes the Department of the Interior to regulate activities affecting virtually all avian species in the United States, and the United States Fish and Wildlife Service has become aggressive in its enforcement. Recently, the United States Army has modified training activities in response to a documented loss of migratory waterfowl killed by ingestion of residue from white phosphorus rounds fired into wetlands (Racine et al., 1992). Protection of sections of rivers under the Wild and Scenic Rivers Act has required the modification of flight corridors for aircraft. The Environmental Conservation Program of the Department of Defense describes natural resource policy on lands under the control of the Department in the United States, its Territories, trusts, and possessions (Department of Defense, 1996). Integrated plans for natural-resource management are developed in accordance with principles of ecosystem management. Activities of the Department of Defense must promote

391

conservation of biological diversity when practicable and consistent with the military mission. The goal of the ecosystem-management program of the Department of Defense is to ensure that military lands support present and future training requirements while preserving and enhancing ecosystem integrity. This goal includes restoring and maintaining native ecosystems, reestablishing and maintaining viable populations of native species, maintaining evolutionary and ecological processes (e.g., disturbance regimes, hydrological process, and nutrient cycles), and managing the sites over time periods compatible with ecosystem dynamics. The challenge lies in balancing these potentially competing resource uses, while recognizing the primacy of the military mission and the goal of ecosystem maintenance. Case studies Several case studies are presented below to illustrate the context, nature, and range of ecosystem management issues faced by military land managers. This discussion is not intended as a comprehensive exposition on the subject. Eglin Air Force Base (Department of the Air Force, 1993) Eglin Air Force Base is located in northwestern Florida, United States. Major ecological associations include sandhills, wetlands and riparian communities, sand pine (Pinus clausa), flatwoods including the Pinus palustris association, pine and mixed hardwood forests, and barrier islands. Eglin is the largest forested military reservation in any country of the North Atlantic Treaty Organization, totaling over 187 500 ha. The primary military mission of Eglin is the development and testing of conventional munitions and sensor tracking systems. Other training activities include ground-troop maneuvers and flight missions. Other uses include the conservation of unique natural resources (of regional, national and international significance), recreation (including fishing, hunting, camping, picnicking, etc.), and the development of forest products. Eglin’s natural-resources management plan recognizes several major issues and management concerns. Land-management activities must, to the extent possible, be compatible with the testing and training mission. However, federal laws, such as the Endangered Species Act, mandate certain management actions that may constrain the mission. For example, Eglin encompasses

392

Stephen DEMARAIS, David J. TAZIK, Patrick J. GUERTIN and Eric E. JORGENSEN

the entire geographical range of several species, and has 89 endangered or potentially endangered species. Among these, the red-cockaded woodpecker (Picoides borealis) is an endangered species of some notoriety on military lands throughout the southeastern United States, because of its impact on land-management decisions. Eglin also harbors over one-third of the estimated 2000 ha of remaining old-growth longleaf pine (Pinus palustris) in the southeastern coastal plain. Training activities on Eglin are compatible with the ecosystem-management goals of the Department of Defense and can be achieved with only minor restrictions or mitigation to minimize training-related disturbance. Physical damage from mechanical equipment is infrequent and on a small scale only. A program is in place to close unnecessary roads, correct erosion problems, and restore forested areas to a natural state. Existing old-growth forests are managed so as to perpetuate them. Canadian Forces Base Shilo (Stewart et al., 1987) Canadian Forces Base Shilo is a 39 511 ha facility in Manitoba, Canada. At the base, the armies of Canada and Federal Republic of Germany train in the use of tank and armored personnel carriers on a 21-day rotation during the five summer months. Each rotation involves 600 troops. The activities and artillery exercises conducted by the Canadians result in damage to plant cover and soils. The major issues of training-related disturbance involve vegetation, soils, and wildlife. Impacts on vegetation and soils include reduction in desirable native species, increased soil compaction, and encroachment of undesirable species such as leafy spurge (Euphorbia esula). Negative impacts on elk (Cervus elaphus) are the primary wildlife concern. Disturbance impacts are mitigated by a variety of approaches. Harrowing, seeding, and fertilizing have been applied to help recovery. Harrowing and a twoyear rest period proved most effective in restoring native species. General range-management practices employed to minimize soil and vegetation damage include use of designated roads only, minimizing sharp turns of vehicles, reducing range fires, rotation of training areas, delay of training start-up until early-season perennials have established growth, and restricting access in areas of recognized ecological value. Mitigation activities for elk have been limited

largely to establishing areas of restricted access and seasonal avoidance of areas where the elk calve and rest, and have their winter range. Hohenfels Combat Maneuver Training Center (Sullivan et al., 1996) Hohenfels Combat Maneuver Training Center occupies 16 200 ha in Bavaria, Germany. The vegetation consists of mixed forest and grasslands in a topography of rolling hills and valleys. Hohenfels has been in operation since 1951, and has been used as a combat maneuver training center (CMTC) since 1989. One of its major training missions currently is training for United Nations operations such as peace-keeping. The Federal Forestry Department has managed CMTC Hohenfels woodlands to maximize environmental protection. Consequently, the CMTC has become a sanctuary for many plant and animal species whose habitats have been lost elsewhere. Training-related disturbance impacts are associated with repeated and prolonged use of heavy vehicles. Impacts include loss of vegetation and associated soil erosion. Hohenfels is home to a variety of threatened and endangered species, making it a site of national, or even European, significance. Mitigation and management efforts are designed to support training while conserving environmental resources. Severely damaged areas are rehabilitated using a variety of measures, including reseeding, replanting, building structures for the control of water and sediment, and improving training-area design. Environmentally sensitive areas are designated as offlimits to training. Communication between trainers and environmental managers is fostered by the development of videographic simulation of alternative training-area designs. Semi-dry and dry meadows contain numerous threatened plant species. This habitat is protected by use of native grasses only in the seed mixtures used for reseeding these areas. Spreading of trees and shrubs is controlled by cutting and by sheep grazing. Shoalwater Bay Training Area (Tunstall, 1993) Shoalwater Bay Training Area occupies 270 000 ha of land together with 14 000 ha of mangroves in central coastal Queensland, Australia, 80 km north of the Tropic of Capricorn. Vegetation is a diverse mix of eucalypt forest on hilly and mountainous areas, with eucalypt and paperbark (Melaleuca) woodland on the plains. Mangroves are extensive on marine sediments, while sand dunes are covered with various forms of heath.

DISTURBANCE ASSOCIATED WITH MILITARY EXERCISES

Shoalwater is managed by the Australian Army as a joint service facility for training exercises up to divisional scale, including vehicle maneuvers. Because of its high conservation status, an Australian Commission of Inquiry has recommended that conservation have equal status with military training. Primary training-related disturbances result from engineering activities, movements of tracked vehicles, and fire. Defensive positions include construction of tank traps and other earthworks, sandbag and timber constructions, vegetation clearing, felling trees for obstacles, barbed wire emplacements, and road demolition. Vehicle traffic results in soil compaction and denudation. Lands adjacent to major camps and defensive positions are typically cleared of woody vegetation using bulldozers and hand tools. Maneuvers by tracked vehicles typically are limited to areas adjacent to roads and tracks, but result in crushing vegetation, soil compaction, and soil displacement. Fire is a natural disturbance event, the frequency of which has increased as a result of training activities. Its unpredictability and potentially dramatic effects make fire a major problem. Although much of the current vegetation (except mangroves) is fire-adapted, an unnaturally high fire frequency could have detrimental effects. Additional damage results from camp sites, tracked vehicles, bombing, naval demolition, and timber harvest. Mitigation and management of disturbance impacts include a rather extensive land-management plan. The plan’s objective is to specify land-management practices appropriate to both military use and conservation. It attempts to contain exercise damage to the minimum necessary to achieve the objectives of the exercise, in order to sustain viability for military use over the long term. Each component of the plan (addressing, for instance, fire, exercises and maneuvers, engineering works, and conservation of the biota) specifies management according to objectives, strategy, priority actions, and performance indicators. The impact of maneuvers involving tracked vehicles is limited by rotational use of training areas. Nontactical movement is limited to roads. Trees are not to be felled indiscriminately, wet soils are avoided, and areas are to be rested when extensively damaged. Areas used for defensive exercises are refurbished so as to minimize environmental damage. Tank traps and other earthworks are developed under strict guidelines, pits and trenches are backfilled and mounded, rubbish is carted off the maneuver area rather than being buried,

393

and all defense stores are recovered. In other tactically significant areas, trees are not to be felled, and tracks are graveled where severe rutting and bogging occurs. Fire is managed by construction and maintenance of perimeter fire breaks, use of prescribed burning to protect life and assets, varying the frequency, timing, intensity, and extent of burns to yield a patchy mosaic, and reducing the frequency and extent of prescribed burning to protect vegetation and wildlife.

CONCLUSIONS

Military lands are widespread and diverse, and include virtually every ecosystem. Therefore, the type and scope of their disturbance are nearly limitless. Nonetheless, there are commonalities that link disturbances occurring on military lands. The common features of disturbance in military lands derive from their use for training maneuvers. Disturbance impacts may result from maneuvers involving tracked vehicles, infantry and live fire, and activities of command and support personnel and engineers, among others. Training maneuvers encompass thousands of activities, any of which could produce persistent disturbance. Historically, tracked-vehicle maneuvers have been investigated in most detail, probably because of their visual impact. Less striking, but potentially of equal importance, are the impacts of less visible activities. Activities such as bivouacking and vehicle fueling may have minimal short-term but significant long-term cumulative impacts. Although a variety of land forms are susceptible to military disturbance, it is noteworthy that maneuvers tend to occur with high frequency in particular favored land forms. Conditions of slope, soil composition, vegetation, and drainage interact to concentrate activities in some areas while adjacent areas remain unaffected. Maneuvers may impact soils and vegetation, thereby altering physical properties, hydrology, structure, and species composition of the surface and sub-surface. Maneuvers may cause retrogression to early and midsuccessional stages. From an ecosystem perspective, these communities are undesirable because they promote common species at the expense of rare latesuccession species. Ecosystem management increasingly is becoming the context for management of military land. In this context, species replacement is an important issue on military lands. As a rule, access to military land is

394

Stephen DEMARAIS, David J. TAZIK, Patrick J. GUERTIN and Eric E. JORGENSEN

controlled and development is carefully planned. Thus, military lands frequently support the last and best examples of intact ecosystems, and the importance of their contribution to regional ecosystem management is significant. The military’s interest in maintaining ecosystems and their processes goes beyond regulatory necessity. Realistic training potential is important, and cannot occur on severely degraded lands. Nonetheless, the area available is usually fixed, with minimal potential for acquisition of new land. Thus, training occurs repeatedly at the same locations. To prevent severe degradation of these locations, restoration-oriented research and management are required. Training impacts and resultant ecosystem restoration occur over large spatial and temporal scales. Thus, documentation of the military’s long-term cumulative impacts will remain a priority. The planning emphasis must remain on longterm management approaches at the landscape and ecosystem level.

REFERENCES Adams, J.A., Endo, A.S., Stolzy, L.H., Rowlands, P.G. and Johnson, H.B., 1982. Controlled experiments on soil compaction produced by off-road vehicles in the Mojave Desert, California. J. Appl. Ecol., 19: 167–175. Barrow, C.J., 1991. Land Degradation. Cambridge University Press, New York, 295 pp. Becher, H.H., 1985. Compaction of arable soils due to reclamation or off-road military traffic. Reclam. Revegetation Res., 4: 155–164. Berlinger, B.P. and Cammack, L.R., 1990. Revegetating rangelands after army maneuvers. Rangelands, 12: 17–20. Butts, K.H., 1990. The Army and the Environment: Report of the Strategic Outreach Program Roundtable Conference. Strategic Studies Institute, United States Army War College, 24 pp. Chancellor, W.J., 1977. Compaction of soil by agricultural equipment. Univ. Calif. Berkeley Div. Agric. Sci. Bull., 1881: 111 pp. Connell, J.H., 1978. Diversity in tropical rainforests and coral reefs. Science, 199: 1302–1310. Conrad, J.C., Riggins, R.E. and Foley, C.C., 1994. Land for Combat Training: Phase I Report. Army Environmental Policy Institute, Information Papers 195: 51 pp. Creswell, L.L., 1994. Endangered Species on Military Training Lands: Cooperation Between the Military Services and the United States Fish and Wildlife Service. Naval War College, Newport, Rhode Island, 104 pp. Deneke, F.J., McCown, B.H., Coyne, P.I., Rickard, W. and Brown, J., 1975. Biological Aspects of Terrestrial Oil Spills, USA CRREL Oil Research in Alaska, 1970–1974. United States Army Corps of Engineers, Cold Regions Research and Engineering Laboratory, 66 pp. Department of Defense, 1991. DoD Directive 6050.16: DoD Policy for Establishing and Implementing Environmental Standards at

Overseas Installations. Department of Defense, Washington, D.C., 5 pp. Department of Defense, 1996. DoD Instruction 4715.3: Environmental Conservation Program. Department of Defense, Washington, D.C., 16 pp. Department of the Air Force, 1993. Natural Resources Management Plan: Eglin Air Force Base, Florida, 1993 to 1997. Department of the Air Force, Washington, District of Columbia, 81 pp. Department of the Army, 1984. Army Field Manual 25–4: How to Conduct Training Excercises. Department of the Army, Washington, D.C., 179 pp. Department of the Army, 1986. Army Field Manual 7–70: Light Infantry Platoon/Squad. Department of the Army, Washington, D.C., 291 pp. Department of the Army, 1988a. Army Field Manual 5–100: Engineer Combat Operations. Department of the Army, Washington, D.C., 166 pp. Department of the Army, 1988b. Army Field Manual 25–100: Training the Force. Department of the Army, Washington, D.C., 45 pp. Department of the Army, 1988c. Army Field Manual 71–1: Tank and Mechanized Infantry Company Team. Department of the Army, Washington, D.C., 222 pp. Department of the Army, 1988d. Army Field Manual 71–2: The Tank and Mechanized Infantry Battalion Task Force. Department of the Army, Washington, D.C., 222 pp. Department of the Army, 1989. Army Training and Evaluation Plan 5–145–31: Mission Training Plan for the Engineer Company. Department of the Army, Washington, D.C., 220 pp. Department of the Army, 1991. Army Training Circular 25–1: Training Land. Department of the Army, Washington, D.C., 27 pp. Department of the Army, 1993. Army Pamphlet 350–38: Standards in Weapons Training. Department of the Army, Washington, D.C., 188 pp. Diersing, V.E. and Severinghaus, W.D., 1984. The Effects of Tactical Vehicle Training on the Lands of Fort Carson, Colorado – An Ecological Assessment. United States Army Construction and Engineering Research Laboratory, Technical Report N-85/03, 46 pp. Diersing, V.E. and Severinghaus, W.D., 1985. Wildlife as an Indicator of Site Quality and Site Trafficability During Army Training Maneuvers. United States Army Construction and Engineering Research Laboratory, Technical Report N-86/03, 19 pp. Dinkins, W.C., Lochmiller, R.L., Bartus, W.S., DeYoung, C.A., Qualls Jr., C.W. and Fulton, R.W., 1992. Cause-specific mortality of white-tailed deer as influenced by military training activities in southwestern Oklahoma. J. Wildl. Dis., 28: 391–399. El-Baz, F., 1992. Preliminary observations of environmental damage due to the Gulf War. Nat. Resources Forum, 16: 71–75. Evinger, W.R., 1995. Directory of U.S. Military Bases Worldwide. Oryx Press, Phoenix, Arizona, 412 pp. Freedman, B., 1989. Environmental Ecology: The Impacts of Pollution and Other Stresses on Ecosystem Structure and Function. Academic Press, London. Freese, D. and Riesbeck, F., 1995. Reclamation of military areas polluted with traces of ammunition and heavy metals, using a new technology and subhydric sediments. In: W.J. van den Brink, R. Bosman and F. Arendt (Editors), Contaminated Soil. Kluwer Academic Publ., Netherlands, pp. 1207–1208.

DISTURBANCE ASSOCIATED WITH MILITARY EXERCISES Genskow, K.D., 1994. Nonpoint Source Pollution: Implications of Clean Water Act Revisions on Army Combat Training and Land Management. Master’s Thesis, Univ. Illinois at Urbana-Champaign, 89 pp. Gese, E.M., Rongstad, O.J. and Mytton, W.R., 1989. Changes in coyote movements due to military activity. J. Wildl. Manage., 53: 334–339. Getz, L.L., Reinbold, K.A., Tazik, D.J., Hayden, T.J. and Cassels, D.M., 1996. Preliminary Assessment of the Potential Impact of Fog Oil Smoke on Selected Threatened and Endangered Species. United States Army Construction and Engineering Research Laboratory, Technical Report 96/38, 44 pp. Gillespie, B.M., 1987. The Impact of Military Maneuvers on Eolian Transport and Soil Compressive Strength in South Central New Mexico. Master of Arts Thesis, University of Wyoming, Laramie, 178 pp. Goodman, S.W., 1996. Ecosystem management at the Department of Defense. Ecol. Appl., 6: 706–707. Goran, W.D., Radtke, L.L. and Severinghaus, W.D., 1983. An Overview of the Ecological Effects of Tracked Vehicles on Major US Army Installations. United States Army Construction and Engineering Research Laboratory Technical Report N-142, 75 pp. Gradwohl, J. and Greenberg, R., 1988. Saving the Tropical Forests. Earthscan Publications, London, 214 pp. Hagarty, E.P., Dee, P.E., Kikkeri, S.R. and Wilcher, J.L., 1993. Remediation of contaminated soil at a military installation site. In: J. Hager, B. Hansen, W. Imrie, J. Pusatori and V. Ramachandran (Editors), Extraction and Processing for the Treatment and Minimization of Wastes. The Materials and Minerals Society, pp. 441–459. Hillel, D., 1980. Fundamental Soil Physics. Academic Press, New York, 486 pp. Hinchman, R.R., McMullen, K.G., Carter, R.P. and Severinghaus, W.D., 1990. Rehabilitation of Military Tracked Vehicles at Fort Carson, Colorado. U.S. Army Construction and Engineering Research Laboratory, Technical Report N-91/01, 62 pp. Hinsenveld, M., 1995. Remediation strategies for contaminated (former) military sites. In: W.J. van den Brink, R. Bosman and F. Arendt (Editors), Contaminated Soil. Kluwer Academic Publishing, Netherlands, pp. 97–98. Johnson, F.L., 1982. Effects of tank training activities on botanical features at Fort Hood, Texas. Southwest. Nat., 27: 309–314. Krzysik, A.J., 1985. Ecological Assessment of the Effects of Army Training on a Desert Ecosystem: National Training Center, Fort Irwin, California. United States Army Construction and Engineering Research Laboratory, Technical Report N-85/13, 139 pp. Krzysik, A.J., 1994. Biodiversity and the Threatened/Endangered/ Sensitive Species of Fort Irwin, California. United States Army Construction and Engineering Research Laboratory Technical Report EN-94/07, 114 pp. Lanier-Graham, S.D., 1993. The Ecology of War: Environmental Impacts of Weaponry and Warfare. Walker, New York, 185 pp. Lathrop, E.W., 1983. Recovery of perennial vegetation in military maneuver areas. In: R.H. Webb and H.G. Wilshire (Editors), Environmental Effects of Off-Road Vehicles; Impacts and Management in Arid Regions. Springer-Verlag, New York, pp. 266–277.

395

Machin, J.L. and Ehresmann, J., 1985. Remediating a fire site. Civ. Eng., 60: 55–56. Marston, R.A., 1986. Maneuver-caused wind erosion impacts, south central New Mexico. In: W.G. Nickling (Editor), Aeolian Geomorphology. Proc. 17th Annual Binghamton Geomorphology Symp., 17: 273–290. McDonald, K.W., 1995. The Effects of Military Maneuvers on Soil Structure Breakdown and Wind Erosion at the National Training Center, Fort Irwin, California. Master of Science Thesis, Western Kentucky University, Bowling Green, 63 pp. Moloney, K.A. and Levin, S.A., 1996. The effects of disturbance architecture on landscape-level population dynamics. Ecology, 77: 375–394. O’Neil, L.J., Waring, M.R., Hughes, H.G., Landin, M.C., Pearson, M.L., Morris, P.A. and Larson, R.J., 1990. Proposed 9th Infantry Division Force Conversion; Maneuver Damage, Erosion and Natural Resources Assessment Yakima Firing Center, Washington. United States Army Waterways Experiment Station, Technical Report EL-90–9, 114 pp. Owens, S., 1990. Defense and the environment: the impacts of military live firing in national parks. Cambridge J. Econ., 14: 497–505. Pearson, M.L., Morris, P.A., Larson, R.J., O’Neil, L.J., Waring, M.R., Hughes, H.G. and Ladin, M.C., 1990. Proposed 9th Infantry Division Force Conversion; Maneuver Damage, Erosion and Natural Resources Assessment Fort Lewis, Washington. United States Army Waterways Experiment Station, Technical Report GL90–13, 115 pp. Peters, R.W. and Miller, G., 1993. Remediation of heavy metal contaminated soil using chelant extraction: feasibility studies. Proc. 48th Industrial Waste Conf., 48: 141–167. Pickett, S.T.A. and White, P.S., 1985. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, San Diego, California, 472 pp. Prose, D.V., 1985. Persisting effects of armored military maneuvers on some soils of the Mojave Desert. Environ. Geol. Water Sci., 7: 163–170. Racine, C.H., Walsh, M.E., Roebuck, B.D., Collins, C.M., Calkins, D., Reitsma, L., Bucjli, P. and Goldfarb, G., 1992. White phosphorus poisoning of waterfowl in an Alaskan salt marsh. J. Wildl. Dis., 28: 669–673. Richter, M. and Franke, C., 1995. Distribution and mobilisation of nitroaromatic compounds in a former military shooting area. In: W.J. van den Brink, R. Bosman and F. Arendt (Editors), Contaminated Soil. Kluwer Academic Publishing, Netherlands, pp. 411–412. Riggins, R.E., Holge, W., Lacey, R.M. and Ward, T.J., 1989. Sediment Control at Army Training Areas Case Study: Hohenfels, Federal Republic of Germany. United States Army Construction and Engineering Research Laboratory Technical Report N-89/08, 25 pp. Sadiq, M., Al-Thagafi, K.M. and Mian, A.A., 1992. Preliminary evaluation of metal contamination of soils from the Gulf Activities. Bull. Environ. Contamination Toxicol., 49: 633–639. Scott, J.M., Ables, E.D., Edwards Jr., T.C., Eng, R.L., Gavin, T.A., Harris, L.D., Haufler, J.B., Healy, W.M., Knopf, F.L., Torgerson, O. and Weeks Jr., H.P., 1996. Conservation of biological diversity: perspectives and the future for the wildlife profession. Wildl. Soc. Bull., 23: 646–657.

396

Stephen DEMARAIS, David J. TAZIK, Patrick J. GUERTIN and Eric E. JORGENSEN

Severinghaus, W.D. and Severinghaus, M.C., 1982. Effects of tracked vehicle activity on bird populations. Environ. Manage., 6: 163–169. Severinghaus, W.D., Riggins, R.E. and Goran, W.D., 1980. Effects of tracked vehicle activity on terrestrial mammals and birds at Fort Knox, Kentucky. Trans. K. Acad. Sci., 41: 15–26. Shaw, R.B. and Diersing, V.E., 1990. Tracked vehicle impacts on vegetation at the Pinon Canyon Maneuver Site, Colorado. J. Environ. Qual., 19: 234–243. Smith, R.L., 1986. Elements of Ecology. Harper and Row, New York, 677 pp. Stanford Biology Study Group, 1971. Destruction of Indochina. In: J.P. Holdren and P.R. Ehrlich (Editors), Global Ecology: Readings Towards a Rational Strategy for Man. Harcourt, Brace, Jovanovich, New York, pp. 146–154. Stewart, J.A., Downs, A.T. and Stones, G.A., 1987. The impact of military training in Canada on indigenous flora and fauna. In: Proceedings: NATO CCMS Seminar Blue Book 159, Preservation of Flora and Fauna in Military Training Areas, Conference Proceedings N-87/09. US Army Construction Engineering Research Laboratory, pp. 107–124. Sullivan, R.G., Hatton, P.J. and Boehm, A., 1996. Environmental Management Programs at Combat Maneuver Training Center Hohenfels. Multimedia CD-ROM. Argonne National Laboratory, Argonne, Illinois. Sweazy, R.M., Rose, F.L. and Baugh, C.L., 1977. Toxic Effects of Military Wastewater Effluent. Water Resources Center, Texas Tech University, Lubbock, Texas, 71 pp. Tazik, D.J., 1991. Effects of Army Training Activities on Bird Communities at the Pinon Canyon Maneuver Site, Colorado. United States Army Construction and Engineering Research Laboratory Technical Report N-91/31, 113 pp. Tazik, D.J., Cornelius, J.D., Herbert, D.M., Hayden, T.J. and Jones, B.R., 1992a. Biological Assessment of the Effects of Military Associated Activities on Endangered Species at Fort Hood, Texas. Special Report EN-93/01, 139 pp. Tazik, D.J., Warren, S.D., Diersing, V.E., Shaw, R.B., Brozka, R.J., Bagley, C.F. and Whitworth, W.R., 1992b. U.S. Army Land

Condition-trend Analysis (LCTA) Plot Inventory Field Methods. United States Army Construction and Engineering Research Laboratory Technical Report N-92/03, 62 pp. Tennesen, M., 1993. Can the military clean up its act? Natl. Wildl., 31: 14–19. Thomas, W., 1995. Scorched Earth: The Military’s Assault on the Environment. New Society Publishers, Philadelphia, Pennsylvania, 227 pp. Thurow, T.L., Warren, S.D. and Carlson, D.H., 1993. Tracked vehicle traffic effects on the hydrologic characteristics of central Texas rangeland. Trans. Am. Soc. Agric. Eng., 36: 1645–1650. Thurow, T.L., Warren, S.D. and Carlson, D.H., 1995. Tracked Vehicle Traffic Effects on the Hydrologic Characteristics of Central Texas Rangeland. United States Army Construction and Engineering Research Laboratory Tech. Man. EN-95/02, 10 pp. Trumbull, V.L., Dubois, P.C., Brozka, R.J. and Guyette, R., 1994. Military camping impacts on vegetation and soils of the Ozark Plateau. J. Environ. Manage., 40: 329–339. Tunstall, B., 1993. Environmental Impact Assessment – Shoalwater Bay Training Area (Draft). Commonwealth Scientific and Industrial Research Organization, Division of Water Resources, Canberra, Australian Capital Territory, 81 pp. United States Bureau of Land Management, 1980. The Effects of Disturbance on Desert Soils, Vegetation, and Community Processes with Emphasis on Off-road Vehicles: A Critical Review. Riverside, California, Desert Planning Staff, 190 pp. Webb, R.H. and Wilshire, H.G., 1980. Recovery of soils and vegetation in a Mojave Desert ghost town, Nevada, U.S.A. J. Arid Environ., 3: 291–303. White, P.S. and Pickett, S.T.A., 1985. Natural disturbance and patch dynamics: an introduction. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, San Diego, California, pp. 3–13. Wilshire, H.G. and Nakata, J.K., 1976. Off-road vehicle effects on California’s Mojave Desert. Calif. Geol., 29: 123–132. Wilson, S.D., 1988. The effects of military tank traffic on prairie: a management model. Environ. Manage., 12: 397–403.

Chapter 16

DISTURBANCE IN URBAN ECOSYSTEMS Herbert SUKOPP and Uwe STARFINGER

INTRODUCTION

Urban ecosystems differ from natural or rural ones in many obvious ways. Human activities, such as building, traffic, or industrial production affect the quality of air, water, and soil which impacts ecosystems in many ways. Plants can be destroyed, their production reduced, animals can get killed or scared away. The results are altered population dynamics, species composition, and energy and matter fluxes in urban ecosystems. In plant ecology, disturbance is frequently defined as a mechanism that limits plant biomass by causing its partial or total destruction (Grime, 1979). In order to include disturbance effects on ecosystems and on animals, we follow White and Pickett (1985) who define disturbance as a relatively discrete event in time that disrupts ecosystem, community, or population structure. This can include the killing, displacement, or damaging of individuals, and create an opportunity for new individuals to become established (Sousa, 1984). The study of urban ecosystems is a relatively recent phenomenon in ecology, because most ecologists have been, and still are, interested mostly in natural ecosystems. Urban ecology as a scientific discipline is being practiced in Europe (mainly western and central Europe) more than in other regions of the world. It was the subject of a few recent books, such as those of Sukopp et al. (1990, 1995), both with an international perspective; those by Gilbert (1991) and Sukopp and Wittig (1993) are more strongly focused on European countries. In North America the ecology of urban forests has been studied in more detail (Rowntree, 1984, 1986). As a consequence, the majority of our data, examples, and references, concern cities in western and central Europe. In this paper, we want to show how the multitude

of human activities influences ecological conditions in cities, and how flora and vegetation react. After describing some typical urban habitats and their disturbance regimes, we will give accounts of plant and animal species in cities and their successional changes, in order to discuss these data in light of general approaches to naturalness and the degree of disturbance.

DISTRIBUTION OF URBAN HABITATS

Even though whole cities can be seen as large ecosystems, especially with regard to energy and matter fluxes, structurally and functionally they form complexes of various interconnected ecosystems (Wittig and Sukopp, 1993; Rebele, 1994). Contrary to common expectations, cities may be quite rich in plants and animals, both quantitatively and qualitatively. Urban ecosystems and the composition of urban plant and animal communities are greatly dependent on human activities causing disturbance. The extent of these impacts varies in time and space. Typically, cities show a mosaic of habitats with increasing degrees of human impact on a gradient from the outskirts to city centers, representing different ages or time spans of human impacts (Table 16.1). A common way to study the specific urban biota and their habitats is by investigation of their history (Aey, 1990) and comparison along rural–urban gradients (McDonnell and Pickett, 1990). Some features of habitats vary clinally with distance from city centers to the suburbs, or with time. The organisms and communities in these habitats react to human influences in various ways, and are consequently different for each structural unit of the city. Understanding of the distribution of habitat types in cities is

397

398

Herbert SUKOPP and Uwe STARFINGER

Table 16.1 Some features of habitat types with different ages and locations on the country–city gradient 1 (examples from Central Europe) Forest

Field

Suburban garden

Inner city block

Climate

forest climate: low daily temperature maxima, high relative humidity

open land climate: higher temperature, lower humidity

mean annual temperature 1.5ºC higher than forest

highest temperature, low humidity

Relief

low, rolling hills

erosion, increased steepness

linear steep banks and dams

flattened

Hydrology

unchanged

increased run-off, higher levels in depressions

lower groundwater levels

lower groundwater levels

Soil

forest soils, brown earth, moist

plowed field soils, more nutrients (N, P, K), higher pH and humus content than forest, moist

surface sealing 50%, garden and rubble soils, nutrient rich, compacted, dry – wet

Vegetation

Quercus–Pinus forests

crops, field-weed communities

fruit trees, ornamentals, Euphorbia peplus–garden-weed communities

Lolium sward, Aegopodium–Urtica community, Hordeum community

Fauna 2

large mammals, tree bats

hare, partridge, field lark, field mouse

rabbit, stone marten

house mouse, Norwegian rat, feral pigeon, house sparrow

1

Sukopp (1990). field lark, Alauda arvensis; field mouse, Microtus spp.; hare, Lepus europaeus; house mouse, Mus musculus; house sparrow, Passer domesticus; Norwegian rat, Rattus norvegicus; partridge, Coturnix coturnix; pigeon, Columba livia; rabbit, Oryctolagus cuniculus; stone marten, Martes foina.

2

an important prerequisite for nature protection and town planning. This information is widely available for many European cities (Starfinger and Sukopp, 1994). To understand present urban biota and their ecosystems it is necessary to see them as a result of historical development. In central Europe, the process of redevelopment of a forest vegetation after the last Ice Age was not completed when human influence began to cause disturbances on a local scale. Large-scale disturbance, however, only began in medieval times with clear-cutting of extensive areas for agriculture. At the same time, human impact on the hydrology of the landscape increased. Historically, towns and cities were almost free of both spontaneously-growing and cultivated plants, due to limited space and the attitude of the inhabitants (Trepl, 1992). Residents of cities have fought nature back to create a cultural, artificial environment as opposed to the more natural environments prevailing outside (Trepl, 1992). Today, however, cities usually consist of a mixture of densely settled areas in the historic centers, remnants of agro-ecosystems, and even near-natural areas in urban forests, parks, and nature reserves.

Urban forests Even in areas adjacent to cities, and more so within the cities, forests are disturbed by urban activities such as building of houses and roads, recreation, emissions, etc. In a comparison of urban with rural forest stands in southeastern Wisconsin (U.S.A.), Sharpe et al. (1986) found that urban sites were disturbed by houses and yards, dumping, footpaths, and other factors that were not evident in the rural sites. Housing areas The vegetation in housing areas is subjected to catastrophic disturbances every time buildings are demolished and rebuilt. The disturbance regime and, consequently, the species composition of housing areas is closely linked to the age of the housing area. Aey (1990) studied the typical properties of soils and flora in housing estates of different ages in L¨ubeck, Germany. The soils of the oldest parts were richer in humus and in nutrients (nitrogen, phosphorus). The plants of the oldest parts were typically native forest plants, which had high requirements for nutrients and humidity. On the more recently disturbed soils of

DISTURBANCE IN URBAN ECOSYSTEMS

the youngest housing area (~25 years old), the flora contained a high proportion of non-native annuals. Streets The ecology of roads and their verges has been mostly studied in rural areas (Ullmann and Heindl, 1989). Some of the effects of car traffic are similar in urban areas, and often they are more pronounced in cities due to high traffic levels. Major disturbances include earth movements during the construction of streets, soil compaction due to trampling and vehicles, eutrophication and rise of pH values, as well as mowing and herbicide application in existing streets (Fig. 16.1). Salinity is a major factor in the soils of urban streets due to de-icing salt, which is widely used in city streets in climates with cold winters. (Today its use is restricted in many European countries.) Urban waste-land Once land is urbanized, it usually remains in urban use. However, some time may pass between the demolition of old buildings and the construction of new ones. In recent decades, an increasing number of inner-city sites have been left vacant after either industrial or traffic (railway) uses were discontinued due to economic or political reasons. The soils, climate, and water regime of these sites are generally strongly altered by human influence, but the vegetation can often develop with relatively little disturbance for some time. Landfills Landfills can be distinguished by the presence or absence of organic material (for example, communal garbage-disposal landfills, and construction debris, respectively). The latter resemble urban vacant lots in many respects. The climate of landfills is primarily influenced by the form of the landfill, which may impact temperatures and precipitation. Blowing dust leads to eutrophication and higher pH values in the soil up to a distance of 100 m. Sewage farms The construction and use of sewage farms marked the beginning of large-scale waste-water treatment in the second half of the 19th century. Sewage farms are usually located in the outer suburbs or areas

399

adjacent to cities. As more efficient methods of wastewater treatment were developed, the use of most sewage farms was discontinued. While in operation, the soils of these farms were permanently moist and nutrients were added in large quantities. Heavy-metal contamination, however, reached high concentrations, so that cultivation of vegetables had to be stopped. The open landscapes of sewage farms with their mosaic of fields, maintenance roads, hedges, etc., however, were rich in species. In one operating sewage farm in Berlin, the number of vascular plant species was slightly higher than in an adjacent agricultural landscape (Sukopp, 1990). Mammals and especially birds were present in high species numbers, including species rare and threatened in other parts of the city. The effects on the soil persist for a long time after the sewage-farm operation ceases. DISTURBANCE AND ABIOTIC FACTORS

Urban climate To maintain a functioning city or town, large inputs of materials and energy are needed. The resulting solid, liquid, and gaseous wastes greatly alter the city and its surroundings (cf. Duvigneaud and Denayer-de Smet, 1977). The result is a specific urban climate, as first described by Kratzer (1937), and subsequently detailed by Landsberg (1981), Oke (1987), and Kuttler (1993). Horbert et al. (1983) have described general characteristics of the urban climate, as compared to the surrounding non-urban areas, by using data from Berlin and numerous published data from other cities: (1) Higher air pollution: in the urban climate, gaseous pollution is 5–25 times higher; condensation nuclei are about 10 times more abundant. (2) Altered radiation: the urban climate has 5–15% fewer hours of sunshine; 20–25% less direct solar radiation (even 50% in winter); ~10% less surface albedo; 12% more reflected radiation leads to an increased net radiation of 11% at noon or 47% in the evening. (3) Wind speeds are reduced by 10–20%, times without wind are increased by 5–20%. (4) The ecologically most important result of these effects is a raised temperature. The difference between temperatures within and outside a city depends on its size. It can reach 9ºC on clear days in Berlin, and 7ºC in Aachen (western Germany). The yearly mean temperature in cities

400

Herbert SUKOPP and Uwe STARFINGER

Fig. 16.1. Soil impacts of a street through a forest in the city of Berlin (Blume et al., 1977, modified).

