Ocean Ecology: Marine Life in the Age of Humans 2020037524, 2020037525, 9780691161556, 0691161550

Citation preview

Ocean Ecol­ogy

Ocean Ecol­ogy Marine Life in the Age of H ­ umans

J. Emmett Duffy Smithsonian Institution

Prince­ton University Press Prince­ton and Oxford

Copyright © 2021 by Prince­ton University Press Prince­ton University Press is committed to the protection of copyright and the intellectual property our authors entrust to us. Copyright promotes the pro­gress and integrity of knowledge. Thank you for supporting ­free speech and the global exchange of ideas by purchasing an authorized edition of this book. If you wish to reproduce or distribute any part of it in any form, please obtain permission. Requests for permission to reproduce material from this work should be sent to permissions@press​.­princeton​.­edu Published by Prince­ton University Press 41 William Street, Prince­ton, New Jersey 08540 6 Oxford Street, Woodstock, Oxfordshire OX20 1TR press​.­princeton​.­edu All Rights Reserved Library of Congress Cataloging-­in-­Publication Data Names: Duffy, J. Emmett, 1960– author. Title: Ocean ecol­ogy : marine life in the age of ­humans / J. Emmett Duffy. Description: Prince­ton : Prince­ton University Press, 2021. | Includes bibliographical references and index. Identifiers: LCCN 2020037524 (print) | LCCN 2020037525 (ebook) | ISBN 9780691161556 (hardcover) | ISBN 9780691190532 (ebook) Subjects: LCSH: Marine ecol­ogy. | Marine ecology—­History. | Marine biodiversity conservation. Classification: LCC QH541.5.S3 D84 2021 (print) | LCC QH541.5.S3 (ebook) | DDC 577.7—­dc23 LC rec­ord available at https://­lccn​.­loc​.­gov​/­2020037524 LC ebook rec­ord available at https://­lccn​.­loc​.­gov​/­2020037525 British Library Cataloging-­in-­Publication Data is available Editorial: Alison Kalett, Abigail Johnson, Whitney Rauenhorst Production Editorial: Terri O’Prey Text Design: Wanda España Cover Design: Wanda España Production: Jacqueline Poirier Publicity: Matthew Taylor Copyeditor: Jennifer McClain Cover image by Florent Rols Photography / Alamy Stock Photo This book has been composed in 11/14 Arno Pro Printed on acid-­free paper. ∞ Printed in China 10 ​9 ​8 ​7 ​6 ​5 ​4 ​3 ​2 ​1

Brief Contents

Chapter 1: Introduction Chapter 2: Life in the Ocean

1 11

Chapter 3: Geography of Marine Life

38

Chapter 4: Introduction to the Anthropocene Ocean

74

Chapter 5: Organisms

112

Chapter 6: Populations

138

Chapter 7: Species Interactions

171

Chapter 8: Ecological Communities

195

Chapter 9: Ecosystems

218

Chapter 10: The Open Ocean

260

Chapter 11: Estuaries and Coastal Seas

296

Chapter 12: Coral Reefs

347

Chapter 13: Ocean 2.0

377

v

Contents

Preface

xv

Chapter 1: Introduction

1

A Framework for Functional Marine Ecol­ogy 3 Plan of the book 4 Some recurrent themes 5 ­Humans in marine ecosystems 6 Ecol­ogy in Practice  6 The central challenges of marine ecol­ogy 6 The ocean and the shore 8 The major patterns 9

Chapter 2: Life in the Ocean

11

The Magnitude of Biodiversity 11 Diversity on Land and Sea 12 Phyloge­ne­tic Classification of Marine Biodiversity 14 The tree of life 18 Phyloge­ne­tic relationships and tree thinking 20 The web of life 21 Functional Organ­ization of Pelagic and Benthic Life 21 Functional groups of pelagic life 22 Functional groups of benthic life 30 Marine Life in the Anthropocene 36 ­Future Directions 37 Summary 37

Chapter 3: Geography of Marine Life

38

A Short History of the Oceans and Continents 38 Climate and Circulation of the World Ocean 40 Geostrophic flow and the central ocean gyres 41 Convergence zones and fronts 42 Thermohaline circulation and the origins of deep ­water 43 Coasts, shallows, and their consequences 44 Major Patterns in the Distribution of Marine Life 44 A conceptual framework for understanding biodiversity 44 The spatial organ­ization of diversity 45 The latitudinal diversity gradient 47 vii

viii

Ocean Ecology

The longitudinal diversity gradient 48 The depth diversity gradient 48 The role of bottom type 48 The Origin of Species 49 The ecol­ogy of speciation 49 Habitat area and geographic range 50 Habitat age 52 Temperature, energy, and metabolic rate 53 Body size 54 Life history and dispersal ability 54 Ecological specialization 55 Ecological opportunity and speciation 55 The Dispersal of Species 56 The Theory and Evidence for Island Biogeography 57 The End of Species: Extinction 58 Integrative Models of Marine Diversification 60 Biogeographic Classifications of the Ocean 61 The ecological geography of the sea 61 Marine ecoregions of the world 62 Large marine ecosystems 63 The Biogeography of Functional Traits 63 The Biogeography of Species Interactions 64 Biogeography of the Anthropocene Ocean 68 Climate warming and re­distribution of global marine fauna 68 Tropicalization 70 The Arctic opening 70 The sixth mass extinction? 72 ­Future Directions 72 Summary 73

Chapter 4: Introduction to the Anthropocene Ocean First, the Good News 74 The ­Great Acceleration 75 Coal and climate change 78 Nitrogen: Detonator of the population explosion 79 The limits to growth 80 The Natu­ral and Cultural History of Homo Sapiens 83 Ecol­ogy for the Anthropocene 84 Culture and the evolution of ­human society 84 Energetics and economics of Homo sapiens 86 The tragedy and triumph of the commons 86 The Anthropocene ocean 87 Ocean Warming 89 Warming effects on communities 90 Sea level rise 91 Ocean Acidification 92 Effects of acidification on organisms 92 Effects of acidification on communities 93

74

Contents

Homo Sapiens: Top Predator of the Ocean 95 The history and extent of fishing 95 The current state of marine fisheries 97 Ecosystem impacts of fishing 98 The ­future of fisheries 99 Marine Biodiversity in the Anthropocene 101 Species decline and extinction 101 Functional consequences of declining biodiversity 102 Marine globalization 103 Evolution in the domesticated ocean 104 Novel ecosystems 105 ­Future Directions 105 Science for solutions 105 Policy for solutions 106 Reasons for cautious optimism 108 Summary 110

Chapter 5: Organisms

112

Building Blocks of Life 113 Ecological stoichiometry 113 Nutrient uptake and use 115 Iron 117 Powering Life 117 Autotrophy 118 Heterotrophy 118 Kinetics of Life 120 Dimensions of Life 121 Mechanics of Life 124 Coding Life 126 Natu­ral se­lection and adaptation 126 Genotype and phenotype 127 Functional Ecol­ogy and the Niche 128 Historical and modern concepts of the niche 128 ­Toward a trait-­based ecol­ogy of marine organisms 129 Functional Ecol­ogy of Marine Primary Producers 131 Functional groups of phytoplankton 131 Functional groups of benthic macrophytes 133 Macroecol­ogy 133 Organisms in the Anthropocene 135 ­Future Directions 136 Summary 137

Chapter 6: Populations

138

Development and Life History 139 The Prob­lem of Larval Dispersal 142 Population Growth: A Brief Review 144 Growth of Age-­and Stage-­Structured Populations: Matrix Approaches 147

ix

x

Ocean Ecology

Demographic Models in Conservation and Management 149 Maximum sustainable yield in fisheries 149 Strategic conservation of vulnerable life stages 151 Life history and the effectiveness of marine reserves 154 Organismal Fitness and Adaptation to the Environment 155 Dispersal, Recruitment, and Metapopulations 158 Tagging and tracking 159 Hydrodynamic simulation of larval movement 161 Larval be­hav­ior 161 Population ge­ne­tic markers of dispersal and connectivity 162 Geochemical tags 164 Macroecol­ogy of Populations 165 Metabolic scaling and life history 165 Abundance and the energetic equivalence rule 166 The macroecol­ogy of range size 167 Marine Populations in the Anthropocene 168 ­Future Directions 169 Summary 170

Chapter 7: Species Interactions

171

Interactions among Species: General Considerations 171 Interactions between Competitors 175 Interactions between Plants and Herbivores 176 Controls on herbivory: Plant traits 178 Controls on herbivory: Herbivore traits 179 Interactions between Prey and Predators 182 Controls on predation: Prey traits 182 Controls on predation: Predator traits 184 Parasitism and disease 184 Predation and community diversity 185 Facilitation and Mutualism 185 Ecological Networks 186 Functional traits as a lens into community organ­ization 187 Traits in interaction networks 189 Emergent properties of ecological networks 189 Ecological Interactions in the Anthropocene 191 Changing species interactions in a changing climate 191 Food web decapitation and trophic skew 193 ­Future Directions 193 Summary 194

Chapter 8: Ecological Communities What Is a Community? 196 Community Dynamics: A Conceptual Framework 198 Ecological se­lection 199 Dispersal and metacommunities 200 Ecological drift 201

195

Contents

Synthesis: Diversity in Ecological Communities 203 Neutral models and their assumptions 203 The unified neutral theory of biodiversity 204 Testing the neutral theory in nature 204 Disturbance and diversity in communities 207 The role of history 207 Dispersal and species richness 209 Metabolic theory and species diversity 210 Space and species diversity 211 Linking Communities to Ecosystems 211 Functional structure of communities 211 Phyloge­ne­tic structure of communities 214 Communities in the Anthropocene 214 Climate change and communities 214 Marine defaunation and trophic skew 215 ­Future Directions 216 Summary 216