DISTURBANCE IN URBAN ECOSYSTEMS

in central Europe is 0.5 to 2.0ºC above that of the surrounding countryside. (5) The annual mean precipitation is increased by up to 20% (Berlin). (6) Due to increased temperature, the relative humidity is between 2% (winter) and 10% (summer) lower; on clear days in Berlin this difference can become 30%. Climatic conditions within a city can vary considerably, depending on type of construction, paving, location in the city and, especially, the distance to large vegetated areas. Depending on these effects, different climatic zones that are more or less concentric can be distinguished. The influence of areas of vegetation on the urban climate were investigated by von St¨ulpnagel et al. (1990). He found a reduction in temperature not only in a green area but also up to 1.5 km away from it. This climatic influence grew with the size of a green area, but was reduced where the area was divided by a road. Soils Soils of urban areas usually show very heterogeneous qualities, because the human impact in cities adds changes in soil qualities to the natural variation present before the city was built. Typically, this impact increases on a rural–urban gradient: even in areas adjacent to cities, soils may be indirectly influenced by emissions and air-borne pollutants. In the forests at the outer fringe of the Berlin metropolitan area, pH values of the soil decreased by 1.1 between 1950 and 1981 (Grenzius, 1984). Even if part of this decrease may be natural, it offers evidence for anthropogenic acidification of soils. Especially in sandy soils, acidification leads to leaching of nutrients. In more central locations in cities the deposition of acidic emissions (SO2 , NOx ) is more than compensated by deposition of dust and fertilization, so that the pH tends to be higher than under natural conditions. Soils of urban forests in New York City were extremely hydrophobic and showed much lower rates of nitrogen mineralization than those of rural sites, suggesting that hydrocarbons may limit the activity of soil microbes. Possible synergistic effects of heavy metal contamination and soil compaction would further reduce the nitrogen mineralization in the urban forest (White and McDonnell, 1988). In the densely built-up parts of the city, many soils are completely destroyed by excavation or are covered

401

with buildings or concrete paving. Consequences of this “surface sealing” for the ecosystem are habitat loss for all plants (except some lichens or mosses) and most soil organisms, and a reduction of groundwater regeneration. The percentage of sealed surfaces in large German cities is between 40 and 60%; individual blocks can have values as high as 98% and most inner city blocks have less than 10% of the area left for vegetation to develop (B¨ocker, 1985). Soil excavation for buildings and compaction by heavy vehicles can destroy soil horizons and mix topsoil with less weathered subsoils. The decomposition processes in the soil of young landfills emit heat. Shortly after the deposition of garbage in German landfills, soil temperatures up to 88ºC were found. Two years later, soil temperatures ranged from 15º to 45ºC (Kunick and Sukopp, 1975). Prominent features of landfill soils are a low bulk density (0.2 kg dm−3 to 0.9 kg dm−3 ), and high levels of organic material (20–30%; Blume et al., 1979). The decomposition of the organic fraction leads to the production of elementary nitrogen, carbon dioxide, and later, under completely anoxic conditions, to hydrogen and methane. These gases not only determine the growing conditions for plants on the landfill itself but also permeate neighboring sites. In addition, soluble inorganic and organic substances are carried away by moving groundwater. In a study of a landfill in Berlin, Blume et al. (1979) showed that lateral diffusion of methane into a neighboring stand of 90-year-old oaks (Quercus robur) damaged and killed trees (Fig. 16.2). The effect reached trees 75 m from the landfill, the trees closest to it being usually most heavily damaged, although no continuous gradient was observed. Treering analyses showed that growth reduction only began after the landfill was covered with topsoil, which caused soil processes to become anaerobic. Individual trees were killed > 80 m away, presumably as an effect of groundwater contaminated with heavy metals (Blume et al., 1979). Damage to herbaceous vegetation was not detected in this study. Schlenther et al. (1996) reported on extensive growth failures in afforestation on a sewage farm in Berlin. Heavy metals and organic pollutants were present in high concentrations, but did not seem to be sufficient to explain damages to the trees. Instead, water retention capacity, a factor connected to organic matter content of the topsoil, and consequently water availability to plants, was a key factor. Heavy metals, however, may become more of a problem in the future when

402

Herbert SUKOPP and Uwe STARFINGER

Fig. 16.2. Impacts of a landfill on the surroundings (Blume et al., 1979, modified). x-axis: distance from center of landfill; y-axis: altitude (m above sea level).

soil development leads to decreased pH values and subsequent higher biological activity in more acidic soils. Two soil types are typical of urban soils: deeply cultivated garden soils with high nutrient content and high water-holding capacity, and soils formed from rubble (e.g., from buildings demolished during war). The latter are alkaline, dry, and well-aerated in the beginning, but develop with humus accumulation (Sukopp et al., 1979b). They are usually low in nitrogen, but have moderate to high contents of phosphorus, calcium, potassium, and other nutrients (Bradshaw and Chadwick, 1980; Blume, 1993). Concentrations of metals, in particular of lead, are often considerably higher in urban areas than those in agricultural soils (Thornton, 1991). Especially in areas of past mining and smelting, metal concentrations may be very high. Although plant toxicity can be observed in some instances, there is little quantitative informa-

tion available on the impacts of metal contaminants in urban soils on plants and animals (Thornton, 1991). The results of all these human impacts on the soil biota in urban areas are poorly known. Generally there seems to be a reduction both in species numbers and biomass of soil organisms. Where soil disturbance and stressed conditions are present at the same time, there may be no permanent residents of the soil at all. In some cases the absence of decomposer organisms will lead to accumulation of litter and poor habitat quality for vegetation (Harris, 1991). Groundwater Groundwater is affected by human activities both quantitatively and qualitatively. Anthropogenic heat sources in buildings, sewage canals, etc., lead to increased temperatures in groundwater near the surface in cities (Balke, 1974; H¨otzl and Makurat, 1981). How the habitat and its communities are affected, however,

DISTURBANCE IN URBAN ECOSYSTEMS

403

Table 16.2 Tree growth of oaks (Quercus robur) under changing groundwater conditions in a city park in Berlin 1 Time

Disturbance

1773–1831

Natural groundwater level

2.2 mm

1832–1901

Groundwater level lowered by 1m

1.1 mm

1902–1945

further lowering, tree cutting

1.4 mm

1946–1978

high groundwater level, additional watering, soil cultivation

3.2 mm

1

Tree-ring width

Sukopp et al. (1979a).

has been little studied. Ecological effects of quantitative changes of the groundwater are better known. In urban areas, there is generally more variation in groundwater level and flow direction than in rural areas (Leuchs and R¨omermann, 1991). Beginning in medieval times, the construction of locks and dams for water mills caused higher groundwater levels, and led to bog formation in the upper reaches of the rivers in Central Europe (Brande, 1986). Later, low-lying areas and bogs were drained to provide new agricultural land. Water usage by growing populations, and the reduction in groundwater regeneration as a result of the sealing of surfaces, began to deplete groundwater levels, a process which accelerated after the Industrial Revolution. Among the consequences of drainage for the vegetation were damage and growth reduction of individual plants, most notably trees (Table 16.2), decline of plant species dependent on groundwater (phreatophytes: Londo, 1976) and changes in the nutrient status of the soils affected. Sandy soils can become more acidic owing to the loss of calcium in the groundwater. In bogs, on the other hand, increased aeration following the lowering of groundwater leads to decomposition and nitrification of the peat, and higher nutrient availability (Sukopp, 1981). In an environmental impact analysis of a planned motorway through the area of an inner city park, Sukopp et al. (1979a) estimated that 450 trees would die because of the lowering of the groundwater level during the construction phase. Surface water Surface water bodies in urban areas suffer from a

variety of man-induced disturbances that result in artificial, straightened, and sealed beds of rivers and lakes, an alteration of the hydrodynamics of rivers, interruption of the flow continuum by impoundments, or the complete destruction of individual water bodies by filling up or by converting natural streams into underground canals. This impedes longitudinal and vertical migrations of animals, and riparian vegetation is often directly destroyed (Schuhmacher, 1991). Main influences on the water itself are eutrophication and pollution, as well as an increase in wave action due to recreational and commercial traffic on the water. This has led to a sharp decline in reed beds in many European countries (den Hartog et al., 1989).

URBAN BIOTA AND ITS REACTION TO DISTURBANCE

Composition of urban floras Some disturbance types cause direct damage to plants. The use of salt for clearing snow and ice can have a detrimental effect on trees along urban streets. Auhagen and Sukopp (1980) found that close to 50 000 street trees (~20% of all street trees) in Berlin were threatened by salt; 90% of street tree mortality in Berlin was due to salt (Ruge, 1978). Sugar maple (Acer saccharum), widely planted in the U.S.A. as a roadside tree, is particularly sensitive to salt, and trees within 10 m of salted roads can be killed (Bradshaw and Chadwick, 1980). Shrubs and herbaceous vegetation are also influenced by salt. High salinities result in open swards and sometimes strips of bare ground, often called “salt burn”, immediately adjacent to streets (Gilbert, 1991). Of the species present in these salt-burned sites, some are known to be salt-tolerant, and in several European countries the invasion of maritime species into upland areas has been reported. Puccinellia distans, originally restricted to coastal salt marshes, is now the most widespread maritime species along roads in Britain (Gilbert, 1991). In Poznan, Poland, the species is dominant along streets, but also colonizes nitrogen-rich sites (Jackowiak, 1982, 1996). Exhaust fumes of cars contain substances (e.g., lead, cadmium) which are found in high concentrations along streets. Wu and Antonovics (1976) found concentrations of lead next to a busy street up to 50 times higher than in a control site. These high concentrations of lead resulted in selection for lead tolerance in populations of

404

Herbert SUKOPP and Uwe STARFINGER

Table 16.3 Examples of threatened species in Berlin and Germany 1 Species

Germany

Berlin Species number

Threatened (%)

Species number

Threatened (%)

Ferns and flowering plants (indigenous and archeophytes)

993

49.2

2728

32.0

Mosses and liverworts

405

75.8

1000

13.9

Mammals

53

54.7

87

49.4

Amphibians

14

78.6

19

57.9

1014

58.5

1300

41.1

Butterflies 1

Auhagen (1991).

Plantago lanceolata. High concentrations of lead and cadmium were reported by Blume et al. (1977) from Berlin, where they were restricted to the uppermost layers of the soil (Fig. 16.1). As pH values also increased significantly adjacent to the street, mobility of the heavy metals was low and no damaging effects on plants were found (Blume et al., 1977). On account of the intensity of human influence on urban ecosystems, many species are in danger of becoming extinct or are already extinct. The proportion of threatened species in various groups of organisms is often considerably higher than in areas with lower anthropogenic influence. In the area of Berlin, more than half of the species of several plant and animal groups are threatened, whereas nationwide this figure is much lower (Table 16.3). The process of species loss over the past 100 years was described by Drayton and Primack (1996) for a park in metropolitan Boston (U.S.A.). They calculated that the rate of decline was 0.36% yr−1 for the native plant species present in 1894, resulting in a total loss of 155 out of the original 422 species in this area of 400 ha. The changes in abiotic factors in cities described above result in changes in the species composition of plants and animals. Together with the formation of new types of habitat – open disturbed grounds as opposed to the closed forests prevailing in the natural landscape – the first set of new plant species arrived. As the native flora had evolved under forested conditions these newcomers, transported deliberately or by chance, proved to be competitive under the conditions now widely present. The overall result of the introduction and naturalization of plants was an increase in the numbers of spontaneously occurring species [J¨ager (1988), for central Europe] which is

especially noticeable in cities. Klotz (1990) found a highly significant regression between plant species number (after logarithmic transformation) and human population in 13 European cities. This was confirmed by Pyˇsek (1993) with a larger data set (77 cities and villages in Central Europe). The latter study showed that this increase is steeper in smaller settlements and levels off at ~1500 species in the largest cites. As population and area are correlated, this is partly an area effect, but also a result of the high habitat diversity in larger cities (Pyˇsek, 1993). Typically, there are just over 500 spontaneously growing species of ferns and flowering plants in small towns in Central Europe, but 1300 or more in big cities (Table 16.4). On the basis of their floristic composition, cities can be divided into concentric zones: 1, densely built-up central areas; 2, partly built-up central areas; 3, inner suburbs; and 4, outer suburbs (Kunick, 1974, 1981). These zones are, of course, not strictly concentric; their distribution and share of the total area varies with peculiarities in the history of individual cities. The floristic composition of these zones reflects the abiotic factors influencing plant growth. The sealing Table 16.4 Approximate numbers of plant species (ferns and flowering plants) growing naturally in Central European towns 1 Town size

Species number

Small and medium towns

530–560

Cities with 100 000–200 000 inhabitants

650–730

Cities with 250 000–400 000 inhabitants

900–1000

Cities with more than 1 million inhabitants >1300 1

Sukopp and Trepl (1993).

DISTURBANCE IN URBAN ECOSYSTEMS

of surfaces with asphalt, concrete, or buildings, for example, varies clinally from 70%–100% in zone 1 to 0%–15% in zone 4. Also, the overheating of cities (see above) is more pronounced in zones 1 and 2 than in 3 and 4. Total vegetation cover, and the number of species recorded as rare in Central Europe, decrease from the urban fringe to the city center. The total number of spontaneously growing plant species, the number of non-native species, and the number of annuals are all higher in the more central areas (Table 16.5). The latter trend becomes more apparent by comparing urban floras to those of adjacent non-urban areas. There are twice as many non-native plant species in the whole city of Berlin (41%) as in surrounding districts (20–25%: Klemm, 1975). Falinski (1971) compared floras of human settlements in Poland and found an increasing percentage of non-native plant species from small forest settlements (20–30% non-natives) to big cities (50–70% non-natives). A similar reaction of the flora to land-use type in an urban area was found in Japan. In a city of the Chiba prefecture within the metropolitan area of Tokyo, Numata (1977) reported different percentages of nonnative naturalized plant species: 49% in residential areas, 32% in upland fields, 14% in rice fields, 13% on a riverside, and 4% on a forest floor. Non-native naturalized species can be assigned to two groups according to the time of introduction to a region: those introduced in prehistoric times, termed archeophytes (from Greek archaios = old), and those introduced in historic times, the neophytes (Greek neos = new). In central Europe, archeophytes are commonly considered to be species introduced before 1500 AD (Schroeder, 1969). These categories are ecologically significant because archeophytes are adapted to habitat types created early in history, including pastures, fields, and their edges; many of the neophytes, on the other hand, occur predominantly in ruderal, industrial, and urban habitats. This was shown in the above example (Kowarik, 1990): the proportion of neophytes is much higher in the flora of Berlin than in adjacent non-urban areas, but no difference was found for archeophytes. The high proportion of non-native species in cities is partly due to the fact that cities are centers of spread, because new species arrive at railway stations or ports and may be cultivated for the first time in (botanical) gardens. On the other hand, anthropogenic changes of growing conditions facilitate their spread; many nonnatives originate from warmer regions and depend on

405 Table 16.5 Characteristics of floristic city zones in Berlin 1 Zone 2 Total vegetation cover (%) Vascular plant species 3

1

2

3

4

32

55

75

95

380

424

415

357

Species rare in Germany 3

17

23

35

58

Non-native plant species (%)

49.8

46.9

43.4

28.5

Annuals (%)

33.6

30.0

33.4

18.9

1

Sukopp et al. (1979b). 1, densely built-up inner city; 2, partly built-up inner city; 3, inner suburbs; 4, outer suburbs. 3 Number of species in a square kilometer. 2

the higher temperatures in cities. Ailanthus altissima (Kowarik and B¨ocker, 1984) and Chenopodium botrys (Sukopp, 1971), for example, are both well established in European cities and originate in warmer areas. Similar urban conditions, as well as the effects of transport of organisms (the latter at least in the cases of plants and birds), lead to relatively uniform species composition in the centers of various cities in the European lowlands. Of 321 non-native ferns and flowering plants found in Braunschweig (Niedersachsen), more than 80% were also found in Berlin, Vienna, and London (Brandes, 1987). Despite considerable differences in climate, one can speak of a common stock of non-native naturalized species in western and central European cities. The main factor limiting vegetation in streets is the sealing of the surface with asphalt or paving stones. In other areas the vegetation is disturbed by human trampling or by vehicles, leading to soil compaction. Nevertheless, the flora of roadsides can be quite rich; in Berlin, in a study of 61 streets, a total of 375 flowering plant species were found growing naturally, a fourth of the total flora of the city (Langer, 1994). Most of these showed low frequency in the sampling areas. One group of 13 species was present in all areas. Most of these were annuals with some resistance to trampling. Most of the widespread plant communities belong to the phytosociological class Plantaginetea, a group which contains plant communities of heavily trodden swards. Succession in urban vegetation Several authors have described the general vegetation succession on anthropogenic sites from pioneer stages to woodland. Eliaˇs (1996) describes five stages for Slovakia (Table 16.6). Gilbert (1991) has given detailed

406

Herbert SUKOPP and Uwe STARFINGER

Table 16.7 Vegetation succession on four types of urban waste-land in Berlin from early (top) to late (bottom) stages 1 Railway ballast

Sand

Brick rubble

Humus rich piled-up topsoil

Conyza canadensis stage

Bromus–Corispermum community

Chenopodium botrys community

Chenopodium strictum

Arrhenatherum-pioneer stage

Berteroa incana community

Oenothera stage

Lactuca–Sisymbrium altissimum community

Dry grassland (Arrhenatherum or Festuca)

Festuca trachyphylla dry grassland

Poa–Tussilago community

Artemisia vulgaris community

Mixed deciduous woodland (e.g. Betula)

Robinia woodland

Chelidonium–Robinia community

Sambucus nigra community

Oak woodland

?

Maple woodland

Maple woodland

1

Modified from Sukopp (1990). Table 16.8 Species richness of spontaneous ferns and flowering plants in urban waste land in Berlin 1

Table 16.6 Successional stages on anthropogenic habitats 1 Stage

Vegetation type

Dominant species Duration (yr)

1st

pioneer

summer and overwintering annuals

1–2

Site type

ha 2

#2

Neo. 2 Ann. 2

Construction site

0.6

172

26

23

Brick rubble site

0.5

158

20

19

2nd

biennial

biennials

1–2

Derelict housing estate

15

325

36

23

3rd

forbs

perennials

5

Derelict railway land

17

332

36

20

4th

grassland

perennial grasses 5–10

Derelict railway goods station

63

417

40

26

5th

woodland

shrubs, trees

Derelict railway goods station

73

395

34

20

1

>10

Modified from Eliaˇs (1996).

1

Kowarik (1986). Abbreviations: ha, area in ha; #, total number of species; Neo., neophytes (% of total); Ann., annuals (% of total) 2

accounts of four similar stages, which he called the Oxford ragwort stage, the tall-herb stage, the grassland stage, and scrub woodland, and stressed the importance of the substratum and the role of chance in determining the succession. Ash (1991) estimated the time needed for woodland to develop on different urban soil types to be between 30 and 40 years for pulverized fuel ash on the one hand, and 100 years for acidic materials such as colliery shale on the other. Examples for urban sites with a relatively long history (several decades) of more or less undisturbed succession have been described in England (Gilbert, 1991) and Germany (e.g., Sukopp, 1990). A successional scheme for waste-lands in Berlin is shown in Table 16.7. An important feature of these inner-city waste-land sites is their richness in neophytes. A major constituent of woodland stages is the North American black locust (Robinia pseudoacacia). It differs from native tree species such as sand birch (Betula pendula) which it may replace, not only in the

fact that it has no specialized herbivores, but also in its symbiosis with nitrogen-fixing Rhizobium bacteria, and in its clonal growth. Due to the accumulation of nitrogen, succession of sites with Robinia is markedly different from sites that contain native species only (Kowarik, 1992). Even if succession can take place on urban wastelands, small-scale disturbances commonly exist in the form of unorganized land uses, such as trampling, “unregulated” garbage disposal, cutting of vegetation for pet food, etc. These influences lead to a small mosaic of successional stages with the possibility of high species numbers, especially in large railway areas (Table 16.8). Vegetation succession on landfills usually starts directly after the deposition ceases. In the first year, plant stands develop from diaspores that were present in the garbage (e.g., Cucumis spp., Cucurbita spp., Solanum spp., etc.). Further succession leads to stands

DISTURBANCE IN URBAN ECOSYSTEMS

407

Table 16.9 Species richness of different species groups in the succession on landfills in Berlin 1 Age (yr)

1

1

1

3

3

4

4

10

10

10

10

10

20

20

20

20

Plot No.

1

2

3

1

2

1

2

4

5

6

7

8

9

10

11

12

Vegetation cover (%)

60

75

40

40

75

80

95

100

95

50

100

95

95

100

100

100

No. of species

33

50

23

19

32

27

25

14

9

12

8

6

21

15

18

26

“Feral” crops and ornamentals

12

14

6

3

5

2

1

Annual and biennial ruderals

4

5

1

4

8

7

5

Field and garden weeds

9

12

7

4

2

6

Grassland species

1

4

3

1

8

1

7

5

1

2

1

1

1

Perennial weeds of settlements

3

6

3

6

6

6

8

7

7

3

3

4

4

1

2

5

4

Nitrophilous forest-edge species

1

4

1

Forest species

1

2

1

Mosses 1

1

1

2

1

1

5

3

3

2

2

3

3

3

2

8

4

9

13

1

1

2

2

1

1

2

2

Kunick and Sukopp (1975).

of short-lived ruderal plants, and eventually species of shrubs and forests invade (Table 16.9). These successions can be used to indicate specific disturbances (Pyˇsek and Hajek, 1996). The presence of ionized substances is indicated by salt-tolerating plants (e.g., Chenopodium ficifolium, C. glaucum, C. rubrum, and Puccinellia distans). Oil derivatives are detected by the decrease of “petroleophobe” species, (Arrhenatherum elatius, Artemisia vulgaris) and the increase of “petroleotolerants” (e.g., Cirsium arvense, Urtica dioica). Heavy-metal contamination can be indicated by damaged tissue, and the presence of specific types of necroses and chloroses. Animal communities Animals are affected by urban activities in a multitude of ways. Noise disturbs many birds more in the form of a singular noise event than as a continuous phenomenon. The fireworks of New Years’ Eve celebrated in German cities were shown to alter considerably the flight movements of crows (Corvus sp.) (J¨adicke and Storck, 1979). The activity of free-ranging cats in a Japanese city was lowest during periods when most people with their dogs were present in the streets (Obara, 1995). Gepp (1977) discussed circumstances causing losses of free-living animals in towns. Street, rail, and air traffic result in the killing of individual animals, especially mammals, birds, amphibians, and insects. Even the structure of urban buildings kills animals

(e.g., when birds fly against windows, or when insects become trapped in rooms). Some species or groups of species are thus eliminated from central areas in cities [e.g., toads (Bufo bufo) and hedgehogs (Erinaceus europaeus)]. On a transect from suburb to center in Graz, Austria, the number of inviduals of flying insects did not vary much, but diversity and total biomass decreased sharply (Gepp, 1977). Restoration of buildings and open space in a densely built-up residential area in Berlin led to a sharp decline in abundance of bird species (H.-G. Braun, pers. comm.). In the metropolitan area of Tokyo, the increase of disturbance during the urbanization process was closely related to the history of species extinctions. Several mammals and large insects disappeared from central parts much earlier than from suburban regions (Obara, 1995). Landfills can be rich in animals, especially species of open ground, such as thermophilous spiders, beetles, and locusts. Birds were studied by Steiof (1987) on six landfills in Berlin with a total area of 165 ha. Of the 44 species found breeding on the landfills, 12 were on the Berlin red data list of endangered species. In the course of succession toward closed stands, the rare species decreased, and ubiquitous animal species became more numerous. Animals of cities have adapted to humans to varying degrees. Some species occur regularly in cities so they can be called urban species [feral pigeon (Columba livia f. domestica), jackdaw (Corvus monedula)]. Some

408

habitat types and their animals disappear as cities develop (e.g., those of nutrient-poor soils). In other situations urban habitat features can take the place of certain natural features; bird species from rocks and cliffs can breed on ledges and walls of buildings, species from caverns can live in cellars or other rooms (Gilbert, 1991). Total numbers of animal species in urban habitats show similar trends to those of plants. At least in many groups of invertebrates, birds, and mammals (except large carnivores) species numbers in cities are higher than in adjacent non-urban areas. The highest species numbers are found in areas of the outer and inner suburbs; in city centers and new housing developments they are considerably lower (Gilbert, 1991; Klausnitzer, 1993). Many of the non-native species of animals and fungi which prefer urban habitats originate from warmer regions (Pisarski and Trojan, 1976; Erkkil¨a and Niemel¨a, 1986).

EVALUATION

The composition of urban ecosystems as opposed to those of rural or near-natural landscapes is the result of a multitude of disturbances which collectively form the human impact. There are different general approaches to the question of how to compare the intensities of disturbance and the changes urban disturbances cause in ecosystems. Chronic disturbances applied to natural communities can produce changes that are in some respect the reverse of succession, changes termed “retrogression” by Whittaker and Woodwell (1973). Retrogression may be quantified using indicators such as weighted averages of “decreasers, increasers, and invaders”, reduction of species diversity from that of undisturbed situations, or community coefficients for disturbed and undisturbed samples. In Europe there is a long tradition of attempts to classify the human impact on the vegetation using different degrees of naturalness. In a review of the literature, Kowarik (1990, 1991) pointed out that there are two principal ways to do this: the concepts of Westhoff (1949), Ellenberg (1963) and others express naturalness as a distance from a pristine, undisturbed ecological condition which existed before human impact began. Jalas (1955), T¨uxen (1956), and Sukopp (1972), on the other hand, tried to evaluate the human

Herbert SUKOPP and Uwe STARFINGER Table 16.10 The hemeroby system (examples ordered with increasing degree of hemeroby) 1 Human influence

Ecosystems

Not present

high mountains

Emissions, minor biomass removal

little-disturbed forests, growing bogs

Tree-cutting, mowing of grass

more intensively managed forests, dry grassland

Plowing, draining, fertilizing

forest plantations of alien species, intensively managed grassland

Deep plowing, intensive fertilization, biocides

fields, gardens, vineyards

Total destruction of vegetation and soil

new landfills, partially paved urban areas

Poisoned or completely sealed surfaces

no vegetation of vascular plants

1

Sukopp (1978); Kowarik (1990).

impact on the vegetation by relating site conditions and vegetation to a potential (or future) undisturbed situation. The difference between the two reference points which Kowarik terms “nature I” and “nature II” becomes apparent in examples where strongly altered sites are left relatively undisturbed for a period of time as in the example of urban waste land described above (p. 406, Table 16.7). Jalas’ concept of hemeroby (Greek: hemeros = tame; bios = life), refined by Sukopp (1972) and Kowarik (1988), is widely applied in Central Europe. It classifies the overall effects of human impact on vegetation in 7 grades (Table 16.10). Kowarik (1988) produced hemeroby spectra for the vascular plant species of Berlin by evaluating their presence in plant communities at different levels of hemeroby. He found the highest overall species richness in situations of intermediate hemeroby, confirming Connell’s “intermediate disturbance hypothesis”. Nonnative species, however, react more positively to human impact: the highest number of them was found in more strongly influenced vegetation of a higher hemeroby level (Fig. 16.3). The strong preference of certain species for certain levels of hemeroby allows their use as indicators of particular disturbance intensities (Kowarik, 1991). CONCLUSIONS

Climate, soil, air, and water in cities suffer from human

DISTURBANCE IN URBAN ECOSYSTEMS

409

Fig. 16.3. Species richness in relation to human impact (increasing from hemeroby level 1 to 9) in the flora of the city of Berlin (from Kowarik, 1991, modified. Levels are not the same as in Table 16.10). Fat solid line, all species; solid line, natives; dotted line, non-natives; dashed line, neophytes; dash-dotted line, archeophytes.

impacts that are usually much stronger than in rural areas. In spite of this multitude of disturbances, cities are quite rich in habitats as well as in plant and animal species. Typically, disturbance of urban ecosystems leads to a decrease in the number of species native to the region and an increase of introduced non-native species. This process is at least partly reversed in the course of succession. Disturbance and the recovery from it during succession are key factors determining the habitat mosaic in urban ecosystems. As an ever-increasing percentage of the world’s population lives in cities, urban biota are under increasing pressure. At the same time, the conservation of urban biota is of great importance, because cities are the primary location where people have contact with nature (Starfinger and Sukopp, 1994). In many parts of the world, knowledge about urban ecosystems and their disturbance regimes is still insufficient.

ACKNOWLEDGMENTS

Lawrence Walker and two anonymous reviewers have commented on and greatly improved the manuscript. Bettina Matzdorf helped with the figures.

REFERENCES Aey, W., 1990. Historical approaches to urban ecology. Methods and first results from a case study (L¨ubeck, West-Germany). In: H. Sukopp, I. Kowarik and S. Hejny (Editors), Urban Ecology. SPB, Academic Publishing, Den Haag, pp. 113–129. Ash, H.J., 1991. Soils and vegetation in urban areas. In: P. Bullock and P.J. Gregory (Editors), Soils in the Urban Environment. Blackwell, Oxford, pp. 153–170. Auhagen, A., 1991. Rote Listen der gef¨ahrdeten Pflanzen und Tiere in Berlin 1991 – Zusammenfassung und Empfehlungen. Landschaftsentwicklung und Umweltforschung, S 6: 5–11. Auhagen, A. and Sukopp, H., 1980. Das Streusalz. Umweltschutzforum Berlin, 47/48: 23–42.

410 Balke, K.D., 1974. Der thermische Einfluß besiedelter Gebiete auf das Grundwasser, dargestellt am Beispiel der Stadt K¨oln. gwf-Wasser/ Grundwasser, 115: 117–124. Blume, H.P., 1993. B¨oden. In: H. Sukopp and R. Wittig (Editors), Stadt¨okologie. G. Fischer, Stuttgart, pp. 154–171. Blume, H.P., Chinnow, D., Hartnick-K¨ummel, C., Hellriegel, T. and Stoll, E., 1977. Bau- und nutzungsbedingte Ver¨anderungen an ¨ Straßenrand-Okosystemen. Z. TU Berlin, 9: 278–322. Blume, H.P., Bornkamm, R. and Sukopp, H., 1979. Vegetationssch¨aden und Bodenver¨anderungen in der Umgebung einer M¨ulldeponie. Z. Kulturtech. Flurbereinigung, 20: 65–79. B¨ocker, R., 1985. Bodenversiegelung – Verlust vegetationsbedeckter Fl¨achen in Ballungsr¨aumen (Surface sealing – loss of vegetation in metropolitan areas. Example Berlin (West) German w. English summary.) Landschaft Stadt, 17: 57–61. Bradshaw, A.D. and Chadwick, M.J., 1980. The Restoration of Land. Blackwell, Oxford, 317 pp. Brande, A., 1986. Mittelalterliche Siedlungsvorg¨ange in Berliner Pollendiagrammen. Cour. Forschungsinst. Senckenberg, 86: 409–414. Brandes, D., 1987. Verzeichnis der im Stadtgebiet von Braunschweig wildwachsenden und verwilderten Gef¨aßpflanzen. Universit¨atsbibliothek der Technischen Universit¨at Braunschweig, 44 pp. den Hartog, C., Kvˆet, J. and Sukopp, H., 1989. Reed. A common species in decline. Aquat. Bot., 35: 1–4. Drayton, B. and Primack, R.B., 1996. Plant species lost in an isolated conservation area in Metropolitan Boston from 1894 to 1993. Conservation Biol., 10: 30–39. Duvigneaud, P. and Denayer-de Smet, S., 1977. L’ecosyst`eme urbain bruxellois. In: P. Duvigneaud and P. Kestemont (Editors), Productivit´e Biologique en Belgique. SCOPE. Travaux de la Section belge du Programme Biologique International Gembloux, pp. 581–599. Eliaˇs, P., 1996. Vegetation dynamics of anthropogenic habitats in ¨ settlements. Verh. Ges. Okol., 25: 219–224. Ellenberg, H., 1963. Vegetation Mitteleuropas mit den Alpen. Ulmer, Stuttgart. Erkkil¨a, R. and Niemel¨a, T., 1986. Polypores in the parks and forests of the city of Helsinki. Karstenia, 26: 1–40. Falinski, J.B., 1971. Synanthropisation of plant cover. II. Synanthropic flora and vegetation of towns connected with their natural conditions, history and function. Mater. Zakłado Fitosocjologie Stosowanej U.W., 27: 1–317. Gepp, J., 1977. Technogene und strukturbedingte Dezimierungsfak¨ toren der Stadttierwelt – ein Uberblick. In: J. Gepp (Editor), Stadt¨okologie. Verlag f¨ur die Technische Universit¨at, Graz, pp. 99– 127. Gilbert, O.L., 1991. The Ecology of Urban Habitats. Chapman and Hall, London, 369 pp. Grenzius, R., 1984. Starke Versauerung der Waldb¨oden Berlins. Forstwissenschaftliches Centralblatt, 103: 131–139. Grime, J.P., 1979. Plant Strategies and Vegetation Processes. John Wiley, New York, 222 pp. Harris, J.A., 1991. The biology of soils in urban areas. In: P. Bullock and P.J. Gregory (Editors), Soils in the Urban Environment. Blackwell, Oxford, pp. 139–152. Horbert, M., Kirchgeorg, A. and von St¨ulpnagel, A., 1983. Ergebnisse stadtklimatischer Untersuchungen als Beitrag zur Freiraumplanung. Umweltbundesamt, Berlin, 187 pp.

Herbert SUKOPP and Uwe STARFINGER H¨otzl, H. and Makurat, A., 1981. Ver¨anderungen der Grundwassertemperaturen unter dicht bebauten Fl¨achen am Beispiel der Stadt Karlsruhe. Z. Dtsch. Geol. Ges., 132: 767–777. Jackowiak, B., 1982. Occurrence of Puccinellia distans in the city of Poznan. Badania fiziograficzne nad Polska Zachodnia, 33: 129–142. Jackowiak, B., 1996. Chorological-ecological model of the spread of Puccinellia distans (Poaceae) in central Europe. Fragm. Florist. Geobot., 41: 551–561. J¨adicke, B. and Storck, H.J., 1979. Großr¨aumige Umlenkung der Schlafplatzfl¨uge von Kr¨ahen durch Silvesterl¨arm. J. Ornithol., 120: 326–327. J¨ager, E., 1988. M¨oglichkeiten der Prognose synanthroper Pflanzenausbreitungen. Flora, 180: 101–131. Jalas, J., 1955. Hemerobe und hemerochore Pflanzenarten: ein terminologischer Reformversuch. Acta Soc. Fauna Flora Fennica, 72: 1–15. ¨ Klausnitzer, B., 1993. Okologie der Großstadtfauna. G. Fischer, Stuttgart, 454 pp. Klemm, G., 1975. Pflanzengeographischer Vergleich neuerer brandenburgischer Lokalfloren. Gleditschia, 3: 35–52 (cited by Kowarik, 1990). Klotz, S., 1990. Species/area and species/inhabitants relations in European cities. In: H. Sukopp, I. Kowarik and S. Hejny (Editors), Urban Ecology. SPB, Academic Publishing, Den Haag, pp. 99– 103. Kowarik, I., 1986. Vegetationsentwicklung auf innerst¨adtischen Brachfl¨achen – Beispiele aus Berlin (West). Tuexenia, 6: 75–98. Kowarik, I., 1988. Zum menschlichen Einfluß auf Flora und Vegetation. Landschaftsentwicklung und Umweltforschung, 56: 1–280. Kowarik, I., 1990. Some responses of flora and vegetation to urbanization in Central Europe. In: H. Sukopp, I. Kowarik and S. Hejny (Editors), Urban Ecology. SPB, Academic Publishing, Den Haag, pp. 45–74. Kowarik, I., 1991. The adaptation of urban flora to man-made perturbations. In: O. Ravera (Editor), Terrestrial and Aquatic Ecosystems: Perturbation and Recovery. Ellis Horwood, London, pp. 176–184. Kowarik, I., 1992. Einf¨uhrung und Ausbreitung nichteinheimischer Geh¨olzarten in Berlin und Brandenburg und ihre Folgen f¨ur Flora und Vegetation. Verh. Bot. Ver. Berlin Brandenburg Beiheft 3: 1–188. Kowarik, I. and B¨ocker, R., 1984. Zur Verbeitung, Vergesellschaftung und Einb¨urgerung des G¨otterbaumes (Ailanthus altissima (Mill.) Swingle) in Mitteleuropa. Tuexenia, 4: 9–29. Kratzer, A., 1937. Das Stadtklima. Vieweg, Braunschweig, 184 pp. Kunick, W., 1974. Ver¨anderungen von Flora und Vegetation einer Großstadt, dargestellt am Beispiel von Berlin (West). Ph.D. Dissertation, Technische Universit¨at Berlin, 472 pp. Kunick, W., 1981. Comparison of the flora of some cities of the Central European Lowlands. In: R. Bornkamm, J.A. Lee and M.R.D. Seaward (Editors), Urban Ecology, 2nd European Ecological Symposium, Berlin, 1980. Blackwell, Oxford, pp. 13– 22. Kunick, W. and Sukopp, H., 1975. Vegetationsentwicklung auf M¨ulldeponien Berlins. Berl. Naturschutzbl¨atter, 56: 141–145. Kuttler, W., 1993. Stadtklima. In: H. Sukopp and R. Wittig (Editors), Stadt¨okologie. G. Fischer, Stuttgart, pp. 113–153.

DISTURBANCE IN URBAN ECOSYSTEMS Landsberg, E., 1981. The Urban Climate, International Geophysics Series, Vol. 28. Academic Press, New York, 275 pp. Langer, A., 1994. Flora und Vegetation st¨adtischer Straßen am Beispiel Berlins. Landschaftsentwicklung und Umweltforschung, S 10: 1–199. Leuchs, W. and R¨omermann, H., 1991. Auswirkungen stadt¨okologischer Gestaltungsmaßnahmen auf die Grundwassersituation. In: H. Schuhmacher and B. Thiesmeyer (Editors), Urbane Gew¨asser. Westarp-Wissenschaften, Essen, pp. 427–445. Londo, G., 1976. Over de Nederlandse lijst van hydro-, freato- en afreatofyten. Gorteria, 8: 25–29. McDonnell, M.J. and Pickett, S.T.A., 1990. Ecosystem structure and function along urban-rural gradients: an unexploited opportunity for ecology. Ecology, 71: 1232–1237. Numata, M., 1977. The impact of urbanization on vegetation in Japan. In: A. Miyawaki and R. T¨uxen (Editors), Vegetation Science and Environmental Protection. Maruzen, Tokyo, pp. 161–171. Obara, H., 1995. Animals and man in the process of urbanization. In: H. Sukopp, M. Numata and A. Huber (Editors), Urban Ecology as the Basis of Urban Planning. SPB, Academic Publishing, Den Haag, pp. 191–201. Oke, T.R., 1987. Boundary Layer Climates. Methuen, London, 435 pp. Pisarski, B. and Trojan, P., 1976. Zoocenozy obszarow zurbanizowanych (Animal associations in urbanized areas). Wiadomosci Ecologiczne, 22: 338–344. Pyˇsek, A. and Hajek, M., 1996. Die Ruderalvegetation der Ablagerungspl¨atze und ihre praktische Ausnutzung zur ¨ Kontaminationsentdeckung. Verh. Ges. Okol., 25: 215–217. Pyˇsek, P., 1993. Factors affecting the diversity of flora and vegetation in central European settlements. Vegetatio, 106: 89–100. Rebele, F., 1994. Urban ecology and special features of urban ecosystems. Global Ecol. Biogeogr. Lett., 4: 173–187. Rowntree, R.A. (Editor), 1984. Ecology of the urban forest. Part I: structure and composition. Urban Ecol., 8: 1–178. Rowntree, R.A. (Editor), 1986. Ecology of the urban forest. Part II: function. Urban Ecol., 9: 227–437. Ruge, U., 1978. Physiologische Sch¨aden durch Umweltfaktoren. In: F.H. Meyer (Editor), B¨aume in der Stadt. Ulmer, Stuttgart, 134– 198. Schlenther, L., Marschner, B., Hoffmann, C. and Renger, M., 1996. Ursachen mangelnder Anwuchserfolge bei der Aufforstung der Rieselfelder in Berlin-Buch – bodenkundliche Aspekte. Verh. Ges. ¨ Okol., 25: 349–359. Schroeder, F.-G., 1969. Zur Klassifizierung der Anthropochoren. Vegetatio, 16: 225–238. Schuhmacher, H., 1991. Limnologische Vorgaben und Bewertungskriterien zur o¨ kologischen Verbesserung urbaner Fließgew¨asser. In: H. Schuhmacher and B. Thiesmeyer (Editors), Urbane Gew¨asser. Westarp-Wissenschaften, Essen, pp. 16–36. Sharpe, D.M., Stearns, F., Leitner, L.A. and Dorney, J.R., 1986. Fate of natural vegetation during urban development of rural landscapes in southeastern Wisconsin. Urban Ecol., 9: 267–287. Sousa, W.P., 1984. The role of disturbance in natural communities. Ann. Rev. Ecol. Syst., 15: 353–391. Starfinger, U. and Sukopp, H., 1994. Assessment of urban biotopes for nature conservation. In: E.A. Cook and H.N. van Lier (Editors), Landscape Planning and Ecological Networks. Elsevier, Amsterdam, pp. 89–115.

411 Steiof, K., 1987. Brutv¨ogel und Deponien-Rekultivierung – ein Beitrag zur Landschaftsbewertung und -planung am Beispiel Berlin. Landschaftsentwicklung und Umweltforschung, 47: 1–107. ¨ Sukopp, H., 1971. Beitr¨age zur Okologie von Chenopodium botrys L. 1. Verbreitung und Vergesellschaftung. Verh. Bot. Ver. Provinz Brandenburg, 108: 3–25. Sukopp, H., 1972. Wandel von Flora und Vegetation in Mitteleuropa unter dem Einfluß des Menschen. Ber. Landwirtsch., 50: 112–130. Sukopp, H., 1978. An approach to ecosystem degradation: opening remarks by session chairman. In: M.W. Holdgate and M.J. Woodman (Editors), The Breakdown and Restoration of Ecosystems. Plenum, New York, pp. 123–127. Sukopp, H., 1981. Grundwasserabsenkungen – Ursachen und Auswirkungen auf Natur und Landschaft Berlins. Wasser – Berlin, Bd. 1. Die technisch-wissenschaftlichen Vortr¨age auf dem Kongress Wasser 1981, Berlin, pp. 239–272. Sukopp, H. (Editor), 1990. Stadt¨okologie. Das Beispiel Berlin. Dietrich Reimer, Berlin, 455 pp. Sukopp, H. and Trepl, L., 1993. Stadt¨okologie und Naturschutz in der Großstadt. Biol. Schule, 42: 187–197. Sukopp, H. and Wittig, R. (Editors), 1993. Stadt¨okologie. G. Fischer, Stuttgart, 402 pp. Sukopp, H., Anders, K., Bierbach, H., Brande, A., Blume, H.P., Elvers, H., Horbert, M., Horn, R., Kirchgeorg, A., L¨uhrte, A. v., Riecke, F., Stratil, H., Trepl, L. and Weigmann, G., 1979a. ¨ Okologisches Gutachten u¨ ber die Auswirkungen von Bau und Betrieb der BAB Berlin (West) auf den Großen Tiergarten. Senator f¨ur Bau- und Wonhungswesen, Berlin, 105 pp. Sukopp, H., Blume, H.P. and Kunick, W., 1979b. The soil, flora, and vegetation of Berlin’s waste lands. In: I.C. Laurie (Editor), Nature in Cities. John Wiley, Chichester, pp. 115–132. Sukopp, H., Hejny, S. and Kowarik, I. (Editors), 1990. Urban Ecology. SPB, Academic Publishing, Den Haag, 282 pp. Sukopp, H., Numata, M. and Huber, A. (Editors), 1995. Urban Ecology as the Basis of Urban Planning. SPB, Academic Publishing, Den Haag, 218 pp. Thornton, I., 1991. Metal contamination of soils in urban areas. In: P. Bullock and P.J. Gregory (Editors), Soils in the Urban Environment. Blackwell, Oxford, pp. 47–75. ¨ Trepl, L., 1992. Stadt-Natur – Okologie, Hermeneutik und Politik. In: Bayerische Akademie der Wissenschaften (Editor), Stadt¨okologie, ¨ Rundgespr¨ache der Kommission f¨ur Okologie, Band 3. Pfeil, M¨unchen, pp. 53–58. T¨uxen, R., 1956. Die heutige potentielle nat¨urliche Vegetation als Gegenstand der Vegetationskartierung. Angew. Pflanzensoziol., 13: 5–42. Ullmann, I. and Heindl, B., 1989. Geographical and ecological differentiation of roadside vegetation in temperate Europa. Bot. Acta, 4: 261–269. von St¨ulpnagel, A., Horbert, M. and Sukopp, H., 1990. The importance of vegetation for the urban climate. In: H. Sukopp, S. Hejny and I. Kowarik (Editors), Urban Ecology. SPB, Academic Publishing, Den Haag, pp. 175–194. Westhoff, V., 1949. Schaakspiel met de natuur. Natuur Landschap, 3. White, C.S. and McDonnell, M.J., 1988. Nitrogen cycling processes and soil characteristics in an urban versus rural forest. Biogeochemistry, 5: 243–262.

412 White, P.S. and Pickett, S.T.A., 1985. Natural disturbance and patch dynamics: an introduction. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, New York, pp. 3–13. Whittaker, R.H. and Woodwell, G.M., 1973. Retrogression and Coenocline Distance. Handbook of Vegetation Science 5. The Hague, pp. 53–73.

Herbert SUKOPP and Uwe STARFINGER Wittig, R. and Sukopp, H., 1993. Was ist Stadt¨okologie? In: H. Sukopp and R. Wittig (Editors), Stadt¨okologie. G. Fischer, Stuttgart, pp. 1–9. Wu, L. and Antonovics, J., 1976. Experimental ecological genetics in Plantago. II. Lead tolerance in Plantago lanceolata and Cynodon dactylon from a roadside. Ecology, 57: 205–208.

Chapter 17

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

INTRODUCTION

The expansion of the human population and its associated agriculture and industry over the last several centuries has brought with it major alterations to landscapes, disturbance regimes, and dispersal of species. While it has become dogma that disturbance of natural systems promotes invasion by non-indigenous species, the causes of invasion and the links between disturbance and biological invasions have yet to be adequately understood. Perturbations to natural or semi-natural systems have in the past been broadly lumped under the heading of disturbance, and several review papers have discussed the relationship between disturbance and biological invasion (Hobbs, 1989; Rejm´anek, 1989; Hobbs and Huenneke, 1992). While each of these reviews provides valuable information regarding the actual or potential role of habitat disruption in promoting invasion, we believe that many important questions about the relationship between disturbance and invasion remain unanswered. For example, although it is clear that some catastrophic disturbances (e.g., logging, clearing for agriculture) are followed by rapid expansion of populations of nonindigenous species, it is not known if invaders become a self-sustaining and long-term part of the “recovering” ecosystem or whether they are successional to native species. Also, the site-specific nature of “natural” disturbances in promoting invasion is unknown, as is the relative importance of disturbance versus propagule availability in promoting species change within communities. All ecosystems experience disturbance on some spatial and temporal scale, yet disturbance is difficult to quantify. Indeed, few studies measure the impact of the disturbing agent in terms of area, spatial

structure, intensity, and temporal distribution of the impacts. We believe that some of the reasons for the lack of clarity surrounding the linkage between disturbance and invasions include non-rigorous use of terminology surrounding the word “disturbance”, difficulty in quantifying or characterizing disturbance regimes in most sites, and a failure to distinguish between anthropogenic and “natural” disturbances and between irregular catastrophic disturbances and recurrent smaller-scale disturbances. Finally, current disturbance patterns may not reflect conditions that promoted the initial invasions, and often information on pattern and process during the invasion process is lacking. There are numerous anecdotal studies where causative mechanisms are invoked but are largely obscure, and where multiple causation is likely. Disturbance is often not independent of associated impacts such as proximity to human habitation or density of nearby populations of invaders, yet these colonization-related processes are generally ignored. Also, many invading species alter the disturbance regime as they become well established, and by so doing can favor further invasion of the site (e.g., D’Antonio and Vitousek, 1992). If this happens rapidly, then the role of disturbance in promoting the initial invasion may be difficult to separate from the effect of the invader on the disturbance regime. In this chapter we review accepted definitions of disturbance, selecting a somewhat conservative one for the purposes of providing as concise and unambiguous a review as possible. We focus on major disturbance types for which we found the best information, and review studies assessing the conditions under which these disturbances promote or inhibit invasion. For terrestrial invasions, we categorize disturbances by

413

414

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

spatial scale and by type (obviously anthropogenic versus more or less natural), confining our review mainly to plant invasions. We also discuss a subset of aquatic plant invasions that occur at the interface between fully terrestrial and fully aquatic systems. These include invasions into riparian corridors, stream margins, and wetlands. We then reverse the issue and examine how invading species themselves alter the disturbance regime. Finally, we assess how altered disturbance regimes caused by introduced species may promote or inhibit further invasions. We use the terms non-native, introduced, and non-indigenous species interchangeably throughout. What is an invasion A daunting problem associated with this review is the lack of a standard scale for defining when an invasion has occurred. Introduced species can be present on a site but may not maintain themselves over several generations or may do so only at a very low density. Are these successful invasions? Most studies use presence/absence data or subjective assessments to conclude whether or not an invasion has occurred. We found numerous examples where introduced species were reported from relatively undisturbed sites, and where we could not judge whether these invasions represented self-replacing populations or were similar in their magnitude or persistence to populations in nearby disturbed habitat. In these cases we considered these invasions to be successful unless otherwise indicated by the authors. Defining the disturbance regime Perhaps the most commonly used definition of disturbance is that of White and Pickett (1985), who stated that a disturbance is “any relatively discrete event in time that disrupts ecosystem, community, or population structure and changes resources, substrate availability, or the physical environment”. However, we believe that this definition is too vague and includes a wide range of changes to populations and ecosystems that might best be described as perturbations or modifications rather than disturbances. For example, their definition can be interpreted to include an increase in the availability of resources without the direct killing of individuals. This could include atmospheric deposition of nitrogen, which several investigators have termed a “disturbance” to natural systems (e.g., Hobbs and Huenneke,

1992). We do not deny that such modifications may promote changes in species composition including invasions; however, we do not consider an increase in resources alone to be a disturbance without a discrete killing event or an event that significantly reduces standing biomass. Accordingly, we prefer to use the definition provided by Sousa (1984): “A disturbance is a discrete, punctuated killing, displacement, or damaging of one or more individuals (or colonies) that directly or indirectly creates an opportunity for new individuals (or colonies) to become established” or for individuals adjacent to the disturbance to garner further resources (our addition). To Sousa’s definition we also add chronic grazing by livestock or wildlife, which may be viewed as a continuous series of damaging events. Our definition excludes: (1) gradual changes in groundwater due to humanrelated alterations of hydrology; (2) gradual increases in aquatic pollutants that might stress individuals and gradually lead to the decline of populations through decreased reproduction; (3) atmospheric pollutants that may weaken plants and make them more susceptible to other stresses or agents of disturbance; and (4) nitrogen deposition, which might change plant performance and the strength of competitive interactions over several years without discrete mortality. We also do not consider the phenomenon of invasion by itself to be a disturbance, although specific invaders may cause death of native organisms and ultimately contribute to altered disturbance regimes. Every site can be characterized as having a natural disturbance regime, where killing events occur with given intensities, sizes, and average return intervals (see Sousa, 1984; White and Pickett, 1985). Characteristics of a disturbance regime are determined by physical site characteristics (e.g., topography, soil structure, slope and soil stability, and altitude) and factors extrinsic to the system, such as the local or regional weather (e.g., frequency of lightning, level of fuel moisture). In addition, the community of species present on a site may influence any aspect of that disturbance regime and the responsiveness of the system to the disturbance. For example, inherent flammability characteristics of individual stems and leaves can influence fire intensity and fire return interval. Size or stature of trees in a forest can influence susceptibility to windthrow. Presumably, the disturbance regime of a particular area changes as the climate changes, and has therefore been in continuous flux through geological time. However, we will confine our discussion to

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS

disturbances or disturbance regimes that are presumed to be representative of the recent history of an area. Anthropogenic disturbances frequently are very different from anything experienced by the natural community occupying a site. Here we will examine both how anthropogenic disturbances influence invasion, and whether or not elements of a “natural” disturbance regime promote or even suppress invasion. Suppression of a natural disturbance regime has been an important part of land management in the 20th century. We include cases where the natural disturbance regime has been altered, and examine how these changes influence invasion.

DISTURBANCE AND INVASION: THE EVIDENCE

For the purposes of this review we will divide disturbances into overt anthropogenic disturbances and disturbances that are natural or near-natural. The former group includes disturbance corridors (roads, trails, pipelines, power-line rights of way), livestock grazing, and clearing for agriculture, timber harvesting or other directed human purposes. These are largely disturbances of a quality or magnitude that do not resemble natural disturbances which might have occurred in that area. Near-natural and natural disturbances include floods, small-scale soil disturbance associated with animal activity, insect outbreaks, grazing by native animals, short-term drought, fire, hurricanes and other storms, and floods. Because the origin of fires is often not reported and previous fire frequencies are often unknown, we chose to treat all fires not clearly associated with logging or other overt forms of land clearing as natural or near-natural disturbances. Terrestrial ecosystems: anthropogenic disturbances Disturbance corridors Numerous articles document the occurrence of introduced plant species along roads and disturbance corridors (Table 17.1). Here we also include a few studies in which the nature of the disturbance at a forest edge was unspecified and could have been catastrophic (agriculture), but where the study focused on the disturbed edge as a source for weed invasion (Brothers and Spingarn, 1992; Brandt and Rickard, 1994). The construction of disturbance corridors creates a new environment for colonization; in addition to elimination

415

of the pre-existing community, construction of a corridor often disrupts the soil profile and structure, eliminates the seed bank, and alters site hydrology and chemistry. The end result is that roadsides and trail edges are novel habitats, and tend to be populated by “weedy” introduced species. However, at least a portion of this occurrence is the result of increased frequency of contact with vehicles or humans, which might inadvertently be spreading propagules along corridors (Chaloupka and Domm, 1986; Lonsdale and Lane, 1994). The importance of propagule arrival through human traffic is rarely compared to or differentiated from the role of disturbance in promoting invasions. It is not surprising to find that introduced species dominate road verges and other disturbance corridors (Table 17.1). What is more interesting is the possibility that these corridors act as entry points for nonindigenous species into natural communities. We found a total of 11 studies, involving 14 community types, that provided data on the occurrence of introduced species along roadsides, and their subsequent penetration into the surrounding relatively undisturbed vegetation. In seven cases, introduced species were largely concentrated near disturbed edges and rarely penetrated into surrounding vegetation (Forcella and Harvey, 1983; Hobbs and Atkins, 1991; Brothers and Spingarn, 1992; Wein et al., 1992; Parker et al., 1993; Brandt and Rickard, 1994). In the other seven cases, corridors were acting to promote invasion of adjacent undisturbed habitat [Tyser and Key, 1988; Burgess et al., 1991; MacDonald et al., 1991; Luken and Goessling, 1995; Zink et al., 1995 (3 cases)]. In the former group of studies, no penetration of undisturbed vegetation was observed even where natural disturbances occurred in the surrounding vegetation. For example, Parker et al. (1993) found that two species of Eurasian weeds were abundant along an abandoned roadside in Wisconsin (U.S.A.) but did not penetrate the intact prairie nearby, in spite of the occurrence there of rodent-caused soil disturbances. Likewise, Brothers and Spingarn (1992) found that introduced species common on disturbed edges of intact forests in Indiana (U.S.A.) did not penetrate the forest remnants, in spite of the occurrence of tree-fall gaps in those habitats. By contrast, Luken and Goessling (1995) in Kentucky (U.S.A.) and MacDonald et al. (1991) in the island R´eunion found that shade-tolerant introduced species along disturbed forest edges could penetrate the understory of adjacent undisturbed forests. Zink et al. (1995) in southern California (U.S.A.) found

416

Table 17.1 Studies examining the occurrence of introduced species on disturbance corridors (roads, trails, pipelines, etc.) Location 1

Invader 2

Findings

Study type 3

Reference

California (coastal habitats)

Eur. grasses, forbs

introduced species most abundant on roadsides

Ane.

Knops et al. (1995)

California (coastal sage scrub, oak woodland, native grassland, chaparral)

Eur. annuals

Eur. annuals dominate pipeline coridor, spread into 3/4 adjacent communities

Obs.

Zink et al. (1995)

Washington (Great Basin Desert)

Salsola iberica

dense on roadsides, does poorly in adjacent desert

Obs.

Brandt and Rickard (1994)

Montana (short-grass prairie)

Centaurea maculosa

dense roadside populations support adjacent grassland populations

Obs.

Tyser and Key (1988)

Montana (alpine/subalpine)

introduced weeds

abundant on roadsides, no penetration of adjacent sites

Obs.

Forcella and Harvey (1983)

Colorado (desert)

annual grasses, forbs

increase with vehicle “tracking”

Obs.

Shaw and Diersing (1990)

New Mexico (montane)

Dipsacus sp.

spreading along roads

Obs.

Huenneke and Thomson (1995)

Arizona (Sonoran Desert)

Eur. annuals

abundant on all disturbance corridors, spreads into adjacent desert

Obs.

Burgess et al. (1991)

Wisconsin (native prairie)

Eur. forbs

confined to roads, not spreading into prairie

Obs.

Parker et al. (1993)

Kentucky (temperate deciduous forest)

Lonicera maackii

grows best along disturbed edges, can penetrate forest

Obs., Exp.

Luken and Goessling (1995)

Indiana (temperate deciduous forest)

many species

abundant along edges, do not penetrate undisturbed forest

Obs.

Brothers and Spingarn (1992)

N.W. Territories: Canada (boreal forest)

Eur. ruderals

Bromus inermis slowly spreading from roads, other species not spreading

Obs.

Wein et al. (1992)

W. Australia (dry forest, shrubland)

Eur. annuals

occur on roads, do not penetrate adjacent vegetation

Obs., Exp.

Hobbs and Atkins (1991); Hester and Hobbs (1992)

Australia, Kakadu (monsoonal forest)

many species

transported by buses, some abundant along roads

Obs.

Lonsdale and Lane (1994)

South Africa (shrubland)

many species

very abundant along roads

Ane., Obs.

Brown and Gubb (1986)

La R´eunion (tropical forest)

many species

abundant along roads, spreading into adjacent forest

Ane.

MacDonald et al. (1991)

North America

1 2 3

Ecosystem type in parentheses following location. Where species names or life forms were not given, invaders are presented in general terms; Eur., European origin. Abbreviations: Ane., anecdotal (no data presented); Obs., observational (with data); Exp., experimental manipulation of vegetation conducted.

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

Other continents

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS

that three of four undisturbed, previously uninvaded communities (native grassland, oak woodland, and coastal sage scrub) bordering a pipeline corridor were being invaded by the introduced species abundant on the corridor. They concluded that the corridor was responsible for giving these species a foothold from which to invade the adjacent communities. The one uninvaded community was dense chaparral. Although we did not attempt an exhaustive review of the literature on the use of corridors by introduced animals, the potential for and impacts of such use could be significant. May and Norton (1996), for example, suggested that roads provide entry corridors for feral cats (Felis catus), dingo (Canis familiaris dingo) and foxes (Vulpes vulpes) into native Australian rainforest, where they are a threat to native animal populations. Livestock grazing Grazing by introduced ungulates is a common component of ecosystems throughout the world. In spite of its prevalence, and documented cases of rangeland degradation and changes in species composition, information on the precise causes of species change is limited. Milchunas et al. (1988) concluded that livestock grazing is not always a disturbance in rangeland ecosystems; this depends on the evolutionary history of the species present on a site and the frequency and intensity of grazing both by wildlife and by livestock. They argued that grazing by native or introduced ungulates may not be a disturbance if the species in a system have evolved with grazing, and can fully compensate for immediate removal of plant material. However, we believe that the removal of biomass results in a change in resources for the remaining or new individuals in most instances, even if this release or change in the resources is rarely measured. Further, we feel that livestock grazing is probably always at a severity or frequency different from those of wildlife grazing and thus represents an alteration in the natural disturbance regime. There have been dramatic changes in rangeland composition in the arid and semi-arid western United States and Australia over the past 150 years (Heady, 1977; Walker et al., 1981; Vavra et al., 1994). In many cases, rangelands have become dominated by introduced species (Heady, 1977; Sparks et al., 1990; Whisenant and Wagstaff, 1991; Miller et al., 1994). In spite of the great spatial extent of this conversion, the causes of most of the large-scale changes remain obscure because they occurred prior to quantitative measurements of the

417

vegetation. For example, in California, native perennial grasses were replaced almost completely by introduced Mediterranean species in the 1800s during a period of drought and over-grazing (Heady, 1977). Yet it is not clear what the relative roles of competition, drought, fire, and grazing were in this conversion (Bartolome and Gemmill, 1981). In the intermountain west in North America, large areas of sagebrush (Artemisia) steppe have been converted from native shrub/perennial grass communities to low-diversity stands of the Eurasian annual grass Bromus tectorum (cheatgrass) during a long period of intense livestock grazing. Recent studies of invasion by B. tectorum into remnants of the Great Basin never grazed by livestock suggest that significant invasion can occur without grazing, and that it is fire rather than grazing that has promoted the conversion to nearly monospecific stands of cheatgrass (Whisenant, 1990a; Svejcar and Tausch, 1991). Several investigators admit that it is difficult to dissect out the roles of fire and grazing in causing changes in arid-zone rangelands (e.g., Sparks et al., 1990). Overall we found only 14 case studies and two reviews that explicitly correlate recent grazing disturbance with biological invasions (Table 17.2). Ten of these provided evidence suggesting at least some increase in the occurrence of introduced species with grazing. Three studies showed that invasion occurred without grazing, and could not be clearly tied to it. One study (Milchunas et al., 1990) showed that introduced species decreased in grazed compared to ungrazed rangeland. Although there is great interest among managers in understanding how the timing of livestock grazing can alter range composition including the abundance of non-native species, we found few quantitative studies in refereed journals that report on this issue. Whisenant and Wagstaff (1991) in Utah found that light and heavy spring grazing promoted the occurrence of introduced annuals in a desert saltbush community, while fall grazing (light or heavy) had no impact on the occurrence of introduced species, and fall-grazed plots were similar to ungrazed control areas. The removal of livestock after decades of livestock grazing often does not result in recovery of native species (e.g., Bartolome and Gemmill, 1981), and exclosure experiments after long-term grazing are difficult to interpret in terms of assessing what the initial grazing impacts were. Despite this, long-term impacts of the cessation of grazing are variable

418

Table 17.2 Relationship between livestock grazing and occurrence of non-native species Location 1

Invader

Findings

Study type 2

Reference

California (annual grassland)

Erodium sp.

promoted by sheep grazing

Exp., Obs.

Rice (1987)

Nevada, Idaho (Great Basin Desert)

Bromus tectorum, Bromus rubens

spreads without grazing

Obs.

Whisenant (1990a); Svejcar and Tausch (1991)

Utah (Great Basin Desert)

Bromus tectorum, other annuals

grazing plus fire promoted invaders

Multi-site, Cor.

Sparks et al. (1990)

Utah (Great Basin Desert)

Bromus tectorum, Halogeton sp.

increased with spring grazing, fall grazing – no effect

Obs., Exp.

Whisenant and Wagstaff (1991)

Colorado (short grass prairie)

Eurasian species

heavy grazing reduced invaders compared to ungrazed

Obs.

Milchunas et al. (1990)

Arizona (desert grassland)

Eragrostis lehmanniana

spreads with or without grazing

Obs.

Anable et al. (1992); McClaran and Anable (1992)

British Columbia (grazing land)

Centaurea diffusa

increases with light or heavy grazing, grazing poorly quantified

Cor.

Myers and Berube (1983)

Intermountain west, U.S.A. (desert)

Eurasian annuals

increase in dominance with heavy grazing

Rev.

Miller et al. (1994)

Mexico (Sonoran Desert)

Pennisetum ciliare

invasion occurs with or without grazing

Obs.

B´urquez and Quintana (1994)

Argentina (subhumid grassland)

exotics

grazing increases exotics

Gra.

Sala et al. (1986)

Argentina (montane)

exotics

grazing caused slight increase in exotics

Chr.

D´ıaz et al. (1994)

Australia (temperate grassland)

exotics

heavy grazing promotes exotics

Cor.

McIntyre and Lavorel (1994)

Australia (temperate grassland)

exotic annuals

grazing promotes annual exotics over native perennials

Rev.

Tr´emont and McIntyre (1994)

Australia (temperate woodlands)

annual grasses and forbs

increased grazing leads to increased exotics

Multi-site, Cor.

Prober and Thiele (1995)

Many areas of world

pines

grazing herbaceous layer promotes pine invasion

Rev.

Richardson and Bond (1991)

North America

1

Ecosystem type in parentheses following location. Abbreviations: Ane., anecdotal (no data presented); Obs., observational (with data); Exp., experimental manipulation of vegetation conducted; Chr., chronosequence; Cor., correlational data; Gra., grazing exclosures; Rev., review. 2

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

Other continents

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS

and site-specific (Hatch et al., 1992; Bartolome and McClaran, 1992). Because of the increase in biomass of introduced species after the cessation of long-term livestock grazing, several investigators have suggested that carefully managed grazing may be useful in controlling introduced annual plants and promoting native perennial species. Overall, studies of grazing impacts in western North America and Australia suggest (and it has become dogma) that there tends to be a decrease in native perennial grasses and an increase in dominance by introduced annual species with heavy grazing [see Miller et al. (1994) for a review of work in North America]. However, there is a general lack of information on the actual mechanisms of conversion. An enormous amount of effort has gone into trying to understand the causes of vegetation change in rangelands and it is clear that grazing has played an important role (Walker et al., 1981; Vavra et al., 1994); but the relative roles of grazing, climatic fluctuation, and fire remain elusive (see Oesterheld et al., Chapter 11, this volume). Catastrophic large-scale disturbances We reviewed 12 studies that detailed the occurrence of introduced species on lands cleared for agriculture, grazing, or forestry (Table 17.3). Of the disturbances considered, land-clearing associated with agriculture is likely to be the most extreme, since it involves disruption of the soil profile as well as total removal of vegetation. We address here only the invasion or persistence of weeds on such sites after abandonment. All studies reported that introduced species were abundant on severely altered landscapes for the first several years after abandonment. This conclusion applies equally to clear-cuts and former agricultural lands, whether in deserts, temperate montane habitats, or tropical forests. Only seven studies provided information on species composition more than seven years after human activities ceased. In three of these studies (Anderson and Marlette, 1986; Hunter, 1991; Brandt and Rickard, 1994) introduced species (grasses) continued to dominate the sites for up to 50 years. All three of these were desert ecosystems that had experienced severe soil disturbance. In two studies (DeFerrari and Naiman, 1994; D´ıaz et al., 1994), one in the Pacific Northwest rainforests and the other in montane grasslands, introduced species were abundant in recently disturbed areas, but declined after more than seven years and were replaced by native species. Two other studies, both in humid habitats in the southeastern

419

United States, found that dominance by persistent nonnative trees increased over time since abandonment (Doren and Whiteaker, 1990a,b; Bruce et al., 1995). Both studies were in humid coastal prairie ecosystems where soil hydrology and structure had been severely altered during farming. Three studies in tropical-island ecosystems provided no information on length of time since disturbance, but reported widespread domination of clear-cut forest patches by introduced woody species (Gade, 1985; Parnell et al., 1989; Savage, 1992). Terrestrial ecosystems: natural/near-natural disturbances Small-scale opening of patches Table 17.4 compares 17 studies that address the relationship between small-scale disturbance and invasion. Two studies examined the role of insect-induced shrub mortality in promoting introduced species, and one study examined the role of death of native bunch grasses in providing canopy gaps for establishment. The other fourteen studies examined colonization by native and introduced species on areas subject to animal disturbance, or mechanical disturbance that simulated small-scale animal disturbance. Surprisingly, we found almost no studies of colonization of non-indigenous species in tree-fall gaps in forested ecosystems. We suspect that there are more studies that are appropriate here, but before the last ten years many investigators did not distinguish between introduced and native colonizers of small gaps or areas of disturbed soil. In the case of canopy gaps not associated with soil disturbance, one study in coastal California found that insect-induced mortality of the native leguminous shrub Lupinus arboreus led to formation of nitrogenrich soil patches which were rapidly colonized by introduced annual grasses and thistles rather than native prairie species (Maron and Connors, 1996). In a second study of insect-induced mortality of native plants in a shortgrass prairie ecosystem in Colorado, grub outbreaks did not affect the richness or abundance of introduced species (Milchunas et al., 1990). In a Californian study (Peart, 1989), mortality of native bunchgrasses increased recruitment of introduced perennial grasses. Our 14 studies of small-scale soil disturbance included a total of 21 different habitats. Disturbance promoted invasion into 16 of these, had no effect on invasion in three others and decreased the success of invaders in only two sites. The most common

420

Table 17.3 Studies examining the occurrence of invasive non-native species in anthropogenic, catastrophic disturbances of terrestrial environments Location 1

Disturbance

Effect on site

Impact on invaders

Study type 2

Reference

Nevada (Mojave Desert)

bombs

total destruction

persistent stands of introduced annual grasses

Obs.

Hunter (1991)

Washington (Great Basin Desert)

agriculture

total destruction

Bromus tectorum domination, even after 47 yr

Obs.

Brandt and Rickard (1994)

Washington (temperate rain forest and montane)

logging/clear cuts

canopy removal

increase within clear-cuts for 3–7 yr, decline thereafter

Multi-site Cor.

DeFerrari and Naiman (1994)

Utah (Great Basin Desert)

agriculture

total destruction

Agropyron desertorum persistent – 50 yr

Obs.

Anderson and Marlette (1986)

Texas (humid prairie)

agriculture

total destruction

Sapium sebiferum increases with time since Multi-site agriculture Obs.

Bruce et al. (1995)

Montana (montane coniferous forest)

logging/clear cuts

canopy removal

up to 60% cover even 7 yr later

Multi-site Obs.

Forcella and Harvey (1983)

Florida (flooded prairie)

agriculture

vegetation destruction, soil alteration

persistent invasion by Schinus terebinthifolius

Obs.

Doren and Whiteaker (1990a,b)

Argentina (montane grassland)

agriculture

total destruction

Eur. annuals common in recently fallow fields, rare in 25 yr old site

Obs.

D´ıaz et al. (1994)

Argentina (inland pampas)

agriculture

total destruction

Eur. weeds dominated up to 5 yr after abandonment

Obs.

D’Angela et al. (1986, 1988)

Mauritius (insular tropical forest)

logging, agriculture

partial and total destruction

exotics dominate all cut forests

Obs.

Parnell et al. (1989)

Rodrigues (insular tropical forest)

logging, agriculture

partial and total destruction

exotics dominate all cut forests

Ane.

Gade (1985)

W. Samoa (insular tropical forest)

logging, agriculture

partial and total destuction

vine-tangles abundant in 2º forest

Obs.

Savage (1992)

North America

1

Ecosystem type in parentheses following location. Abbreviations: Ane., anecdotal (no data presented); Obs., observational (with data); Exp., experimental manipulation of vegetation conducted; Chr., chronosequence; Cor., correlational data; Gra., grazing exclosures; Rev., review. 2

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

Other continents

Location 1

Invader

Disturbance

Effects of disturbance

Study type 2

Reference

North America California (coastal prairie)

Eur. annuals

exotics dominate dead shrubs patches

Obs., Exp.

Maron and Connors (1996)

California (coastal grassland)

Eur. grasses

insect-induced shrub mortality canopy death

Exp.

Peart (1989)

California (coastal grassland) California (coastal grassland, dune, scrub) California (serpentine grassland)

Eur. grasses Carpobrotus edulis Bromus mollis

gopher mounds gopher mounds + mechanical gopher mounds

increases establishment of perennial invaders enhanced establishment of invaders promotes invasion into grassland; no effect on dune and coastal scrub promotes invasion of native grassland

Exp. Obs., Exp.

Peart (1989) D’Antonio (1993)

Obs., exp.

California (valley grassland) California (coastal shrubland, Sierran foothill) Oregon (coastal prairie)

Erodium sp. Cytisus scoparius

Washington (coastal prairie) Washington (coastal prairie) Nevada (Great Basin Desert) Montana (short grass prairie)

Cytisus scoparius Senecio vulgaris Bromus rubens Centaurea maculosa Eur. species

gopher mounds gopher mounds + mechanical mechanical (simulated rodent) mechanical mechanical gophers ground squirrels, other mammals insect-induced death of natives bison wallows

Hobbs et al. (1988); Hobbs and Mooney (1991) Rice (1987) Bossard (1991)

reduced invasion promotes invasion enhanced invasion increased spread of invader into native prairie no effect on invasion

Exp. Exp. Obs. Obs.

Parker (1996) Bergelson et al. (1993) Hunter (1991) Tyser and Key (1988)

Obs.

Milchunas et al. (1990)

colonized by 3 ruderal species

Obs.

Wein et al. (1992)

mechanical

enhanced invasion

Exp.

mechanical

enhanced invasion

Exp.

Hobbs (1989); Hobbs and Atkins (1988) Burke and Grime (1996)

mechanical

no effect

Exp., Obs.

Belsky (1986)

Colorado (shortgrass prairie) N.W. Territories: Canada (boreal forest) Other continents W. Australia (Eucalyptus woodland, heathland) United Kingdom (grassland) Tanzania (savanna)

Senecio jacobaea

Eur. ruderals

Avena fatua, Ursinia sp. herbaceous exotics herbaceous exotics

Obs., Exp. promotes invasion reduced invasion in coastal site, enhanced it Exp. in Sierra site promotes invasion of prairie Exp.

McEvoy and Rudd (1993)

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS

Table 17.4 Studies reporting relationship between small-scale disturbance and presence or abundance of invasive non-native species

1

Ecosystem type in parentheses following location. Abbreviations: Ane., anecdotal (no data presented); Obs., observational (with data); Exp., experimental manipulation of vegetation conducted; Chr., chronosequence; Cor., correlational data; Gra., grazing exclosures; Rev., review. 2

421

422

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

invaders to be promoted were European annuals or biennials. In one of the studies that showed no impact of disturbance on invasion, D’Antonio (1993) found in coastal California that herbivory on seedlings of an aggressive non-indigenous species was so great in two of her three study habitats that it overrode any effect of disturbance in promoting invasion. The two studies which showed a negative effect of disturbance on invasion both involved invasion of the shrub Scotch broom (Cytisus scoparius) into the western United States. Bossard (1991) found that disturbance reduced establishment of broom in one site in California because seed predators and herbivores were attracted to disturbed microsites and removed most of the broom seeds from them. Parker (1996) found that localized disruption of the cryptogamic crust in a Washington coastal-prairie site decreased rather than increased establishment of Scotch broom. Gap dynamics are known to be important in promoting community- or landscape-scale coexistence in a wide range of communities from prairies to tropical rainforests, and it is likely that small-scale disturbances are important for the persistence within the landscape of short-lived, well-dispersed non-indigenous species. However, as in the case of anthropogenic-disturbance corridors (roads, trails), the role that small-scale disturbance plays in invasion of adjacent undisturbed habitat by introduced species is probably dependent on habitat and species. Most of the studies we found suggested that the introduced species studied could invade away from soil disturbances into “undisturbed” adjacent habitat. Hunter (1991) provided evidence that disturbance by gophers (Thomomys bottae) increases abundance of red brome (Bromus rubens) in ungrazed sites in the Nevada desert, but this species also appears to be able to invade undisturbed habitat where gopher disturbances are lacking (Beatley, 1966; Svejcar and Tausch, 1991). Peart (1989) reported that, although establishment of the three introduced grass species he studied was enhanced by gopher disturbances, two of the species were able to colonize undisturbed prairie, particularly areas dominated by annual grasses. Reproduction of individuals on gopher mounds was high, and may be an important source of seeds for colonization of surrounding areas. D’Antonio (1993) found that seedlings of the introduced succulent Carpobrotus edulis required rodent disturbance to establish in a California coastal grassland; once established, however, this long-lived species could grow out over and suppress herbaceous individuals within the

adjacent grassland. By contrast, Hobbs and Mooney (1991) and Hobbs et al. (1988) found that, although gopher disturbances promoted invasion of California serpentine grasslands by the Eurasian grass Bromus mollis, this species did not spread into undisturbed portions of these sites. Indeed, further disturbance of an initial gopher disturbance led to the disappearance of B. mollis from the local patch. Likewise, McEvoy and Rudd (1993) found that small-scale soil disturbance led to localized outbreaks of the Eurasian biennial Senecio jacobaea in an Oregon coastal prairie, but that it could not invade the dense adjacent undisturbed vegetation. Any habitat with burrowing organisms is likely to have a “natural” regime for occurrence of local-scale disturbances. The importance of size, frequency, and pattern of small-scale disturbances in relationship to invasion rates and persistence of invaders has rarely been examined. Bergelson et al. (1993) demonstrated experimentally that the abundance of an introduced annual forb was greater when disturbances were large (30 cm diameter vs. 5 or 15 cm) and randomly arrayed over the landscape. Peart (1989), however, found no effect of gap size on colonization success by introduced grasses in a California prairie, whereas McConnaughay and Bazzaz (1987) suggested that seed size of the colonizer determines the importance of gap size. In addition, there are no examples to show how alteration of “natural” local-scale disturbance regimes have interacted with invasion. Fires Numerous studies have examined the influence of fire on the susceptibility of arid and semi-arid ecosystems to invasion (Table 17.5). Most do not quantify fire frequency, intensity, or size, but simply compare invasion in burned versus unburned sites, or before versus after fire. In 18 out of 24 studies involving either controlled burns, multi-site comparisons, or long-term observations after wildfire, fire was found to promote or enhance invasion by non-native species. In at least three studies (Hobbs and Atkins, 1990; Trabaud, 1990; Wein et al., 1992), dominance by introduced species was short-lived after fire, and native species became dominant within a few years. We identified 12 studies that reported invasion occurring without fire or any apparent change in fire regime, but where fires promoted the spread or increase in density of the species in question. The ecosystem types most readily invaded without fire are desert habitats such as the Great Basin of western

Location 1

Invader

Impacts of fire

Study type 2

Reference

California (Mojave Desert)

Bromus rubens, Schismus barbatus

invades without fire; enhanced by fire

Multi-site, Cor.

Brown and Minnich (1986)

California (Sierran foothills)

Eur. annual grass and forbs

grasses reduced by fire, forbs enhanced

Controlled burn

Parsons and Stohlgren (1989)

California (maritime chaparral)

Carpobrotus edulis

germination reduced by fire but invasion promoted by fire

Obs., Exp.

Zedler and Scheid (1988); Hickson (1988); D’Antonio et al. (1993)

California (Sierran foothills)

Cytisus scoparius

promoted by fire

Exp.

Bossard (1991)

Oregon (Great Basin Desert)

Bromus tectorum

maybe invasion without fire; enhanced by fire

Obs., Ane.

Klemmedson and Smith (1955)

Idaho (Great Basin Desert)

Bromus tectorum + other annuals

invades small amount without fire; enhanced by fire

Multi-site, Cor.

Whisenant (1990a)

Utah (Great Basin Desert)

Bromus tectorum + other annuals

maybe enhanced by fire (sites also heavily grazed)

Multi-site, Cor.

Sparks et al. (1990)

Nevada (Great Basin Desert)

Bromus rubens

invades without fire; promoted by fire

Cor.

Beatley (1966); Hunter (1991)

Nevada (Great Basin Desert, island in lake)

Bromus tectorum, Bromus rubens

invades without fire

Obs.

Svejcar and Tausch (1991)

Intermoutain west U.S.A. (Great Basin Desert)

Taeniatherum asperum

invades without fire some sites; other sites it needs fire to invade

Ane.

Young and Evans (1971)

Arizona (Sonoran Desert grassland)

Eragrostis lehmanniana

germination enhanced by fire; spread maybe enhanced by fire; can spread without fire

Exp., Controlled burn

Ruyle et al. (1988); Cable (1971); Anable et al. (1992)

North America

Sonora, Mexico (Sonoran Desert)

Pennisetum ciliare

invades slowly without fire; enhanced by fire

Ane.

B´urquez and Quintana (1994)

South Dakota (shortgrass prairie)

Bromus japonicus

invades without fire; suppressed by fire

Exp., Controlled burn

Whisenant (1990b); Whisenant and Uresk (1990)

Hawaii (seasonal submontane woodland)

perennial C4 grasses

some invade without fire, all promoted by fire

Multi-site, Obs.

Hughes et al. (1991); Smith and Tunison (1992)

N.W. Territories, Canada (boreal forest)

Eur. ruderals

large natural fire promoted temporary invasion by 3 species

Obs.

Wein et al. (1992)

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS

Table 17.5 Studies examining the relationship between fire and invasions by non-native species in terrestrial ecosystems (excluding riparian)

Other continents Hyparrhenia rufa

invades without fire; promoted by fire

Obs.

Bilbao (1995)

annual grasses

fire promotes short-lived flush of intr. annuals

Obs.

Trabaud (1990)

W. Australia (shrublands, heathland)

Eur. annual grasses

decreased invasion shrubland; no effect in heathland

Exp.

Hobbs and Atkins (1991) continued on next page

423

Venezuela (llanos) France

424

Table 17.5, continued Location 1

Invader

Impacts of fire

Study type 2

Reference

W. Australia (Eucalyptus woodland)

ann. forbs, grasses

enhanced by fire

Obs.

Bridgewater and Backshall (1981)

W. Australia (Banksia woodland)

Eur. annuals

increased by fire but for short time only

Controlled burn

Hobbs and Atkins (1990)

W. Australia (Eucalyptus woodland)

Eur. annuals

decreased by fire

Controlled burn

Hester and Hobbs (1992)

E. Australia (grassland)

Eur. annuals

enhanced by fire (site heavily grazed by livestock)

Obs.