Chapter 9: Ecosystems

218

History of the Ecosystem Concept 219 Evolution of the ecosystem concept 223 Ecosystems as complex adaptive systems 224 Primary Production 225 Light and photosynthesis 225 Nutrient uptake and use 226 Herbivory 226 Control of Biomass Distribution and Productivity in Marine Ecosystems 228 The green world hypothesis 228 Bottom-up control of biomass and productivity by resources 229 Top-­down control of biomass and productivity by consumers 231 Trophic cascades in the ocean 232 Detritus-­consumer interactions 233 Functional Structure of Marine Ecosystems 236 Organismal traits in ecosystems 236 The size spectrum 237 The macroecol­ogy of trophic interactions 240 Biodiversity and the Functioning of Ecosystems 241 Biodiversity and ecosystem functioning: Theory 241 Biodiversity and ecosystem production: Empirical evidence 242 Biodiversity and ecosystem stability 245 Alternative Stable States and Regime Shifts in Complex Adaptive Ecosystems 249 Empirical evidence for regime shifts in marine ecosystems 251 Mechanisms of marine regime shifts 251 Applications of Marine Ecosystem Modeling in Fisheries and Management 252 Models of the Global Ecosystem 255

xi

xii

Ocean Ecology

Marine Ecosystems in the Anthropocene 257 Eutrophication 257 Defaunation and trophic skew 258 ­Future Directions 258 Summary 259

Chapter 10: The Open Ocean

260

Physical Forcing of Pelagic Ecosystems 261 The global distribution of ocean productivity 261 Vertical structure of the pelagic ­water column 262 The spring bloom 265 High-­nitrogen low-­chlorophyll (HNLC) regions 266 Organisms and Traits 266 The phytoplankton: Major functional types 266 Grazers: Major functional types 268 Grazing 269 Structure and Organ­ization of Pelagic Communities 269 Specialization and resource partitioning 270 Nonequilibrium dynamics 270 Chaos 271 Functioning of Pelagic Ecosystems 272 Pelagic food webs: The microbial loop 272 The biological pump and the global carbon cycle 273 Trophic control in pelagic ecosystems 274 The Deep Sea 276 Adaptations to life in the deep sea 276 Pelagic-­benthic coupling 278 Deep-­sea biodiversity 279 Chemosynthetic Ecosystems: Vents and Seeps 281 Hydrothermal vents 282 Cold seeps 283 Macroecol­ogy of the Open Ocean 284 Controls on biodiversity in the open ocean 284 Global controls on microbial diversity 286 Macroecol­ogy of open-­ocean ecosystem pro­cesses 287 Deep-­Sea Fisheries 288 The Open Ocean in the Anthropocene 289 Climate and the Anthropocene ocean 289 Ocean acidification 292 High-­seas fisheries 292 ­Future Directions 293 Summary 294

Chapter 11: Estuaries and Coastal Seas The Edge of the Sea 296 Interacting ocean and continents 296 Estuaries 298

296

Contents

Coastal life and communities 300 Coastal ecosystem pro­cesses 302 Rocky Shores 304 Geomorphology and environment 305 Organisms and traits 305 Community organ­ization and key interactions 306 Ecosystem pro­cesses and ser­vices 309 Rocky shores in the Anthropocene 309 Sediment Bottoms 310 Geomorphology and environment 310 Organisms and traits 311 Community organ­ization and key interactions 313 Ecosystem pro­cesses and ser­vices 314 Sediment bottoms in the anthropocene 315 Seagrass Meadows 317 Geomorphology and environment 317 Organisms and traits 318 Community organ­ization and key interactions 319 Ecosystem pro­cesses and ser­vices 321 Seagrass meadows in the Anthropocene 322 Salt Marshes 325 Geomorphology and environment 326 Organisms and traits 327 Community organ­ization and key interactions 328 Ecosystem pro­cesses and ser­vices 330 Salt marshes in the Anthropocene 330 Mangrove Forests 331 Geomorphology and environment 332 Organisms and traits 332 Community organ­ization and key interactions 333 Ecosystem pro­cesses and ser­vices 334 Mangrove forests in the Anthropocene 334 The Seascape: Interactions among Habitats 335 Coastal Ecosystems in the Anthropocene 337 Climate change and the coast 337 Decline of foundation species 338 Trophic skew 338 Nonnative and invasive species in coastal ecosystems 338 Coastal fisheries 340 Eutrophication and hypoxia 340 Multiple stressors in coastal ecosystems 341 Coastal anthropogenic biomes: Urbanized estuaries 343 ­Future Directions 344 Summary 346

xiii

xiv

Ocean Ecology

Chapter 12: Coral Reefs

347

Geomorphology and Environment of Coral Reefs 348 Geomorphology 348 Abiotic environment 351 Organisms and Traits: Functional Diversity in Reef Ecosystems 352 Biodiversity: Foundation species 352 Diversity and functional ecol­ogy of primary producers 354 Diversity and functional ecol­ogy of consumers 355 Community Organ­ization and Key Interactions 357 Origin and maintenance of diversity in reef communities 357 Herbivory in reef ecosystems 360 Trophic cascades in reef ecosystems 361 Disease in reef ecosystems 362 Phase shifts and alternative stable states on coral reefs 363 Regional variation in coral reef dynamics 366 Ecosystem and Biogeochemical Pro­cesses 367 Coral reef production and nutrient cycling 367 Coral reef fisheries 368 Coral Reefs in the Anthropocene 369 Coral reefs in a warming ocean 371 Ocean acidification 372 Coral reef fisheries 373 The ­future of coral reefs 373 Strategic coral reef conservation and management 374 ­Future Directions 374 Summary 375

Chapter 13: Ocean 2.0 The Earth Is a Complex Adaptive System 377 Biodiversity Is as Impor­tant as Climate 378 ­Humans Are Now the Force of Nature 378 Apocalypse Not 379 But What about Nature? 379 Rays of Hope 381

Glossary

383

Lit­er­a­ture Cited Photo Credits Index

431

393

429

377

Preface

When I entered grad school, I was required to enroll in physical oceanography. Completing the course meant resurrecting the calculus I’d learned in college and promptly forgotten. One day I confessed to my professor that I was struggling with the complex math. He seemed surprised, and said “But physics is s­imple—­there are clear and consistent laws that can be described by equations. What’s ­really complex is your field, ecol­ogy.” He was right. The difference between physics and biology stems from the extravagant variety and idiosyncrasy of life, Earth’s most striking feature as well as the most formidable challenge to understanding how it works. The rules of life seem far fewer, and the exceptions more numerous, than ­those of physics. This book aims to put that diversity at the center of our understanding of the world ocean. It is or­ga­nized around the conviction that biodiversity—­the variety of life forms—is the heart of functioning ecosystems, and aspires to outline a coherent framework for understanding its d­ rivers, dimensions, and functional roles in ecosystems. This way of thinking is relatively new to ecol­ogy. Research on ecological structure—­the distribution and abundance of species—­has historically taken a separate path than that of systems ecol­ogy, meaning the emergent biogeochemical fluxes of energy and materials through a land-­or seascape. Th ­ ese paths are the provinces of community ecol­ogy and ecosystem ecol­ogy, respectively—­two separate nations, with distinct cultures, languages, and lit­er­a­tures, and l­ ittle traffic across the gulf between. But that gulf may be yielding—­a sort of revolution over the last two de­cades is building a bridge. As a postdoc in the early 1990s, I was captivated by the first experiments exploring how species diversity influences ecosystem productivity, a topic that seemed a perfect marriage between the fascination of natu­ral history and a profound scientific and societal challenge. I spent much of the next de­cade thinking about this prob­lem and exploring it with experiments. Over that time I was struck by repeatedly hearing the same criticisms: that the experiments proliferating around this theme ­were simplistic (which largely was true, at least in the beginning) and w ­ ere a distraction from more impor­ tant pro­cesses driving ecosystems in the real world (which arguably was not). In fact, t­ here are several reasons to suggest that biodiversity may in fact be more, not less, impor­tant in wild nature than in experiments. A de­cade l­ater I believe that suggestion has been vindicated. Growing evidence shows that the kinds and numbers of species are just as impor­tant as major environmental gradients in mediating the way ecosystems work. Of course, the environment ultimately controls what kinds of organisms live in a system, so this dichotomy is somewhat artificial. But the biological composition of ecosystems also owes a lot to idiosyncrasies of evolutionary history and dispersal (or lack thereof), so similar environments in dif­fer­ent regions can differ substantially in ecosystem pro­cesses. For example, prior to Eu­ro­pean colonization, northern North Amer­i­ca lacked earthworms, which allowed accumulation of leaf litter and greatly altered soil properties. The functional characteristics, or traits, of organisms determine how they affect ecosystem pro­ cesses. Focusing on the roles of traits in ecological pro­cesses opens the door (with a very long hallway still ahead) to a unified functional ecology—­from the biophysical pro­cesses that drive cellular metabolism, through interactions among species, to flows of materials through ecosystems. The central link and the central challenge to such unification is the diversity of life’s functional capacity. Although clearly central to ecol­ogy, the challenge in integrating biological diversity into a unifying

xv

xvi

Ocean Ecology

framework is idiosyncrasy, the seemingly infinite and unpredictable ways that organisms make a living. So addressing the specifics of major marine ecosystem types is unavoidable in understanding them. Among the most comprehensive treatments of marine ecological pro­cesses, from individual to ecosystem level, at the gradu­ate level was Ivan Valiela’s book of that name, first published in 1984. In the preface to the second edition (1995), Valiela highlighted the impossibility of keeping track of the vigorous flood tide of scientific lit­er­a­ture. That was over two de­cades ago (and he was hardly the first—­even Charles Elton bemoaned the difficulty of keeping up in 1927). The situation is far more daunting t­ oday. One book and one author ­can’t pretend to cover the ­whole of marine ecol­ogy more than superficially. Instead I propose a way of thinking about ocean life and a foundation of basic knowledge on which the rest can be built. I hope this w ­ ill give students a framework for connecting the parts. While working on this book, I’ve had to contemplate the possibility that scientific books may be fading into obsolescence in the internet age. I hope not. If anything, books—­the right kinds of books—­are more valuable than ever in the torrent of overwhelming information (not to mention disinformation), fragmentation, and eroding attention span endemic to the online media ecosystem, what­ever its other benefits. I grew up in an age when books ­were central, even revered. Many of them changed my life. I can only hope that this one proves to be useful for someone e­ lse.