Lunt (1990)

South Africa (fynbos)

Pinus radiata

invasion very slow without fire; enhanced by fire

Obs.

Richardson and Brown (1986)

South Africa (fynbos)

Pinus halepensis

invasion very slow without fire; enhanced by fire

Obs.

Richardson (1988)

South Africa (fynbos)

Banksia

enhanced by fire

Obs.

Richardson et al. (1990)

1

Ecosystem type in parentheses following location. Abbreviations: Ane., anecdotal (no data presented); Obs., observational (with data); Exp., experimental manipulation of vegetation conducted; Chr., chronosequence; Cor., correlational data; Gra., grazing exclosures; Rev., review. 2

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS

North America and the Sonoran Desert of Mexico and Arizona. Whisenant’s (1990a) study of invasion by cheatgrass of the Snake River plains in Idaho is an excellent example of invasion in the absence of fire (and livestock grazing), but clear enhancement of dominance by recurrent fire. In addition to deserts, we found examples of fire promoting invasion of Mediterranean ecosystems such as the fynbos of South Africa and maritime chaparral of California. Invasion by introduced species occurs only exceptionally after chaparral fires in California and has been reported only for maritime chaparral (Hickson, 1988; Zedler and Scheid, 1988; D’Antonio et al., 1993). Invasion here does not appear to be due to an alteration in natural disturbance regimes, but rather to the more open form of the maritime chaparral with a less diverse assemblage of native post-fire species (Tyler, 1994). Numerous investigators have now shown that natural fire in the South African fynbos promotes invasion of these sites by introduced pines and Australian species of Acacia and Hakea (Table 17.5). Indeed, Richardson and Bond (1991) reported numerous cases from throughout the world where range extension by pines is promoted by fire. Because of the longevity of pines, these post-fire invasions are likely to represent long-term successional changes in natural communities. This is also true for invasion of maritime chaparral in California by Carpobrotus edulis, which is an aggressive competitor against native species (D’Antonio and Mahall, 1991); its invasion likely represents a long-term successional change at these sites. In only four studies was fire found to cause a decrease in abundance of introduced species. Hester and Hobbs (1992) reported that fire and manipulation of fire regimes may be a useful management tool to control non-native species in Australia: most of the native species at their sites responded positively to fire, whereas the introduced species present on the sites were negatively impacted. By contrast, in a management-oriented study in California (Parsons and Stohlgren, 1989), fire caused a decrease in abundance of introduced annual grasses, but an increase in abundance of introduced forbs, with no net positive effect on native species. The season of burning was found to be an additional factor influencing the likelihood of invasions by non-indigenous species (Parsons and Stohlgren, 1989; Hobbs and Atkins, 1990; Whisenant, 1990b). Thus, where the life forms of the available suite

425

of introduced species are diverse, their responses to fire are likely to be variable. Overall, we found few studies to support the role of fire in promoting invasion into most ecosystems which have a long history of fire, such as the mixed conifer forests of western North America, California chaparral, prairies of the midwestern United States, heathland and open Eucalyptus forest in Western Australia, and African savannas. Indeed, to generalize about the importance of fire would require an extensive review of all studies documenting changes in species composition in response to fire, and most such studies do not even mention whether the responding species are introduced or native. Virtually all of the studies we report on here suggesting that fire events or recurrent fire promoted long-term presence of invaders were in deserts, or ecosystems such as oceanic islands (see Hughes et al., 1991) where the role of fire as a selective force in plant evolution appears to have been minimal. The mountain fynbos of South Africa was the only ecosystem that is being invaded on a very large scale by fire-promoted species, and where fire has been a recurrent phenomenon prior to invasion (we do not consider the invasion by C. edulis into maritime chaparral to be a large-scale invasion, because this form of chaparral is restricted). Although fire suppression has been an important part of land management in many regions of the world, there are few published reports of how a decrease in fire frequency might promote invasion by nonindigenous species. Bruce et al. (1995) suggested that fire suppression in Texas coastal prairies might be allowing invasion by Chinese tallow tree, Sapium sebiferum. Their study sites, however, were all severely disturbed by agriculture and hydrological modification prior to invasion. Although fire suppression in Californian chaparral has caused variation in fire size and intensity, and in germination frequencies of various native species (Moreno and Oechel, 1991), there is no evidence that it has favored invasion by non-indigenous species. However, more studies are needed in this area. Storm damage and other large-scale disturbances We found almost no quantitative studies examining the relationship between canopy damage caused by storms and invasion of otherwise intact natural communities by non-indigenous species. Williams (1993) found that closed-canopy forests in Virginia in the eastern United States were invaded by the shadeintolerant Asian tree Paulownia tomentosa only after

426

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

these forests had been severely damaged in August of 1969 by Hurricane Camille. Likewise, native forests on Tahiti have been almost completely transformed to monospecific stands of the New World tree Miconia calvescens; accounts suggest that the rise to dominance by Miconia followed severe hurricane damage to the native forest canopy (J. Schwartz, pers. commun.). Most accounts of invasion following hurricane damage are similarly anecdotal. We found no information on the role of landslides in promoting or limiting invasion, although we believe that this phenomenon is important. For example, in coastal California, landslides are rapidly colonized by Pampas grass, Cortaderia jubata, an invader from South America (D’Antonio, pers. observ.). Summary of terrestrial plant invasions and disturbance In spite of the common belief that disturbance is a precursor to invasion in plant communities, we found numerous examples where invasion could occur without disturbance. This was particularly true in desert habitats. Crawley (1987) found for the British flora that invading species were more common in plant communities where the average cover was low – which is certainly true of desert habitats compared to grasslands or forest. In addition to the studies outlined above, there are several other reports of plant species that can invade without any apparent disturbance. A detailed list has been provided by Rejm´anek (1989). Several of the reports of species invading forested sites without any obvious disturbance are on islands (e.g., Lorence and Sussman, 1986; Huenneke and Vitousek, 1990; Rejm´anek, 1996). Are these also “open” or “undersaturated” systems? While this is possible, we recommend interpreting such studies with caution. In many of the reports of invasions into undisturbed island ecosystems, sites adjacent to the invaded forests have undergone catastrophic disturbance, are full of nonnative species, and serve as reservoirs for dispersal into undisturbed sites. Also, islands have a long history of habitat modification, including elimination of native avifauna (e.g., Olson and James, 1982; Steadman and Kirch, 1990; James, 1995) and alteration of native invertebrate communities, which may have reduced the resistance of the native biota to invasions (e.g., Lake and O’Dowd, 1991; D’Antonio and Dudley,

1995). Overt disturbance may not be necessary to promote invasions on islands today, but present-day communities may lack mechanisms that previously might have made them more resistant to invasion. Few studies have actually attempted to determine the mechanisms through which disturbance might influence invasion. An understanding of these mechanisms would require elucidation of factors within undisturbed communities that make them more or less resistant to invasion. Disturbance might influence invasion by: (1) eliminating predators or herbivores that would otherwise reduce or eliminate the invaders; (2) reducing competitive pressure from pre-established species; (3) stimulating germination of seeds of invaders; and/or (4) altering resources to levels that favor the invaders rather than pre-existing individuals. Where mechanisms have been investigated they have been found to be complex and site-specific. For example, D’Antonio et al. (1993) investigated the mechanisms through which fire appears to promote invasion of maritime chaparral in California by the South African succulent Carpobrotus edulis. They found that high temperatures killed C. edulis seeds, but that invasion occurred anyway because of the existence of microsites where soil temperatures were not elevated during fire, and because seeds of this fleshy-fruited species are readily dispersed into burned sites by abundant native frugivores. Thus, fire intensity and fuel distribution should have a significant influence on whether or not invasion will occur. In addition, fire altered soil conditions and favored invasion by promoting C. edulis seedling growth, so that herbivores could no longer crop back new invading plants. In a grassland site, D’Antonio (1993) found that soil disturbance by rodents promoted invasion by C. edulis because it stimulated seed germination and decreased seedling competition between C. edulis and Eurasian grasses. Rodent-caused soil disturbance had no overall net effect on invasion by C. edulis in a dune and a coastal-scrub site where herbivory by rabbits and deer consistently eliminated emerging seedlings: disturbances were not large enough to decrease the density of herbivores and thereby enhance invasion. Predicting when and why disturbance promotes invasion will thus require many more such mechanistic studies because currently we have very little understanding of which mechanisms are more likely to be responsible for community resistance to invasion and how disturbance regime interacts with these.

DISTURBANCE AND BIOLOGICAL INVASIONS: DIRECT EFFECTS AND FEEDBACKS

Riparian and wetland invasions The relationship between disturbance and invasion by non-indigenous species into aquatic ecosystems appears to be substantially different from that in many terrestrial ecosystems, primarily owing to the pervasive role that disturbance has in structuring community composition. Hydrological dynamics are of fundamental importance to aquatic systems, and may constitute direct-disturbance impacts, as in destructive flooding following spates or snowmelt (likewise, disruption of flows, such as the physical damming of channels by landslides, beaver dams, or human structures, alters the relationship between hydrology and biota). While the degree to which surface hydrology modifies habitat and community structure is the basic difference between aquatic and terrestrial systems, many other disturbances or perturbations are shared with upland systems, including fire, forest blowdown, or grazing. In defining disturbance in riparian and wetland habitats, we again restrict our discussion mainly to physical processes which remove tissue and open space for colonization. Others have broadened the definition to include natural physical stresses, such as sustained drought (Stanley et al., 1994) and chronic changes (usually increases) in nutrient supplies, sediment deposition, and chemical pollution. For the purpose of this review, we consider most of these to be non-episodic perturbations which constitute changed conditions for organisms rather than punctuated mechanisms structuring these systems. Riparian ecosystems Riparian ecosystems are considered disturbanceprone; nearly every unregulated watercourse experiences substantial, often catastrophic, variation in discharge, which can remove established biota from in-stream and floodplain environments. Indeed, most stream ecologists consider that native aquatic and riparian organisms depend upon periodic flooding for creation and maintenance of suitable habitat, removal of excess deposited material, and reduction in density of competitors (Fisher et al., 1982; Resh et al., 1988; Poff, 1992). Riparian vegetation alongside unregulated or regularly disturbed rivers is early-successional in character, generally composed of fast-growing, shortlived, or disturbance-tolerant species which can reestablish populations before the next scouring event (Hupp and Osterkamp, 1985). These are traits frequently shared with invasive, non-indigenous plants in terrestrial systems. In at least 18 of the 38 studies we reviewed which examined the status of introduced

427

plants in riparian corridors, natural flood regimes in some manner promoted the growth of invasive species (Table 17.6). Another six studies found that exotic riparian species were favored on bare soils if space was opened by physical disturbance other than the flood regime, and in most of the remaining studies the role of substrate scouring was not considered. Studies in which natural disturbance favored invasion are divided between cases of non-indigenous species tending to remain within the most frequently disturbed near-stream environment, and studies which document some expansion into less disturbed assemblages following initial establishment in the riparian zone. In the Ardour River in France, Tabacchi (1995) showed that introduced species comprised 20% of the active streambank vegetation, but were poorly adapted to tolerate drought away from the channel. Older arms of the river included fewer non-native species, and these were more sensitive to local loss from flooding. Likewise, the Mediterranean tree Fraxinus ornus is dispersed downstream by water flow and grows abundantly (including regrowth from damaged boles) in the high-energy channel habitat; but 65 years after the introduction of this tree into new channels, it had not expanded into adjacent riparian vegetation (Th´ebaud and Debussche, 1991). DeFerrari and Naiman (1994) and Planty-Tabacchi et al. (1996) reported that nonindigenous species were abundant in riparian zones and in particular on gravel bars within stream beds in Washington and Oregon (and also in France), but that only infrequently did these species penetrate the intact mesic forest away from the stream beds. Microstegium vimineum is an introduced, shade-tolerant annual grass, which was reduced by 50% during flooding in a North Carolina floodplain, then doubled its previous abundance the following year through high seed set. Still, it has not invaded the native riparian woodland nearby (Barden, 1987). In Great Britain, Japanese knotweed (Fallopia japonica = Reynoutria japonica), found naturally on sparsely vegetated lava flows, is a riparian ruderal species that appears to be replaced by other vegetation within 50 years (Palmer, 1994). In most of these cases, the invaders tend not to alter the character of the riparian ecosystem substantially, nor to displace native species. Other non-indigenous streamside plants, or even the same species in a different environment, may be less benign. Numerous studies show that, while initial establishment of plants has occurred in regularly disturbed riparian areas, invasion often proceeds

428

Table 17.6 Non-indigenous plants associated with riparian ecosystems, and relationships with disturbance and flow regimes Location 1

Setting

Invader

Findings

Study type 2

Reference

New York (temperate forest)

woodland/urban streams and canals

Lythrum salicaria

establishes in disturbed soil on dikes/ditches, spreads into streams/marshes

Obs.

Stuckey (1980)

Pennsylvania (temperate forest)

woodland marsh and stream

Lythrum salicaria

fluctuating water levels favor germination on open soils; spreads into marsh

Obs.

Thompson (1991)

Virginia (temperate forest)

woodland floodplain

Paulownia tomentosa

hurricane opens space for germination, but declines with succession

Obs.

Williams (1993)

North Carolina (temperate forest)

woodland floodplain

Microstegium vimineum

flood and fire open space for germination; poor colonization in established vegetation

Obs.

Barden (1987)

Texas (semi-arid scrub)

newly formed river delta at reservoir

Tamarix sp.

explosive germination on stabilized soils

Obs.

Robinson (1965)

Ohio (temperate forest)

restored riparian marsh, Typha spp. agricultural

Rocky Mts. (arid montane)

springs and streams in rangelands

Eleagnus angustifolia established on stabilized levees, subsequently invaded riparian areas

Obs.

Knopf and Olson (1984)

Colorado (arid montane)

high plains riparian areas, grazing, and some agriculture

Salix spp. (Eurasian)

Exp., Obs.

Shafroth et al. (1994)

Utah (Great Basin Desert)

many former Populus riparian zones

Tamarix sp., natural high salinity tolerated better by exotics; Eleagnus angustifolia physical disturbance may not be required for establishment

Cor.

Carman and Brotherson (1982)

Southwest U.S.A.

desert riparian areas

Tamarix sp.

flooding promotes germination, but establishment requires sustained moisture

Obs.

Horton et al. (1960)

Arizona (Mojave Desert)

Populus riparian, water diverted for agriculture and urban use

Tamarix sp.

declining water level favors Tamarix, which can use unsaturated soil water unlike native Populus and Salix

Exp.

Busch et al. (1992)

Southwest U.S.A.

desert riparian, Tamarix sp. agricultural, rangelands

water diversion and impoundment reduces flood frequencies, and is associated with expansion of Tamarix

Obs.

Everitt (1980); numerous others

Arizona (Sonoran Desert)

Populus/Platanus riparian, grazed

Cynodon dactylon

Bermuda grass stabilizes substrate, provides flood refuge for macrophytes

Cor.

Dudley and Grimm (1994)

Arizona (Sonoran Desert)

mixed riparian forest, nature reserve

Tamarix sp., Paspalum sp., Cynodon dactylon

flood caused greater mortality to Tamarix than native Populus and Salix; Paspalum and Cynodon increased rapidly

Obs.

Stromberg et al. (1993)

North America

invasive Typha declines as planted and natural riparian Obs. vegetation close space

continued on next page

Carla M. D’ANTONIO, Tom L. DUDLEY and Michelle MACK

cuttings and seedlings survived wide range of hydrological conditions; expands by vegetative growth in moist sites

Niswander and Mitsch (1995)

Location 1

Setting

Invader

Findings

Study type 2

Reference

Arizona (Mojave Desert)

mesquite riparian area

Tamarix sp.

fire reduces Tamarix more so than Prosopis, but regenerates rapidly

Cor.

Busch (1995)

Southwest U.S.A.

desert riparian areas; diversions for agriculture, etc.

Tamarix sp. and dams eliminated flooding; Tamarix germinates more Eleagnus angustifolia successfully than Populus, Eleagnus highly shade-tolerant so establishes under canopy

Obs.

Howe and Knopf (1991)

Rocky Mts. U.S.A. (Great Basin/montane)

riparian

Eleagnus angustifolia Eleagnus phenology more flexible than Populus; tolerates drought and shading; takes advantage of ecological variation

Obs.

Shafroth et al. (1995)

Colorado (semi-arid montane)

cottonwood riparian, dams for agriculture

Tamarix sp.

stabilized flow associated with Tamarix increase and Populus decline, channel narrows and deepens

Obs.

Snyder and Miller (1992)

Pacific Northwest (temperate forested montane coniferous forest) riparian areas, minor logging

many species

24–30% of riparian species are aliens, associated with natural flooding; some invasion of mature vegetation, but most species in “young” communities

Cor.

Planty-Tabacchi et al. (1996)

California (Great Basin)

groundwater pumped for export

annual grasses

lowered water associated with decline in native riparian plants; increase in exotic annuals

Cor.

Manning (1992)

California (sage scrub chaparral)

ephemeral channel, agriculture and urban development

Schinus molle

road and agriculture run-off maintains continuous seepage; allows slow establishment of pepper tree

Obs.

Howard and Minnich (1989)

California (coastal mesic forest)

permanent river in nature preserve

Rana catesbiana, macrophytes

reduced flow and scouring during drought favor plant Cor. build-up, then favoring bullfrog increases; both decline with return to natural flooding

Northern Territory, Australia

arid watercourses

Tamarix aphylla

rare flood/wet year dispersed seed and favored establishment, may displace native river red gum (Eucalyptus camaldulensis)

Obs.

Griffin et al. (1989)

South Australia

woodland riparian

Pittosporum undulatum

native invasive species; increases during drought, although native Tristaniopsis resists flooding better

Obs.

Melick and Ashton (1991)

South Australia

woodland riparian

many spp.

fire, clearing, and fragmentation all area associated with increase in exotics

Obs.

Recher et al. (1993)

Western Australia

seasonal tropical riparian/marshlands

Mimosa pigra

colonizes “undisturbed” marsh, invades then into riparian, and ultimately uplands, alters vegetation and associated wildlife

Cor.

Braithwaite et al. (1989)

Great Britain

floodplains, farmland

Fallopia japonica

establishes on soils opened by flood scour, replaced by Obs. other vegetation in 300 kg m−2 for Sequoia forests) (Cannell, 1982). The rate of aboveground net primary production typically ranges from 0.4 to 2.5 kg m−2 yr−1 . The time between stand-replacing disturbances is commonly on the order of centuries, with typical ages for mature trees of 50 to >2000 years. The current extent of temperate forests in the northern hemisphere is probably about 20% less than in preagricultural times (Williams, 1994), and most of the remaining forests have been moderately to heavily affected by human activities. These human-related changes include widespread alteration of fire regimes, conversion of old-growth forests to younger forests,

conversion of diverse forests into intensively managed plantations, and introduction of exotic species. The patterns of forest disturbances have been substantially altered, even where human influences have been relatively slight. The basic features of natural disturbances in temperate forests are briefly mentioned in this chapter, as many additional details are available in other chapters. The focus then shifts to human-related disturbances (particularly forest harvesting), highlighting the similarities and differences between these and more natural disturbances. Ideally, these subjects might be woven together with a theme of natural vs. humanrelated disturbances, but the wide range of scales and impacts in both categories minimize any useful generalizations.

PATTERNS AND PROCESSES OF NATURAL DISTURBANCES

Foresters and ecologists have always been aware that disturbances can shape the development of forests, but a deeper appreciation of the pervasive importance of disturbances has developed only in the past few decades. Across the United States, for example, the annual rate of tree mortality averages about 18% of the rate of forest growth (on a volume basis: United States Department of Agriculture Forest Service, 1981). This mortality is driven by many processes, and patterns range from death of single trees to widespread disturbances across landscapes. Old-growth forests commonly occupied much of the temperate-forest region before widespread industrialization, and old-growth forests typically experienced

453

454

high rates of mortality. For example, gaps in the canopies of old-growth forests of amabilis fir (Abies amabilis), western hemlock (Tsuga heterophylla), and mountain hemlock (Tsuga mertensiana) in coastal British Columbia account for 18% of the ground area, and another 60% of the forest area is close to a gap – at a distance less than the canopy radius of a single tree (Lertzman and Krebs, 1991). In addition, in North America before European settlement, natural disturbance regimes led to substantial areas of younger forests within landscapes dominated by older forests (Whitney, 1994). White and Pickett (1985) developed some terms for describing the major features of this broad array of disturbances, including distribution (in space, and along environmental gradients), frequency or return interval, rotation period (time needed for entire area to be disturbed at least once), predictability, size, intensity, and severity. They also noted that occurrence of one disturbance, such as a drought, could increase the likelihood of another (such as an insect outbreak). These features can be grouped broadly in terms of major disturbances (called stand-replacing disturbances, where most of the dominant plants are removed) and minor disturbances (which leave the majority of dominant plants alive: Oliver and Larson, 1996).

WIND

Most forests of the world experience wind events that damage trees (see Webb, Chapter 7, this volume, for more detail). These wind disturbances span scales from removal of branches on some trees by moderate winds (often associated with heavy loading of branches with snow or ice) to removing most of the dominant trees across landscapes by hurricanes. The risk of wind damage to an individual tree (or stand) tends to increase with age, as a result of increasing size and resistance to the wind (Oliver and Larson, 1996). The resistance of a tree to wind damage depends in part on the wind regime experienced during the development of the tree; trees exposed to strong winds develop stem forms (taper) that accommodate greater stress from swaying canopies. Changes in stand structure, such as creation of gaps, may increase the wind velocities experienced by individual trees, leading to either breakage of stems or uprooting of the entire tree. Uprooted trees create pit and mound

Dan BINKLEY

microtopography (covering up to 30% of a forest area), mixing soil horizons and providing raised microsites that may favor tree regeneration (Pritchett and Fisher, 1987). Many case studies document the widespread impacts of winds. Whitney (1994) tabulated estimates of major wind damage for forests in the eastern United States before European settlement. The return interval (reciprocal of the annual proportion of an area devastated by windstorms) averaged between 500 and 1300 years for the northeastern and midwestern U.S.A. This range overlaps with the return interval for fire, which ranged from 130 to 14 000 years, aside from an interval of 80 years for jack pine (Pinus banksiana). Franklin (1988) cited unpublished data from across the Pacific Northwest of the United States on mortality of trees in old-growth stands. Winds were responsible for 80% of mortality in coastal forests of Sitka spruce (Picea sitchensis) and western hemlock (Tsuga heterophylla), compared with 40% of the mortality forests of Douglas-fir (Pseudotsuga menziesii) and western hemlock in the Cascade Mountains, and 20% of mortality in drier, less dense forests of ponderosa pine (Pinus ponderosa) to the east. Large windstorms can create mosaics of disturbance across large landscapes. For example, a windstorm in 1921 damaged forests across 25 000 ha of the Olympic Peninsula of Washington (U.S.A.), with severity of damage ranging from slight to almost complete removal of dominant trees on different portions of the landscape (Oliver and Larson, 1996). This storm had two major effects: accentuating landscape heterogeneity by creating a mosaic pattern of old forests mixed with regenerating forests, and insuring some continued homogeneity into the future by synchronizing the regeneration time across large portions of the landscape. The condition of stands may play a large role in determining the effects of high winds. Diseased or damaged trees tend to be less wind-firm, and some species are more prone to wind damage than others. For example, Hurricane Hugo in 1989 toppled mature loblolly pine (Pinus taeda) more readily than similarsized longleaf pine (P. palustris) (Gresham et al., 1991). Wind disturbance may lead to a synergistic interaction (sensu White and Pickett, 1985) with other disturbances, particularly with fire and insect outbreaks. Accumulations of large quantities of dead wood can greatly increase the risk of subsequent fires. A strong windstorm in the Rocky Mountains of

DISTURBANCE IN TEMPERATE FORESTS OF THE NORTHERN HEMISPHERE

Colorado (U.S.A.) in 1939 is thought to have provided the conditions necessary for an outbreak of spruce bark beetles (Dendroctonus rufipennis), and the associated blue-stain fungus (Leptographium engelmannii) that devastated almost 300 000 ha over the following decade (Veblen et al. 1991; see section “Insects and Diseases” below). Nutrient supplies may increase following major wind disturbances, as a result of decreased uptake by plants, perhaps lower supplies of labile carbon for microbial immobilization, and higher rates of mineralization of organic matter. I know of only one study that examined net nitrogen mineralization in a wind-generated age sequence of temperate forests. The high-mortality zones of fir waves1 of both Fraser fir (Abies fraseri) in North Carolina (U.S.A.) and balsam fir (Abies balsamea) in New York (U.S.A.) showed rates of net nitrogen mineralization that were similar to those in adjacent mature stands (Sasser and Binkley, 1989). More information would be needed to develop a general picture. Wind disturbances generally leave some vegetation relatively intact, and some root systems or broken stems may sprout. Disturbances to the soil are generally slight, except where roots have been upturned. Revegetation is relatively rapid, and nutrient losses are probably minimal.

INSECTS AND DISEASES

Insect populations are generally present in forests at low-to-moderate levels (Schowalter, Chapter 9, this volume). Low populations may have substantial effects on trees, and landscape-scale disturbances typically result when populations increase by orders of magnitude above background levels. These major disturbances typically have a synergistic component – population irruptions often depend in part on weather patterns and on susceptibility of host trees (which may in turn relate to weather patterns or other stresses). Forest diseases, such as root-rot fungi, generally affect smaller portions of a landscape than do major insect outbreaks, but also show large synergistic effects. In addition, synergies are common between insects and diseases. Bark beetle infestations are the indirect cause of tree death in many instances; the direct cause is the proliferation of blue 1

455

stain fungal hyphae developing from spores transported by the beetles. Many temperate regions are experiencing unusually severe outbreaks of insects and disease pathogens as a result of human-related introductions to new forest areas. Exotic bark beetles (Scolytus multistriatus) introduced the exotic Dutch elm disease (Ceratocystis ulmi) which decimated American elm (Ulmus americana) over much of its native range in North America. Similarly, the chestnut blight (Endothiella parasitica) from Asia was introduced to forests of the eastern United States early in this century, and within a few decades American chestnut (Castanea dentata) was eliminated as a dominant component of these forests. Gypsy moths (Porthetria dispar) have greatly influenced rates of defoliation (particularly for oaks – Quercus spp.) in some forests of the same region, but with much lower rates of mortality. Nutrient losses in stream water may be increased after disturbance by insects or diseases, but the increases are relatively small. For example, Swank (1988) noted that defoliation of a hardwood forest (up to one-third of the canopy removed through the growing season for several years) raised stream-water concentrations of nitrate nitrogen from background levels of about 10 g °−1 to about 40 g °−1 . Although the response of stream-water nutrient concentrations was dramatic, the increase in loss of nitrogen was quite small (2000 years (Pyne et al., 1996). Information is spottier for other temperate regions. Heavily settled parts of Europe have very low fire frequencies; fire regimes in Russia and other countries of the former Soviet Union vary substantially among regions (with changes in settlement patterns and fire-control policies), and the relatively low frequency of fires in temperate forests in China was overshadowed by the 3×106 ha of forest consumed by the Black Dragon fire in 1987 (Pyne et al., 1996).

Conifer forests (northwestern USA) 50–750 1

Based on Heinselman (1981); Martin (1982); Agee (1993); Clark and Robinson (1993); Oliver and Larson (1996). 2 This broad range of forest types and fire regimes is probably representative of similar forest types in other temperate regions. 3 Note that any particular hectare may burn more or less often than the long-term average across the landscape.

Forests with short return intervals for fire commonly experienced surface fires that did not kill the dominant vegetation, but probably shaped the composition of the understory and the long-term dynamics of the forests. Return intervals of a century or more typically led to high-intensity, stand-replacing fires (Fig. 18.1). At a continental scale, the 48 contiguous United States contain about 250×106 ha of temperate forests, and about 15×106 ha of forests burned each year in the first third of the 20th Century (MacCleery, 1992), for an average return period on the continental scale of about 15–20 years. This fire pattern was unusual, reflecting a substantial role of humans in igniting fire, but little effort at fire suppression. Since the 1960s, only about 1–2×106 ha have burned annually, for a

Surface fires Low-intensity, frequent surface fires were characteristic minor disturbances of pine forests (Pinus palustris, P. taeda) in the southeastern part of North America, and in western forests of ponderosa pine (P. ponderosa). As a result of these fires, accumulations of material (and nitrogen) in the forest floor (litter layer, O horizon) were low, there were lower densities of trees per

DISTURBANCE IN TEMPERATE FORESTS OF THE NORTHERN HEMISPHERE

hectare (particularly for ponderosa pine forests), and high production and diversity of herbaceous species. The biomass of pine forests in the southeastern United States is typically 15 000–30 000 kg ha−1 , with 50– 150 kg nitrogen ha−1 ; repeated burning reduces forestfloor biomass to 10 000–15 000 kg ha−1 and nitrogen content to 20–35 kg ha−1 (Binkley et al., 1992). The loss of nitrogen to the atmosphere during surface fires averages about 5 kg nitrogen per megagram of fuel consumed (Fig. 18.2). Ponderosa pine forests with natural fire regimes in

457

the southwestern United States have a tree density of about 30 to 150 trees ha−1 , compared with 700 to >2000 trees ha−1 following decades of fire suppression (Covington and Moore, 1994). Low-density stands have major understory communities (containing 50 to 200 species in a hectare) that contribute substantially to ecosystem productivity, whereas in high-density, unburned stands of ponderosa pine productivity is concentrated primarily in the overstory trees (Covington, 1994). Scorched pines may be more susceptible to attacks by bark beetles (Agee, 1993), another example of synergy between disturbance factors. Stand-replacing fires

Fig. 18.2. Nitrogen loss is proportional to fuel consumption (about 5 kg N lost per megagram of fuel consumed) for both surface fires (upper) and slash fires (lower). Pita1 (loblolly pine) from Kodama and Van Lear (1980); Pita2 from Richter et al. (1982); Pita3 from Schoch and Binkley (1986); Pipo1 (ponderosa pine) from Nissley et al. (1980); Pipo2 from Covington and Sackett (1984); Laoc (larch (Larix occidentalis)Douglas-fir stand) from Jurgensen et al. (1981); data for slash fires from Little and Ohmann (1988).

Stand-replacing fires vary widely, in the amounts of fuel consumed, temperature regimes (aboveground and belowground), and subsequent environmental effects. Once forests have accumulated enough fuel, the occurrence, intensity, and size of stand-replacing fires depend greatly on weather conditions. The scale of stand-replacing fires ranges from several hectares up to hundreds of thousands (or even millions) of hectares. Within a single burned area, variations in fuels, topography, and winds typically lead to a variety of fire intensities and subsequent implications for ecosystem recovery. These ranges in scale are illustrated well by the work of W. Romme and his colleagues in Yellowstone National Park (Figs. 18.3, 18.4). The extent of fires (and the resultant stand size) appeared to depend strongly on weather conditions. The proportion of an area burned during 20-year intervals (Fig. 18.4) showed great variation, again largely resulting from weather patterns. The very dry summer of 1988 led to fires that burned (to varying degrees) about one-third of the 800 000 ha Park. Climate varies on time scales that are shorter than the life-span of many temperate forests. For example, Johnson and Larsen (1991) examined the fire return interval in an area of 500 km2 of the Canadian Rocky Mountains. Prior to 1730, the climate was warm and dry, with an average fire return interval of about 50 years for forests consisting predominantly of lodgepole pine. Cooler and moister conditions lengthened the return interval to 90 years. Few data are available for the distribution of fire sizes in temperate forests. One of the best data sets comes from the National Forests of the Sierra Nevada range in California (McKelvey and Busse, 1996). Between 1900 and 1992, about 20 to 30%

458

Dan BINKLEY

Fig. 18.3. Map of lodgepole pine/subalpine fir (Abies lasiocarpa) stands in an area of 73 km2 in Yellowstone National Park (Wyoming, U.S.A.), as reconstructed for 1778. Number = stage of stand development (1, post-fire; 2, seedling/sapling; 3, immature; 4, mature; 5, transition; 6/7, old-growth; 8, meadows; 9, riparian; 10, steep slopes); approximate stand ages in brackets. Average patch size of historic fires varied substantially (from Romme, 1982), for both the pre-European settlement period (prior to 1870) and during a period of intensive fire suppression (1900–1960).

of the landscape at elevations between 500 m and 1200 m burned, compared with 5–10% of the landscape between 1500 m and 2200 m. The majority of the areas burned were consumed in fires of modest size (50% (Keenan and Kimmins, 1993). Burning of organic matter after harvesting can remove very large quantities of nutrients (Fig. 18.2), sometimes more than the quantities removed in biomass products. More careful application of intensive site-preparation treatment usually increases the growth of crop trees, but whether this increased growth results from increased site fertility or simply from reduced competition with non-crop tree vegetation (summarized in Walstad and Kuch, 1987) remains largely undetermined. Where sitepreparation treatments include ditches to drain excess

462

water and improve soil aeration, substantial increases in total ecosystem productivity result (Binkley, 1986). The removal of nutrients in biomass may lead to lower productivity following the harvesting disturbance, but empirical data are surprisingly limited. Nutrient removals depend strongly on the types of products removed from a site; stem-only harvests typically remove about half the quantities of nutrients that are lost in whole-tree operations (Binkley, 1986). Forest fertilization is a widespread practice, both at the time of regeneration and later in the forest rotation. The disturbances associated with forest harvest often lead to increased rates of soil erosion, much of the increase depending on characteristics of road construction, and the degree of disturbance of the forest floor (Binkley, 1986; Binkley and Brown, 1993). In general, erosion rates after harvesting are too low to affect the productivity of the next generation of forest, but often high enough to reduce water quality (especially in association with poor road design). The leaching losses of nutrients typically increase as a result of harvest disturbances, but these increases are always far smaller than the nutrient losses in harvested biomass or through fire (Binkley, 1986). Forest-management programs may be designed to ensure the maximum yield of wood production. Natural sources of mortality are minimized by control of competition (including tree harvest to reduce competitiondriven mortality), and active suppression of fire and insects. Intensively managed forests probably lead to harvest of >80% of the wood produced by the forest, with other disturbance factors accounting for 50 yr, depending on the proportion of trees removed and the parameter of interest. Leaf area and productivity of trees may recover as quickly as 6 years after complete clearcutting for some hardwoods (Zavitkovski and Newton, 1971; Covington and Aber, 1980), or more typically 20–70 years for many conifers (Switzer et al., 1966; Turner and Long, 1975; Long and Smith, 1992; Ryan et al., 1997). These rates are probably similar to those following some natural disturbances (such as windthrow), but faster than some (such as severe wildfire). Silvicultural regimes may also be aimed to

Dan BINKLEY

accelerate the development of old-growth structure in younger forests (Newton and Cole, 1987) by selectively removing smaller competing trees to increase the growth of larger trees. Silvicultural regimes may include green tree retention as a part of clear-cutting operations to increase the similarity between clearcutting and natural disturbances (such as fires) that tend to leave some surviving trees (Rose and Muir, 1997; Franklin et al., 1997). The effects of harvesting on the diversity of the subsequent plant community depends on many factors, such as post-harvest site preparation (using fire, heavy machinery, herbicides), planting of selected species, and management of the density (and overall structure) of the population of regenerating trees (Keenan and Kimmins, 1993). Overall, harvesting tends to foster increased diversity of species other than trees for a period that typically lasts until full canopy closure, after which diversity may be reduced owing to heavy shade. This suggestion of a general trend has so many exceptions, however, that it may not be useful. Structural diversity of the forest may be increased by harvesting where single trees or small patches are harvested, or greatly reduced where large cutting units are regenerated with even-aged plantations. Generalization about the effects of disturbances on wildlife is not possible. Individual species may be favored or harmed by disturbances (Patton, 1992; Harris and Harris, 1997). Species relying on high productivity and diversity of understory plants typically increase in population after disturbances that reduce habitat suitability for species with strong affinities for old-growth forests. The response of a species to disturbance in one portion of its range may not apply to other situations or areas. Given this breadth of wildlife responses, useful insights are likely to be developed only for specific situations, and then only when considered in a landscape context (Keenan and Kimmins, 1993). Another difference between natural stand-replacing disturbances and forest management is the landscape scale of forestry operations. Forest cutting operations tend to have well-defined boundaries, whereas many natural disturbances (such as fire) have irregular and fuzzy boundaries (Hunter, 1990). In the past, most forestry practices have emphasized objectives such as yield of wood products, and, in some operations, maintenance of diverse habitats across landscapes. Spies et al. (1994) provide one of the few quantitative assessments of the effects of forest practices on

DISTURBANCE IN TEMPERATE FORESTS OF THE NORTHERN HEMISPHERE

463

Fig. 18.6. Distribution of the sizes of patches in a 2600 km2 landscape in Oregon, in 1972 and 1988, showing substantial reductions in the average size of the interior closed-canopy forests (>100 m from forest edge) (A). The reduction in large patches corresponded with an increase in the total number of patches in the landscape (B) (from Spies et al., 1994).

landscape-scale patterns. These authors used Landsat imagery to characterize changes between 1972 and 1988 in a landscape in Oregon (U.S.A.) covering 2600 km2 . In 1972, much of this area contained oldgrowth forests (on both private and public lands), with mature, closed-canopy forests occupying 71% of the landscape. Logging over the next 16 years lowered the coverage of closed-canopy forests to 58%. The declines in forest cover were greater on privately owned lands (declining from 50% in 1972 to 28% in 1988) than on publicly owned, non-wilderness lands (where forest cover declined from 79% to 68%). These net changes were the balance between removals by logging (moving landscape units from closed forest to other

conditions), and addition by the succession of formerly cleared areas into closed-canopy conditions. Over the 16-year period, the average rate of conversion from closed-canopy forest was 1.7% yr−1 , and the rate of reestablishment of closed-canopy conditions was 1.3%. These changes in forest cover included major changes in the patch size of the landscape, and in the amount of edges between closed forest and other land types. With “edge” defined as 100 m of closed forest adjacent to another land type, the amount of edge increased from 9% to 13% of the landscape in the publicly owned land. The amount of interior closed forest (more than 100 m from an adjacent land type) declined from 60% to 42% on publicly owned lands, and from 33% to

464

12% on privately owned lands. Across the landscape, the amount of the total interior forest in large patches declined (Fig. 18.6), and the number of individual patches (typically of smaller size) increased. Once forest harvesting has generated a landscape pattern of edge and interior space, these patterns tend to be very persistent and difficult to erase (Wallin et al., 1994). Overall, this example demonstrated that forest harvesting creates novel (unprecedented) patterns of forest conditions at landscape scales, as compared with natural disturbance regimes. The implications of these novel changes are not at all clear. Recent research has provided new evidence of the effects of forest harvesting, including better insights about the effects of silvicultural activities on a broad range of forest features, about the spatial and temporal scales of impacts, and about the case-specific nature of many interactions. Current research and practice in forestry will continue to expand the knowledge base on the effects of harvesting disturbances, including retention of live trees (aggregated and dispersed across a harvesting unit), varying rotation lengths, mixtures of species, and interactions of wildlife, plant diversity, and wood production [see the broad range of chapters in Kohm and Franklin (1997)]. The application of this new knowledge to land management activities will, of course, depend on social and political choices.

CONCLUSIONS

Disturbances are fundamental to temperate forests, across scales of time (from seasons to millennia) and space (from single trees to landscapes). The impact of disturbances depends on the interactions of many ecological factors, and these complex interactions prevent any simple characterization of likely changes in forests. Human-related disturbances have come to dominate most of the temperate-forest landscapes, through harvesting trees (and influencing revegetation), suppressing fire, introducing exotic species (including pests and pathogens), and altering landscape patterns (size and distribution of stands). Forests are generally robust relative to these disturbances, but the character of the forests after human disturbance typically differs substantially from those following natural disturbances. Major differences include increased production of wood, narrower distributions of age-classes across landscapes, and altered diversity of species and stand structures within the forests.