Acknowl­edgments I owe a special debt to my PhD adviser, Mark Hay, who influenced me more than anyone outside my nuclear ­family. I arrived at the University of North Carolina’s Institute of Marine Science as a PhD student only three years a­ fter Mark started as an assistant professor. The timing was fortuitous—­his initial flurry of grant-­writing was bearing fruit and t­ here was suddenly a lot to do and few p­ eople to do it. Mark was a locomotive of energy and intellect, and it was exhilarating trying to keep up. We traveled all over the world together. ­Those ­were g­ reat years. I ­didn’t fully appreciate ­until much ­later how fundamentally my worldview was ­shaped by the immersion, literally and figuratively, in so many regions of the ocean. One of Mark’s many gifts was pushing me through long practice to unlearn the numbing, jargon-­laden style so common in scientific writing. I have tried to employ that lesson ­here. Dennis Taylor took a chance and hired me in my first permanent job during his brief but transformational tenure as director of the V ­ irginia Institute of Marine Science, as he set it on the course ­toward a world-­class institution. When I arrived at VIMS as a green assistant professor in 1994, my first assignment was a new gradu­ate course in marine ecol­ogy cotaught with another new but more se­nior arrival, Hugh Ducklow. This was a stroke of luck. I’m a marine benthic community ecologist with a natu­ral history bent, and Duck, as most call him, is a biological oceanographer with expertise in microbial ecol­ogy, ecosystem modeling, and a quantitative, global outlook. The two of us somehow welded together the poles of biological oceanography and benthic community ecol­ogy into a unique course, and we taught it together for the next 13 years. I not only gained a deep knowledge of the ocean as a global ecosystem but also learned from Hugh how to teach. To this day the way I approach the subject, including a strong focus on primary lit­er­a­ture, owes much to his example during ­those early days as a deer in the headlights. Science is a community. It is rooted in personal interactions, diverse perspectives, debate, competition, collaboration, rivalries, and friendships. The knowledge and ideas that arise from that ecosystem are themselves like living entities that grow and evolve such that it is often difficult to trace where they came from. My thinking has been influenced, no doubt in ways I d­ on’t fully recognize, by the many colleagues, mentors, and students that have been my community over the years. Th ­ ere are too many to name individually, since I ­will surely overlook someone impor­tant, but in addition to ­those mentioned above, I must thank several formal and informal mentors, including Nancy Knowlton, Mark Bertness, Rick Grosberg, John Graves, and Seth Tyler. I also gratefully acknowledge several

Preface

reviewers who provided thoughtful feedback on early chapters of this book and helped steer it clear of a treacherous fate. ­These include Mark Hay, Jennifer Ruesink, and especially Mark Carr and his students, who gave me a timely and invaluable real­ity check on how to serve the needs of young ­people starting out in the field. Michael Tennenbaum gave me the opportunity of a lifetime and challenged me to raise the bar as high as I can. And I am grateful to three institutions that sponsored the deep experiences of ocean life that I hope are reflected in the following pages: the V ­ irginia Institute of Marine Science, the Smithsonian Institution, and the National Science Foundation. This is contribution 77 from the Smithsonian’s MarineGEO and Tennenbaum Marine Observatories Network. Alison Kalett, my editor at Princeton University Press, first approached me with the idea for this book, expertly parried my initial skepticism, and kept me on course with sage advice at several key junctures in its (long) evolution—I am so happy we both stuck with it, and it has been a pleasure working with the staff of Princeton University Press. Thank you to Carmen Ritter, who drafted the glossary, and to Eleanor Cole, Leah Harper, and Michelle Rossman who shouldered much of the burden of organ­ izing figures and permissions. Fi­nally, and in many ways most importantly, it’s hard to imagine where I would be now or what I would be d­ oing without my ­family. I had the good fortune that my childlike curiosity survived into adulthood and found a place to take root, thanks largely to my supportive parents. My spouse and son, Liz and Conor, have been the joy of my life and contributed to this book more than any of us know through the seemingly endless years of its gestation. “I could drink a case of you, darlin’.”

xvii

Ocean Ecol­ogy

1

Introduction

T

wo and a half billion years ago life on Earth entered a new age. At that time the early biosphere consisted entirely of bacteria-­like cells, churning through an anoxic, sulfurous ocean. Somewhere in this brew, a group of cells evolved a new form of energy industry, powered by the abundant sunlight and producing molecular oxygen as a by-­product. The mutants began to spread, filling the ocean and atmosphere with oxidizing f­ree radicals and gradually poisoning the planet for the anaerobic cells that had reigned since the beginning of life. By mastering this feat of oxygenic photosynthesis, the cyanobacteria came to rule the world, relegating the previously dominant anaerobes to fringe habitats and fundamentally transforming the geochemistry of planet Earth. The G ­ reat Oxygenation Event, as it is known, was prob­ably the most far-­reaching global change in Earth’s history. And it had another momentous consequence: one cell’s poison became another’s food. A second group of opportunistic microbes evolved to exploit the growing supply of molecular oxygen, in the pro­cess developing a far more power­ful molecular engine. This innovation, aerobic metabolism, drove the evolution of complex, multicellular life, which proliferated vigorously. Some 300,000 years ago that lineage spun off Homo sapiens. It’s been a long road from the G ­ reat Oxygenation Event to the current Anthropocene epoch. Perhaps 100 million years into the oxygen era, another evolutionary upheaval roiled the ­waters. In the blink of an eye, geologically speaking, global temperatures plunged and glaciers spread from the poles nearly to the equator, creating a “snowball earth” not seen before or since. The trigger, it now appears, was the proliferation of advanced eukaryotic algae, which produced dimethyl sulfide in such quantities that it nucleated a global cloud bank and tipped the earth system into a winter that lasted millions of years (Feulner et al. 2015). Nor ­were such upheavals confined to Earth’s microbial youth. Around 55,000 years ago—­during a previous episode of rapid climate change—­a band of early h­ umans ventured out from their East African homeland and spread rapidly into Eu­rope and across Asia. With l­ittle more than fire and crude stone tools, our ancestors overhunted, outcompeted, or displaced most other large land animals within a few centuries of arrival on ­every continent and island they encountered (Sandom et al. 2014). And our impacts have accelerated ever since. Within the lifetime of el­derly ­people living t­ oday, ­humans have transformed the earth almost as profoundly as during the entire span since our departure from Africa more than 50,000 years ago. Like ­those first photosynthetic bacteria long before, Homo sapiens is now altering the composition of the atmosphere and climate in profound ways whose outcome we can only dimly foresee. Why start a book about marine ecol­ogy with such ancient history? ­There are two reasons. First, ­these examples offer a vivid lesson about the central role of living organisms in the working of the earth system as a ­whole. The common thread ­running through them is that life has repeatedly and fundamentally transformed the atmosphere, the ocean, and the climate from the beginning of time. Consider what our planet’s atmosphere and climate would be like in the absence of life (­table 1.1).

1

2

Ocean Ecology

TABLE 1.1 ​Influence of life on the geochemical composition of Earth’s atmosphere Planetary atmospheres: Their composition Venus Carbon dioxide Nitrogen Oxygen Argon Methane Surface temperatures (°C) Total pressure (bars)

96.5% 3.5%

Mars 95%

Earth without life 98%

Earth as it is 0.03%

2.7%

1.9%

79%

trace

0.13%

0.0

21%

70 ppm

1.6%

0.1%

1%

0.0

0.0

1.7 ppm

0.0 459 90

−53 0.0064

240 to 340 60

13 1.0

Source: Lovelock (1995).

As late as the 1960s the scientific consensus was that Earth’s atmosphere was produced passively by gases diffusing from the planet’s interior, and that living organisms w ­ ere passengers rather than ­drivers. This view was turned on its head by the atmospheric chemist James Lovelock and the evolutionary biologist Lynn Margulis, first in a series of journal articles (Lovelock and Margulis 1974) and culminating in Lovelock’s famous (or infamous) book Gaia: A New Look at Life on Earth (Lovelock 1979), which argued that the atmosphere is “a dynamic extension of the biosphere itself.” The Gaia hypothesis had its flaws. But it got right that Earth’s atmosphere, and by extension global climate, are fundamentally influenced by metabolism of the living biosphere. ­These facts are now universally accepted (Kasting and Siefert 2002, Holland 2006). Although each of t­ hese transitions in earth history could prob­ably be linked to some environmental change, their far-­reaching impacts ­were not predictable effects of environmental forcing. Instead, they involved intrinsic biological pro­cesses that spun off with their own momentum as a result of evolution and interactions among species (Falkowski et al. 2008). Similarly, at a much more local scale, organisms strongly influence one another’s distributions, and thus the organ­ization of communities, in ways only loosely related to the nonliving environment. The experiments on marine rocky shores in the 1960s that demonstrated this revolutionized ecol­ogy (Paine 1994, Bertness, Bruno, et al. 2014). We w ­ ill see throughout this book that the pervasive consequences of biological interactions play out at local scales in shaping the organ­ization of communities, just as they do at the scale of oceans and eons. Biodiversity, the interacting variety of living organisms, transformed the earth system. So the first lesson from ­these vignettes is that life is not just a passenger but also a driver of environmental processes—­biology feeds back in power­ful ways to shape the earth’s abiotic environment. The second theme illustrated by t­hese examples is that life goes on. Even changes that seem catastrophic—­that are catastrophic to t­ hose that experience them—­eventually ­settle the ecosystem into a new groove around the altered conditions. Nature abhors a vacuum, as Aristotle first put it, and we can bet on the resilience of life to adapt and produce a productive, structured system in place of what was lost. This is not much comfort to t­ hose who depended on the former system, such as the Precambrian anaerobes sidelined in the G ­ reat Oxygenation Event, or modern socie­ties that rely on predictable crop harvest and prices, or even ­those of us who simply love nature as it is. Change is hard. But it is inevitable. ­These two themes offer both a cautionary tale and a ray of hope for the sobering challenges we face in the current Anthropocene age. The caution stems from the accelerating impact of humanity

Chapter 1 Introduction

on the distribution, abundance, and even existence of species, which ­will have far-­reaching and still poorly understood consequences for the earth’s ecosystems that support us. The hopeful message is that nature is resilient, with an inherent, vigorous capacity to recover even from major disturbances. This is reassuring since disturbances are a constant feature of the modern ocean. We can help it bounce back if we understand it. To be effective stewards of the earth and ocean, we need to understand how its parts fit together to make the system as a ­whole work. Building that understanding is the mission of this book.