Dan BINKLEY REFERENCES Agee, J.K., 1993. Fire Ecology of Pacific Northwest Forests. Island Press, Washington, DC., 493 pp. Archibold, O.W., 1995. Ecology of World Vegetation. Chapman and Hall, London, 510 pp. Barrett, J.W., 1995. Regional Silviculture of the United States. Wiley, New York, 643 pp. Barrett, S.W. and Arno, S.F., 1982. Indian fires as an ecological influence in the northern Rockies. J. For., 80: 647–651. Binkley, D., 1986. Forest Nutrition Management. Wiley, New York, 290 pp. Binkley, D. and Brown, T., 1993. Management Impacts on Water Quality of Forests and Rangelands. USDA Forest Service GTRRM-239, 114 pp. Binkley, D., Richter, D., David, M.B. and Caldwell, B., 1992. Soil chemistry in a loblolly-longleaf pine forest with interval burning. Ecol. Appl., 2: 157–164. Campbell, R.E., Baker Jr., M.B. and Ffolliott, F., 1977. Wildfire effects on a ponderosa pine ecosystem: an Arizona case study. USDA Forest Service Research Paper RM-191, Ft. Collins, Colorado. Cannell, M.G.R., 1982. World Forest Biomass and Primary Production Data. Academic Press, London, 389 pp. Christensen, N.L., 1987. The biochemical consequences of fire and their effects on the vegetation of the Coastal Plain of the southeastern United States. In: L. Trebaud (Editor), The Role of Fire in Ecological Systems. SPB Academic Publishing, The Hague, pp. 1–21. Christensen, N.L. and Muller, C.H., 1975. Effects of fire on factors controlling plant growth in Adensotoma chaparral. Ecol. Monogr., 45: 29–55. Clark, J.S. and Robinson, J., 1993. Paleoecology of fire. In: P.J. Crutzen and J.G. Goldammer (Editors), Fire in the Environment: The Ecological, Atmospheric, and Climatic Importance of Vegetation Fires. Wiley, New York, pp. 193–214. Covington, W.W., 1994. Implications for ponderosa pine/bunchgrass ecological systems. In: Sustainable Ecological Systems: Implementing an Ecological Approach to Land Management. USDA Forest Service GTR-RM-247, Ft. Collins, Colorado. Covington, W.W. and Aber, J.D., 1980. Leaf production during secondary succession in northern hardwoods. Ecology, 61: 200–204. Covington, W.W. and Moore, M.M., 1994. Postsettlement changes in natural fire regimes and forest structure: ecological restoration of old-growth ponderosa pine forests. In: R.N. Sampson and D.L. Adams (Editors), Assessing Forest Ecosystem Health in the Inland West. Haworth Press, New York, pp. 153–181. Covington, W.W. and Sackett, S., 1984. The effect of a prescribed fire in Southwestern ponderosa pine on organic matter and nutrients in woody debris and forest floor. For. Sci., 30: 183–192. Crams, J.S. and DeBano, L.F., 1965. Soil wettability: a neglected factor in watershed management. Water Resourc. Res., 1: 283–286. Darby, H.C., 1956. The clearing of woodland in Europe. In: W.L. Thomas (Editor), Man’s Role in Changing the Face of the Earth. University of Chicago Press, Chicago, pp. 183–216. Dyck, W.J., Mees, C.A. and Comerford, N.B., 1989. Medium-term effects of mechanical site preparation on radiata pine productivity in New Zealand – a retrospective approach. In: W.J. Dyck and

DISTURBANCE IN TEMPERATE FORESTS OF THE NORTHERN HEMISPHERE C.A. Mees (Editors), Research Strategies for Long-term Site Productivity, Bulletin 152. Forest Research Institute, Rotorua, New Zealand. Ewel, J., Berish, C., Brown, B., Price, N. and Raich, J., 1981. Slash and burn impacts on a Costa Rican wet forest site. Ecology, 62: 816–829. FAO, 1995. Forest Resources Assessment, 1990: Global Sythesis. UN Food and Agriculture Organization, Rome. Franklin, J., 1988. Pacific Northwest forests. In: M.G. Barbour and W.D. Billings (Editors), North American Terrestrial Vegetation. Cambridge University Press, Cambridge, pp. 103–130. Franklin, J., Berg, D.R., Thornburgh, D.A. and Tappeiner, J.C., 1997. Alternative silvicultural approaches to timber harvesting: variable retention harvest systems. In: K.A. Kohm and J.F. Franklin (Editors), Creating a Forestry for the 21st Century. Island Press, Washington, D.C., pp. 111–149. French, R.A., 1963. The making of the Russian landscape. Adv. Sci., 20: 44–56. French, R.A., 1983. Russians and the forest. In: J.H. Bater and R.A. French (Editors), Studies in Russian Historical Geography. Academic Press, London. Gregory, S.V., 1997. Riparian management in the 21st Century. In: K.A. Kohm and J.F. Franklin (Editors), Creating a Forestry for the 21st Century. Island Press, Washington, D.C., pp. 69–85. Gresham, C.A., Williams, T.M. and Lipscomb, D.J., 1991. Hurricane Hugo wind damage to southeastern U.S. coastal forest tree species. Biotropica, 23: 420–426. Harris, E. and Harris, J., 1997. Wildlife Conservation in Managed Woodlands and Forests. Wiley, New York, 342 pp. Heinselman, M.L., 1981. Fire intensity and frequency as factors in the distribution and structure of northern ecosystems. In: Fire Regimes and Ecosystem Properties. USDA Forest Service GTRWO-26, Washington, DC, pp. 7–57. Hungerford, R., 1980. Microenvironmental response to harvesting and residue management. In: Environmental Consequences of Timber Harvesting in Rocky Mountain Coniferous Forests. USDA Forest Service General Technical Report INT-90, Ogden, Utah, pp. 37–74. Hunter Jr., M.L., 1990. Wildlife, Forests, and Forestry. Regents/ Prentice Hall, Englewood Cliffs, 370 pp. Johnson, E.A. and Larsen, C.P.S., 1991. Climatically induced change in fire frequency in the southern Canadian Rockies. Ecology, 72: 194–201. Jurgensen, M.F., Harvey, A.E. and Larsen, M.J., 1981. Effects of Prescribed Fire on Soil Nitrogen Levels in a Cutover Douglasfir/western Larch Forest. USDA Forest Service Research Paper INT-275, Ogden, Utah. Kauffman, J.B., Christensen, N.L., Goldammer, J.G., Justice, N.L., May, J., Pyne, S.J., Stocks, B.J., Trabaud, L.V., Trollope, W., Weiss, K. and Williams, M., 1993. Group report: the role of humans in shaping fire regimes and ecosystem properties. In: P.J. Crutzen and J.G. Goldammer (Editors), Fire in the Environment: The Ecological, Atmospheric, and Climatic Importance of Vegetation Fires. Wiley, New York, pp. 375–388. Keenan, R.J. and Kimmins, J.P., 1993. The ecological effects of clearcutting. Environ. Rev., 1: 121–144. Kimmins, J.P., 1992. Balancing Act: Environmental Issues in Forestry. UBC Press, Vancouver, 244 pp. Knight, D.H., Yavitt, J.B. and Joyce, G.D., 1991. Water and nitrogen

465

outflow from lodgepole pine forest after two levels of tree mortality. For. Ecol. Manage., 46: 215–225. Kodama, H. and Van Lear, D.H., 1980. Prescribed burning and nutrient cycling relationships in young loblolly pine plantations. South. J. Appl. For., 4: 118–121. Kohm, K.A. and Franklin, J.F. (Editors), 1997. Creating a Forestry for the 21st Century. Island Press, Washington, D.C., 475 pp. Lertzman, K.P. and Krebs, C.J., 1991. Gap-phase structure of a subalpine old-growth forest. Can. J. For. Res., 21: 1730–1741. Little, S. and Ohmann, J., 1988. Estimating nitrogen lost from forest floor during prescribed fires in Douglas-fir/western hemlock clearcuts. For. Sci., 34: 152–164. Long, J. and Smith, F., 1992. Volume increment in Pinus contorta var. latifolia: the influence of stand development and crown dynamics. For. Ecol. Manage., 53: 53–64. MacCleery, D.W., 1992. American Forests: A History of Resiliency and Recovery. USDA Forest Service FS-540, Washington, DC., 59 pp. Martin, R.E., 1982. Fire history and its role in succession. In: J.E. Means (Editor), Forest Succession and Stand Development Research in the Northwest. Oregon State University, Corvallis, pp. 92–99. McKelvey, K.S. and Busse, K.K., 1996. Twentieth-Century fire patterns on Forest Service lands. In: Sierra Nevada Ecosystem Project: Final Report to Congress, Vol. II, Assessments and Scientific Basis for Management Options. University of California, Centers for Water and Wildland Resources, Davis, pp. 1119– 1138. Newton, M. and Cole, E.C., 1987. A sustained-yield scheme for old-growth Douglas-fir. West. J. Appl. For., 2: 22–25. Nissley, S., Zasoski, R. and Martin, R., 1980. Nutrient changes after prescribed surface burning of Oregon ponderosa pine stands. In: Proceedings Sixth Conference on Fire and Forest Meteorology. Society of American Foresters, Washington, DC, pp. 214–219. Nyland, R.D., 1996. Silviculture: Concepts and Applications. McGraw-Hill, New York, 633 pp. Oliver, C.D. and Larson, B.C., 1996. Forest Stand Dynamics. Wiley, New York, 520 pp. Olson, J., 1981. Carbon balance in relation to fire regimes. In: Fire Regimes and Ecosystem Properties. USDA Forest Service GTRWO-26, Washington, DC., pp. 327–378. Ovington, J.D. (Editor), 1983. Temperate Broad-leaved Evergreen Forests. Ecosystems of the World 10. Elsevier, Amsterdam, 241 pp. Patton, D.R., 1992. Wildlife Relationships in Forested Ecosystems. Timber Press, Portland, 392 pp. Pritchett, W.L. and Fisher, R.F., 1987. Properties and Management of Forest Soils. Wiley, New York, 494 pp. Pyne, S.J., 1993. Keeper of the flame: a survey of anthropogenic fire. In: P.J. Crutzen and J.G. Goldammer (Editors), Fire in the Environment: The Ecological, Atmospheric, and Climatic Importance of Vegetation Fires. Wiley, New York, pp. 245–266. Pyne, S.J., Andrews, P.L. and Laven, R.D., 1996. Introduction to Wildland Fire. Wiley, New York, 769 pp. Rackham, O., 1986. The History of the British and Irish Countryside. Dent, London. Richter, D., Ralston, C. and Harms, W., 1982. Prescribed fire: effects on water quality and forest nutrient cycling. Science, 215: 661–663.

466 R¨ohrig, E. and Ulrich, B. (Editors), 1991. Temperate Deciduous Forests. Ecosystems of the World 7. Elsevier, Amsterdam, 635 pp. Romme, W.H., 1982. Fire and landscape diversity in subalpine forests of Yellowstone National Park. Ecol. Monogr., 52: 199–221. Romme, W.H. and Despain, D.G., 1989. Historical perspective on the Yellowstone fires of 1988. BioScience, 39(10): 695–699. Rose, C.R. and Muir, P.S., 1997. Green-tree retention: consequences for timber production in forests of the western Cascades, Oregon. Ecol. Appl., 7: 209–217. Ryan, M.G., Binkley, D. and Fownes, J.H., 1997. Age-related decline in forest productivity: pattern and processes. Adv. Ecol. Res., 27: 213–262. Sasser, C.L. and Binkley, D., 1989. Nitrogen mineralization in highelevation forests of the Appalachians. II. Patterns with stand development in fir waves. Biogeochemistry, 7: 147–156. Schmidt, T.L., Spencer Jr., J.S. and Hansen, M.H., 1996. Old and potential old forest in the Lake States, USA. For. Ecol. Manage., 86: 81–86. Schoch, P. and Binkley, D., 1986. Prescribed burning increased nitrogen availability in a mature loblolly pine stand. For. Ecol. Manage., 14: 13–22. Shvidenko, A. and Nilsson, S., 1997. Are the Russian forests disappearing? Unasylva, 48: 57–64. Sierra Nevada Ecosystem Project, 1996. Status of the Sierra Nevada, Vol. I, Assessment Summaries and Management Strategies, Sierra Nevada Ecosystem Project, Final Report to Congress. University of California, Centers for Water and Wildland Resources, Davis, California. Smith, D.M., Larson, B.C., Kelty, M.J. and Ashton, P.M.S., 1997. The Practice of Silviculture: Applied Forest Ecology. Wiley, New York, 537 pp. Spies, T.A., Ripple, W.J. and Bradshaw, G.A., 1994. Dynamics and pattern of a managed coniferous forest landscape in Oregon. Ecol. Appl., 4: 555–568. Stocks, B.J. and Simard, A.J., 1993. Forest Fire Management in Canada. Disaster Manage., 5: 21–27. Swank, W.T., 1988. Stream chemistry response to disturbance. In: W.T. Swank and D.A. Crossley Jr (Editors), Forest Hydrology and Ecology at Coweeta. Springer-Verlag, New York, pp. 339–357. Swanson, F.J., 1981. Fire and geomorphic processes. In: Fire Regimes and Ecosystem Properties. USDA Forest Service GTR WO-26, Washington, DC, pp. 421–444. Switzer, G.L., Nelson, L.E. and Smith, W.H., 1966. The

Dan BINKLEY characterization of dry matter and nitrogen accumulation by loblolly pine (Pinus taeda L.) Soil Sci. Soc. Am. Proc., 30: 114–119. Turner, J. and Long, J.N., 1975. Accumulation of organic matter in a series of Douglas-fir stands. Can. J. For. Res., 5: 681–690. United States Department of Agriculture Forest Service, 1981. An Assessment of the Forest and Range Land Situation in the United States. Forest Resource Report 22, Washington, DC. Veblen, T.T., Hadley, K.S., Reid, M.S. and Rebertus, A.J., 1991. The response of subalpine forests to spruce beetle outbreak in Colorado. Ecology, 72: 213–231. Wallin, D.O., Swanson, F.J. and Marks, B., 1994. Landscape pattern resonse to changes in pattern generation rules: land-use legacies in forestry. Ecol. Appl., 4: 569–580. Walstad, J.D. and Kuch, P.J. (Editors), 1987. Forest Vegetation Management for Conifer Production. Wiley, New York, 523 pp. Wells, S.G., 1987. The effects of fire on the generation of debris flows in southern California. Geol. Soc. Am. Rev. Eng. Geol., 7: 105–114. White, P.S. and Pickett, S.T.A., 1985. Introduction. In: S.T.A. Pickett and P.S. White (Editors), The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, pp. 3–13. White, W.D. and Wells, S.G., 1981. Geomorphic effects of the La Mesa fire. In: The La Mesa fire symposium. Los Alamos National Laboratory LA-9236-NERP, pp. 73–90. Whitney, G., 1994. From Coastal Wilderness to Fruited Plain: A History of Environmental Change in Temperate North America from 1500 to the Present. Cambridge University Press, Cambridge. Wilbur, R.B., 1985. The effects of fire on nitrogen and phosphorus availability in a North Carolina coastal plain pocosin. Dissertation, Duke University. Williams, M., 1989. Americans and Their Forests. Cambridge University Press, Cambridge. Williams, M., 1994. Forests and tree cover. In: W.B. Meyer and B.L. Turner II (Editors), Changes in Land Use and Land Cover: A Global Perspective. Cambridge University Press, Cambridge, pp. 97–124. Wright, H.A. and Bailey, A., 1982. Fire Ecology: United States and Southern Canada. Wiley, New York. Zavitkovski, J. and Newton, M., 1971. Litterfall and litter accumulation in red alder stands in western Oregon. Plant Soil, 35: 257–268.

Chapter 19

“Ecology becomes a more complex but far more interesting science when human aspirations are regarded as an integral part of the landscape.” (Ren´e Dubos, 1980)

ANTHROPOGENIC DISTURBANCE AND TROPICAL FORESTRY: IMPLICATIONS FOR SUSTAINABLE MANAGEMENT G.S. HARTSHORN and J.L. WHITMORE

INTRODUCTION

Anthropogenic disturbance in tropical-forest landscapes spans a huge spatial range as well as an intensity gradient. For example, log extraction by heavy machinery damages many standing trees and the fragile soil (Johns et al., 1996). At the other extreme, mega-scale clearing of tropical forests in Brazil for commercial plantations (Moran, 1981; McNabb et al., 1997), cattle ranching (Uhl and Buschbacher, 1987; Serrao et al., 1996), and surface mining (Parrotta et al., 1997a) have produced extensive deforested landscapes (World Resources Institute, 1992). Even islands of intact forest in a sea of disturbance are influenced by the severity and type of disturbance around them (Lovejoy and Bierregaard, 1990; Bierregaard et al., 1992; Turner, 1996). The several thousands of years during which indigenous peoples living in tropical-forest regions (G´omezPompa et al., 1987; Bush et al., 1992) were long believed to have had negligible effects on those forests and their biodiversity. Except in urban centers such as those developed by the Mayan culture in northern Mesoamerica, it has been thought that indigenous peoples were present at low population densities and subsisted by harvesting local resources (particularly game, fish, and fruit) and practicing shifting cultivation (Fig. 19.1) of staple crops [manioc (Manihot spp.), maize (Zea mays)]. Even in wet forest areas, shifting cultivation with an ample fallow period (>20 yr) allowed natural rebuilding of the patch of felled forest. Recent paleoecological evidence (charcoal, pollen, and maize phytoliths) from the Dari´en area of Panama – a famous wilderness area rich in endemic species

(Gentry, 1982) – indicates a 4000-year history of human disturbance and settlement (Bush and Colinvaux, 1994). The data are consistent with depopulation and abandonment of most agricultural areas following the conquest by the Spanish, Portuguese, and other Europeans. The well-developed forests appear to have regrown within 350 years (Budowski, 1970). Nevertheless, the extraordinary biodiversity of the Dari´en area seems not to have been seriously affected by 4000 years of indigenous agriculture. Disturbance in tropical-forest landscapes has become an important focus for ecology and management over the past few decades, largely because disturbance is increasingly viewed as a normal process in most forest ecosystems (Pickett and White, 1985). Foresters and land stewards have long managed forests and trees not only for commercial production of timber, but often to assist or enhance site recovery after disturbance (Parrotta et al., 1997b). Recognition of disturbance regimes has become integral to improved understanding of tropical-forest dynamics and potential land uses. Natural disturbance regimes are the basis for developing more ecologically sound models for managing complex tropical forests (Hartshorn, 1995). A great deal of ecological research and environmental awareness about tropical forests has been generated over the last 30 years. But often it has fallen short in offering solutions to local issues. It is only at this local level that real solutions can be implemented. Local issues must be addresssed one by one and resolved as unique problems, rather than viewing them from afar and trying to prescribe global approaches. The critical questions concern the interactions and synergies of disturbance as they influence the growth,

467

468

G.S. HARTSHORN and J.L. WHITMORE

Fig. 19.1. Traditional shifting cultivation of subsistence crops where a small patch of forest is cut and burned (upper photo). Note that the clearings are few and well-spaced in the matrix of forest. Hand-held aerial photo near San Carlos de R´ıo Negro, Amazonian Venezuela. The lower photo shows typical slash-and-burn advancement of the agricultural frontier into tropical forests. Note that older fields are not abandoned and new fields are not isolated clearings in the forest matrix. Hand-held aerial photo in the Chapare of Amazonian Bolivia (Hartshorn).

TROPICAL FORESTRY AND DISTURBANCE

reproduction, and mortality of the trees comprising forest communities. Not only does disturbance, sensu lato, play a prominent role in natural forests, but it also is of considerable importance to tree plantations. Most of the tree species commonly favored in plantation forestry are early or mid-successional – that is, they require some degree of prior site disturbance to thrive. Tree plantations usually require site preparation, weeding during the early years, and thinning after crown closure, all of which constitute programmed disturbance (Homfray, 1936; Evans, 1992). This chapter focuses on anthropogenic disturbances in tropical-forest ecosystems (including natural forests and woodlands, as well as tree plantations) and their implications for sustainable management of these important natural resources. The tropics are considered broadly here, corresponding to that part of the planet that has supported tropical forests and woodlands over the past few centuries (cf. Richards, 1996). More important than latitude is the moisture regime of tropical-forest and woodland ecosystems, which may span a range of mean annual precipitation from well below 1000 mm to over 10 000 mm. Seasonality of rainfall is usually a principal determinant of forest type as well as the choice of tree species for plantation forestry (Evans, 1992). Inter-annual variations in rainfall seasonality are often significant factors in the dynamic biological processes and ecological interactions in tropical forests. Definitions Global concerns about tropical deforestation, compatibility of forestry and conservation, and sustainability of forest management practices, inter alia, require that the terminology used should be as clear as possible. Therefore, we offer several brief definitions to aid in understanding what we mean when we use terms like “the tropics,” “forest management,” or “enrichment planting.” The tropics and subtropics include most of the land within the equatorial latitudes 23.5º north and south, including higher altitudes and drier zones. The tropics are not only the lowland, wetter areas such as the classic “rainforests” of the Amazon and Congo basins, but also the much more extensive seasonally dry areas peripheral to the core of wetter forests. The subtropics (roughly from 12–13º to 23–25º latitude) are more prone to tropical cyclones (including hurricanes), tend to be more seasonal in rainfall and temperature,

469

and have generally lower species richness than the inner tropics (Holdridge, 1967; Migenis and Ackerman, 1993). Thus, the scope of this chapter is necessarily broad, including a wide variety of ecosystems, soil types and climatic regimes, as well as native and introduced tree species. Forestry is the science, business, and art of creating, conserving, and managing forests and forest lands for the continuing or sustainable use of these resources (Ford-Robertson, 1971). Forestry is one of the key tools in management of lower-latitude forest landscapes. Without appropriate forestry practices in response to or in coordination with anthropogenic disturbances, these forest resources will not survive, as has been repeatedly demonstrated throughout the forested tropics over the last few decades. The determination of which forestry technique (or techniques) might be appropriate in a specific forest depends on several factors, such as the present condition of the resource, past land use or abuse, disturbance frequency, commercial pressures, poverty levels, ratio of land to population, land tenure, and management objectives. Forest management is the practical application of scientific, social, and economic principles to the administration and working of forest land for specified objectives (Ford-Robertson, 1971). Particularly since World War II, deforestation in the tropics has become an increasingly serious problem. Most of the harvesting of tropical hardwoods and the conversion of forests to other land uses has occurred in the absence of forest management (Wadsworth, 1997). It should be noted that prior to 1950 the temperate forest region suffered far greater deforestation than did the tropics (Dubos, 1980), often also in the absence of forest management. Silviculture is applied forest ecology – that is, application of the theory and practice of controlling forest establishment, composition, structure, and growth, for commercial or other objectives (Ford-Robertson, 1971; Smith et al., 1997). Extensive silviculture relies mainly on natural regeneration, sometimes with enrichment planting (Spurr, 1979). Ecologically-sound examples in the tropics exist in Trinidad (Clubbe and Jhilmit, 1992), Surinam (Vega Condori, 1987), Peru (Hartshorn and Pariona, 1993, 1997), Malaysia and Ghana (Moad and Whitmore, 1994). Intensive silviculture on the other hand usually involves tree planting. Productive commercial plantations in the tropics include: eucalypts in Congo (formerly Zaire) and Brazil; pine plantations by the company AMCEL in the Brazilian Amazon, at Usutu in Swaziland, and in Fiji; and Swietenia in

470

G.S. HARTSHORN and J.L. WHITMORE

Table 19.1 Ten top ranked tropical countries for area of natural forests and for area of plantation forests 1 Country

Bangladesh

Total area (km2 )

Natural forests 1990 (km2 )

Rank

Plantation forests 1990 (km2 )

142 776

7 690

Bolivia

1 098 582

493 170

7

2350 280

Brazil

8 506 663

5 611 070

1

49 000

Colombia

1 138 339

540 640

5

1260

Rank

8 3

Cuba

114 494

17 150

2450

7

India

3 287 588

517 290

6

132 300

1

Indonesia

2 042 034

1 095 490

3

61 250

2

Mexico

1 972 546

485 860

8

1090

678 034

288 560

Peru

1 285 215

679 060

4

Sudan

2 505 809

429 760

10

Thailand

513 998

127 350

Venezuela

912 050

456 900

Myanmar

Vietnam Zaire (Congo) 1

332 569

83 120

2 344 113

1 132 750

2350

9 2

8

1840 2030

10

5290

5

2530

6

14 700

4

420

Data from World Resources Institute (1996).

Indonesia and Fiji (Evans, 1992). Silvicultural systems turn the harvest of wood into a tool for establishing the next forest. Plantation silviculture has a much longer history than natural forest management in the tropics. Selective logging is a non-technical term currently in vogue. It is also called selective cutting, creaming, or high-grading (Ford-Robertson, 1971), and refers to commercial harvest of logs, usually only the very best, with no regard for the future of the forest. The selection system, on the other hand, is a silvicultural system that harvests some trees in all size classes either singly, in small groups or strips, in order to promote natural regeneration and stand rebuilding on a nearly continuous basis. The objective is maintenance of an uneven-aged stand – that is, sustainable forest management close to the natural condition – while extracting wood products (Burns, 1983; Society of American Foresters, 1994). Due to the similarity of terms, selective logging and the selection system of harvest tend to be confusing, but they obviously represent quite different practices. A plantation is a forest crop or stand raised artificially, either by sowing of seed, or planting of seedlings or vegetatively-grown planting stock (FordRobertson, 1971). In 1990, there were 30.8×106 ha of tropical timber plantations (FAO, 1995); this statistic

excludes non-timber tree crops such as cacao (Theobroma cacao) or coffee (Coffea arabica). Tropical timber plantations are dominated by three favorites – eucalypts (Eucalyptus spp.), pines (Pinus spp.) and teak (Tectona grandis). In comparison to the FAO estimate of 1987×106 ha of tropical forests in 1990 (Lanly, 1995), tropical tree plantations cover very little (1.4%) of the tropical-forest landscape (Table 19.1). Despite this huge discrepancy in proportional areas of timber plantations and of natural forests, plantation forestry is the principal source of industrial wood. Plantation silviculture for environmental or protection objectives is also practiced. Examples include eucalypts for windbreaks in Sudan, Prosopis for reclamation in the coastal desert of Peru, Casuarina for quarry reclamation in Kenya, pines in Malawi to protect catchments, and a wide variety of agroforestry plantings to promote better crop production or protect sources of irrigation water. When local communities depend on the forest for products, such plantations tend to yield fuelwood and other wood-based benefits, but only as secondary goals (Evans, 1992). Enrichment planting involves disturbance of the existing forest cover, usually in the form of a line cut through secondary or degraded forest, to facilitate planting of preferred species. After establishment, the

TROPICAL FORESTRY AND DISTURBANCE

canopy is further opened to promote faster growth of the planted trees. The intervening understory vegetation is usually altered only minimally in order to reduce the competition from weedy species, while keeping costs down. Coppice is a regeneration method in which standing trees are cut and subsequent crops originate mainly from adventitious or dormant buds on living stumps, but also as suckers from roots and rhizomes (Burns, 1983). This system is only appropriate for those species that coppice naturally after disturbance. Eucalyptus globulus in the Andes is often regenerated by coppicing. Agroforestry is a collective name for all land-use systems and practices where woody perennials are deliberately grown on the same land-management unit as crops and/or domestic animals, either in spatial mixture or in time sequence, assuming there are both ecologic and economic interactions between the woody and the non-woody components (Khurana and Khosla, 1993). For the past few decades, agroforestry has been a critical component in the efforts to lessen the disturbance effects of traditional agricultural practices (e.g., slash and burn) in the tropics. Many agroforestry practices are based on some of the more successful models developed by indigenous peoples over the centuries (or millennia) that integrate tree planting with agriculture (Wint, 1978; Denevan et al., 1987; Vayda, 1987). Old-growth forest is usually considered to be synonymous with virgin or primary forest, and to be undisturbed by humans (Clark, 1995). This has become increasingly contentious as widespread evidence of human occupation or use has been found even in remote tropical forests (Horn and Sanford, 1993). Although many tropical forests may have been disturbed by indigenous peoples, the scale and intensity of disturbance is difficult to quantify. Most old-growth forests have well-developed ecosystem functions (e.g., tight nutrient cycling, gap dynamics) and harbor impressive biodiversity in spite of previous disturbance. Secondary forest traditionally refers to forest regrowth after catastrophic disturbance. In the past few decades, however, there has been increasing use of this term to describe logged forest (e.g., Brown and Lugo, 1990). Corlett (1994) has asked the key question, “How much disturbance is necessary to make a forest secondary?” Logged forests usually still maintain significant functions as ecosystems and for conservation of biodiversity, thus they are excellent

471

examples of the difficulty in differentiating old-growth from late secondary forest types along a disturbance gradient. In contrast, early to late secondary forests can have significantly lower biodiversity than is seen in old-growth forests (Young and Wang, 1989; Zou et al., 1995). One hears a lot about tropical deforestation, particularly in the temperate zone. There certainly are a series of local forest management problems in tropical countries which merit immediate attention, and which, when viewed nationally or regionally, constitute a serious global problem. Each of these local problems is different, with solutions unique to each locality, and must be resolved at the local level. Unfortunately, this chapter by necessity must address the issue broadly. But the far more interesting endeavor is to analyze and solve a particular forest management problem on the ground. Too much global analysis, with too little local application of current knowledge to specific problems, is a major failing. Armchair resource management might be a good term for this phenomenon; it has become an important issue in dialogue between developed and less-developed countries.

DISTURBANCE EFFECTS

The causes and effects of disturbance in tropical forests are numerous and complex. A review of natural disturbance phenomena in tropical forests is beyond the scope of this chapter. There is a considerable body of scientific literature on specific aspects, such as scale (cf. Nelson et al., 1994; van der Meer and Bongers, 1996), magnitude (cf. Janzen, 1988; Murphy and Lugo, 1995), frequency (cf. Hartshorn, 1992; Walsh, 1996), and fragmentation (cf. Vitousek, 1988; Lovejoy and Bierregaard, 1990; Turner, 1996). For more general reviews of natural disturbances in tropical forests, the interested reader is referred to Johns and Skorupa (1987), Arriaga (1988), Appanah (1993), Inoue et al. (1993), Alvarez-Buylla (1994), Boose et al. (1994), Bennett and Dahaban (1995), Denslow (1995), Thiollay (1996) and Walker et al. (1996). In this section we review the interactions between anthropogenic disturbances and tropical forests. Impacts to soils and soil organisms The physical effects of disturbance are often most pronounced on soils. Common phenomena such as

472

logging, slash-and-burn agriculture, and conversion to pasture may have profound effects on both physical and chemical properties of soils (Malmer and Grip, 1994; McNabb et al., 1997). Physical effects include compaction, increased bulk density, reduced organic matter, and greater erosion. Chemical effects include loss of nutrients, higher acidity, and lower base saturation. Timber harvesting – particularly the use of heavy machinery to extract logs – can cause serious compaction, rutting, and erosion of tropical soils. While the felling of one or a few trees and the creation of canopy openings are analogous to natural tree falls, the extraction of logs usually causes appreciable damage to soils and remaining trees. Estimates of logging damage range widely. In a detailed quantitative study of direct and indirect logging damage in Sarawak, Nussbaum et al. (1995) calculated that, of a total area of 300 ha surveyed, 5% was used for log landings, 25% for skid trails, 30% was occupied by debris piles, 20% was disturbed forest, and 20% was left as residual undisturbed patches. Damage estimates are comparable in the eastern Amazon (Uhl and Vieira, 1989). Congdon and Herbohn (1993), in northern Queensland (Australia), found that the effects of selective logging were still apparent after 25 years, disturbed soils having higher bulk densities and pH, and lower cation exchange capacity, and concentrations of Kjeldahl nitrogen and available phosphorus; their data suggest that recovery from selective logging is dependent on soil fertility and on the intensity of disturbance. Mycorrhizas play an integral role in most tropical forests, greatly enhancing nutrient uptake by trees through mutualistic associations (Janos, 1980). In west Malaysian forests, mycorrhizal infection of tree seedlings was reduced by 25% after selective logging and 75% after heavy logging (Alexander et al., 1992). Mycorrhizas are also important to the establishment of successional species in abandoned pastures or agricultural fields (Fischer et al., 1994). In the Ivory Coast, soil mycoflora varied with major soil types, but showed clear resilience to drought in 1982–83, recovering quickly after the return of rains (Maggi et al., 1990). Impacts on plants The abrupt opening of the forest canopy, whether by natural tree falls or the felling of timber trees, increases light levels reaching the forest floor below the canopy

G.S. HARTSHORN and J.L. WHITMORE

gap as well as in the forest understory bordering the gap (Williams-Linera, 1990; Denslow, 1995). For shadetolerant understory plants and seedlings, the abrupt exposure to direct sunlight may cause sun-scalding of leaves or even death. In an experimental study of the possible causes of mass mortality of understory seedlings, Lovelock et al. (1994) found that fatalities are due to a combination of photoinhibition with moisture stress. Shade-tolerant species show a greater degree of photoinhibition in forest gaps at midday than do shade-intolerant species. As some of these understory plants may be potential canopy species, their loss can influence the development of the future forest on that site. In theory, one might suppose that harvesting of a highly prized species would threaten or endanger that species; however, there is limited documentation of such a direct relationship. The true mahoganies (Swietenia spp.) are a noteworthy example. Littleleaf mahogany (S. mahagoni), native to the Caribbean region, has been commercially unavailable for several decades, yet this species is not threatened with biological extinction. Large-leaf mahogany (S. macrophylla), now the focal timber species harvested in the southern Amazon (Ver´ıssimo et al., 1995), also has extensive populations in northern Mesoamerica that have been exploited for at least three centuries (Lamb, 1966; Hartshorn et al., 1984), without this species being seriously threatened with extinction. Genetic erosion in such cases is potentially a problem, but again there are only limited data to support this concern (but see Hall et al., 1994, 1996; Murawski et al., 1994). Intensive harvesting of all size classes of a species can lead to local disappearance, as with Myristica malabarica and Syzygium gardneri in southern India (Daniels et al., 1995). The key to not overharvesting a species is the setting and enforcement of minimum cutting limits which allow individual trees to reach sexual maturity (i.e., flowering and fruiting) before harvesting. Even seasonality can be important, as in the case of large-leaf mahogany, which is mostly logged in the dry season, but whose seeds are not dispersed until the end of the dry season. Thus, a much reduced seed crop is available to provide natural regeneration after logging. The silvicultural method must then firmly rely on regeneration established prior to harvest, which may or may not be present, and could be destroyed during harvest. Fire is one of the most common agents of disturbance in the tropics. Except in the very wettest

TROPICAL FORESTRY AND DISTURBANCE

473

Fig. 19.2. Tropical forest landscape converted to grassland by annual burning. Note the skeleton trees still standing on the burned slopes along the Lae–Wau road, Papua New Guinea (Hartshorn).

regions, fire is an integral component of slash-and-burn agriculture – used by colonists as well as indigenous shifting cultivators. In seasonally dry ecosystems, fire is typically an annual event (Fig. 19.2), often extending into intact forest. Usually the later in the dry season that fire occurs, the more severe are the effects, killing more standing trees and decreasing the proportion of coppicing by the survivors (Uhl and Kauffman, 1990; Sampaio et al., 1993). The great Borneo fires of 1982–83 were exacerbated by an El Ni˜no drought and fueled by logging slash (Leighton and Wirawan, 1986). Beaman et al. (1985) estimated that the Borneo fires destroyed about 106 ha of forest in Sabah. Tree mortality attributable to the logging ranged from 6 to 12%; post-logging drought caused an additional 12– 18% mortality, while drought and fire together caused 38–72% mortality (Woods, 1989). Impacts on animals Selective logging seems to have only modest effects

on animal diversity (Johns and Skorupa, 1987; Johns, 1992; Bennett and Dahaban, 1995; Frumhoff, 1995; Johns, 1996; Thiollay, 1996). Although mobile animals may migrate away from areas of active logging, they seem to return fairly quickly once logging ceases and to re-establish normal population numbers and guild structures within 3–8 years (Lambert, 1992). Far more serious is the improved access which logging roads offer to hunters and colonists (Fig. 19.3). Commercial or subsistence hunting of bush-meat can seriously deplete popular animal species to the point of causing local extinctions (Wilkie and Finn, 1990; Redford 1992). Heavily-hunted large frugivores are often key agents of seed dispersal for tropical forest trees (Johns, 1987; Terborgh, 1995). When large, previously undisturbed tracts of forest are modified, large animals such as jaguars (Panthera onca) and tapirs (Tapirus spp.) disappear. In eastern Ecuador, Canaday (1996) found that insectivorous birds of the interior forest were more likely to be absent from disturbed forests and non-forest

474

G.S. HARTSHORN and J.L. WHITMORE

Fig. 19.3. Forest access roads under construction for intensive logging of lowland tropical rainforest, Gulf Province, Papua New Guinea (Hartshorn).

habitats than non-insectivores. Canaday’s disturbance gradient spanned about 3.5 km, and included a coffee plantation on the edge of the study forest. Feinsinger et al. (1988) analyzed the assemblage of hummingbird species along a disturbance gradient in a Costa Rican cloud forest. They found that hummingbird species interactions are most organized in the mature-phase forest, less organized in small tree-fall gaps, and almost undefined in large gaps. The pattern is attributed to greater availability of food (floral nectar) in the large gaps, yet the hummingbird assemblages in these disturbed patches within old-growth forest were not at all comparable to the more “weedy” hummingbird assemblages in anthropogenic old-field habitats. Thus, the authors suggested that responses of consumers to disturbance mosaics in old-growth forests may often be subtle and complex. Many tropical bird species that are generally rare throughout their range are vulnerable to forest fragmentation and disturbance (Terborgh et al., 1990; Hagan and Johnston, 1992). Fragmentation in tropical forests and/or in temperate forests is a concern

because of the effect on neotropical migratory birds, who spend time in both. Selective logging affects species diversity and abundance of butterflies in tropical forests (Hill et al., 1995). In Maluku, Indonesia, species richness, abundance, and evenness of butterflies were all significantly higher in unlogged forest than in selectively logged forest. Six butterfly species with restricted geographic distributions were found only in the unlogged forest, constituting a complex butterfly community there. The authors suggested that the distributional pattern of tropical butterfly species may be used as an indicator of forest disturbance. In Sabah, moths show significant loss of diversity (especially at higher taxonomic levels) with forests having been disturbed or converted to plantations (Holloway et al., 1992). Eggleton et al. (1995) studied the species richness of termites in five forest plots in southern Cameroon with differing disturbance levels. Severe disturbance resulted in a large reduction of termite species, whereas there was little difference in termite species richness between

TROPICAL FORESTRY AND DISTURBANCE

slightly disturbed and old-growth forests. Soil-feeding termites dominated in the old-growth and regenerating forests, but were greatly reduced in the severely disturbed forests. Wood-feeding termites appeared to be more resilient to disturbance than soil-feeding termites. Interestingly, the authors found no evidence of secondary invasion of disturbed forests by savannaassociated termite species. In Sumatra, Indonesia, a population of phytophagous beetles (Epilachna vigintioctopunctata) exploded during abnormally low rainfall (1982–83 El Ni˜no), but reproduction was suppressed by normal rainfall (Inoue et al., 1993). The beetle population was limited by food shortage at the end of the favorable dry period, and by high mortality during normal rainy periods. Impacts on ecosystems Along with clearing for agriculture and pasture establishment, logging is well known to be a serious disturbance factor in most tropical forests. In addition to the obvious effects on forest structure, major anthropogenic disturbances such as these also affect ecosystem processes. In a study of three different levels of selective logging in Sabah, Malaysia, Douglas et al. (1992) found a four-fold increase in stream sediment yield after a logging road was built across the upper catchment, and an increase of 5-fold to 18-fold in sediment yield after the 0.54 ha catchment was logged. One year after logging the sediment yield had declined to 3.6 times the amount from a control catchment, indicating fairly rapid partial recovery. In an experimental study of water yields from paired catchments in tropical rainforest in Sabah subjected to different methods of clear-felling, cutting and burning secondary vegetation led to 50% more runoff, while clear-felling and mechanical logging followed by burning increased runoff by about 60% (Malmer, 1992). Extensive surface runoff caused surface and gully erosion along tractor tracks (Malmer, 1996). The ecological effects of selective logging are often downplayed because the resulting opening of the canopy is similar or identical to that caused by natural tree falls. However, selective logging is clearly additional to the background level of natural tree falls. Changes in local wind patterns and eddy effects may erode the gap edge by causing additional tree falls. In Kibale forest in Uganda, heavy logging resulted in large areas of herbaceous tangle, attracting elephants which suppressed forest regeneration by damaging young

475

trees, thus perpetuating the herbaceous dominance of the understory (Struhsaker et al., 1996). Of more serious concern is the potential degradation of the general forest canopy by fairly intensive logging of most of the large trees. Ecologists have long recognized the importance of edges, or ecotones. There have been few quantitative studies of synergies between forest fragmentation and edge effects, especially in the tropics. In a study of forest fragments ranging from 1.4 to 590 ha in Queensland, Australia, Laurance (1991) showed that forest fragments had higher canopy and subcanopy damage than non-fragmented forests, as well as exceptional abundance of heavy lianas, climbing rattans and weedy species. Although the most striking changes occurred within 200 m of an edge, change was detectable up to 500 m inside forest fragments. A similar swamping of the margins of forest isolates has been documented in the Brazilian Amazon (Lovejoy and Bierregaard, 1990; Bierregaard et al., 1992). By far the most pervasive and serious disturbance facing tropical forests in general is the extensive conversion of forests to non-forested landscapes (Fig. 19.4). In tropical America, a primary reason for this conversion has been the establishment of pastures for beef cattle. The magnitude and severity of the conversion process have been so great that some scientists have questioned the possibility of forest recovery (e.g., G´omez-Pompa et al., 1972). The overwhelming trend of forest conversion to other land uses, and the widespread use of fire, have generated much scientific concern about the role of tropical deforestation in the atmospheric build-up of carbon dioxide and the concomitant effects on global climate change. In contrast, the potential effects of global climate change on tropical ecosystems have been largely ignored. Of particular concern in global-change scenarios is the increasing frequency and severity of droughts in the tropics, as has been documented in Panama (Leigh et al., 1990; Condit et al., 1996) and Sabah (Walsh, 1996). Fluctuations in rainfall seasonality affect plant phenology, so that scarcity of fruit may cause local famine among fruit-eating animals (Hartshorn, 1992). Far too little is known about the effects of climate change on tropical forests (cf. Colinvaux et al., 1996; Fuller and Prince, 1996). Recovery and restoration During the last three decades it has been popular to de-

476

G.S. HARTSHORN and J.L. WHITMORE

Fig. 19.4. Highly fragmented tropical dry forest in the western Central Valley, Costa Rica (Hartshorn).

scribe the tropics as fragile (e.g., Farnworth and Golley, 1974; Ayling, 1991). Scientific evidence, however, does not support the claim of general tropical fragility. The idea of resilient natural systems, even disturbed tropical forests, appears to be a more appropriate interpretation (Dubos, 1980; Orians, 1982; Lugo, 1995). One needs to understand tropical forest resiliency better, for whether forests are resilient or fragile will be the key to how they can be managed. Hurricane damage provides a good example of the resistance and resilience of different types of tropical forests to these devastating events (cf. Walker et al., 1996). In structurally complex rainforest on Guadeloupe (Imbert et al., 1996), high canopy trees served to shield smaller trees, and clusters of tall trees protected forest structure from hurricane damage. In contrast, the structurally uniform mangrove forest showed patchy damage related to species susceptibility. Floristic composition and forest structure may be the principal determinants of the effects of hurricane damage. Hurricanes remove considerable amounts of foliage from surviving trees, increasing

the light intensity in the understory 2–3-fold (Turton, 1992). Some of the best research is being done in the eastern Amazon, particularly in the Brazilian State of Par´a where vast areas were converted to cattle pastures, though many were subsequently abandoned. Researchers noted that areas used intensively for cattle ranching before abandonment were very slow to recover. Nepstad et al. (1996) carried out a comparative study of tree establishment in abandoned pasture and in experimental gaps in adjacent old-growth forest. Treeseedling emergence and coppicing were both about 20 times more in the forest than in abandoned pasture. Principal limitations were the low numbers of tree seeds dispersed into the pasture and much higher seed and seedling predation in the pasture than in the forest. Even physically protected transplants fared poorly in the abandoned pasture, as a result of inimical conditions such as higher air temperatures, lower humidity, and greater soil moisture stress. However, those lands pastured with fewer cattle, or less abused, tended to recover to secondary forest rather rapidly.