A Framework for Functional Marine Ecol­ogy Ecol­ogy is about understanding how living organisms interact with their environment. Its goals can be summarized in a few overarching questions: How does the environment influence the distribution and diversity of life? How do interactions among organisms modify t­ hose patterns? How do the groups of species that emerge from t­ hese interactions influence pro­cessing of energy and materials in ecosystems? And, fi­nally, how do all ­those pro­cesses feed back to change the environment? Nearly ­every topic in ecol­ogy is a more specific derivation of one of t­ hese questions. But despite this superficial simplicity, ecol­ogy is the most difficult science. This is ­because it is, in essence, the study of every­thing. Ecol­ogy begins with energy from a star 150M km away, and follows that energy as it flows through biomolecules that mediate life pro­cesses, and interactions among species, to the metabolism of megacities built by technologically advanced h­ umans. In the modern era, answering t­ hese questions requires e­ very tool we can muster, from molecular probes of genomes to satellites that quantify biomass at a planetary scale. B ­ ecause ecol­ogy is about interactions, every­thing is connected to every­ thing ­else. Organ­izing this sprawling domain poses big challenges. We need a framework to get hold of it. This book aims to advance such a framework for marine ecosystems. The basic argument is ­simple: ecosystems emerge from interactions among four major features of nature (figure 1.1), which are predictable in broad outline from biophysical princi­ples. First, ecol­ogy begins with geomorphology: the shape of the rotating earth and the arrangement and composition of land masses on its surface. ­These f­ actors constrain circulation of the ocean and atmosphere, which in turn create the climate and provide the physical template within which the other pro­cesses act. Second is the abiotic environment, largely defined by that geomorphic template and interacting with the sun’s incoming radiation. The environment encompasses the solar and other energy that drives chemical and biological pro­cesses, the distribution of temperature that sets rates of ­those pro­cesses, and the distribution of ­water and chemical ele­ments needed to create living biomass. Third is biodiversity, the constantly changing communities of living organisms that evolve and assem­ble in response to geomorphology and environmental forcing, and feed back to modify them. This network of diverse, interacting organisms is the heart of the dynamic earth system, comprising a sort of engine (Falkowski et al. 2008) that drives the fourth feature, biogeochemistry, Biogeochemistry the fluxes of energy and materials through the system. This framework is not revolutionary. Anyone can see that the environment influences the distribution and abundance of organisms. Physical forcing of communities and ecosystems was central to the deBiodiversity velopment of ecol­ogy as a science (Elton 1927, MacArthur 1972) and especially in biological oceanography, which is arguably the most integrative field of ecol­ogy (Mann and Lazier 1996, A. R. Longhurst Geomorphology Environment 2007). It’s also widely recognized that the kinds of organisms in a system influence fluxes of ­matter and energy (Chapin et al. 2000). Where this book departs from many previous treatments of ecol­ogy is its emFigure 1.1. ​This book’s framework for thinking about ecol­ogy. phasis on the links between the four components, the bidirectional

3

4

Ocean Ecology

Oceanography Environment

Biology

Ecosystem processes

Marine community ecology Environment

Biodiversity

This book Biodiversity

Environment

Ecosystem processes

Figure 1.2. ​A simplified view of relationships among the environment, biological communities, and ecosystem pro­cesses in dif­fer­ent traditions of marine science.

nature of ­these links, and the central role of biodiversity in the functioning of the ocean’s ecosystems (figure 1.2). The vignettes that opened this chapter illustrate that environment, biodiversity, and biogeochemistry feed back intimately and strongly in both directions, and biodiversity is at their center (see figure 1.1). We ­will see that in modern times, as in early earth history, the kinds of organisms pre­sent are often just as influential as climate and resource supply in shaping the biomass, productivity, and cycling of materials through ecosystems, and that organisms fundamentally change—­even create—­the physical environment. Each of the links among t­ hese components is a two-­way street. In modern parlance, an ecosystem is a complex adaptive system. This ecological web is impor­tant to all of us ­because the earth is transitioning to a new geologic epoch, the Anthropocene (chapter 4). ­Human civilization grew up during the ~13,000-­year Holocene epoch that followed the retreat of the last major ice sheets. The Holocene has been a period of climatic stability, which we take for granted. That stability is coming to an end, but it’s less clear what ­will replace it. The profound changes in Earth’s early history wrought by an evolving biosphere hold lessons for us in this new era.

Plan of the book Following the functional framework outlined above, the book begins by introducing the main players in the ocean’s ecosystems, approaches to organ­izing their diversity, and key features of their biology that drive ecosystem pro­cesses (chapter 2). We then consider geomorphology: the physical template defined by the earth’s configuration of land and ocean, and how it has influenced the evolutionary history and current distribution of marine life (chapter 3). Having thus set the stage, we turn to the current Anthropocene epoch, and how marine ecol­ogy is changing in the modern ocean (chapter 4). The next set of chapters develops the key features and pro­cesses at successive levels of organ­ ization, from individual organisms to ecosystems. Building the bridges between ­these levels—­from environment, through biodiversity, to biogeochemistry—­requires mechanistic theory. At the cellular level, the metabolic theory of ecol­ogy seeks to explain major features of organismal function based on physical princi­ples, focusing on how metabolism varies with energy input, temperature, and the fundamental organismal trait of body size (chapter 5). ­These princi­ples help link environmental ­drivers through activities of individual organisms to their abundance and distribution in communities. Organisms adapt to their environment via growth, reproduction, and ge­ne­tic change within populations, which can be described mathematically (chapter 6). The pool of species produced by this adaptive pro­cess over evolutionary time is then filtered to a set that co-­occurs within a local area—­the

Chapter 1 Introduction

community—­via the pro­cesses of dispersal and deterministic interactions (ecological se­lection), with a background of random demographic drift (chapters  7, 8). Neutral theory provides a null model against which to explore the importance of biological interactions in structuring communities. Lastly, at the broadest scale of the ecological hierarchy, interacting communities of organisms influence the fluxes of materials and energy through ecosystems (chapter 9). The final set of chapters applies the concepts developed in the e­ arlier sections to the major ecosystem types of the world ocean, including the open pelagic ocean and the deep-­sea floor (chapter 10), coastal and estuarine systems (chapter 11), and coral reefs (chapter 12). Each of t­ hese chapters is structured around the themes we began with, building on a template of geomorphology, major physical forcing, the characteristic functional types of organisms that dominate ­under ­those conditions, and the ecosystem pro­cesses that emerge from their interactions. Each chapter concludes with a discussion of how its themes are changing in the Anthropocene ocean, and a consideration of the way forward, asking: What are the major unanswered questions that need attention? What are the current challenges that we need to overcome? How can we apply what w ­ e’ve learned to practical prob­lems in marine ecol­ogy?

Some recurrent themes Several themes emerge from this conceptual framework that recur throughout the book. The first has already been mentioned: the central importance of biological diversity. The essence of life is its tendency ­toward continuous, self-­generated, adaptive change (Szathmary and Maynard Smith 1995). A consequence of life’s responses to the physical environment acting over eons of earth history has been proliferation into a spectacular range of forms, functionally differentiated and interacting—­that is, biodiversity (chapter 2). That biological diversity is often as impor­tant as under­lying abiotic pro­cesses in determining patterns of biomass, productivity, trophic structure, even the composition of the atmosphere and ocean. As one example, phytoplankton—­the tiny single-­celled algae that dominate open-­ ocean ­waters—­are more heavi­ly grazed, channel less biomass to detritus, and store less carbon in sediments than higher plants like seagrasses with their complex, largely inedible support structures (Cebrián and Lartigue 2004). This dichotomy in traits of primary producers is largely responsible for the profound differences in how pelagic versus benthic ecosystems work (Steele 1985, 1991). A second, related theme is that functional biology, the physics and chemistry of how organisms work, is the link between dif­fer­ent scales and components of ecosystems. Differences in the elemental composition and nutritional quality among types of plants strongly influence how energy and nutrients flow up the food chain, constraining every­thing ­else that happens in the ecosystem (Cebrián et al. 2009). In the ocean, the availability of one ele­ment in par­tic­ul­ ar, nitrogen, varies systematically among types of plants and strongly affects the biomass of herbivores that eat them, and hence all food web pro­cesses. Such relationships illustrate that ecosystem structure and biogeochemical fluxes are closely linked to the functional biology of species. That functional diversity is ­shaped by environmental forcing (chapter 3), but it also evolves idiosyncratically and, as the opening vignettes emphasize, can feed back to change the environment substantially. Focusing on function emphasizes the connections between environmental conditions, diverse assemblages of interacting species, and the fluxes of energy and m ­ atter through ecosystems that result from their activities. This approach contrasts with many previous approaches to ecol­ogy, which ­either intentionally ignore species (Mann and Lazier 1996) or treat explanation of species richness as an end in itself, largely divorced from their functional characteristics (Ricklefs 1987, Hubbell 2001, Vellend 2016) (see figure 1.2). I am not criticizing ­those approaches—­explaining diversity and ecosystem pro­cesses are challenging and impor­tant jobs in their own right, and I draw heavi­ly on ­those syntheses. But the frontier that we focus on ­here is the links between community composition and biogeochemical fluxes. I attempt to keep that frontier in view throughout.

5

6

Ocean Ecology

The third recurrent theme is a special case of the importance of biology in the earth system mentioned above. This is the pervasive impact of the global keystone species Homo sapiens, which has made us a force of nature with a vastly outsized influence on Earth’s ecosystems, including the world ocean (chapter 4). The ocean and especially the coastal zone is a very dif­fer­ent world than the one that my grandparents ­were born into. Some estimates suggest ­we’ve lost two-­thirds of the world’s salt marsh and seagrass cover since mea­sure­ments began (Lotze et al. 2006, Waycott et al. 2009) and a third of global mangrove area (Spalding 2010), and coral reefs face existential threats (Hughes, Barnes, et al. 2017). Like it or not, we are now the stewards of our planetary ecosystem. The f­ uture of all species, including our own, depends on our actions, and the next few de­cades ­will make or break their fate and our own. This is far more than an academic issue. We depend intimately on the ocean. Over a third of the world’s population lives in coastal regions and islands that make up less than 5% of Earth’s land area. The global ocean economy was valued by the OECD (2017) “very conservatively” at $1.5 trillion, and provided 31 million full-­time jobs in 2010, mainly in capture fisheries but with a substantial component in coastal tourism. Marine fisheries in 2013 provided 17% of the global population’s supply of animal protein, and are especially impor­tant in coastal and small-­island states of the developing world (FAO 2016). ­These and other ser­vices are degrading in the face of overfishing, pollution, and climate change, potentially threatening food security and h­ uman and environmental health (Millennium Ecosystem Assessment 2006b).

­Humans in marine ecosystems ­ umans have left a mark on ecosystems nearly from the day our ancestors ventured out of Africa milH lennia ago. In coastal regions, even small populations with s­ imple technology quickly depleted the most easily accessible marine animals (Wing and Wing 2001). But ­human populations and impacts have exploded since the mid-­twentieth-­century “­Great Acceleration” (chapter 4), altering all the major components of the earth system: the environment, biodiversity, biogeochemical fluxes, and even geomorphology in the case of coral reefs, river deltas, and heavi­ly populated estuaries (figure 1.3). Humanity’s rise has reached the status of a major historical event in earth history, comparable in scale to glaciations or tectonic movements. This is mainly a consequence of mobilizing huge quantities of fossil carbon formerly sequestered beneath Earth’s surface and using it to power widespread land use change to support a growing h­ uman population and standard of living (Steffen et al. 2007). ­Humans have changed the rules of the game. Our understanding of how the natu­ral world works was historically based on patterns in nature that developed over the long sweep of evolutionary time. Th ­ ose expectations may no longer be valid b­ ecause the systems we live in ­today are out of whack, far from the quasi equilibrium driven by the geography and environment of evolutionary history. Adapting the science of ecol­ogy to the human-­ dominated world is a central challenge for both basic and applied science.

Ecol­ogy in Practice The central challenges of marine ecol­ogy This book is motivated by the premise that major advances in marine ecol­ogy ­will be made at the interfaces between traditions and approaches—­theory and natu­ral history, benthic and pelagic, community and ecosystem, molecular and earth system—­and between ­these historically basic fields of science and the applied fields of fisheries and environmental management and social sciences. Linking ­these approaches and perspectives brings a fuller toolbox than any individual tradition has been accustomed to using, and benefits from a range of advances in technology and data science. It is early days for this integration, but I hope that readers, especially students, are inspired to think in such terms about how to bridge bound­aries.

Chapter 1 Introduction

A

B

C

Figure 1.3. ​Transformation of the geomorphic landscape by humanity. (A) The human-­built Palm and World Islands in the Arabian/Persian Gulf off Dubai, begun in 2001. (B) Rice terraces in Yunnan, China. (C) Mountains transformed by strip mining in Arizona, USA.

Complexity and contingency are inescapable in ecol­ogy, more so than in the physical sciences. That complexity is both exhilarating and intimidating, and stems primarily from biological diversity. The first spark of life has flourished into a variety of forms that we are still struggling mightily to get a ­handle on. Physics has a canon of field theories, chemistry has its periodic ­table, but it sometimes seems that biology has 10 million special cases. Or maybe 100 million; w ­ e’re still not sure. Fortunately, the situation is not so overwhelming as it may seem. Two features of life provide ­handles for taming this complexity. First, organisms are biophysical entities, composed of the same chemical ele­ments and subject to the same physical laws as the rest of the universe (chapter 5). While the diversity of organisms is g­ reat, the ways they work are constrained within well-­defined limits. This opens the door to general theory in ecol­ogy. Second, ­because all species radiated from a single origin, they share many characteristics, and the more recently a species arose, the more similar it is to its siblings in form and function. This means that, functionally, t­ here are fewer types of life forms than ­there are species. This phyloge­ne­tic legacy often allows generalizations about how w ­ hole lineages of organisms function. All diatoms, for example, require silica to build their opaline shells, and all sea turtles need to come ashore to lay eggs. Nevertheless, inherent tensions remain in ecol­ogy. Perhaps the central one is between holistic and reductionist approaches. Getting to know wild organisms and understanding what they do, that is, natu­ral history, is the raw material of ecol­ogy at all levels. Personally, I have always loved identifying organisms, watching them, and figuring out how they make a living. Th ­ ose details are critical to understanding organisms and their interactions, which are in turn critical to managing and conserving life. But seeing the forest that emerges from ­these trees—­the ecosystem—­can also require backing away from the details. Considering a system as a w ­ hole reveals that it is more than the sum of its parts. That is, ecosystems have emergent properties (chapter 9). I take it as given that the purpose of science is to seek and synthesize general rules of cause and effect about how the world works. My approach to ecol­ogy, which guides the organ­ization of this book, seeks the generalities, the rules that transcend the details of natu­ral history to explain why disparate organisms are built and behave as they do and how they interact to produce the emergent features of communities and ecosystems. In this context, the good news about the dizzying complexity of ecol­ogy is that it provides nearly

7

8

Ocean Ecology

unlimited opportunities for seeking and testing ­those general rules. The best example of such a fundamental law is the theory of natu­ral se­lection—it is s­ imple enough to state briefly in plain language and to be understood by most anyone, yet power­ful enough to capture the central mechanism that has produced the living world (chapters 5, 6). I strive throughout the book to link such under­lying princi­ples. Some arise from basic math and physics, such as the laws of thermodynamics and mass balance. Some can be expressed by ­simple equations, such as logistic population growth. O ­ thers are more multifaceted, such as the metabolic theory of ecol­ogy and ecological stoichiometry (chapter 5). My treatment of such theory usually includes minimal mathe­matics. This does not reflect a lack of appreciation for rigorous theory, but rather that I d­ on’t think intuitively in mathematical terms and, in my experience, neither do most students in ecol­ogy. I refer interested readers to other excellent sources for an entrée into the mathe­matics of ecol­ogy (Gotelli 2008). How to balance holistic versus reductionist approaches? The art of ecol­ogy, in practice, is resolving the tension between generality and specificity, which amounts to an optimization prob­lem. The solution depends on the question at hand. How much do we have to know about the system in order to characterize it at the necessary resolution? Optimizing the trade-­off between generality and specificity depends on the goal and involves both empirical considerations and values. For conservation of an endangered species, it’s impor­tant to know every­thing we can about its natu­ral history, so we strive for specificity—­studying the minute details of species biology and distribution, which are key to effective conservation and management. In contrast, if we want to answer the general question of why habitat fragmentation consistently reduces diversity across a wide range of ecosystems, we need to identify general princi­ples. What is the sweet spot between the explanatory power of a model—­ meaning any proposed explanation, w ­ hether expressed mathematically or not—­and the effort and resources required to evaluate it? In statistics, this optimization is a formal pro­cess that compares alternative models by weighing explanatory power against the cost in numbers of par­ameters used (Burnham and Anderson 2002). In ecol­ogy generally, the approach is more informal but the goal is the same. Since time, funding, and ­human capacity are ­limited, striking a balance between the explanatory power of our model and the cost of constructing that model becomes a serious, practical issue. Stated briefly, the central challenge is to achieve Einstein’s somewhat cheeky recommendation: “Every­thing should be made as s­ imple as pos­si­ble, but not simpler.” How do we achieve that balance? ­There are no shortcuts. Historically, ecol­ogy was the province of field naturalists working with their wits and ordinary tools at hand. The field has evolved rapidly, and answering the questions posed above now requires bringing to bear the full panoply of approaches in modern biology, from molecular reconstruction of evolutionary history to large-­and small-­scale experimental manipulation of living communities in their natu­ral habitats, to biogeographic analy­sis, microelectronic physiology, and remote sensing applied to the physical d­ rivers that act on scales of global spaces and eons—­simulation models that strain even the colossal computational power of modern computers. Nevertheless, the single most critical tools for understanding how living ­things interact with their environment remain the h­ uman brain and senses. To deeply understand the ocean and its life, you have to plunge in and open your eyes. Only with the understanding that comes from careful observation ­will the shiny products and data streams of the modern age mean anything of importance.