TROPICAL FORESTRY AND DISTURBANCE

Puerto Rico, on the other hand, has had appreciable abandonment of agricultural lands over the past 60 years. Aide et al. (1995) found that forest recovery in pastures is greatly delayed in comparison with recovery after natural disturbances. Species richness and density of woody species were quite low during the first 10 years of pasture succession; woody biomass did not increase substantially for 15 years after abandonment. As is often the case, the presence of grasses and the dominance of ferns significantly retard the establishment of secondary forest. When disturbance occurs to such an extent that patches of natural forest become island remnants without a vegetation matrix, recovery becomes of critical importance. What is the threshold beyond which remnant patches lose a significant amount of their biodiversity and ecosystem functions? Islands of forest habitat not only lose significant numbers of species, but they also are more susceptible to invasion by pioneer or exotic species (Vitousek, 1988; Lovejoy and Bierregaard, 1990; Bierregaard et al., 1992). Restoration of forest landscapes from remnant islands has serious management implications. Colonizing species can facilitate the successional or restorative process. Particularly important is the role of pioneer species such as Cecropia. Early successional phases are characterized by extremely rapid tree growth, biomass accumulation, and leaf turnover. Near San Carlos de R´ıo Negro, Venezuela, it was found that maximum net photosynthesis, leaf nitrogen content and specific leaf area all peaked within 3 years and declined significantly over the first 10 years of succession (Ellsworth and Reich, 1996). Changes in species composition and in resource availability combine to produce the common pattern of decreasing leaf nitrogen concentrations and photosynthetic rates during early forest succession. At Los Tuxtlas, Mexico, the dominant pioneer tree, Cecropia obtusifolia, has fast rates of seed-bank turnover, as a result of high seed predation and pathogen attacks (Alvarez-Buylla and Garcia-Barrios, 1991). The patchy establishment of robust, unpalatable shrubs in degraded or abandoned pastures often serves to enhance succession by providing attractive fruits for volant seed dispersers (which defecate or regurgitate seeds of other forest species), as well as more favorable microhabitats for tree establishment. Some examples include Cordia multispicata in the Brazilian Amazon (Vieira et al., 1994) and Melastomataceae in the pine savanna of Belize (Kellman, 1979). Probably more

477

frequently than is currently understood, the causes of disturbance may have synergistic or even multiplicative effects. For example, it is hypothesized that two scales and types of disturbance – fire and gaps – in the context of soil patchiness, control the pattern of forest islands in savannas (San Jos´e et al., 1991). Multiplicative disturbances can be devastating. Drought, hurricanes, fires, logging, El Ni˜no events, and others are usually studied singly, which may not lead to a very solid understanding of tropical-forest resilience or fragility.

TROPICAL FOREST MANAGEMENT

In the temperate region, forestry is a fairly simple operation compared to the tropics; and it has not changed much in the last two centuries. The writings of Gifford Pinchot (e.g., 1899; 1903), based on over 100 years of European experience, are rather similar to management guidelines of today. Sustainability and ecosystem management are not new ideas. On the other hand, wishes and goals of public landowners have certainly changed, which is part of the political process. The same forest-management principles, and some of the same techniques, are used to attain these new goals. In the tropics, however, with vastly more complex flora, fauna, and socio-economic conditions, successful forest-management techniques have been much slower to evolve (Vanclay, 1992; Hartshorn, 1996). Colonial foresters, well versed in temperate forestry, tried repeatedly to transplant those practices to the tropics, with nearly 150 years of failures to show for it (see, for example, volumes of the Indian Forester dating back to the mid-1800s). Given the difficulties encountered in managing the existing forests, their tendency was to “simplify” or “Europeanize” the complexity of the tropical forest. In other words, they replaced the diverse native forest with monoculture plantations, generally using better-known exotic species. To this day, the most successful tropical forestry, from the perspective of timber production, involves monoculture plantations of very few exotic species (Ewel, 1991; Evans, 1992). Just as in the temperate region, tropical goals are changing; wood production – important as it may be – is only one of many benefits that are now required from tropical forests. Indigenous management systems were, and future management systems for tropical forests need to be, based on programmed disturbance using knowledge

478

of the local forest ecosystem (cf. G´omez-Pompa and Bainbridge, 1995). It is unlikely that many of the currently valid indigenous management systems will be actively pursued far into the 21st century, no matter how successful they have been. Sad as it may be, these societies, that have been resilient over millennia, are not likely to fare well unless major and costly efforts are made to encourage their survival. In spite of past attempts at this, they are disappearing one by one. Forest-based indigenous cultures remaining 100 years from now will be more like museum pieces than management systems. Much the same can be said for old-growth tropical forests, and future research and management projects need to respond to these realities. Research programs concentrating on pristine forests tend to identify and describe in detail the many problems in trying to maintain them in an “undisturbed” state. However, researchers rarely get involved in management approaches that can help to solve these problems. Forestry, forest management, silviculture, and related disciplines have the responsibility of converting existing knowledge to solve such problems, wherever the political will and socio-economic realities allow. Forest management aspirations may well be shown in healthy secondary forests that are productive according to 21st century needs and goals, with a certain portion (10–20%) of forests preserved in old-growth natural areas. Although it is now reasonably well documented that disturbance is a central factor in the dominance– diversity relations of tropical forests and that maximum diversity should occur at intermediate levels of disturbance (Janzen, 1970; Connell, 1978; Clark and Clark, 1984), there has not been much success in using the growing knowledge about disturbance to develop more ecologically sophisticated models for forest management (Roberts and Gilliam, 1995; Uhl et al., 1997). Two exceptions are the strip-cut model (Hartshorn and Pariona, 1993, 1997) and medium-scale disturbance to promote natural regeneration of mahogany (Snook, 1996). The former is based on the relatively high proportion of shade-intolerant canopy tree species that require gaps for successful establishment and rapid growth (Hartshorn, 1978; Gorchov et al., 1993). The latter takes advantage of mahogany’s excellent natural regeneration in larger disturbances caused naturally by hurricanes, fires, and river meanders. This feature of mahogany offers promise for regeneration of this valuable species using silvicultural techniques based

G.S. HARTSHORN and J.L. WHITMORE

on artificial disturbance. A recent paper by Whitman et al. (1997) describes a mahogany-logging operation in northern Belize as “not silviculturally sustainable” because its disturbance may be insufficient to promote adequate mahogany regeneration. Unless a stand has plenty of advanced regeneration prior to harvest, “gentler” harvest techniques may leave openings too small for regeneration of the desired species. Post-disturbance management objectives that call for continuation of forest cover normally require forestry practices, more specifically forest management and silviculture. Managers are seldom willing simply to allow forest to return naturally without an effort to speed up the process, control species composition, or otherwise influence the establishment of a postdisturbance forest. Rarely are they patient enough to wait the decades required by nature, or able to prevent further disturbance during those decades. This is true for a variety of management objectives, whether wood production, biodiversity values, or watershed protection. Managerial responses to disturbance follow an intensity gradient, just as disturbances do. At one extreme is the low-budget, extensive approach to allow nature to heal the wound by natural regeneration. This can be appropriate in zones of low human population density, where site abandonment for decades is a feasible management tool, and/or where biodiversity values constitute the primary management objective. A slightly more intensive method would be to intervene with techniques that encourage desired species and discourage undesired species. A more intensive approach would be enrichment planting in the secondary forest, whereby lines are cut approximately every 5 m, and seedlings (or seeds) of desired species are planted along the cut lines. Clearing of the lines is then done for several years to allow enough sunlight for the young plants to thrive (Moad and Whitmore, 1994). Several steps up the intensity gradient might involve some form of clear-cuts as a harvest/regeneration technique. Ideally, silvicultural techniques to regenerate new forests are soundly based on knowledge of the local forest and its components. The responses of forests to disturbance offer solid clues as to how new forests can be created, using either natural (extensive) or artificial (intensive) methods. The methods are many (cf. Evans, 1992; Pinard et al., 1995; ter Steege et al., 1996; Wadsworth, 1997), and are far broader than a choice between a clear-cut or a selection system. They may apply to gaps in healthy forests, restoration of

TROPICAL FORESTRY AND DISTURBANCE

disturbed or unhealthy forests, healing of catchments, afforestation of non-forest land, or a variety of other landscape situations. On sites where disturbance has led to clearing of the site, but has not yet resulted in secondary forest regrowth, where food or land is scarce due to heavy human populations, and/or where medicinals or other forest benefits are desired by local inhabitants, agroforestry methods can be used to increase the productivity of the site in terms of food, fiber, fuelwood, or other products. There are a wide variety of agroforestry approaches, most of which have evolved from indigenous practices (Lojan Idrobo, 1992; Moad and Whitmore, 1994; Denevan et al., 1987). The agroforestry approach to tropical land use carries with it many advantages. While it often departs from the natural ecosystem more than some would wish, it is far better than many of the more destructive options, and some agroforestry systems can be quite similar to the natural, original forest. Agroforestry can create forests in areas long devoid of trees, as in the successful restoration of native Prosopis to northern coastal Peru, where irrigation assists with establishment on shifting sand dunes. According to Valdivia and Cueto (1979), Prosopis produces beans (362 kg ha−1 yr−1 ), livestock fodder (90 kg tree−1 yr−1 ), and honey (40 kg hive−1 yr−1 ). In general, agroforestry systems have the following important attributes: (1) they tend to be very peopleoriented, with benefits principally targeting local residents; (2) they can help fund, or justify, regeneration of new forests; (3) they employ appropriate technologies; (4) they offer multiple benefits; and (5) they are wildlife-friendly – usually by providing habitat. Agroforestry is generally practiced in rural, small operations controlled by land-owners, but there have been several cases of industrial agroforestry (Whitmore and Burwell, 1986). A variety of private firms have found agroforestry to be advantageous in their landmanagement projects. Some instances of disturbance are so severe that unorthodox methods of restoration may be required. Scale of disturbance obviously contributes to the severity, with direct consequences for either natural successional processes or human restoration efforts. In the latter case, the use of monocultures or mixedspecies plantations of fast-growing trees may be appropriate techniques, whether the goal is to restore the native ecosystem as rapidly as possible, or to restore productivity to the site for human benefit (Lugo and Liegel, 1987; Parrotta, 1992; Wormald, 1992; Parrotta

479

et al., 1997a,b). Some planted tree species can mature and complete the site-restoring cycle in 10–15 years. Whether native or exotic, species must be selected that will rapidly restore site conditions to a point where either the goal of restoration or that of productivity may be met. Indeed, the question arises, is a native species really native to a severely degraded site? At what point must a manager rely on a plantation of certain robust exotics to help restore a site to where native species can re-occupy it? Some exotic weed pests become so dominant on a disturbed site that site recovery is impaired, and restoration can be economically out of the question. Obviously, species that might become uncontrollable weeds must be avoided. Fortunately, current silvicultural knowledge allows good predictability for a wide range of tree species (Brown and Lugo, 1994). Having the technical method worked out, however, is usually not sufficient. The cost of restoring large areas may well be prohibitive, especially in areas where labor costs are high. Tax or other incentives to promote industry involvement in restoration may succeed in some countries (e.g., Brazil, Costa Rica, Indonesia). In cases where the management objective is to restore the disturbed site to a natural condition, abandonment may permit nature to run its course. In cases where recovery is too slow in human terms, or where the site has been badly abused to the point that natural recovery is impeded, then soil restoration and planting of tree seedlings may be required. Restorative plantations often improve microclimate conditions on a site to the point that native vegetation can then get a foothold. Interestingly, it seems to matter little whether native or exotic species are employed in this nurse-crop approach; indeed the natives might not thrive under harsh conditions where exotics often will. This is one of the reasons why certain exotics are widely planted. Obviously, exotics that reproduce vigorously on a given site should not be used, as they can become dominant weeds (Evans, 1992; Brown and Lugo, 1994; Parrotta et al., 1997a). Plantations can be used where the management goal is to produce wood for later harvest, whether for fuelwood, poles, or other products for local use, or for pulp, veneer, or lumber for local use or export. In such cases, any of several techniques and species can be employed, often quite successfully, and often while addressing secondary goals such as site improvement, biodiversity, water harvest values, ecotourism, or wildlife habitat. In the past, most such plantations have

480

involved exotic species, often pines, eucalypts, teak, or Leucaena (Evans, 1992). Recent studies (Espinoza and Butterfield, 1989; Russo and Sandi, 1995; Butterfield, 1996) have shown that plantation success with native species can also be promising. Mixed-species (multiple species) plantations have proven successful in many tropical and subtropical countries (Wormald, 1992). Production forestry by plantations can represent an extreme and intensive use of a disturbed site. Special cases are worth noting, where disturbed sites are in need of more intense management in order to restore them to productivity for the benefit of local people. The higher montane regions of the tropics, such as the Andean or Himalayan chains, are often heavily populated with communities totally dependent on fuelwood (Nepal–Australia Forestry Project, 1980; Lojan Idrobo, 1992; United Nations Conference on Environment and Development, 1992; Sharma and Chaudhry, 1997). The original forests are depleted, and, because of steep slopes, soil erosion is a serious problem (Young, 1994). However, plantations often can stabilize the site while offering fuel and other amenities, such as fodder. Reforestation efforts in lowlatitude arid zones, where fuelwood is often the vital forest product for local inhabitants and the resource is frequently depleted when demand exceeds supply, are made more difficult by slow tree growth rates and frequent disturbance by fire and grazing (Evans, 1992; Moad and Whitmore, 1994; Schroth et al., 1996). Perhaps of lesser magnitude is disturbance by hurricanes in Caribbean island nations. High population density and dry ecosystems are common, but forests that have evolved under such a disturbance regime usually respond and recover on their own, albeit slowly. When recovery is too slow, intervention may be required. In each of these cases, montane zones, arid zones, and hurricane-prone forests, plantation forestry is likely to be an important tool in addressing disturbances such as fire, wind, overgrazing, and excessive harvest of fuelwood. Managing native tropical forests for wood harvest using an uneven-aged silvicultural system requires considerable planning, as well as a gentle approach. Low-impact harvesting techniques are gaining in acceptance, particularly where they protect and/or facilitate natural regeneration of preferred species. Such techniques generally employ inventory of harvestsize stems, a survey of advance regeneration, cutting of woody vines (lianas, bush ropes) a year or more before harvest, careful and parsimonious placement of

G.S. HARTSHORN and J.L. WHITMORE

extraction routes, directional felling, and follow-up to assure adequate regeneration of the managed forest. Plantation harvest typically involves clear-cutting as part of the site preparation for the next generation of planted trees. Clear-cutting is an extreme form of disturbance, especially when the site is burned, but can be an aid for regeneration of some species. Certain species of wildlife are also favored by clear-cuts. While large clear-cuts can be ecologically unsound as well as visually unpleasant, smaller clear-cuts can, in some cases, serve as a valuable management tool. Even where patch clear-cuts are used to favor the regeneration of a desired species, it may be necessary to leave advance regeneration to form part of the new forest. The problem in northern Belize as described by Whitman et al. (1997) indicates a need for patch clearcuts to provide better regeneration of mahogany. Managers need to keep two things in mind: (1) If clear-cuts are used, they should be silviculturally justifiable and kept small; and (2) forest land needs to be managed for multiple benefits, not just for one or two, even though one of these benefits will surely be considered the dominant one in a given landmanagement plan.

CONCLUSIONS

Disturbance is an integral cause of tropical forest dynamics and regeneration. Tropical foresters and land stewards increasingly are using disturbance to improve ecological forest management, site restoration, and the selection of tree species for plantation forestry. Disturbance of tropical-forest landscapes spans broad spatial and intensity gradients, which are not easily categorized in discrete units. Disturbance parameters such as scale (e.g., from gap to landscape), frequency (from daily to centuries), magnitude (from patch to regional), severity (from extensive to intensive), and patchiness (from mosaic to uniform) all contribute to the impacts and effects. Yet there is very little documentation of their synergistic effects. Disturbance effects are most noticeable in soils and plants, but much less in animals because of their mobility. Common phenomena such as logging, slash-andburn agriculture, and conversion of forests to pasture can degrade or destroy forests, but also may have profound effects on soils. When carelessly performed, mechanized logging is particularly destructive to soils and to many of the remaining trees. Less obvious

TROPICAL FORESTRY AND DISTURBANCE

are the potentially damaging effects of global climate change, particularly its effects on the seasonality or intensity of rainfall. There is increasingly strong evidence of a multi-decade drying trend exacerbated by increasing severity of drought. Seasonally dry areas are quite susceptible to fire, causing very serious burns in logged forests, and many fires are anthropogenic in origin. Fragmentation also has serious effects on forest remnants, with edge effects detectable up to a few hundreds of meters into the remnant. Selective logging has modest effects on much wildlife, although logging roads often permit commercial or subsistence hunters to decimate newly accessible game animals. Responses to disturbance are mostly individualistic, although some broad generalizations are possible. Mycorrhizas are the most studied of soil organisms; they show a decline in abundance due to selective logging and other disturbances, but are fairly resilient to drought. Even though wind disturbance such as hurricanes can wreak havoc with subtropical forests, the partial defoliation of the canopy can trigger gregarious flowering of understory trees and shrubs. Forest types differ markedly in their resistance and ability to recover from hurricanes, largely related to the capacity of the species to sprout new branches or stems. Animals vary greatly in their responses to disturbances. Species and guilds of the forest interior tend to be more susceptible to loss than groups typical of disturbed habitats. Ecosystems are stressed by largescale or high-intensity disturbances. Severely altered systems such as extensive pastures in tropical-forest landscapes may be extremely slow to recover to forest. Robust colonizing species often play key roles in facilitating the successional process. Although tropical forests appear to be fairly resilient to disturbance, there must be adequate seed sources and dispersal agents for natural regeneration of the new forest to succeed. Repetitive or too-frequent disturbance can degrade tropical forests to a state where it may be very difficult to restore well-developed forest habitat. There are many elements involved in managing a tropical-forest resource successfully, whether globally (conceptually) or locally (in practice), whether for wood production or for other products/benefits. The well-being of tropical forests in general is of vital interest to people everywhere. But that well-being is directly tied to the well-being of the people who reside in or near a given forest. One cannot address one without the other. The well-being of the local, forest-

481

dependent people requires a productive and healthy forest (Whitmore, 1992). To be sustainable, forests must be managed by nature or nurture. The form of management will depend on the objectives. Silviculture (applied forest ecology), is a principal tool to maintain a healthy, vigorous and productive forest (Smith et al., 1997). Silviculture leads to renovation of the forest, and often requires disturbance (tree removal) to accomplish its goal. Harvest in a managed forest should be conducted in a way to promote regeneration of the new forest. Silvicultural practices include restoring biodiversity, and this is a correct and proper primary goal of forest management in some forests. If 10–20% of a nation’s forests have this as the primary goal, 80–90% of the forest resource is then available in which other goals can take precedence (Whitmore, 1992). There is a place for plantation silviculture, just as there is for intensive cultivation of corn (Zea mays) or rice (Oryza sativa). Preference should be given to native species, more than has been the case to date, but exotics may be better for a given objective and set of conditions. Given widespread deforestation and conversion of forest to other uses, it makes no sense to fell a healthy forest in order to establish plantations. Highly productive plantations of tropical tree species will likely out-compete temperate-zone sources of wood during the next century. Temperate-zone forestry experience does not transfer well to the tropics, where conditions are so different. However, one thing that has been learned in temperatezone forestry merits attention in the tropics. After the primary forest is cut in an area, one often studies the new, secondary forest to develop best management practices. Then, when that secondary forest is harvested, one expects these best management practices still to be valid for the third forest. But the third forest is usually quite unlike the second one, just as the second forest was unlike the primary forest. The third forest will need its own studies and best management practices. In the tropics, where rotations can be much shorter, this lesson is even more critical than in the temperate zone. There is also a tendency to confuse forest management (e.g., selection harvest) with lack of forest management (e.g., selective logging). Many forests are logged with no attempt at management, and the discipline of forest management is credited with “another failure” (Wadsworth, 1997). This is one of the many blind spots to be overcome before

482

proceeding with sustainable approaches to tropicalforest management. Swietenia macrophylla is certainly far less abundant now than it used to be in localized areas such as northern Costa Rica. Many tropical forest ecosystems have been converted to agriculture, affecting virtually all native species, not just mahogany. Assuming that the mahogany there is gone as a result of international trade may sidestep the real issue, and lead to false conclusions and ineffective solutions. Repeated attempts to list S. macrophylla in the Convention on International Trade in Endangered Species have prompted accelerated harvest of the resource to the point of extermination. Landowners in tropical nations perceive the Convention on International Trade in Endangered Species as a threat by foreign nations who want to control their property, just as landowners in the United States perceive conservation groups or government agencies as impinging on their rights. The key word here is perception. The intentions of those supporting the Convention on International Trade in Endangered Species, while considered highly suspect by some, are not the issue. Relegating local forest management to international decision-making bodies, when it excludes local decision-makers or managers from the process, cannot lead to rational, local resolution of naturalresource issues. One must ask how useful knowledge can be shared, so that logical solutions can be reached benefitting the planet as well as local inhabitants. In an interesting paper contrasting science and environmental ethics, Sarukh´an (1996) reviewed the work of others in this area and concluded that ecology must be solidly based on science, that most scientists hope to contribute to the preservation of “our common heritage, the Earth”, and that there is an absolute need to assist poorer nations to reach sustainable development comparable to that on the rest of the planet. He warned against belief in a nature that does not really exist, and against using pseudo-scientific knowledge as a cornerstone in one’s thinking – as do many environmentalist groups. Tropical-forest management has not progressed much, despite considerable ecological research over the past few decades. Several subjects are still in need of research. Research on restoration of productivity to degraded or disturbed sites is one of the most pressing needs (Brown and Lugo, 1994). More study is needed on the suitability of native species for plantations, the optimum utilization of tropical forests and tropical

G.S. HARTSHORN and J.L. WHITMORE

woods, and of non-wood products. However, much of what is already known is not now being used; such implementation would be sufficient to resolve many forest-management problems in land use and ecology. Research alone will serve no purpose without solid programs of extension, education, and technology transfer (Whitmore, 1992). Plantation silviculture with exotics continues to be the main approach to tropical forestry oriented to wood products. Apart from plantations, most harvesting of tropical wood is still based on the cutting of forests not managed in any systematic fashion. Much more attention will need to be devoted to integrating the principles of disturbance ecology into transdisciplinary efforts to manage complex tropical forests, secondary forests, and tropical tree plantations sustainably.

REFERENCES Aide, T.M., Zimmerman, J.K., Herrera, L., Rosario, M. and Serrano, M., 1995. Forest recovery in abandoned tropical pastures in Puerto Rico. For. Ecol. Manage., 77: 77–86. Alexander, I., Ahmad, N. and See, Lee Su, 1992. The role of mycorrhizas in the regeneration of some Malaysian forest trees. Philos. Trans. R. Soc. London Ser. B, 335: 379–388. Alvarez-Buylla, E.R., 1994. Density dependence and patch dynamics in tropical rain forests: matrix models and applications to a tree species. Am. Nat., 143: 155–191. Alvarez-Buylla, E.R. and Garcia-Barrios, R., 1991. Seed and forest dynamics: a theoretical framework and an example from the Neotropics. Am. Nat., 137: 133–154. Appanah, S., 1993. Mass flowering of dipterocarp forests in the aseasonal tropics. J. Biosci., 18: 457–474. Arriaga, L., 1988. Natural disturbance and treefalls in a pine–oak forest on the Peninsula of Baja California, Mexico. Vegetatio, 78: 73–79. Ayling, R.D., 1991. Deforestation, research and the Tropical Forestry Action Plan. For. Chron., 67: 539–541. Beaman, R., Beaman, J., Marsh, C. and Woods, P., 1985. Droughts and forest fires in Sabah in 1983. Sabah Soc. J., 8: 10–30. Bennett, E. and Dahaban, Z., 1995. Wildlife responses to disturbances in Sarawak and their implications for forest management. In: R. Primack and T. Lovejoy (Editors), Ecology, Conservation, and Management of Southeast Asian Rainforests. Yale University Press, New Haven, Connecticut, pp. 66–86. Bierregaard Jr., R., Lovejoy, T.E., Kapos, V., Santos, A.A. and Hutchings, R.W., 1992. The biological dynamics of tropical rainforest dynamics. BioScience, 42: 859–866. Boose, E.R., Foster, D.R. and Fluet, M., 1994. Hurricane impacts to tropical and temperate forest landscapes. Ecol. Monogr., 64: 369–400. Brown, S. and Lugo, A.E., 1990. Tropical secondary forests. J. Trop. Ecol., 6: 1–32. Brown, S. and Lugo, A.E., 1994. Rehabilitation of tropical lands: a key to sustaining development. Restoration Ecol., 2: 97–111.

TROPICAL FORESTRY AND DISTURBANCE Budowski, G., 1970. The distinction between old secondary and climax species in tropical Central American lowland forests. Trop. Ecol., 11: 44–48. Burns, R.M. (Editor), 1983. Silvicultural Systems for the Major Forest Types of the United States, Agriculture Handbook 445. U.S. Department of Agriculture Forest Service, Washington, DC., 191 pp. Bush, M.B. and Colinvaux, P.A., 1994. Tropical forest disturbance: paleoecological records from Darien, Panama. Ecology, 75: 1761–1768. Bush, M.B., Piperno, D.R., Colinvaux, P.A., de Oliveira, P.E., Krissek, L.A., Miller, M.C. and Rowe, W.E., 1992. A 14 300yr paleoecological profile of a lowland tropical lake in Panam´a. Ecol. Monogr., 62: 251–275. Butterfield, R., 1996. Early species selection for tropical reforestation: a consideration of stability. For. Ecol. Manage., 81: 161–168. Canaday, C., 1996. Loss of insectivorous birds along a gradient of human impact in Amazonia. Biol. Conserv., 77: 63–77. Clark, D.A. and Clark, D.B., 1984. Spacing dynamics of a tropical rain forest tree: evaluation of the Janzen-Connell model. Am. Nat., 124: 769–788. Clark, D.B., 1995. Abolishing virginity. J. Trop. Ecol., 12: 735–739. Clubbe, C.P. and Jhilmit, S., 1992. A case study of natural forest management in Trinidad. In: F.R. Miller and K.L. Adams (Editors), Wise Management of Tropical Forests 1992, Proceedings of the Oxford Conference on Tropical Forests 1992. Oxford Forestry Institute, Oxford, UK, pp. 201–209. Colinvaux, P.A., Liu, K.-B., Oliveira, P.D., Bush, M.B., Miller, M.C. and Steinitz Kannan, M., 1996. Temperature depression in the lowland tropics in glacial times. Climatic Change, 32: 19–33. Condit, R., Hubbell, S.P. and Foster, R.B., 1996. Changes in tree species abundance in a Neotropical forest: impact of climate change. J. Trop. Ecol., 12: 231–256. Congdon, R.A. and Herbohn, J.L., 1993. Ecosystem dynamics of disturbed and undisturbed sites in north Queensland wet tropical rain forest. I. Floristic composition, climate and soil chemistry. J. Trop. Ecol., 9: 349–363. Connell, J.H., 1978. Diversity in tropical rain forests and coral reefs. Science, 199: 1302–1310. Corlett, R.T., 1994. What is secondary forest? J. Trop. Ecol., 10: 445–447. Daniels, R.J.R., Gadgil, M. and Joshi, N.V., 1995. Impact of human extraction on tropical humid forests in the Western Ghats in Uttara Kannada, South India. J. Appl. Ecol., 32: 866–874. Denevan, W.M., Padoch, C. and Flores Paitan, S., 1987. Swiddenfallow agroforestry, Iquitos, Peru. In: A.E. Lugo (Editor), People and the Tropical Rain Forest. US Man and the Biosphere Program, pp. 8–9. Denslow, J.S., 1995. Disturbance and diversity in tropical rain forests: the density effect. Ecol. Appl., 5: 962–968. Douglas, I., Spencer, T., Greer, T., Bidin, K., Sinun, W. and Meng, W.W., 1992. The impact of selective commercial logging on stream hydrology, chemistry and sediment loads in the Ulu Segama rain forest, Sabah, Malaysia. Philos. Trans. R. Soc. London Ser. B, 335: 397–406. Dubos, R., 1980. The Wooing of Earth: New Perspectives on Man’s Use of Nature. Charles Scribner’s Sons, New York, 183 pp. Eggleton, P., Bignell, D.E., Sands, W.A., Waite, B., Wood, T.G. and Lawton, J.H., 1995. The species richness of termites (Isoptera)

483 under differing levels of forest disturbance in the Mbalmayo Forest Reserve, southern Cameroon. J. Trop. Ecol., 11: 85–98. Ellsworth, D.S. and Reich, P.B., 1996. Photosynthesis and leaf nitrogen in five Amazonian tree species during early secondary succession. Ecology, 77: 581–594. Espinoza, M. and Butterfield, R., 1989. Adaptabilidad de 13 especies nativas maderables bajo condiciones de plantaci´on en las tierras bajas h´umedas del Atl´antico, Costa Rica. In: R. Salazar (Editor), Manejo y Aprovechamiento de Plantaciones Forestales con Especies de Uso M´ultiple. Actas de la Reuni´on IUFRO (S1.07–09), Antigua, Guatemala, April 1989, pp. 159–172. Evans, J., 1992. Plantation Forestry in the Tropics: Tree Planting for Industrial, Social, Environmental and Agroforestry Purposes. Clarenden Press, Oxford, UK, 403 pp. Ewel, J.J., 1991. Diversity: yes, we got some bananas. Conserv. Biol., 5: 423–425. FAO, 1995. Forest resources assessment 1990: global synthesis. FAO Forestry Paper 124. FAO, Rome. Farnworth, E.A. and Golley, F.B. (Editors), 1974. Fragile Ecosystems: Evaluation of Research and Applications in the Neotropics. Springer-Verlag, New York, 258 pp. Feinsinger, P., Busby, W.H., Murray, K.G., Beach, J.H., Pounds, W.Z. and Linhart, Y.B., 1988. Mixed support for spatial heterogeneity in species interactions: hummingbirds in a tropical disturbance mosaic. Am. Nat., 131: 33–57. Fischer, C.R., Janos, D.P., Perry, D.A., Linderman, R.G. and Sollins, P., 1994. Mycorrhiza inoculum potentials in tropical secondary succession. Biotropica, 26: 369–377. Ford-Robertson, F.C. (Editor), 1971. Terminology of Forest Science, Technology, Practice & Products: English-language Version. Joint FAO/IUFRO Committee on Forestry Bibliography and Terminology. Society of American Foresters, Washington, DC, 349 pp. Frumhoff, P.C., 1995. Conserving wildlife in tropical forests managed for timber. BioScience, 45: 456–464. Fuller, D.O. and Prince, S.D., 1996. Rainfall and foliar dynamics in tropical southern Africa: potential impacts of global climate change on savanna vegetation. Climatic Change, 33: 69–96. Gentry, A.H., 1982. Patterns of Neotropical plant species diversity. Evol. Biol., 15: 1–84. G´omez-Pompa, A. and Bainbridge, D.A., 1995. Tropical forestry as if people mattered. In: A.E. Lugo and C. Lowe (Editors), Tropical Forests: Management and Ecology. Springer-Verlag, New York, pp. 408–422. G´omez-Pompa, A., V´asquez-Yanes, C. and Guevara, S., 1972. The tropical rain forest: a nonrenewable resource. Science, 177: 762–765. G´omez-Pompa, A., Flores, J.S. and Sosa, V., 1987. The ‘Pet kot’: a man-made tropical forest of the Maya. Interciencia, 12: 10–15. Gorchov, D.L., Cornejo, F., Ascorra, C. and Jaramillo, M., 1993. The role of seed dispersal in the natural regeneration of rain forest after strip cutting in the Peruvian Amazon. Vegetatio, 108: 339–349. Hagan, J.M. and Johnston, D.W. (Editors), 1992. Ecology and Conservation of Neotropical Landbirds. Smithsonian Institution Press, Washington, 609 pp. Hall, P., Chase, M.R. and Bawa, K.S., 1994. Low genetic variation but high population differentiation in a common tropical forest tree species. Conserv. Biol., 8: 471–482. Hall, P., Walker, S. and Bawa, K.S., 1996. Effect of forest

484 fragmentation on genetic diversity and mating system in a tropical tree, Pithecellobium elegans. Conserv. Biol., 10: 757–768. Hartshorn, G.S., 1978. Tree falls and tropical forest dynamics. In: P. Tomlinson and M. Zimmermann (Editors), Tropical Trees as Living Systems. Cambridge University Press, New York, pp. 617– 638. Hartshorn, G.S., 1992. Possible effects of global warming on the biological diversity in tropical forests. In: R. Peters and T. Lovejoy (Editors), Global Warming and Biological Diversity. Yale University Press, New Haven, Connecticut, pp. 137–146. Hartshorn, G.S., 1995. Ecological basis for sustainable development in tropical forests. Annu. Rev. Ecol. Syst., 26: 155–175. Hartshorn, G.S., 1996. Tropical forests: are conservation and forestry compatible? Selbyana, 17: 1–5. Hartshorn, G.S. and Pariona, A.W., 1993. Ecologically sustainable forest management in the Peruvian Amazon. In: C. Potter and J. Cohen (Editors), Perspectives on Biodiversity: Case Studies of Genetic Resource Conservation and Development. AAAS Press, Washington DC, pp. 151–166. Hartshorn, G.S. and Pariona, A.W., 1997. Sustainable tropical forestry. In: G.K. Meffe and C.R. Carroll (Editors), Principles of Conservation Biology, 2nd Edition. Sinauer Associates, Sunderland Massachusetts, pp. 603–608. Hartshorn, G.S., Nicolait, L., Hartshorn, L., Bevier, G., Brightman, R., Cal, J., Cawich, A., Davidson, W., DuBois, R., Dyer, C., Gibson, J., Hawley, W., Leonard, J., Nicolait, R., Weyer, D., White, H. and Wright, C., 1984. Belize Country Environmental Profile: A Field Study. Tropical Science Center, San Jose, Costa Rica, 151 pp. Hill, J.K., Hamer, K.C., Lace, L.A. and Banham, W.M.T., 1995. Effects of selective logging on tropical forest butterflies on Buru, Indonesia. J. Appl. Ecol., 32: 754–760. Holdridge, L.R., 1967. Life Zone Ecology. Tropical Science Center, San Jos´e, Costa Rica, 206 pp. Holloway, J.D., Kirk-Spriggs, A.H. and Khen, C.V., 1992. The response of some rain forest insect groups to logging and conversion to plantation. Philos. Trans. R. Soc. London Ser. B, 335: 425–436. Homfray, C.K., 1936. Notes on Thinning in Plantations. Bengal Forest Bulletin No. 1. Silvicultural Series No. 1. Government of Bengal, Forest Department, Bengal Government Press, Alipore, 44 pp. Horn, S. and Sanford Jr., R.L., 1993. Holocene fires in Costa Rica. Biotropica, 24: 354–361. Imbert, D., Labbe, P. and Rousteau, A., 1996. Hurricane damage and forest structure in Guadeloupe, French West Indies. J. Trop. Ecol., 12: 663–680. Inoue, T., Nakamura, K., Salmah, S. and Abbas, I., 1993. Population dynamics of animals in unpredictably-changing tropical environments. J. Biosci., 18: 425–455. Janos, D.P., 1980. Vesicular-arbuscular mycorrhizae affect lowland tropical rainforest plant growth. Ecology, 61: 151–162. Janzen, D.H., 1970. Herbivores and the number of tree species in tropical forests. Am. Nat., 104: 501–528. Janzen, D.H., 1988. Tropical dry forests: the most endangered major tropical ecosystem. In: E.O. Wilson and F. Peter (Editors), Biodiversity. National Academy Press, Washington DC, pp. 130– 137. Johns, A.D., 1987. The use of primary and selectively logged

G.S. HARTSHORN and J.L. WHITMORE rainforest by Malaysian hornbills (Bucerotidae) and implications for their conservation. Biol. Conserv., 40: 179–190. Johns, A.D., 1992. Vertebrate responses to selective logging: implications for the design of logging systems. Philos. Trans. R. Soc. London Ser. B, 335: 437–442. Johns, A.D. and Skorupa, J.P., 1987. Responses of rain forest primates to habitat disturbance – a review. Int. J. Primatol., 8: 157–191. Johns, A.G., 1996. Bird population persistence in Sabahan logging concessions. Biol. Conserv., 75: 3–10. Johns, J.S., Barreto, P. and Uhl, C., 1996. Logging damage during planned and unplanned logging operations in the eastern Amazon. For. Ecol. Manage., 89: 59–77. Kellman, M., 1979. Soil enrichment by neotropical savanna trees. J. Ecol., 67: 565–577. Khurana, D.K. and Khosla, P.K., 1993. Agroforestry for Rural Needs, Vol. II, IUFRO Workshop Proceedings, P1.15–00: 22–26 Feb. 1987, New Delhi. Indian Society of Tree Scientists, Dr. Y.S. Parmar University of Horticulture and Forestry, Solan, India, 811 pp. Lamb, F.B., 1966. Mahogany of Tropical America: Its Ecology and Management. University of Michigan Press, Ann Arbor, Michigan, 220 pp. Lambert, F.R., 1992. The consequences of selective logging for Bornean lowland forest birds. Philos. Trans. R. Soc. London Ser. B, 335: 443–457. Lanly, J.P., 1995. The status of tropical forests. In: A.E. Lugo and C. Lowe (Editors), Ecological Studies, Vol. 112, Tropical Forests: Management and Ecology. Springer-Verlag, New York, pp. 18– 32. Laurance, W.F., 1991. Edge effects in tropical forest fragments: application of a model for the design of nature reserves. Biol. Conserv., 57: 205–219. Leigh Jr., E.G., Rand, A.S. and Windsor, D.M. (Editors), 1990. Ecolog´ıa de un Bosque Tropical: Ciclos Estacionales y Cambios a Largo Plazo. Smithsonian Tropical Research Institute, Panama, 468 pp. Leighton, M. and Wirawan, N., 1986. Catastrophic drought and fire in Borneo tropical rain forest associated with the 1982–1983 El Ni˜no southern oscillation event. In: G. Prance (Editor), Tropical Rain Forests and the World Atmosphere. Westview Press, Boulder, Colorado, pp. 75–102. Lojan Idrobo, L., 1992. El Verdor de los Andes: a´ rboles y arbustos nativos para el desarrollo forestal altoandino. Proyecto Desarrollo Forestal Participativo en los Andes, Quito, Ecuador, 217 pp. Lovejoy, T.E. and Bierregaard Jr., R., 1990. Central Amazonian forests and the minimum critical size of ecosystems project. In: A.H. Gentry (Editor), Four Neotropical Forests. Yale University Press, New Haven, Connecticut, pp. 60–71. Lovelock, C.E., Jebb, M. and Osmond, C.B., 1994. Photoinhibition and recovery in tropical plant species: response to disturbance. Oecologia, 97: 297–307. Lugo, A.E., 1995. Tropical forests: their future and our future. In: A.E. Lugo and C. Lowe (Editors), Ecological Studies, Vol. 112, Tropical Forests: Management and Ecology. Springer-Verlag, New York, pp. 3–17. Lugo, A.E. and Liegel, L.H., 1987. Comparison of plantations and natural forests in Puerto Rico. In: A.E. Lugo (Editor), People and the Tropical Forest. US Man and the Biosphere Program, Washington, DC, pp. 41–44. Maggi, O., Persiani, A.M., Casado, M.A. and Pineda, F.D., 1990.