The ocean and the shore The very dif­fer­ent traditions of biological oceanography (of pelagic systems) and what is usually called marine ecol­ogy (of benthic systems) have themselves been s­ haped strongly by the environment. Studies of pelagic and benthic systems use dif­fer­ent methods, which have led to dif­fer­ent cultures of research and to focus on dif­fer­ent questions. Planktonic communities are mixed suspensions of tiny organisms, and directly observing them in their natu­ral habitats is very challenging. Access to the open ocean requires big ships, and constraints of both shipboard work and the small sizes of

Chapter 1 Introduction

Figure 1.4. ​Distribution of primary producer biomass across the world ocean’s surface. The image is a composite average of sea surface chlorophyll concentrations since the SeaWIFS satellite was launched in 1997. Chlorophyll concentrations range from very low (deep blue) in the open ocean to highest (red) in the most productive coastal areas.

plankton ­favor a focus on aggregate properties, such as bulk chlorophyll as a mea­sure of producer biomass. Importantly, both the dynamic, physically forced nature of pelagic systems and t­ hose logistical constraints have encouraged a culture of multidisciplinary collaboration in oceanography. Biological oceanography thus focuses primarily on questions about ecosystem-­level pro­cesses involving broad functional groups on large, regional scales. For example, how does physical forcing interact with nutrient concentrations to explain global and regional patterns of phytoplankton biomass and production (chapter 10)? The triumph of this approach is vividly illustrated by the view from space of the distribution of the ocean’s primary producers (figure 1.4), which can now be reproduced surprisingly accurately by models (Follows et al. 2007). Benthic marine ecol­ogy has had a dif­fer­ent flavor, traditionally practiced by lone naturalists and steeped in place-­based knowledge. Benthic ecosystems are dominated by seaweeds and sessile invertebrates that are attached and vis­i­ble to the naked eye, fostering a focus on detailed observation and experimentation. Benthic marine ecol­ogy thus developed around questions about interactions among species on organismal scales: How do competition and predation influence local diversity and species composition? Which species are most impor­tant to community organ­ization and why? Much of the pioneering work on pro­cesses of community organ­ization in ecol­ogy as a ­whole emerged from experimental research in marine benthic communities, particularly t­hose of rocky intertidal shores (Paine 1994, Bertness, Bruno, et al. 2014). In recent years, the approaches of biological oceanography and benthic community ecol­ogy have begun to converge, to the benefit of both. I hope this book helps advance that convergence.

The major patterns Marine ecol­ogy was born early in the twentieth ­century from the need to understand and manage fisheries (Petersen 1918). The natu­ral focus was what controls the productivity of plants, the forage animals that sustain fish, and the fish themselves. With the benefit of detailed modern knowledge (see figure 1.4), we have been able to address more specific questions: Why is primary production concentrated at high latitudes and along coastlines? Why do some regions of the ocean

9

10

Ocean Ecology

rich in nitrogen support so l­ittle phytoplankton biomass? How impor­tant are predators compared with inorganic nutrients in controlling plant biomass, and what f­ actors influence the relative strengths of t­ hese controls? How can coral reefs, with ­little vis­i­ble plant biomass, support such an abundance of large fish? Conversely, why do many coastal systems, such as seagrass beds, support luxuriant plant biomass that goes largely uneaten? The answers to t­ hese questions require understanding how environment, resources, and traits of plants and consumers interact. We address them in this book, and we offer impor­tant questions that have not yet been answered. As change in the ecosystems accelerates during the Anthropocene, a new set of questions is redirecting our attention from the mythical pristine systems that traditionally occupied marine ecol­ogy to ­those that we see plainly all around us, transformed by the footprint of humanity. How common are alternative stable states in marine ecosystems and how can we recognize them? What features of the human-­dominated seascape are favorable versus detrimental for marine life, and how do we maximize the favorable aspects? Increasingly, the most pressing questions are at the interface of natu­ral and social science, involving the ecol­ogy of the keystone species Homo sapiens: What policy mea­ sures most effectively mold h­ uman be­hav­ior to sustainable practices, locally and globally? How do we take advantage of h­ uman nature to achieve desirable ends? And how do we coevolve our social contracts and our relationships with nature t­oward a world that is just and fair for all? Answering ­these questions w ­ ill require entrepreneurial, solution-­oriented attention. This book aspires to provide a foundation for addressing them.

2

Life in the Ocean

The Magnitude of Biodiversity Life is our planet’s most striking feature, what uniquely distinguishes it from the rest of the universe, and what drives the functioning of Earth’s ecosystems. The diversity of life is staggering: roughly 1.9 million dif­fer­ent kinds of organisms—­species—­have been described scientifically, at least 226,000 of which live in the ocean (Appeltans et al. 2012). But t­ hose are only a fraction of the number that remain unknown to science. Estimates of the total number of species on Earth differ widely, based variously on expert opinion, proportions of undescribed species in field samples, extrapolation from rates of new species descriptions, and macroecological scaling laws. Most estimates are in the range of 5–11 million eukaryotic species worldwide, 0.7–2.2 million of them marine (Costello et al. 2012, Appeltans et al. 2012, Pimm et al. 2014). The uncertainty is especially high for the ocean ­because most of it is unexplored and populated by tiny organisms that are difficult to sample. A survey of a small beach in Scotland sequenced DNA from sediments and detected 182 species of nematodes (roundworms) alone along an 800 m transect, which compares with ~450 species of nematodes previously described from the entire British Isles (Fonseca et al. 2010). More than 95% of the sequences did not match sequences in publicly available databases, and many are new to science. Similarly, DNA barcoding of invertebrate communities that colonized on settling plates at two well-­studied sites on the east coast of the USA turned up over 2000 species, many of them new to science, and 22 phyla (Leray and Knowlton 2015). Discovery of a new species of mammal or bird routinely generates press around the world, whereas we are still finding new groups of marine invertebrates at the deepest branches of the animal kingdom (i.e., phyla) (figure 2.1). Since 1983 t­ hese have included the Loricifera, tiny creatures nestling among sand grains; and Cycliophora, which make their living among the mouthparts of Norwegian lobsters (Kristensen 2002). And then ­there is the invisible world of microbes, which are key players in all marine ecosystems and whose almost unfathomable diversity is only now coming into view with the aid of modern genomics (chapter 10). Scaling laws applied to a comprehensive database of molecular sequences suggest ­there may be as many as a trillion species of microbes worldwide (Locey and Lennon 2016). Biological diversity, often abbreviated as biodiversity, is usually defined as the variety of life forms at all levels, including ge­ne­tic variation within species, among species, and among ecosystem types. A detailed account of the diversity of marine organisms is well beyond the scope of this book, but understanding how ecosystems work requires some knowledge of biodiversity. In this chapter we take a brief tour of the tree of life’s major branches, their functional characteristics, and approaches to organ­izing that functional diversity. We review the evolutionary origins of diversity in chapter 3 and the pro­cesses that maintain diversity within communities in chapter 7. The influence of biodiversity on ecosystem pro­cesses is discussed in detail in chapters 8 and 9, and is a theme we return to regularly.

11

12

Ocean Ecology

A

B

Figure 2.1. ​New discoveries among the deep branches of life in the ocean. (A) Nanaloricus mysticus, phylum Loricifera, described in 1983. (B) A member of the phylum Cycliophora in its natu­ral habitat, clinging to the mouth-­hairs of a Norwegian lobster.

Diversity on Land and Sea Most animal species are found on land ­because most species of animals are insects, which are absent from the ocean. Beetles alone account for nearly a quarter of all eukaryotic species described so far. So species diversity is lower in the sea than on land. However, the ­great majority of terrestrial organisms, in both abundance and number of species, come from only three major groups: seed plants, insects, and vertebrates. Therefore, t­ hese three taxa, along with microbes, dominate the workings of terrestrial ecosystems. Terrestrial global change biology, for example, focuses largely on the biology of seed plants. But among higher taxa, meaning the deep branches of the tree of life, the ocean is vastly more diverse than the land (­table 2.1). Roughly 35 animal phyla are currently recognized, though the number changes occasionally as new phyloge­ne­tic analyses suggest that a taxon is more or less unique than previously believed. Only one of ­these phyla, the obscure velvet worms (Onychophora), lives only on land, and even this is an artifact of extinction in the sea since fossils remarkably similar to onycophorans are known from the 500-­million-­year-­old marine sediments of the Burgess Shale. In contrast, nearly half the known phyla of animals occur only in the sea, and t­ hese include not only microscopic creatures like loriciferans and cycliophorans but also many of the ocean’s major players and ecosystem engineers (figure 2.2). The echinoderms, for example, are exclusively marine and include the familiar sea urchins—­impor­tant herbivores in benthic ecosystems worldwide—­and the original keystone predator, the sea star Pisaster ochraceous. The sponges (Porifera) are sessile filter-­ feeders that arose from single-­celled ancestors on a branch possibly separate from that of all other Metazoa (multicellular animals). ­These unique organisms include a few freshwater species, but in the ocean sponges have radiated into a diverse and abundant fauna. Sponges dominate many benthic communities, include large and impor­tant habitat-­forming species, and dominate water-­ column biogeochemical pro­cesses in such systems by pumping huge volumes of seawater as they

Chapter 2 Life in the Ocean

­TABLE 2.1 ​Diversity of marine and terrestrial animals at the deepest branches of life (phyla) Exclusively marine

Marine & nonmarine

Exclusively nonmarine

Brachiopoda (lamp shells)

Acanthocephala

Micrognathozoa

Chaetognatha (arrow worms)

Annelida (segmented worms)

Onychophora (velvet worms)

Ctenophora (comb jellies)

Arthropoda ( joint-­legged animals)

Cycliophora

Chordata (chordates)

Dicyemida

Cnidaria (coelenterates)

Echinodermata

Ectoprocta (bryozoans)

Echiura

Entoprocta

Gnathostomulida

Gastrotricha

Hemichordata (acorn worms)

Mollusca (mollusks)

Kinorhyncha

Nematoda (roundworms)

Loricifera

Nematomorpha (horse­hair worms)

Phoronida

Nemertea (ribbon worms)

Placozoa

Platyhelminthes (flatworms)

Priapulida

Porifera (sponges)

Sipuncula (peanut worms)

Rotifera (wheel animals)

Xenacoelomorpha

Tardigrada (­water bears)

Total phyla: 17

16

2

Source: Norse (1993).

feed on suspended bacteria. ­There is nothing like them on land. Arrow worms (Chaetognatha) are an exclusively marine phylum of small, fanged predators common throughout the blue ­waters of the open ocean. The list goes on. At the higher taxonomic levels, then, the ocean is far more diverse than the land. As is true of animals, the functional diversity of primary producers is much greater in the ocean than on land (figure 2.3). Nearly all terrestrial primary producers belong to a single clade derived in the distant past from green algae (the Embryophyta or “land plants”), whereas marine autotrophs come from all three of life’s major domains—­Archaea, Bacteria, and Eukarya—­and comprise at least eight major divisions or phyla among eukaryotes alone. Marine primary producers range from photosynthetic bacteria of several kinds, through the smallest known eukaryotic organism Ostreococcus tauri (Prasinophyceae)—­a tiny planktonic cell with a single mitochondrion and chloroplast (Courties et al. 1994)—to the ­giant kelp Macrocystis pyrifera, which can reach a length of > 45 m and creates the physical habitat that supports rich coastal ecosystems in cooler w ­ aters of both the North and South Pacific (figure 2.3B). ­These divergent lineages are often lumped together as “algae,” a disparate group that includes all aquatic phototrophs that are not flowering plants. Even flowering plants inhabit the margins of the ocean, having secondarily invaded the sea from terrestrial ancestors, and they are dominant producers along the coastal fringes (chapter 11).