TROPICAL FORESTRY AND DISTURBANCE Edaphic mycoflora recovery in tropical forests after shifting cultivation. Acta Oecol., 11: 337–350. Malmer, A., 1992. Water-yield changes after clear-felling tropical rainforest and establishment of forest plantation in Sabah, Malaysia. J. Hydrol., 134: 77–94. Malmer, A., 1996. Hydrological effects and nutrient losses of forest plantation establishment on tropical rainforest land in Sabah, Malaysia. J. Hydrol., 174: 129–148. Malmer, A. and Grip, H., 1994. Converting tropical rainforest to forest plantation in Sabah, Malaysia. Part II. Effects on nutrient dynamics and net losses in streamwater. Hydrol. Processes, 8: 195–209. McNabb, K.L., Miller, M.S., Lockaby, B.G., Stokes, B.J., Clawson, R.G., Stanturf, J.A. and Silva, J.N.M., 1997. Selection harvests in Amazonian rainforests: long-term impacts on soil properties. For. Ecol. Manage., 93: 153–160. Migenis, L.E. and Ackerman, J.D., 1993. Orchid–phorophyte relationships in a forest watershed in Puerto Rico. J. Trop. Ecol., 9: 231–240. Moad, A. and Whitmore, J.L., 1994. Tropical forest management in the Asia-Pacific region. J. Sustainable For., 1: 25–63. Moran, E.F., 1981. Developing the Amazon. Indiana University Press, Bloomington, Indiana, 292 pp. Murawski, D.A., Gunatilleke, I.A.U.N. and Bawa, K.S., 1994. The effects of selective logging on inbreeding in Shorea megistophylla (Dipterocarpaceae) from Sri Lanka. Conserv. Biol., 8: 997–1002. Murphy, P.G. and Lugo, A.E., 1995. Dry forests of Central America and the Caribbean. In: S.H. Bullock, H.A. Mooney and E. Medina (Editors), Seasonally Dry Tropical Forests. Cambridge University Press, New York, pp. 9–34. Nelson, B.W., Kapos, V., Adams, J.B., Oliveira, W.J., Braun, O.P.G. and do Amaral, I.L., 1994. Forest disturbance by large blowdowns in the Brazilian Amazon. Ecology, 75: 853–858. Nepal–Australia Forestry Project, 1980. Tree Species Trials in Nepal: Some Early Results. Department of Forestry, Australian National University, 144 pp. Nepstad, D.C., Uhl, C., Pereira, C.A. and Da Silva, J.M.C., 1996. A comparative study of tree establishment in abandoned pasture and mature forest of eastern Amazonia. Oikos, 76: 25–39. Nussbaum, R., Anderson, J. and Spencer, T., 1995. Effects of selective logging on soil characteristics and growth of planted dipterocarp seedlings in Sabah. In: R. Primack and T. Lovejoy (Editors), Ecology, Conservation, and Management of Southeast Asian Rainforests. Yale University Press, New Haven, Connecticut, pp. 105–115. Orians, G.H., 1982. The influence of tree-falls in tropical forests in tree species richness. Trop. Ecol., 23: 255–279. Parrotta, J.A., 1992. The role of plantation forests in rehabilitating degraded tropical ecosystems. Agriculture, Ecosystems and Environment, 41: 115–133. Parrotta, J.A., Knowles, O.H. and Wunderle Jr., J.M., 1997a. Development of floristic diversity in 10-year-old restoration forests on a bauxite mined site in Amazonia. In: J.A. Parrotta and J.W. Turnbull (Editors), Catalyzing Native Forest Regeneration on Degraded Tropical Lands. For. Ecol. Manage., 99: 21–42. Parrotta, J.A., Turnbull, J.W. and Jones, N., 1997b. Catalyzing native forest regeneration on degraded tropical lands. In: J.A. Parrotta and J.W. Turnbull (Editors), Catalyzing Native Forest Regeneration on Degraded Tropical Lands. For. Ecol. Manage., 99: 1–7.

485 Pickett, S.T.A. and White, P.S. (Editors), 1985. The Ecology of Natural Disturbances and Patch Dynamics. Academic Press, New York, 472 pp. Pinard, M.A., Putz, F.E., Tay, J. and Sullivan, T.E., 1995. Creating timber harvest guidelines for a reduced-impact logging project in Malaysia. J. For., 93: 41–45. Pinchot, G., 1899. A Primer of Forestry: Part I – the Forest, USDA Division of Forestry, Bulletin 24. Government Printing Office, Washington DC, 88 pp. Pinchot, G., 1903. A Primer of Forestry: Part II – Practical Forestry, USDA Bureau of Forestry, Bulletin 24. Government Printing Office, Washington DC, 78 pp. Redford, K.H., 1992. The empty forest. BioScience, 42: 412–422. Richards, P.W., 1996. The Tropical Rain Forest: An Ecological Study, 2nd Edition. Cambridge University Press, Cambridge, UK, 575 pp. Roberts, M.R. and Gilliam, F.S., 1995. Patterns and mechanisms of plant diversity in forested ecosystems: implications for forest management. Ecol. Appl., 5: 969–977. Russo, R. and Sandi, C.L., 1995. Early growth of eight native timber species in the humid tropic region of Costa Rica. J. Sustainable For. Manage., 3: 81–84. Sampaio, E.V.S.B., Salcedo, I.H. and Kauffman, J.B., 1993. Effect of different fire severities on coppicing of caatinga vegetation in Serra Talhada, PE, Brazil. Biotropica, 25: 452–460. San Jos´e, J.J., Fari˜nas, M.R. and Rosales, J., 1991. Spatial patterns of trees and structuring factors in a Trachypogon savanna of the Orinoco Llanos. Biotropica, 23: 114–123. Sarukh´an, J.K., 1996. Science, society and environmental ethics. Voices of Mexico, 37: 109–115. Schroth, G., Kolbe, D., Pity, B. and Zech, W., 1996. Root system characteristics with agroforestry relevance of nine leguminous tree species and a spontaneous fallow in a semi-deciduous rainforest area of West Africa. For. Ecol. Manage., 84: 199–208. Serrao, E., Nepstad, D.C. and Walker, R., 1996. Upland agricultural and forestry development in the Amazon: sustainability, criticality and resilience. Ecol. Econ., 18: 3–13. Sharma, S. and Chaudhry, S., 1997. Forestry, agriculture, and people’s participation in the Central Himalaya. J. Sustainable For., 4: 63–73. Smith, D.M., Larson, B.C., Kelty, M.J. and Ashton, P.M.S., 1997. The Practice of Silviculture: Applied Forest Ecology, 9th Edition. Wiley, New York, 537 pp. Snook, L.K., 1996. Catastrophic disturbance, logging and the ecology of mahogany (Swietenia macrophylla King): grounds for listing a major tropical timber species in CITES. Bot. J. Linn. Soc., 122: 35–46. Society of American Foresters, 1994. Silviculture Terminology, with Appendix of Ecosystem Management Terms. Prepared by the Silviculture Working Group of the Society of American Foresters, Washington DC, 14 pp. Spurr, S.H., 1979. Silviculture. Sci. Am., 240: 76–91. Struhsaker, T.T., Lwanga, J.S. and Kasenene, J.M., 1996. Elephants, selective logging and forest regeneration in the Kibale forest, Uganda. J. Trop. Ecol., 12: 45–64. ter Steege, H., Boot, R.G.A., Brouwer, L.C., Caesar, J.C., Ek, R.C., Hammond, D.S., Haripersaud, P.P., Hout, P. v.d., Jetten, V.G., van Kekem, A.J., Kellman, M.A., Khan, Z., Polak, A.M., Pons, T.L., Pulles, J., Raaimakeers, D., Rose, S.A., Sanden, J.J. v.d. and

486 Zagt, R.J., 1996. Ecology and Logging in a Tropical Rain Forest in Guyana: With Recommendations for Forest Management. Tropenbos Series 14, Wageningen, 123 pp. Terborgh, J., 1995. Wildlife in managed tropical forests: a neotropical perspective. In: A.E. Lugo and C. Lowe (Editors), Tropical Forests: Management and Ecology. Springer-Verlag, New York, pp. 331– 341. Terborgh, J., Robinson, S.K., Parker III, T.A., Munn, C.A. and Pierpont, N., 1990. Structure and organization of an Amazonian forest bird community. Ecol. Monogr., 60: 213–238. Thiollay, J.-M., 1996. Distributional patterns of raptors along altitudinal gradients in the northern Andes and effects of forest fragmentation. J. Trop. Ecol., 12: 535–560. Turner, I., 1996. Species loss in fragments of tropical rain forest: a review of the evidence. J. Appl. Ecol., 33: 200–209. Turton, S., 1992. Understorey light environments in a north-east Australian rain forest before and after a tropical cyclone. J. Trop. Ecol., 8: 241–252. Uhl, C. and Buschbacher, R., 1987. Potential productive capacity of abandoned pasture lands in the Brazilian Amazon. In: A.E. Lugo (Editor), People and the Tropical Forest. US Man and the Biosphere Program, Washington DC, pp. 35–36. Uhl, C. and Kauffman, J.B., 1990. Deforestation, fire susceptibility, and potential tree responses to fire in the eastern Amazon. Ecology, 71: 437–449. Uhl, C. and Vieira, I.C.G., 1989. Ecological impacts of selective logging in the Brazilian Amazon: a case study from the Paragominas region of the state of Par´a. Biotropica, 21: 98–106. Uhl, C., Barreto, P., Ver´ıssimo, A., Vidal, E., Amaral, P., Barros, A.C., Souza Jr., C., Johns, J. and Gerwing, J., 1997. Natural resource management in the Brazilian Amazon: an integrated approach. BioScience, 47: 160–168. United Nations Conference on Environment and Development, 1992. Managing fragile ecosystems: sustainable mountain development. Agenda 21, Chapter 13. United Nations, New York. Valdivia, S. and Cueto, L., 1979. Settlement and rural development in the eriazas zones of the north coast of Peru. In: G. De Las Salas (Editor), Proceedings, Workshop on Agroforestry Systems in Latin America, 26–30 March 1979, Turrialba, Costa Rica. CATIE and UNU, pp. 163–169. van der Meer, P.J. and Bongers, F., 1996. Patterns of tree-fall and branch-fall in a tropical rain forest in French Guiana. J. Ecol., 84: 19–29. Vanclay, J.K., 1992. Species richness and productive forest management. In: F.R. Miller and K.L. Adams (Editors), Wise Management of Tropical Forests 1992, Proceedings of the Oxford Conference on Tropical Forests 1992. Oxford Forestry Institute, Oxford, UK, pp. 1–9. Vayda, A.P., 1987. Shifting cultivation and patch dynamics in an upland forest in East Kalimantan, Indonesia. In: A.E. Lugo (Editor), People and the Tropical Forest. US Man and the Biosphere Program, Washington DC, pp. 16–17. Vega Condori, L., 1987. Crecimiento de cedro Cedrela odorata manejado en fajas de rastrojo y en asocio inicial con cultivos. CONIF INFORMA, No. 10. Bogota, Colombia, 20 pp.

G.S. HARTSHORN and J.L. WHITMORE Ver´ıssimo, A., Barreto, P., Tarifa, R. and Uhl, C., 1995. Extraction of a high-value natural resource in Amazonia: the case of mahogany. For. Ecol. Manage., 72: 39–60. Vieira, I.C.G., Uhl, C. and Nepstad, D.C., 1994. The role of Cordia multispicata Cham. as a ‘succession facilitator’ in an abandoned pasture, Paragominas, Amazonia. Vegetatio, 115: 91–99. Vitousek, P.M., 1988. Diversity and biological invasions of oceanic islands. In: E.O. Wilson and F. Peter (Editors), Biodiversity. National Academy Press, Washington DC, pp. 181–189. Wadsworth, F.H., 1997. Forest Production for Tropical America, Agriculture Handbook 710. USDA Forest Service, Washington DC, 563 pp. Walker, L.R., Silver, W.L., Willig, M.R. and Zimmerman, J.K. (Editors), 1996. Special issue: long term responses of Caribbean ecosystems to disturbance. Biotropica, 28: 414–613. Walsh, R.P.D., 1996. Drought frequency changes in Sabah and adjacent parts of northern Borneo since the late nineteenth century and possible implications for tropical rain forest dynamics. J. Trop. Ecol., 12: 385–407. Whitman, A.A., Brokaw, N.V.L. and Hagan, J.M., 1997. Forest damage caused by selection logging of mahogany (Swietenia macrophylla) in northern Belize. For. Ecol. Manage., 92: 87–96. Whitmore, J.L., 1992. Advances in sustainable forest management of natural tropical forest, and in plantation of native species. In: Juan Razali Wan Mohd, Shamsudin Ibrahim, S. Appanah and Mohd. Farid Abd. Rashid (Editors), Proceedings, International Symposium on Harvesting and Silviculture for Sustainable Forestry in the Tropics. Oct. 5–9, 1992. FRIM, Kuala Lumpur, Malaysia. pp. 1–9. Whitmore, J.L. and Burwell, B., 1986. Industrial agroforestry. Unasylva, 38: 28–34. Wilkie, D.S. and Finn, J.T., 1990. Slash-burn cultivation and mammal abundance in the Ituri Forest, Zaire. Biotropica, 22: 90–99. Williams-Linera, G., 1990. Origin and early development of forest edge vegetation in Panama. Biotropica, 22: 235–241. Wint, S.H., 1978. Report on man-made forests in Burma. In: Forest News for Asia and the Pacific, Volume II. FAO, Bangkok, pp. 5–8. Woods, P., 1989. Effects of logging, drought, and fire on structure and composition of tropical forests in Sabah, Malaysia. Biotropica, 21: 290–298. World Resources Institute, 1992. World Resources 1992–93. Oxford University Press, New York, 385 pp. World Resources Institute, 1996. World Resources 1996–97. Oxford University Press, New York, 365 pp. Wormald, T.J., 1992. Mixed and pure forest plantations in the tropics and subtropics. FAO Forestry Paper 103. FAO, Rome, 152 pp. Young, K.R., 1994. Roads and the environmental degradation of tropical montane forests. Conserv. Biol., 8: 972–976. Young, S.S. and Wang, Z.J., 1989. Comparisons of secondary and primary forests in the Ailao Shan region of Yunnan, China. For. Ecol. Manage., 28: 281–300. Zou, X., Zucca, C.P., Waide, R.B. and McDowell, W.H., 1995. Longterm influence of deforestation on tree species composition and litter dynamics of a tropical rain forest in Puerto Rico. For. Ecol. Manage., 78: 147–157.

Chapter 20

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA ´ Claudio M. GHERSA and Rolando J.C. LEON

INTRODUCTION

The processes and patterns of changes through ecological time at a site are called succession. During this process, the physical environment continually changes along with plant composition, generating a dynamic system characterized by complex biotic–abiotic interactions. Odum (1969) and Whittaker (1975) codified many of the features of a classical successional model for progressive community development. They suggested that species diversity, community complexity, biomass, and floristic stability increased with successional time. The concept of succession and the features of a progressive model are applicable to human-disturbed systems (Vitousek and Walker, 1987). Peet (1992) stated that community change is often categorized as either episodic or gradual; it is episodic when the change is discontinuous and generated by exogenous factors (e.g., tillage), and it is gradual, when change is continuous and generated by endogenous factors (e.g., competition). He also stated that the first type of community change is generally viewed by ecologists as disturbance, whilst the second type is viewed as succession. Because he believed that episodic and gradual changes could be confounded depending on the scale of observation, he thought that it was not possible fully to separate disturbance from succession. Therefore, both should be included in any treatment of vegetation dynamics. These considerations are especially important for agroecosystems where an annual agriculture cycle is superimposed on natural colonization processes (Soriano, 1971); each year begins with tillage, followed by sowing and crop production, and finishes with a harvest. As a consequence, a weed community (a weed is considered

here as a plant growing in a cultivated area that is not harvested or grazed) must be responsive to several patterns of change in the physical environment: (1) seasonal variation; (2) agricultural cycles; and (3) long-term environmental trends, such as increasing soil erosion or climate change. Very little is known about the significance of succession for the functional properties of agroecosystems. Agroecosystems are characterized by the establishment and management of a modified and simplified plant community, often comprising exotic species. This changes the ecosystem by altering the composition and activities of associated herbivore, predator, symbiont, and decomposer communities (Swift and Anderson, 1992). The composition, diversity, structure, and dynamics of agroecosystems may differ in many respects from those of the original ecosystem that dominated the landscape before the onset of agricultural activities. Little information exists on how shifts in weed flora affect function in these systems, and even less information exists enabling one to test whether a succession can really exist in this highly disturbed environment. Most information is skewed by the perception that weeds, pests, and diseases are invaders (Williamson, 1996) which do not follow the generally accepted stages of succession and which have only negative effects on ecosystem function. In agroecosystems, land is grazed, burned, or cultivated in a cyclic way. This means that land is exposed to regular disturbances and has periods with low soil cover, but with high levels of resource availability. Because during these periods nutrient absorption is low and mineralization of organic matter is high, a great proportion of the mineralized nitrogen from organic matter is lost by leaching or denitrification

487

488

(Tivy, 1990; Smith et al., 1992; Swift and Anderson, 1992). These conditions with high resource availability and a simplified biotic system make the crop–weed community susceptible to invasion. Each invasion of the community creates a new scenario of instability, which Williamson (1987) calls “press” perturbation – a situation in which the structural and functional properties of the community are modified by new species rather than by the extinction of ones already present. Forcella and Harvey (1983) showed that both species richness and equitability of the alien arable weeds in the north-western United States have increased since the turn of the century. This suggests that alien arable weeds are following the expected features for a classical successional model, but there is no information on how the invasion of new species has modified the functional properties of the community. Decomposer organisms play a major role in driving succession in both natural ecosystems and in agroecosystems (Peet, 1992; Schulze and Mooney, 1993; Coleman and Crossley, 1996). In agroecosystems, organic matter is produced in pulses, and a great proportion of its decomposition occurs while the soil has no plant cover. Therefore, a short successional process is started after the crop cycle is ended, which depends on organic matter produced previously. This is called heterotrophic succession (McNaughton and Wolf, 1984) and may be considered a retrogressive process, controlled by the available substrate. Swift and Anderson (1992) hypothesized that the relationship between the number of plant species and ecosystem functions follow a hyperbolic pattern, in which a plateau is reached with a low species number. They believed that this is the point at which the plants take control of the decomposer system. They also considered that species composition is important because different plants can give different physical or chemical signals to the ecosystem. Plants differing in physical structure create different spatial interactions, changing, for example, the volume of the resource space to be exploited. The chemical signals originate both from the productive capacity of the plant (i.e., the input of carbon and energy to the system, the ability to compete for water and nutrients) and the patterns of synthesis of chemicals (e.g., allelopathic molecules, ratios of carbohydrate to lignin). In spite of the fact that decomposition of soil organic matter is considered as a critical factor in ecosystem stability (Smith et al., 1992), the relationship between the number of plant species and composition is often

´ Claudio M. GHERSA and Rolando J.C. LEON

overlooked when agronomists and soil scientists try to understand nutrient availability for crops. Disturbance is expected to stop successional processes and modify the diversity and complexity of the community, depending upon its spatial and temporal dimensions (Glenn-Lewin and van der Maarel, 1992). Recovery of agroecosystems from disturbance is considered as secondary succession, and the dominant mechanisms are population processes (Peet, 1992). Many different kinds of disturbance have been studied (White, 1979; Pickett and White, 1985), but few studies focus on changes occurring in the species composition of weeds at different scales of time and space, and even fewer try to understand the relationships between the processes governing those changes. Recently, Swift et al. (1996) proposed a set of hypothetical models addressing the linkage between agricultural intensification and total agroecosystem biodiversity (Fig. 20.1). These models predict that the end result of the intensification of agricultural activities will be a system with very low diversity (e.g., intensive cereal production). In Fig. 20.1A, Model IV is an application of the intermediate disturbance hypothesis (Connell, 1978), in the sense that biodiversity remains high and even increases with agricultural intensification, until a critical stage of intensification is reached. Increase in intensity after this stage will produce rapid declines in biodiversity. Both model IV and model II in Fig. 20.1A allow some intensity of disturbance without losing appreciable diversity. The “hump-backed model” proposed by Grime (1973), and a similar one proposed by Tilman (1982), may also describe successional patterns and the relation between disturbance and diversity (Fig. 20.1B) According to both models species density (the average number of species in a unit area) will increase over successional time to a maximum, after which it will decrease. In both models, disturbance reduces the level of stored live and dead biomass, and allows for release of resources to the soil. Intermediate levels of disturbance thus facilitate maximum diversity, because biomass accumulation is stopped, competition is relaxed, and soil fertility is high. Grime (1979) also considered that life-history strategies replace one another during successional time, starting with ruderal species, going through competitors, and ending with late-successional stress-tolerant species. This last strategy appears when most of the soil resources available for plants during the competitor stage are stored in biomass or necromass of the flora. Grime used his model to discuss maintenance

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA

489

we present data from croplands of the rolling pampa at regional or landscape and field or patch scales which reveal changes in weed species composition, morphotypes, and physiotypes. We then discuss ideas about the mechanisms that control successional changes in agroecosystems.

DESCRIPTION OF LANDSCAPES AND CROP HISTORY OF THE STUDY SITE

Fig. 20.1. (A) Hypothetical relationships between agricultural intensification and total agroecosystem biodiversity. Note that the x-axis is non-quantitative. The four curves illustrate four different scenarios, representing differential effects of agricultural management on total biodiversity with differing implications for conservation. Adapted from Swift et al. (1996). (B) Hypothetical relationships between biodiversity and increasing crop and litter or resources, and stress or disturbance. Vertical arrows indicate probable management intensity and disturbance level for the croplands in the rolling pampa. Curve (a) represents the scenario of Connell, (b) that of Grime and Tilman. Adapted from Connell (1978), Grime (1973) and Tilman (1982).

of monocultures in agricultural systems, and argued that dominance and reductions in species richness are attainable, if practices would allow the development of a large standing crop and minimize the frequency of cropping. Adequate models and essential data do not exist for understanding how patterns of spatial heterogeneity, at large or small scales, regulate population growth or the build-up of communities during secondary succession in agroecosystems. Such understanding is crucial for the design of cultural practices and systems that generate environments which are, at least temporarily, unsatisfactory to certain weed species. In this section,

The pampas [also described elsewhere in this series (Soriano, 1991)] occupy a vast area of Argentina, including the Province of Buenos Aires, and parts of the Provinces of Entre R´ıos, Santa F´e, C´ordoba, La Pampa, and San Luis. This area corresponds to a subregion of the R´ıo de la Plata grasslands, which extends over 70 million hectares of Argentina, Brazil, and Uruguay. The size and shape of opal phytoliths in the soil show that grassland vegetation and the types of grasses composing the vegetation have been invariable throughout the period of pedogenesis (Tecchi, 1983). The entire region has been developed, especially during this century, for the livestock industry and agriculture. The area in Argentina dedicated to cropland increased from ~6 million hectares during the first 5 years of this century, to 26 million hectares in 1984 (FAO, 1986). This activity destroyed most of the natural grasslands in the arable areas of the pampas. The flora of the pampas comprises about 1000 species of vascular plants, including several that have been introduced (Parodi, 1947). Although the pampas are considered to be of uniform physiognomy and topography, several subunits are recognized on the basis of geomorphology, drainage, geology, physiography, soils, and vegetation (Soriano, 1991). All the information and the discussion of successional changes in arable land included in this chapter refer exclusively to the subunit called the rolling pampa, which is the main cropland of Argentina (hatched area in Fig. 20.2). This area is gently rolling (Fig. 20.3) and is located between 34ºS and 36ºS latitude and 58ºW and 62ºW longitude. It is bounded by the R´ıo de la Plata and the R´ıo Paran´a on the northeast, by the R´ıo Salado on the southwest, and by the R´ıo Matanza on the southeast (Soriano, 1991). The climate is temperate and humid, without a dry season, but with a hot summer. The annual average rainfall is ~1000 mm and the mean annual temperatures range between 16ºC and 17ºC.

490

´ Claudio M. GHERSA and Rolando J.C. LEON

Fig. 20.3. Summer view of a successional grassland growing in the well-drained Argiudol soil type in a gently rolling landscape in San Antonio de Areco (Buenos Aires Province). The dominant grasses are: Bothriochloa laguroides, Paspalum dilatatum, and Stipa papposa. Other frequent species are Briza subaristata, Eragrostis lugens, Melica brasiliana and Sporobolus indicus. Croplands are visible in the distant background. Trees at the roadsides and around houses are planted.

The representative soil is a Mollisol, the most common type being Argiudol. Wheat (Triticum aestivum), maize (Zea mays), and linseed (Linum usitatissimum) were the most important crops during the first stages of the cropping boom, which commenced in 1875; the area harvested for these crops rose to ~4.5, 3.0, and 1.4 million hectares, respectively, at the start of World War II. In the mid-1930s, crops of sunflower (Helianthus annuus) became significant. After the mid 1950s, grain sorghum (Sorghum bicolor) became widespread and the area dedicated to soybeans (Glycine max) increased from 0.1 million hectares in 1972 to 2.5 million hectares in 1985. Minimum-tillage and zero-tillage practices increased greatly after 1980. Irrigation and fertilization are nearly nonexistent, but recent increases in commodity prices, and the need of farmers to augment revenue, are rapidly changing this situation. Potential vegetation and plant communities of arable land It is difficult to reconstruct the composition and structure of the original vegetation of this region. By Fig. 20.2. Geographic location of the rolling pampa.

491

Number of species

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA

Community Fig. 20.4. Temporal pattern of increase of species producing high quantities of secondary metabolites (alkaloids, terpenoids, and cyanophoric glucosides), relative to the change in total number of species; C, cropland; G, grassland. Adapted from Su´arez (1997).

the time of the first floristic surveys (Parodi, 1930), virgin grasslands were already rare. In the rolling pampa (Fig. 20.2) vegetation structure corresponds to a prairie in humid years and to pseudo-steppe during dry periods (Fig. 20.3). Winter temperatures are rarely a serious limitation, whereas drought in summer frequently inhibits growth of most species. The species that characterize the dominant community in the fertile arable soils are: Bothriochloa laguroides, a grass with short rhizomes which grows only during the high-temperature period of the year; Stipa neesiana, a bunch-grass up to 50 cm high; and three grasses forming small tufts, Aristida murina, Piptochaetium montevidense, and Stipa papposa. Other common grasses are Melica brasiliana, Paspalum dilatatum, and Piptochaetium bicolor. Agrostis montevidensis, Briza subaristata, Bromus unioloides, Danthonia montevidensis, Eragrostis lugens, Panicum bergii, Paspalum notatum, Poa bonariensis, Schizachyrium spicatum, Setaria parviflora, and Stipa hyalina form a set of less frequent grass species. Shrubs and suffruticose plants are poorly represented. The most frequent species in this category are Baccharis articulata, B. coridifolia, B. notosergila, B. trimera, Eupatorium subhastatum, Heimia salicifolia, Hedeoma multiflora, Margyricarpus pinnatus, and Vernonia rubricaulis. Small broad-leaved herbs and sedges including Adesma bicolor, Berroa gnaphalioides, Carex bonariensis, Chaptalia spp., Chevreulia sarmentosa, Conyza spp.,

Facelis retusa, Hypochaeris spp., Micropsis spathulata, Oxalis spp., Phyla canescens, Polygala australis, Tragia geraniifolia, Verbena spp., and Vicia spp. are interspersed among the grasses (Soriano, 1991).

CHANGES IN THE DIFFERENT AGROECOSYSTEMS

Successional changes in cropped land Regional-landscape scale A clear pattern can be discerned based on the floristic composition in maize croplands in the rolling pampa, using the data of Parodi (1926, 1930) and two phytosociological surveys carried out in an area of approximately 2.5×106 ha by Le´on and Suero (1962) and Su´arez et al. (1995). Weeds present in maize fields of the rolling pampa are in a non-equilibrium state similar to that observed in Denmark by Haas and Streibig (1982) and in the northwestern United States by Forcella and Harvey (1983), in which total species richness and species equitability have increased since the turn of the century. The number of species in the original grasslands of the rolling pampa growing in well-drained soils was ~222. It was dramatically impoverished by early agricultural activities to ~53 by 1926 (32 of the original flora plus 21 new) (Parodi, 1926) (Fig. 20.4). This

´ Claudio M. GHERSA and Rolando J.C. LEON

492

B

Species number

Species number

A

Community

Community

Fig. 20.5. (A) Temporal changes in morphotypes of species in the grassland community (G) and the maize crop weed community (C); solid bar, dicots; shaded bar, monocots. (B) Temporal changes in the number of species of different origin; hatched bar, native; shaded bar, exotic; solid bar, cosmopolitan. Adapted from Ghersa et al. (1996).

impoverishment in the flora was not unexpected, as the native habitat for the grassland species was lost by soil tilling. Thereafter, the agricultural landscape was continually invaded by weeds. In 1960, the total number of species in maize crops was 79 (Le´on and Suero, 1962) – 34 from the original grassland and 45 exotic to the grassland; at the present time, the total number of species present in maize crops is 99 (Su´arez et al., 1995), 54 from the original grassland and 45 exotic (Fig. 20.4). Species number increased during the period from 1926 to 1960, and again from 1960 to 1995, at rates of 1.32 and 1.50 species per year, respectively. The average net rate of species increase per year since 1926 is 1.01 (Ghersa et al., 1996). Through this process, a great proportion of the floristic richness of the rolling pampa has been restored, even though the native perennial grasses have nearly disappeared (Fig. 20.5). The cultivated landscape is characterized as a mosaic, where patches of bare ground are interspersed among areas experiencing different frequencies and intensities of disturbance of the soil surface. The ratio of the landscape area to the disturbance area is crucial in determining the dynamics of the landscape as a whole (Prentice, 1992; Harvey and May, 1997). Cropping activities continually provide propagules of weed species (Radosevich et al., 1997), relax competition by reducing plant density, and mineralize nutrients in the soil organic matter. Thus, communities in arable land can be characterized as

highly invasible (Crawley, 1987). This can explain why species richness has increased over time. Nevertheless, early in the century, most of the species already present in the rolling pampa were excluded from the weed community in maize croplands; but with time an increasing number of the species belonging to the original grassland reinvaded the weed community (Fig. 20.5B). This observation prompts the question as to whether species that were excluded early in the century by cultivation have evolved and adapted to the disturbance regime of maize croplands, or whether the disturbance regime and the environment as a whole have changed, and now are suitable for the weed species originally excluded. Although there is no way of directly answering these questions, some insight can be generated by considering how groups of species distinguished by origin (native–exotic), morphotypes (dicotyledon–monocotyledon), and physiotypes (C3 – C4 metabolic pathways, or production of secondary metabolites: essential oils, coumarins, and alkaloids), have changed over time. The greatest increase in species richness in the rolling pampa was registered for native and exotic dicotyledons. Weeds with high production of secondary metabolites also increased in relation to the total number of species (Fig. 20.4), but the ratio of C3 to C4 species has remained at approximately 1 from 1930 to 1990 (Ghersa et al., 1996). Because the greatest load of herbicide applied in the region was used to control broad-leaved species in cereal crops (Hall

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA

et al., 1992), a reduction in dicotyledonous weeds in relation to grasses would have been expected (Fryer and Chancellor, 1970). At a regional scale, this was not true. The new species to invade the region were frequently dicotyledons (Fig. 20.5A). In agricultural lands, C4 species should increase in abundance with respect to C3 , because they are better adapted to water and nitrogen stress caused by soil deterioration (Baker, 1974). This could be particularly important in the conditions of the rolling pampa, where cropping is carried out in rain-fed systems with low fertilization. In contrast, the C3 :C4 ratio remained unchanged through the study period, despite the changes that occurred in agroecosystems (Hall et al., 1992) and in the specific composition of the weed community. Climate may be governing the invariant C3 :C4 ratios (Stowe and Teeri, 1978; Fowler, 1981). The seasonal change in the rolling pampa, from a hot summer with a negative water balance, to a temperate winter with positive water balance, allows the existence of both C3 and C4 physiotypes by curtailing any selective advantage experienced by one physiotype in a particular scenario. Holzner (1982) has suggested that climate is a primary factor determining the geographic distribution of weeds, and that anthropogenic and ecological variables occupy subordinate levels, relevant to fitness at the patch scale. If one accepts that the C3 :C4 ratio is controlled by climate, then what is the explanation for the higher relative increase in dicotyledonous species in spite of the negative environments created anthropogenically by herbicides? In the original grassland, dicotyledonous species were more numerous than monocotyledons, and they still remain more frequent in old pastures (Fig. 20.5A). Nevertheless, if one accepts that plant density in cropland is lower than that in undisturbed grassland, and that the crop is the dominant species in the community (all the weed communities that we have described were in well-managed fields, and most weeds were small and subordinated, usually covering less than 20% of the ground), then one could speculate that the relative increase in dicotyledons relative to monocotyledonss is generated by differences in competitive ability. Monocotyledons are well adapted to high-density environments, with high levels of irradiance. In contrast, dicotyledons are better adapted to lower densities and lower irradiance levels (Koner, 1993). Since most of the light is intercepted by the

493

dominant crop species, dicotyledons should have a competitive advantage. Production of secondary metabolites increases when plants are grown in environments imposing biotic or abiotic stress (Coley et al., 1985; Herms, 1992; Wink, 1993). Species producing high levels of secondary metabolites are well represented in mature stages of succession (Grime, 1979). This can be interpreted as a successional trend imposed by species interactions, as in any secondary succession, or by degradation of the land caused by cropping. In the first case, competition and density-dependent mortality increase over time, favoring stress tolerant strategies in response to increases in live and dead biomass. In the second interpretation, stress conditions should appear by reduction in water and nutrient availability for plant growth caused by impoverishment of the physical and chemical properties of the soil. Field and patch scale A patch of vegetation can be defined as an area small enough for all the individual plants growing in it to have strong interactions. This means that the area of a patch can extend only to as much as a few square meters in herbaceous systems or to a thousand square meters in tree systems (Prentice, 1992). In arable land, interactions may be either reduced or enhanced by cultural activities. Cultural activities disperse pests and diseases, facilitate herbivory and source–sink relations, and relax competition. For this reason we are considering that a field under a particular cropping system is similar to a patch in a natural system. This similarity should be high when field area is small and it decreases with extension of field area. At the field or patch scale, it is less probable to find stability in the community than at the regional or landscape scale. This means that changes in species abundance and richness over short time intervals are more readily expected. At a regional scale, changes in climate and soils control vegetation dynamics, but at the patch scale disturbance has a greater impact. Disturbance related to cropping activities should generate non-equilibrium systems (White, 1979). The pattern of succession should appear only if one looks at long records of successive disturbance cycles (Soriano, 1971; Delcourt et al., 1983). Species richness of weeds increased since 1930, and there is a particular pattern of change in relation to land-use history and system of tillage. In a survey (Le´on and Suero, 1962) of fields with maize crops grown under conventional

´ Claudio M. GHERSA and Rolando J.C. LEON

494 Table 20.1 Species richness and crop yield for soybean and maize crops in the rolling pampa Crop 1

Maize Maize Soybean 4 Soybean 5

1 2 3 4 5

Year

1960 1990 1990 1990

Number of fields surveyed

Field yield (kg ha−1 )

15

3360

15

3030

11 10

Relative yield 2

Species richness 3 Field

Community

1.05

11.2

44

0.95

15.7

48

7940

1.12

22.0

67

5840

0.83

15.7

49

15

3470

1.08

13.8

45

26

2910

0.73

10.5

51

13

1630

1.13

16.1

44

6

1260

0.87

14.3

33

The first record of each crop system corresponds to high-yield fields, and the second to low-yield fields. Relative yield was calculated as the ratio of the yield of a given field to the average yield of all fields surveyed in that year. Field, average number of species recorded per field; community, total number of species recorded in fields with the same community. Conventional tillage. Reduced tillage.

tillage (ploughing and harrowing), values of species richness at the level of field and community were higher in fields producing a low yield of grain (relative to the mean production for the year of the survey) than in fields with a high yield. The opposite relation was registered in the 1990 survey (Table 20.1; Su´arez et al., 1995). In addition, richness of weed species was analysed in soybean crops grown under conventional tillage and under reduced tillage. In this case, species richness at point and community level increased when a conventional tillage system was used. There was an increase in both field species richness and relative yield of soybean seed in fields cultivated with conventional tillage, but community richness was lower when relative yields were high. In reduced tillage systems, relative yield of soybean seed and species richness increased. In contrast, community richness was higher when relative yields were high (Table 20.1). In the 1990 surveys, field and community richness of weeds in maize were higher than those in soybean. This means that the communities of weed species were distinct in spite of the fact that all surveys were conducted in fields with the same soil types. All these differences in weed species richness show how heterogeneity at the scale of fields (50–150 ha) induced anthropogenically contributes to community and landscape diversity (McNaughton, 1983). At a regional scale this heterogeneity provides for a diversity of species, allowing succession to continue (Prentice, 1992).