13

14

Ocean Ecology

A

B

C

Figure 2.2. ​Animals from exclusively or primarily marine phyla. (A) The North Pacific ochre sea star Pisaster ochraceous (Echinodermata). (B) The arrow worm Spadella cephaloptera (Chaetognatha). (C) Antarctic sponges (Porifera).

The invisible life of the ocean—­the microbes—­are in a class by themselves. The term microbes refers to prokaryotes and single-­celled eukaryotes (figure 2.4). Microbes make up the bulk of the marine biosphere’s living biomass and species diversity and by far the greatest functional diversity of metabolic pathways that underlie biogeochemical pro­cesses. Biological oceanography has been transformed in the last few de­cades by the discovery of the vast abundance, diversity, and ecological importance of t­ hese very small organisms. The most striking example is the picoplanktonic cyanobacterium Prochlorococcus, the most abundant photosynthetic organism on Earth, making up 30%– 60% of total phytoplankton biomass throughout the oligotrophic open ocean. But microbes also include a plethora of ancient lineages of Bacteria, Archaea, eukaryotes invisible to microscopy, parasitic flagellates that infect nearly all major taxa in the plankton, and an abundant community of even smaller viruses (box 2.1). Most of the organisms in this microbial menagerie are smaller than the phytoplankton captured in nets, and thus w ­ ere underestimated or missed entirely by traditional sampling methods. This world first came to light in 2003 when shotgun genome sequencing was applied to w ­ ater samples collected from the Sargasso Sea, producing more than 1800 species of planktonic microbes, many of them only distantly related to previously known lineages (Venter et al. 2004).

Phyloge­ne­tic Classification of Marine Biodiversity Species can be classified ­either phyloge­ne­tically, according to common ancestry (who they are), or functionally (what they do). The two approaches coincide to some degree b­ ecause closely related species also tend to be functionally similar—­for example, all diatoms are unicellular, photosynthetic

A

B

C

D

E

Figure 2.3. ​The diversity of marine primary producers. (A) The picoplanktonic cyanobacterium Prochlorococcus, dominant primary producer of the oligotrophic ocean. (B) The ­giant kelp Macrocystis pyrifera, world’s largest marine plant. (C) A crustose coralline alga, characteristic of highly grazed habitats. (D) The seagrass Zostera marina, most widespread marine macrophyte in shallow w ­ ater. (E) The reef coral Acropora palmata, former foundation species of shallow reefs throughout the tropical West Atlantic. Reef corals derive most of their nutrition from symbiotic algae, so are functionally plants.

A

D

B

E

C

F

Figure 2.4. ​The diverse microbial life of the ocean. (A) ­Giant xenophyophore protists on the deep-­sea floor, providing habitat for brittle stars. (B) A planktonic radiolarian. (C) The dinoflagellate Ornithocercus. (D) A planktonic foraminiferan. (E) A tintinnid ciliate. (F) Viruses at the surface of a host cell.

Ocean Ecology

Box 2.1. ​Invisible ocean: Discovery of a new microbial world Early in the twenty-­first ­century the developing genomic revolution blew the lid off the classical view of the ocean’s pelagic ecosystems, revealing a staggering diversity and abundance of tiny, previously unknown organisms. The first milestone came with a comprehensive inventory of microscopic organisms in seawater using metagenomics,

that is, characterization of ge­ne­tic material from bulk environmental samples rather than by sampling individual species. In the early 2000s researchers used the new technique of whole-­genome shotgun sequencing to census the microbial communities in surface w ­ aters from the Sargasso Sea off Bermuda. In this approach, a

(A)

(B)

(C)

(D)

(E)

(F)

(G)

90,000

60,000 Diversity

Number of OTUs

16

30,000

0

0

6 5 4 3 2 1

Pico- Nano Micro Meso nano

400 100 200 300 500 Cumulative number of DNA sequence reads (million)

Figure B2.1.1. (A) A diatom Chaetoceros bulbosus (Stramenopila), with chloroplasts shown in red (scale = 10 μm). (B) A heterotrophic dinoflagellate with kleptoplasts in red (scale = 20 μm). (C) An Acantharian with endosymbiotic cells of the haptophyte alga Phaeocystis in red (scale = 50 μm). (D) A colonial network of Rhizarians with captive dinoflagellate symbionts (scale = 50 μm). (E) A copepod with gut colonized by parasitic dinoflagellates in red (scale = 100 μm). (F) Cross section of a dinoflagellate infected with parasitoid alveolates (blue spots) (scale = 5 μm). (G) Rarefaction curves of number of operational taxonomic units (OTUs) with increasing number of samples, and Shannon diversity (inset) for each size fraction of plankton (­after de Vargas et al. 2015).

17

Chapter 2 Life in the Ocean

genomic sample is broken up into random fragments and sequenced; then powerful computer algorithms reassemble the fragments into coherent genome sequences by matching them along areas of overlap. The Sargasso Sea census revealed at least 1800 operational taxonomic units (OTUs, or putative species), including 148 previously unknown bacterial types, and over a million new genes (Venter et al. 2004). Molecular explorers then turned to deep ocean environments, first in the North Atlantic and waters around hydrothermal vents, finding that bacterial communities in these habitats were up to two orders of magnitude more diverse than any previously reported. Most samples were dominated by a few common and abundant types, as is true of nearly all ecological communities (chapter 7), but beyond these were a long list of thousands of species represented by only a few sequences each. Many of the latter were radically different from previously known lineages and apparently very ancient, and were dubbed the “rare biosphere” (Sogin et al. 2006). The next major exploration of the ocean’s invisible biosphere targeted pelagic communities worldwide, in both the surface and deeper ocean. The Tara Oceans expedition launched a sailing schooner that sampled 68 sites around the world. The voyage yielded more than 40 million gene sequences from the ocean’s smallest organisms (viruses, prokaryotes, and small eukaryotes)

(H)

Apicomplexa Rickettsiales Spirotrichea Oceanospirillales Other Alveolata

and revealed no end in sight for the diversity of marine microbial life (figure B2.1.1). Samples of the intermediate plankton size range, from the smallest eukaryotic protists to the smallest animals, produced ~150,000 putative species, about a third of which could not be assigned to any known higher taxon (de Vargas et al. 2015). As is also common among larger animals and plants, community composition of the more than 35,000 species most strongly followed gradients in temperature, but in contrast to larger organisms there was little clear differentiation among geographic regions (Sunagawa et al. 2015). Ocean metagenomic studies have since moved beyond the description of diversity to catalog the types and frequencies of interactions involving the microbial menagerie—the ocean’s “interactome” (Lima-Mendez et al. 2015). The Tara expedition samples revealed a rich range of interactions (figure B2.1.2). Among the surprises was the commonness of parasitism, even among the tiniest protists and copepods: the largest number of interactions involved Syndiniales, a group of parasitic and symbiotic dinoflagellates that infected nearly all taxa of both protists and metazoans (mostly copepods) in these samples. Viruses were also abundant and highly host-specific. These results suggest a much more intricate and specialized interaction web than previously suspected, implying that parasitism exerts a profound and previously underappreciated impact on plankton dynamics and therefore on biogeochemical cycles.

Chaetognatha Syndiniales

Cyanobacteria Acantharea Bacillariophyta

Arthropoda

Prymnesiophyceae MAST Polycystinea

Positive associations

Negative associations

Dinophyceae

Figure B2.1.2. The interactome of the open pelagic ocean. Figures show links among the top 15 taxa as colored segments, where links connecting two segments indicate co-occurrence (left) or mutual exclusion (right). Thickness of the connector reflects the number of links between the species within each category, and color reflects the partner taxon with the most total links (after Lima-Mendez et al. 2015).

125-91392_Duffy_OceanEcology_5P.indd 17

3/16/21 9:51 PM

18

Ocean Ecology

(A) Monophyletic clade D

E

G

C

H F

(B) Polyphyletic clade

J

K I

B

D

E

G

C

H F

J

K I

B A

(C) Paraphyletic clade D

E

G

C

H F

J

K I

B A

A

Figure 2.5. ​Schematic phyloge­ne­tic trees, showing (A) a natu­ral group (monophyletic clade) of species defined by shared lines of evolutionary descent, and (B, C) two types of nonnatural groups.

organisms with a siliceous shell. But the converse is less reliable: functionally similar species are not always closely related. As noted above, unicellular, photosynthetic organisms (phytoplankton) include a highly diverse collection of prokaryotic and eukaryotic lineages that diverged early in earth history. ­Here we review the major branches in the tree of life before turning to functional approaches for classifying marine organisms.