The 1960 and 1995 phytosociological surveys in maize crops (Le´on and Suero, 1962; Su´arez et al., 1995) can be used to classify species according to their constancy, and to observed changes in composition of their groupings in trying to understand successional changes. It can be considered that species present in a high proportion of stands (high constancy values) are well adapted to the regional climate and soil conditions. Those having low constancy values are only adapted to particular local conditions, or are newly invading species. In the 1960 survey Amaranthus quitensis, Anoda cristata, Chenopodium album, Datura ferox, Digitaria sanguinalis, Echinochloa colonum, Paspalum distichum, Physalis viscosa, and Setaria geniculata had constancy values greater than 50% (Group I). Euphorbia lasiocarpa, E. ovalifolia, Polygonum aviculare, Portulaca oleracea, Tagetes minuta, Xanthium cavanillesii, and X. spinosum had constancy values between 20 and 50% (Group II). Species of less than 20% constancy (Group III) added up to 64% of the total number of species, and are not presented here. In the 1990 surveys, Group I remained unchanged except for the loss of three species: P. viscosa and S. geniculata moved to Group II, and P. distichum, to Group III, and the gain of four species: E. lasiocarpa, P. oleracea, and T. minuta, from Group II, and Sorghum halepense from Group III. In the 1990 survey Group II comprised Bidens subalternans, Coronopus didymus, Dichondra microcalyx, Oxalis chrysantha, Physalis viscosa, Setaria geniculata, Sonchus oleraceus, Stellaria media,

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA

495

Fig. 20.6. Multivariate analysis for Tagetes minuta chemiotypes (production and composition of terpenoids) for plants collected from different provinces and grown in Buenos Aires under experimental conditions: principal components analysis by the geographic origin of each plant collection. S, Salta; J, Jujuy; B, Buenos Aires; M, Mendoza; R, R´ıo Negro; A, San Juan provinces. Adapted from Gil et al. (1999).

and Veronica persica. All of these species had low constancy values in the previous survey. Group III in the 1990 survey included 69% of the total species. This can be seen as a replacement of species caused by a successional trend responding to climate and/or soil changes. There is no information suggesting that climate has effectively changed during this period, but it is known that the soil changed. In Argentina very little fertilizer is applied, and tillage frequency has increased since 1970 (Hall et al., 1992). Changes in constancy of some weed species shows that the soil environment in cultivated land is becoming unsuitable for some species, whilst it is becoming better for others. An increase in species richness in the 1990 survey, particularly in Groups II and III, suggests that there is a slow process whereby new habitat is created, and species adapted to it are invading. Tagetes minuta was one of the species that increased its constancy in the course of 30 years. It is known to produce terpenoids and tiophenes that are biologically active as biocides, repellents, and attractors for several insects (Soule, 1993). This South American species was present in the area of the city of Buenos Aires, and in Chile, where it was collected in 1724 and described as a medicinal plant used by American natives (Dillenius, 1732; Cabrera, 1967). Although distributed in a wide geographic area, it was not present in the original

grassland community of the rolling pampa (Parodi, 1926), but now is one of the dominant species in maize fields. Recent studies show that T. minuta obtained from different locations in Argentina differ in the production and composition of secondary metabolites, thus showing some degree of allopatric speciation. When the collections from different provinces were grown in a single controlled environment, the plants from the rolling pampa (Buenos Aires Province) had the greatest variability in the amount and quality of secondary metabolites compared to the rest of the collections. The range of variation observed for the rolling pampa collection was nearly the same as that observed for all the collections together (Fig. 20.6). Production of secondary metabolites by plants from the rolling pampa was also sensitive to changes in stress related to competition and the presence of nematodes (Gil et al., 1999). The variability and the plasticity in the rolling pampa populations may have appeared as an adaptive adjustment to the agricultural environment. Variability found in many weed populations has been explained by gene flow, enhanced by cultural practices, among populations that had undergone some process of allopatric speciation (Ghersa et al., 1994). Through this process, populations may gain phenotypic plasticity, as in the case of T. minuta plants from the rolling pampa, which allocate energy to defence only under biotic

496

stress. This strategy probably has allowed this species to survive and invade an agricultural environment characterized by pest outbreaks and an alternation of resource abundance (when the soil is ploughed) and restriction (when most of the resources are monopolized by the crop). Differences in point richness and community richness of weeds are partly due to differences in the soil factors that control weed germination, according to observations by de la Fuente (1997) and Su´arez (1997). These authors conducted reciprocal seed plantings in fields with high and low levels of soil degradation. Seeds came from the weeds present in the community associated with each level of degradation. They also observed germination response of weed species to manipulated soil conditions. Differences in soil temperature in the 5-cm surface layer differentially affected weed germination. The differences in soil temperature were caused by changes in the soil and in the accumulation of litter. Surfaces of less degraded soils were darker and absorbed more radiation than did more degraded soils. The difference in color was caused by higher organic matter content and lower clay content of the soils subjected to less cultivation and erosion. In the degraded soils, clay content in the plough layer increased because the eroded organic horizon was mixed with the clay-rich B horizon. Litter on the soil surface was related to the reduced tillage systems. Promotion or reduction in seed germination of weed species was in accordance with differences observed in point and community richness. In soybean crops grown with conventional tillage, for example, constancy of Bidens subalternans was 4% when the soil was highly degraded, as against 19% when the soil was less degraded. Experiments on how seed germination of this species is regulated by soils with characteristics similar to those in the fields surveyed revealed that the changes in species constancy were associated with differences in both seed germination and seedling emergence (Fig. 20.7). When changes in the soil environment are not related to the mixture of soil horizons caused by erosion and tillage, the most important changes for the seeds are caused by the effect of tillage on litter accumulation. The effect of tillage system, fertilization, and crop residues on carbon balance and respiration, as well as on the distribution of organic matter and microfauna in the soil, was evaluated in various studies conducted in different sites of the rolling pampa (Pilatti et al., 1988;

´ Claudio M. GHERSA and Rolando J.C. LEON

Alvarez et al., 1995a–e; Alvarez et al., 1996). These studies showed that the main changes are related to soil temperature and stratification in mineralization. Litter accumulation on the soil surface following reduced tillage diminishes the average maximum and daily range of soil temperatures. In conventional mouldboard ploughing systems, both organic matter and decomposer activity are distributed throughout the 15– 20 cm plough layer, whereas in the undisturbed and reduced-tillage systems, activity is concentrated in the top 5-cm layer of soil, and decreases exponentially with depth. In spite of these differences in stratification of mineralization activities, emission of carbon dioxide and the overall carbon dynamics in bulk soil were not changed by tillage system or nitrogen fertilization during the 5 years of observation (Alvarez et al., 1996).

Fig. 20.7. Cumulative seedling emergence of Bidens subalternans in the following treatments: high soil degradation, light surface (solid triangles); low soil degradation, dark surface (solid squares), and high soil degradation, artificial dark surface (solid circles). Lines represent regression models fitted to the observed values. From de la Fuente (1997).

Successional changes in pastures We consider pastures separately from croplands because they differ in frequency and intensity of anthropogenic disturbance. When sown, pastures are like any cropland, but subsequently succession occurs under the less disturbed conditions of grazing. Succession in these conditions is fairly rapid at the beginning, but then slows down, depending on the intensity and

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA

Field and patch scale Successional studies of the rolling pampa (Le´on and Oesterheld, 1982) show that fields sown as pastures,

Relative cover (%)

Years Fig. 20.8. Changes in cover of planted (solid line) and spontaneous (dotted line) species in pastures of the rolling pampa. From Le´on and Oesterheld (1982).

Diversity

Regional and landscape scales Phytosociological surveys carried out in 1995–1997 in 45 fields with old pastures (>15 years since sowing) made it possible to compare their floristic composition with one of the natural grasslands described by Parodi (1930). Surprisingly, both the 1930 and the 1995– 1997 communities had ~220 species (Fig. 20.5A). Although changes in species composition occurred, because total species number of the old pastures is composed of 147 of the grassland original species and 75 new ones (Fig. 20.5B), a successional climatic stability with regard to diversity persists. The average rate of increase of new species was 1.15 species yr−1 . The structure of the community remained unchanged, but replacement occurred in about 25% of rare and subordinate species. The most important species that were replaced included Cenchrus myosuroides, Desmanthus sp., Trifolium argentinense, and T. polymorphum. The new species in the community are weeds such as Crepis sp., Hypochaeris radicata, Senecio burchelii, planted ornamentals like Leucanthemum vulgare, and forage species such as Agropyron elongatum, Festuca arundinacea, Lotus tenuis, Medicago sp., Melilotus alba, Phalaris aquatica, and Trifolium sp. It is important to note that the rate of replacement in a grassland community through the last 70 years is quite similar to that observed during the same period for the maize weed community. This rate of species replacement has been observed in other grasslands as a response to grazing or anthropogenic disturbances (Sala et al., 1986; Swift et al., 1996). An important question remains unanswered: do the new introductions perform the same function as do the species that they replaced [probably not], and what are the consequences of changes on the overall behavior of the system?

with similar management practices and soil types, experience a reduction in the cover of the planted species and increases in species richness and diversity (as measured by the Shannon–Weaver index: Whittaker, 1977; see Figs. 20.8, 20.9). At late stages, the dominant species are those of the native community (Le´on et al., 1984). Apparently, the combined effect of cattle grazing and planted species does not inhibit or delay succession, which appears to be driven by soil and climatic conditions.

Number of species

frequency of grazing. Vegetation is relatively stable so long as abiotic stress or biotic factors inhibit or at least delay further succession (van Andel et al., 1993). As soon as the density of herbivores is reduced below the level of carrying capacity by an external factor, the rate of vegetation succession increases. Moderate grazing generates a patchy system, where some patches are heavily grazed while others are untouched. Succession proceeds in the ungrazed patches (Bakker et al., 1983).

497

Years Fig. 20.9. Changes in species richness and diversity in pastures of the rolling pampa. Shannon–Weaver index of diversity (solid line), species richness (dotted line). From Le´on and Oesterheld (1982).

Successional changes in wastelands, relicts, and corridors The rolling landscape of this region is shaped by geomorphological processes, highly distorted by human activities. Anthropogenic modification of the landscape takes place in different ways. The region

498

´ Claudio M. GHERSA and Rolando J.C. LEON

Fig. 20.10. Succession in wasteland and a fence corridor in the rolling pampa. The dominant woody species at road borders are Baccharis articulata and B. notosergila, and, near the fence, Gleditsia triacanthos and Parkinsonia aculeata

is divided by fences to prevent animals from grazing grain crops, as well as by roads, railways, and electric or telephone lines. Together, these generate a network of corridors of semi-natural (non-cropped) vegetation. Because the average size of fields ranges between 50 and 100 ha, even narrow corridors (1– 300 m wide) cover a significant proportion of the landscape. Moreover, the region is characterized by small areas of abandoned land near corrals, silos, houses, and railway stations, as well as on the outskirts of small towns and cities. These small areas differ in size and frequency, and range from a few square meters to a few hectares. In these areas, the structure of the community is modified because poorly represented shrubs and trees from the semi-natural community invade disturbed and abandoned areas. The grass community is dominated by perennial grasses, such as Cortaderia selloana and Paspalum quadrifarium, forming large tussocks mixed with weeds like Dipsacus fullonum and Sorghum halepense. The main woody species are native taxa such as Acacia bonariensis, Baccharis sp., Discaria longispina, Aloysia gratissima, Parkinsonia aculeata, and Sphaeralcea bonariensis, and exotic species such as Broussonetia papyrifera, Gleditsia triacanthos, Ligustrum sp., Melia azedarach,

and Morus alba. Gleditsia triacanthos is also invading riparian zones along the many rivers and creeks of the rolling pampa landscape. These areas are important habitats for wildlife, woody species, and perennial grasses from the original community. Small mammals such as rodents and armadillos, as well as flying and walking birds, use these corridors to escape cropping activities and predation. The absence of agricultural activities, together with sporadic fire and dispersal of tree seeds by birds that perch on fences (Montaldo, 1993), are the main factors enhancing the invasion of corridors, waste land, and relicts by woody species (Fig. 20.10). Together, these factors drive succession away from the original grassland. Facelli and Le´on (1986) and Mazia et al. (1996) carried out experiments in the inland pampa to evaluate if the grassland could be invaded by woody species. Their results support the idea that, when soil tillage is absent, and some removal of above-ground biomass of grass takes place, Gleditsia triacanthos, Prosopis caldenia, and Ulmus pumila can become established. The successional process, driven by the invasion of woody species of the waste land and corridors, generates areas with dense populations of shrubs and trees, many of them having thorns. This creates

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA

problems of visibility, particularly at road intersections and bridges, and impedes access to fences. There is also the perception that these areas are habitat for unwanted mammals such as foxes, mice, and armadillos, as well as for weeds and diseases. For this reason, land owners and government are forced to remove trees to keep fences and roads clean. Nevertheless, some patches have escaped this removal and are now small forests.

CONCLUDING REMARKS

We have shown that, at the regional scale, there has been an increase in species number in the spontaneous vegetation of croplands in the rolling pampa. The increase has occurred without loss of the dominant weeds, and some functions (e.g., the C3 and C4 photosynthetic pathways, and production of secondary metabolites). This corresponds to the “press” perturbation model (Williamson, 1987). In contrast, total species number in the old pastures was quite stable over time, in spite of the fact that some species replacement occurred. In these old pastures, the dynamics of the community fits better with a classical model for secondary succession, in which changes are driven by competitive exclusion. Less frequent soil tillage in pastures than in cropland allows for high accumulation of biomass in pastures and increases in plant density, so that competition among plants is increased. Nevertheless, in both the croplands and the pastures, total number of species and the structure of the community are apparently constrained by soil and climate. At the regional scale, the rate at which agricultural land gained species was surprisingly similar to the rate at which new species invaded pastures, replacing the original species of the rolling pampa grassland (Parodi, 1930). This could just be a coincidence, or the same process could be regulating the invasion of new species in the highly disturbed cropped field and the less disturbed pastures. It is possible that, in both the cropped land and the pastures, the arrival of propagules of new species and soil and climatic constraints are governing the dynamics of the weed flora at similar rates. In cropland, the press perturbation caused by the dominant crop, which is added each year to the spontaneous weed community, creates a “crop climatic environment” allowing for an over-representation of

499

dicotyledonous species. In the old pastures dicotyledons are also over-represented, probably because of the effects of grazing. In spite of the fact that woody species have in the past been poorly represented in the rolling pampa, the successional process observed in the wastelands and relicts unequivocally suggests that, should agriculture stop, the grassland would change to some kind of woodland or a savanna type vegetation. Data at the regional scale do not correspond to any of Swift’s models, all of which predict a regressive succession, in which diversity decreases as intensity of disturbance increases (Fig. 20.1). After agriculture expanded over the grasslands in the rolling pampa at the beginning of this century, species richness and landuse intensity increased together (Hall et al., 1992). Although fields of the rolling pampa are not irrigated, and still have a relatively low load of pesticides and fertilizers as compared to Europe and the United States, today’s agriculture in the rolling pampa is intensively managed. On the other hand, the situation in the 1930s was characterized by low-intensity management. Pastures with disturbance intensity lower than that experienced by fields with annual crop species have a higher species richness. This would be in agreement with a decay in diversity in relation to an increased disturbance intensity. Considering the patch or field scale, our data could fit Connell’s Model IV because: (1) community richness of the weed species grew with an increase in the period under maize production (low relative grain yield is related to high richness of the weeds); (2) the opposite occurred in the 1990 survey, (low relative yields of maize are related to low richness of the weeds); and (3) in soybean croplands, richness under conventional tillage is higher than under reduced tillage (Table 20.1). Low yields are directly related to soil degradation, and this in turn is related to the intensity of disturbance. According to Grime’s (1979) concepts of life strategies, the replacement of life strategies over time in the agroecosystem we studied probably follows a similar trend to that in natural succession. This is based on the assumption that availability of soil resources follows a similar trend in natural and agricultural ecosystems, but that the change is caused by different mechanisms. In the natural system, soil is depleted by biotic consumption, thus stress-tolerant strategies are favored in mature successional stages. In croplands, soil resources are scarce as a consequence of erosion, and because nutrients are exported with the harvested crops. If the increase in constancy of species that produces a

500

high level of secondary metabolites (Fig. 20.4) reveals an increase in stress tolerance, then soil degradation may be forcing replacement of competitors by stresstolerant species. Recent invasion of cultivated lands by native species of the original grassland, which in early stages of the cropping history were excluded from the maize weed community, further supports this idea. If the original grassland was mature, species of that community would have been stress-tolerant. When soils were highly fertile, native species were excluded by crops and by weeds with competitive strategies. Now that the soils represent different degrees of degradation, native species that are stress-tolerant can reinvade. The overall effect of increased representation of the stresstolerant strategy in the flora of the agroecosystem is unknown, but stability should increase over time, as indicated by increase in species number and/or species replacement. A weed community dominated by plants producing terpenoids, thyophenes, coumarins, latex, and alkaloids should have an important biological impact, and could negatively affect pest outbreaks (Wink, 1993). Stress-tolerant weed species themselves are more difficult to control, as evident from the stability of the major weeds present in maize crops, despite the efforts invested in controlling them. Both of these attributes (i.e., stress-tolerance, and production of secondary metabolites) have consequences to pest and weed management of agroecosystems. A tradeoff may be possible; the beneficial effect of stresstolerant weeds in pest control may be balanced against the negative effect on crop yield. This trade-off differs between the landscape and the field level. There are several processes operating simultaneously in the agroecosystems of the rolling pampa. There are changes in the landscape related to human activities, creating variability in “gamma” diversity. Abandoned land, roads, corridors, wasteland, and a diversity of crops and farming practices generate a large diversity of habitats. This structural complexity at the regionallandscape level creates conditions suitable for invasion processes, whereby the dynamics and composition of the weed community become insensitive to changes in agricultural practices over time. Habitat diversity creates refuges for native and exotic plant species, impeding species extinction and allowing the development of a population to a threshold size, above which it can withstand the environmental stochasticity of the agricultural landscape and sustain recolonization of empty patches (Mack, 1995)

´ Claudio M. GHERSA and Rolando J.C. LEON

Therefore, large-scale invasion processes are regulated by climatic constraints to succession, and are important in maintaining species diversity. This diversity functions as the genetic and species bank from which weed invasions on the patch or field scale and species evolution can be nourished continually. At the field or patch scale, the composition of the weed flora associated with a crop is dynamic and responsive to changes in agricultural practices over time. Community dynamics and structure are controlled by the crop understorey environment, and by the physical and chemical properties of the soil. Most of the hypotheses describing how agricultural disturbances affect biodiversity are based on differences in diversity among plots differing spatially in the intensity and frequency of the disturbance (Pimentel et al., 1992; Swift and Anderson, 1992). It is clear that disturbance reduces biodiversity, and this effect can be thought to end in a monoculture when disturbance is extreme. Alternatively, if changes in biodiversity are observed following each type of disturbance over time, the effect of disturbance may be quite different. Successional and evolutionary processes will tend to reduce the original differences among levels of disturbance. For example, in the rolling pampa in 1930, the grassland community had 169 more species than did the maize community. In 1996, the difference was reduced to 123 species, and in areas cultivated with maize the number of species that belonged to the original grassland community increased from 25 to 46. It is notable that successional changes in the stresstolerant weed community in the arable land of the rolling pampa can lead to stability similar to that which can be expected for a mature successional stage. The successional trend observed for the rolling pampa is similar to that observed for weeds in Europe and in the United States. Not only is succession not stopped by agricultural activities, but it follows the expected trend for natural systems: regional diversity increases with time. At the regional and patch scale, physiotype (C3 :C4 ratio) is more stable than species number, morphotype, or chemiotype. Unfortunately, weed researchers have often overlooked the longterm ecological processes involved in colonization, succession, and equilibrium. This oversight can lead to misinterpretation of the factors causing species shifts and limit predictive ability concerning future weed invasions, species losses, or evolution of herbicideresistant populations. Clearly, it will be a long time before accurate predictions are commonplace. Until the

SUCCESSIONAL CHANGES IN AGROECOSYSTEMS OF THE ROLLING PAMPA

key factors regulating successional changes in agricultural lands can be identified, adequate management of weed or other pest populations will not be possible.

REFERENCES Alvarez, R., Santanatoglia, O.J. and Garcia, R., 1995a. In situ decomposition of 14 C-labelled wheat as affected by soil disturbance. Agrochimica, 39: 343–347. Alvarez, R., Santanatoglia, O.J. and Garcia, R., 1995b. Soil respiration and carbon inputs from crops in a wheat–soybean rotation under different tillage systems. Soil Use Manage., 11: 45–50. Alvarez, R., Santanatoglia, O.J. and Garcia, R., 1995c. Effect of temperature on soil microbial biomass and its metabolic quotient in situ under different tillage systems. Biol. Fertil. Soils, 19: 227–230. Alvarez, R., Santanatoglia, O.J., Daniel, P.E. and Garcia, R., 1995d. Respiration and specific activity of soil microbial biomass under conventional and reduced tillage. Pesqui. Agropecu. Bras., 30: 701–709. Alvarez, R., Diaz, R.A., Barbero, N., Santanatoglia, O.J. and Blotta, L., 1995e. Soil organic carbon, microbial biomass and CO2 -C production from three tillage systems. Soil Tillage Res., 33: 17–28. Alvarez, R., Russo, M., Prystupa, P. and Blotta, L., 1996. Flujo de C-CO2 a la atm´osfera y contenido de materia org´anica del suelo bajo labranza convencional y siembra directa. Actas XV Congreso Argentino de Ciencia del Suelo, pp. 189–190. Baker, H.G., 1974. The evolution of weeds. Ares, 5: 1–24. Bakker, J.P., De Bie, S., Dallinga, J.H., Tjaden, P. and de Vries, Y., 1983. Sheep grazing as a management tool for heathland conservation and regeneration in the Netherlands. J. Appl. Ecol., 20: 541–560. Cabrera, A.L., 1967. Flora de la Provincia de Buenos Aires. Colecci´on Cient´ıfica INTA, Tomo IV, 443 pp. Coleman, D.C. and Crossley, D.A., 1996. Fundamentals of Soil Ecology. Academic Press, San Diego, 205 pp. Coley, P.D., Bryant, J.P. and Chapin III, F.S., 1985. Resource availability and plant antihervibore defence. Science, 230: 895–899. Connell, J.H., 1978. Diversity in tropical rain forests and coral reefs. Science, 199: 1302–1310. Crawley, M.J., 1987. What makes a community invasible? In: A.J. Gray, M.J. Crawley and P.J. Edwards (Editors), Colonization, Succession, and Stability. Blackwell Scientific Publications, London, pp. 429–455. de la Fuente, E., 1997. Identificaci´on de las estrategias adaptativas de las malezas en un sistema de producci´on de soja con labranza cero como base para el dise˜no de alternativas de manejo. MSc Thesis, University of Buenos Aires, 92 pp. Delcourt, H.R., Delcourt, P.A. and Webb, T., 1983. Dynamic Plant Ecology: the spectrum of vegetational change in space and time. Quant. Sci. Rev., 1: 153–175. Dillenius, J.J., 1732. Tagetes Multiflora, Minuta Flore. Plantarum Bariorum. Lam 280. Hortus Elthamensis, 2: 374–375. Facelli, J.M. and Le´on, R.J.C., 1986. El establecimiento espont´aneo de a´ rboles en la Pampa. Phytocoenologia, 14: 263–274.

501

FAO, 1986. FAO Production Yearbook 1985, Vol. 39. FAO Stat. Ser. 70. Food and Agricultural Organization, Rome, 330 pp. Forcella, F. and Harvey, S.J., 1983. Relative abundance in an alien weed flora. Oecologia (Berlin), 59: 292–295. Fowler, N., 1981. Competition and coexistence in a North Carolina grassland. II The effects of the experimental removal of species. J. Ecol., 69: 843–854. Fryer, J.D. and Chancellor, R.J., 1970. Evidence of changing weed populations in arable lands. Proc. 10th Br. Weed Control Conf., 3: 958–964. Ghersa, C.M., Roush, M.L., Radosevich, S.R. and Cordray, S.M., 1994. Coevolution of agroecosystems and weed management. Bioscience, 44: 85–94. Ghersa, C.M., Martinez-Ghersa, M.A. and Su´arez, S.A., 1996. Spatial and temporal patterns of weed invasions: Implications for weed management and crop yield. In: H. Brown, G.W. Cussans, M.D. Devine, S.O. Duke, C. Fernandez-Quintanilla, A. Helweg, R.E. Labrada, M. Landes, P. Kudsk and J.C. Streibig (Editors), Proceedings Second International Weed Control Congress, 2: 41–47. Gil, A., Ghersa, C.M. and Leicach, S., 1999. Essential oil yield and composition of Tagetes minuta accessions from Argentina. Biochem. Syst. Ecol., in press. Glenn-Lewin, D.C. and van der Maarel, E., 1992. Patterns and processes of vegetation dynamics. In: D.C. Glenn-Lewin, R.K. Peet and T.T. Veblen (Editors), Plant Succession. Theory and Prediction. Chapman and Hall, London, pp. 11–59. Grime, J.P., 1973. Competitive exclusion in herbaceous vegetation. Nature, 242: 344–347. Grime, J.P., 1979. Plant Strategies and Vegetation Processes. Wiley, New York, 222 pp. Haas, H. and Streibig, J.C., 1982. Changing patterns of weed distribution as a result of herbicide use and other agronomic factors. In: M.H. LeBaron and J. Gressel (Editors), Herbicide Resistance in Plants. Wiley, New York. Hall, A.J., Rebella, C.M., Ghersa, C.M. and Culot, P.H., 1992. Crop systems of the pampas. In: C.J. Pearson (Editor), Field Crop Ecosystems. Ecosystems of the World 18. Elsevier, Amsterdam, pp. 413–449. Harvey, P.H. and May, R.M., 1997. Case studies of extinction. Nature, 385: 776–777. Herms, D.A., 1992. The dilemma of plants: to grow or defend. Q. Rev. Biol., 67: 283–335. Holzner, W., 1982. Weeds as indicators. In: W. Holzner and M. Numata (Editors), Biology and Ecology of Weeds. Dr. W. Junk Publishers, The Hague, pp. 187–190. Koner, Ch., 1993. Scaling from species to vegetation: the usefulness of functional groups. In: E.-D. Schulze and H.A. Mooney (Editors), Biodiversity and Ecosystem Function. Springer Verlag, Berlin, pp. 117–137. Le´on, R.J.C. and Oesterheld, M., 1982. Envejecimiento de pasturas implantadas en el norte de la depresion del salado. Un enfoque sucesional. Rev. Fac. Agron., 3: 41–49. Le´on, R.J.C. and Suero, A., 1962. Las comunidades de malezas en los maizales y su valor indicador. Rev. Argent. Agron., 29: 23–28. Le´on, R.J.C., Rusch, G.M. and Oesterheld, M., 1984. Pastizales pampeanos-impacto agropecuario. Phytocoenologia, 12: 201–218. Mack, R.N., 1995. Understanding the processes of weed invasions: the

502 influence of environmental stochasticity. In: Weeds in a Changing World. BCPC Symposium Proceedings, 64: 65–74. Mazia, C.N., Chaneton, E.J., Le´on, R.J.C. and Ghersa, C.M., 1996. Tree species colonization in pampean grasslands and forest plant communities. Proc. Annu. Meet. Ecol. Soc. Am., 77: 290. McNaughton, S.J., 1983. Serengeti grassland ecology: the role of composite environmental factors and contingency in community organization. Ecol. Monogr., 53: 291–320. McNaughton, S.J. and Wolf, L.L., 1984. Ecolog´ıa General. Omega, Barcelona, 713 pp. Montaldo, N.H., 1993. Dispersi´on por aves y e´ xito reproductivo de dos especies de Ligustrum (Oleaceae) en un relicto de selva subtropical en la Argentina. Rev. Chil. Hist. Nat., 66: 75–85. Odum, E.P., 1969. The strategy of ecosystem development. Science, 164: 262–270. Parodi, L.R., 1926. Las malezas de los cultivos en el partido de Pergamino. Rev. Fac. Agron. Vet. Univ. Buenos Aires, 5: 75–188. Parodi, L.R., 1930. Ensayo fitogeogr´afico sobre el partido de Pergamino. Estudio de las praderas pampeanas en el norte de la Provincia de Buenos Aires. Rev. Fac. Agron. Vet. Univ. Buenos Aires, 271. Parodi, L.R., 1947. La estepa pampeana. La vegetaci´on de la Rep´ublica Argentina. Geograf´ıa de la Rep´ublica Argentina. An. Soc. Argent. Est. Geograf., 8: 143–207. Peet, R.K., 1992. Community structure and ecosystem function. In: D.C. Glenn-Lewin, R.K. Peet and T.T. Veblen (Editors), Plant Succession, Theory and Prediction. Chapman and Hall, London, pp. 103–151. Pickett, S.T.A. and White, P.S. (Editors), 1985. The Ecology of Natural Disturbance and Patch Dynamics. Academic Press, Orlando, 472 pp. Pilatti, M.A., de Orellana, J.A., Priano, L.J., Felli, O.M. and Grenon, D.A., 1988. Incidencia de manejos tradicionales y conservacionistas sobre propiedades f´ısicas, qu´ımicas y biol´ogicas de un argiudol en el sur de Santa Fe. Cienc. Suelo, 6: 19–30. Pimentel, D.A., Stachow, U., Takacs, D.A., Burbaker, H.W., Dumas, A.R., Meaney, J.J., O’Neil, J.A.S., Onsi, D.E. and Corzilius, D.B., 1992. Conserving biological diversity in agricultural and forestry systems. Bioscience, 42: 354–364. Prentice, I.C., 1992. Climate change and long-term vegetation dynamics. In: D.C. Glenn-Lewin, R.K. Peet and T.T. Veblen (Editors), Plant Succession. Theory and Prediction. Chapman and Hall, London, pp. 293–339. Radosevich, S.R., Holt, J. and Ghersa, C.M., 1997. Weed Ecology. Implications for Management. Wiley, New York, 589 pp. Sala, O.E., Oesterheld, M., Le´on, R.J.C. and Soriano, A., 1986. Grazing effects upon plant community structure in subhumid grasslands of Argentina. Vegetatio, 67: 27–32. Schulze, E.-D. and Mooney, H.A., 1993. Ecosystem function and biodiversity. In: E.-D. Schulze and H.A. Mooney (Editors), Biodiversity and Ecosystem Function. Springer Verlag, Berlin, pp. 497–510. Smith, J.L., Papendick, R.I., Bezdicek, D.F. and Lynch, J.M., 1992. Soil organic matter dynamics and crop residue management. In: F. Blaine Metting Jr (Editor), Soil Microbial Ecology. Marcel Dekker, pp. 65–94. Soriano, A., 1971. Aspectos r´ıtmicos o c´ıclicos del dinamismo de la comunidad vegetal. In: R.H. Mej´ıa and J.A. Moguilevski (Editors), Recientes Adelantos en Biolog´ıa. pp. 441–445.

´ Claudio M. GHERSA and Rolando J.C. LEON Soriano, A., 1991. R´ıo de la Plata grasslands. In: R.T. Coupland (Editor), Natural Grasslands. Ecosystems of the World 8. Elsevier, Amsterdam, pp. 367–407. Soule, J.A., 1993. Tagetes minuta: a potential new herb from South America. In: J. Janick and J.E. Simon (Editors), New Crops. Wiley, New York, pp. 649–653. Stowe, L.G. and Teeri, J.A., 1978. The geographic distribution of C4 species of the dicotyledoneae in relation to climate. Am. Nat., 112: 609–623. Su´arez, S.A., 1997. Comunidad de Malezas en la Pampa Ondulada como ´ındice de Biodiversidad y de Rendimiento de los Cultivos. MSc Thesis, University of Buenos Aires, 80 pp. Su´arez, S.A., Le´on, R.J.C., Ghersa, C.M. and Burkart, S., 1995. Cambios flor´ısticos en las comunidades de maleza del maiz relacionados con el deterioro del ambiente. In: Actas Primeras Jornadas Cient´ıficas del Medio Ambiente. Asociaci´on de Universidades Grupo Montevideo, Montevideo (Uruguay). Swift, M.J. and Anderson, J.M., 1992. Biodiversity and ecosystem function in agricultural systems. In: E.-D. Schulze and H. Mooney (Editors), Biodiversity and Ecosystem Function. Springer, Berlin, pp. 15–42. Swift, M.J., Vandermeer, J.H., Ramakrishnan, P.S., Anderson, J.M., Ong, C.K. and Hawkins, B.A., 1996. Biodiversity and agroecosystem function. In: H.A. Mooney, J. Hall Cushman, E.A. Medina, O.E. Sala and E.-D. Schulze (Editors), Functional Roles of Biodiversity. A Global Perspective. Wiley, New York, pp. 261–298. Tecchi, R.A., 1983. Contenido de silicofitolitos en suelos del sector sudoriental de la pampa ondulada. Cienc. Suelo, 1: 75–82. Tilman, D., 1982. Resource Competition and Community Structure. Princeton University Press, Princeton, New Jersey, 296 pp. Tivy, J., 1990. Agricultural Ecology. Longman Scientific and Technical, New York, 288 pp. van Andel, J., Bakker, J.P. and Grootjans, A.P., 1993. Mechanisms of vegetation succession: a review of concepts and perspectives. Acta Bot. Neerl., 42: 413–433. Vitousek, P.M. and Walker, L.R., 1987. Colonization, succession and resource availability: ecosystem level interactions, In: A.J. Gray, M.J. Crawley and P.J. Edwards (Editors), Colonization, Succession and Stability. Blackwell Scientific Publications, London, pp. 207– 224. White, P.S., 1979. Pattern, process, and natural disturbance in vegetation. Bot. Rev., 45: 229–299. Whittaker, R.H., 1975. Communities and Ecosystems, 2nd Edition. MacMillian, New York, 203 pp. Whittaker, R.H., 1977. Evolution of species diversity in land communities. Evol. Biol., 10: 1–67. Williamson, M., 1996. Biological Invasions. Chapman and Hall, London, 256 pp. Williamson, M.H., 1987. Are communities ever stable? In: A.J. Gray, M.J. Crawley and P.J. Edwards (Editors), Colonization, Succession and Stability. Blackwell Scientific Publications, London, pp. 353– 372. Wink, M., 1993. Production and application of phytochemicals from an agricultural perspective. In: T.A. van Beek and H. Breteler (Editors), Phytochemistry and Agriculture. Clarendon Press, London, pp. 171–213.

Chapter 21

PHYSICAL ASPECTS OF SOILS OF DISTURBED GROUND R.E. SOJKA

INTRODUCTION

Humanity’s presence on earth has forced the selective adoption of both anthropocentric and naturalistic perspectives of soil as an ecosystem component. From the anthropocentric perspective, soil is an ecosystem component used by humans for specific purposes (e.g., to grow forests and crops; support structures or roadways; and as a filtration medium). The naturalistic perspective sees soil primarily as the natural foundation or backdrop for other ecological systems and processes, and philosophically excludes many soil-management technologies and scenarios, favoring only soil uses and management practices that derive from natural ecosystem processes. The naturalistic perspective is more willing to concede that soil, like other ecosystem elements, may at times respond to perturbations counter to human needs and aesthetics. The role of environmental managers and scientists is to know when and how firmly to embrace the validity of either or both outlooks. That requires an appreciation of the properties of ecosystem components, and how those properties affect a given management objective. Familiarity with fundamental soil properties is essential to understanding the physical aspects of soils of disturbed ground, regardless of the interpreter’s perspective. This chapter presents a summary of essential soilscience concepts necessary to begin understanding the interactive role of soil in a disturbed ecosystem. The emphasis is on soil physical properties and processes. However, soil is a biologically and chemically dynamic system with strong interactions, interdependencies, and feedback among all its compartments, phases, and functions. Thus, some fundamental chemical and biological concepts relevant to soil physical status are also briefly outlined. The framework of fundamental

concepts is used to explain the role of soil physical status in several important kinds of land disturbance. A detailed analysis of all aspects of soil physical perturbation from all conceivable kinds of physical land disturbance is beyond the scope of the chapter and the expertise of the author. But application of principles to several key types of ecological disruption in which soil physical disturbance is important provide a conceptual framework that can be extended to other scenarios.

THREE-PHASE SOIL MODEL

The essential physical aspects of soil are often represented by a simple three-phase model (Fig. 21.1). The three phases are solid, liquid, and gas. The proportion, arrangement, and constitution of each phase dictates soil properties and functionality within a given ecosystem or for a given use. Typically, and perhaps surprisingly, half the volume of soil beneath one’s feet is composed of air and water.

Fig. 21.1. The three-phase conceptual model of soil, showing typical liquid, gas, and solid composition, including the distribution of mineral vs. organic solids in a productive soil from the United States “corn belt”.

503

504

R.E. SOJKA

Table 21.1 Limits of soil size separates for various classification schemes System 1

USDA-NRCS

Particle size range (mm) Very coarse sand

Coarse sand

Medium sand

Fine sand

Very fine sand

Silt

Clay

2.0–1.0

1.0–0.5

0.5–0.25

0.25–0.1

0.1–0.05

ISSS

2.0–0.2

DIN, BSI, MIT

2.0–0.6

ASTM

2.0–0.42

Corp, Bureau

4.76–2.0

Highway

2.0–0.42

0.6–0.2 2.0–0.42

0.05–0.002