The tree of life ­ ecause all life forms arose from a single origin early in Earth’s history, they are related by a pedigree, B or ­family tree of species, known as a phylogeny (figure 2.5). A phylogeny summarizes the lines of descent and thus the evolutionary relationships among the entities at the tips, ­whether they be species or genes (D. A. Baum et al. 2005). The common ancestry of all species has two implications for ecol­ogy. First, the pattern of ancestral relationships defines a natu­ral, hierarchical classification of life. This classification begins with domains or kingdoms, which represent the earliest branching trunks of the evolutionary tree and extend to closely related species as the twigs at its tips. Reconstructing that pattern of branching and using it to classify life is the province of phyloge­ne­tic systematics (Nei and Kumar 2000, Wiley and Lieberman 2011). The second consequence of life’s common ancestry is that evolutionary divergence among species is roughly proportional to the time since they ­were separated, and therefore recent common ancestry of species implies some similarity in their biology. Indeed, it is surprisingly common for ecologically impor­tant traits to be conserved at very deep branches in the tree of life. For example, all peracarid crustaceans—­the amphipods, isopods, and their kin, which include over 10,000 species—­have direct development, hatching into benthic juveniles without a planktonic stage. This conserved development mode has large consequences for demography, population ge­ne­tic structure, ecological interactions, and evolutionary pro­cesses. Similarly, feeding mode (e.g., carnivory, herbivory, planktivory) is typically conserved at the level of taxonomic f­ amily or even higher levels in many fishes and invertebrates. Such evolutionary conservatism can have far-­reaching consequences for ecosystems. A global meta-­analysis of marine herbivory shows that the strongest predictors of grazer impacts on plants are producer taxonomic order and morphology (which are related), whereas, surprisingly, t­ here is l­ ittle or no effect of latitude, temperature, or nutrients (Poore et al. 2012). This phyloge­ne­tic conservatism of ecological traits has generated an active field of comparative methods and phyloge­ne­tic community ecol­ogy (Webb et al. 2002, Srivastava et al. 2012). Despite (or perhaps ­because of) this pro­gress, it is impor­tant to emphasize that ancestry is not destiny—­many taxa of organisms show strong phyloge­ne­tic conservatism in their traits, but ­these are nevertheless poor predictors of ecological interactions (Losos 2011). Trait-­ ecology relationships need to be tested, not assumed.

Chapter 2 Life in the Ocean

ta

hil e

s

r

op r th er m

ta eo ha rc ya

Ox S id ulfu ize r rs

Hy pe

na

Cre

Archaea r Eu

Gr no een n-s Gra ulf mur po ba siti cte ve ria ba cte ria Bacteria

eo cha

Halophiles Methanogens

Proteobacteria Entamoebae Slime molds Animals Fungi

Microsporidia

nts es Pla llat nads e Flag omo h Tric

ia ter ac b ia les o a an ter Cy bac tog o o v Fla herm T

Diplomonads

Eukarya

Figure 2.6. ​The tree of life, showing its three main branches. T­ hese contain 92 named phyla of Bacteria, 26 phyla of Archaea, and the eukaryotic supergroups.

With the advent of modern DNA sequencing and metagenomics, the tree of life has been revealed in much finer resolution. Life is now recognized as having three main trunks or domains, the Bacteria, Archaea, and Eukarya (figure 2.6), which diverged very early in earth history. One of the most impor­tant events in late twentieth-­century biology was the discovery, hidden among the Bacteria, of a formerly unrecognized deep branch of life, the Archaea—­organisms superficially similar to Bacteria in being tiny cells lacking nuclei, but other­wise as dif­fer­ent genet­ically and metabolically from Bacteria as ­either is from animals or plants. Many of the genes and metabolic pathways of Archaea are closer to t­ hose of eukaryotes than of Bacteria and, uniquely among living organisms, their cell membranes contain ether lipids. Archaea include many of the key players in ocean ecosystems, occurring throughout the ocean ­water column and sediments, but they are especially impor­tant in chemically extreme environments, such as low-­oxygen zones, sulfide-­rich sediments, hydrothermal vents, and chemical hot springs (Woese et al. 1990, Offre et al. 2013). ­These extremophile Archaea greatly expand the metabolic capacity of the biosphere over that attributable to Bacteria and eukaryotes. They include organisms that produce methane and that inhabit hot springs and nearly salt-­saturated ­waters, and cells that metabolize metal ions, methane, and even elemental hydrogen. Bacteria and Archaea are the oldest life forms on earth. In addition to their lack of nuclei and other intracellular infrastructure, they share the curious ability to transfer genes among unrelated organisms, which shuffles ge­ne­tic material, and associated physiological and metabolic capacity, among lineages. Such horizontal gene transfer among Bacteria is one reason for the frightening spread of antibiotic re­sis­tance among ­human pathogens in hospitals. It has also played a central role in developing the collective metabolic capacity of the ocean microbiome (Falkowski et  al. 2008). The promiscuity of horizontal gene transfer prob­ably also facilitated a key event in life’s early history when entire bacterial cells w ­ ere engulfed by, and incorporated as permanent symbionts within, cells of Archaea. The symbiotic bacteria eventually became completely integrated into their host cells—we now know them as mitochondria—­enabling a power­ful new form of energy metabolism that greatly increased efficiency of energy production. Th ­ ese early cellular chimaeras w ­ ere the ancestors of the first eukaryotes, and their energetic superiority enabled them to flourish and diversify as the third domain of life (Lane 2015). Another such symbiotic event involved capture of photosynthetic cyanobacteria by early eukaryotes, which ­were permanently integrated as chloroplasts. Over subsequent eons, many genes of the originally free-­living symbionts w ­ ere lost, while o­ thers ­were transferred permanently to the host’s genome, blurring the distinction between host and symbiont. Such horizontal gene transfer is a

19

20

Ocean Ecology

standard mode of evolution among Bacteria and Archaea, and was long thought to be restricted to ­these relatively s­ imple organisms. In contrast, the ge­ne­tic material of eukaryotes is isolated within a membrane-­bound nucleus, and their cells are far more complex structurally than ­those of Bacteria and Archaea. Yet genomic studies provide growing evidence that horizontal gene transfer from prokaryotic symbionts to eukaryotic hosts has been common and has had major consequences (Boto 2010), for example, in allowing certain herbivorous insects to tap new food sources and become major crop pests (Moran et al. 2008). The Eukarya include virtually all living ­things with which most ­people are familiar and are traditionally divided into four kingdoms: plants, animals, fungi, and protists. Another major outcome of the molecular revolution of recent de­cades has been recognition that single-­celled eukaryotes, formerly lumped into the grab bag of “Protozoa,” constitute a highly diverse collection of lineages, as dif­fer­ent from one another as ­humans are from jellyfish (figures 2.4, 2.6). Recognition of this deep diversity required the erection of a large number of new phyla of single-­celled organisms, collectively known as protists (Caron et al. 2012, Forster et al. 2016). ­These developments highlight the point that the living world is and always has been microbial—­large, complex animals like Homo sapiens are anomalies.

Phyloge­ne­tic relationships and tree thinking How is the tree of life reconstructed? Classification of life by genealogical relationships is based on two premises: (1) the natu­ral unit of classification is a monophyletic group or clade (from the Greek klados, “branch”), that is, a group that contains an ancestral species and all its descendants; and (2) ­these monophyletic groups are defined in practice by shared derived characters (synapomorphies), which provide evidence of common ancestry. A phyloge­ne­tic tree is thus a set of hierarchically nested groups, or clades (see figure  2.5). By contrast, early taxonomic classifications grouped species by characteristics that are superficially similar but not inherited from common ancestors (e.g., Linnaeus’s taxon “Vermes” for many unrelated wormlike animals). The Vermes is an example of a polyphyletic group, one that includes descendants of multiple unrelated ancestors. Another example is “algae,” which includes members from the Bacteria and several unrelated protist groups. The second type of artificial taxonomic group is the paraphyletic taxon, which includes some but not all descendants of a common ancestor. An example is “fishes,” which excludes the terrestrial vertebrates that evolved from fishes and are nested within that clade. The traditional group “Protozoa” is also paraphyletic ­because it excludes the animals and fungi that evolved from protists. How do we identify shared derived characters? ­Because the ancestors of most life forms are long extinct and cannot be directly observed, genealogical relationships must be reconstructed using the indirect evidence still available to us. This comes from fossils, which are fragmentary and generally available only for groups with mineralized skele­tons, and from the DNA sequences of living organisms. DNA preserves a rec­ord of ancestry, but it is obscured by evolutionary pro­cesses over the long period of time since divergence. Thus, some ingenuity is required to extract the signal of common ancestry from the noise of subsequent random mutations and adaptations. Reconstructing a phyloge­ ne­tic tree starts with collecting data on characters, a generic term for any observable feature of an organism, which can include morphology, biochemistry, or be­hav­ior. The most valuable characters are in DNA sequences, which provide rich quantities of data and for which models of evolution are well worked out. In essence, phylogeny reconstruction compares the sequences of bases in the DNA of species ­under the assumption that the differences among them have accumulated via random mutation over time. Mutated sequences that are shared among species, and that differ from ­those of other species, are considered evidence of common ancestry. To analyze that evidence, models of evolutionary change in sequences are applied to the distribution of character states (base sequences,

Chapter 2 Life in the Ocean

in the case of DNA) among species, which groups them into hierarchical patterns of genealogical relationships. A central princi­ple guiding phyloge­ne­tic reconstruction is parsimony: given a choice between evolutionary scenarios, the one that requires the fewest evolutionary changes in the observed data is preferred.

The web of life Divergence of lineages from a common ancestor over Earth’s history has created a branching pattern of relationships among lineages and species—­a tree of life. But t­ hose branches are not forever separate from one another. In some cases, lineages fuse with one another partially and exchange genes even among rather distantly related evolutionary branches, as mentioned above, forming what is more accurately a web of life than a tree. Such horizontal gene transfer is common among prokaryotes, but it also occurs more often than previously appreciated among eukaryotic lineages (Andersson 2005). In some types of organisms, exchange of ge­ne­tic material across species, known as hybridization, is fairly common. For example, hybridization among species is routine in oak trees, which have nevertheless retained their integrity as species for millions of years. Among marine animals, hybridization and ge­ne­tic introgression among species appears to be especially widespread in reef corals, some of which are better thought of as syngameons—­complexes of partially interfertile species that regularly exchange genes (van Oppen et al. 2001, Ladner and Palumbi 2012). Hybridization is impor­tant for a ­couple of reasons. First, genes obtained by hybridization increase the functional capacity of a lineage, with impor­tant consequences for its physiology, species interactions, and ecosystem pro­cesses. Second, the distinctness of individual species is often maintained as much by their use of dif­fer­ent habitats and geographic separation as it is by their intrinsic be­hav­ior and physiology. But in the Anthropocene, ­human activities are transporting and mixing species that ­were formerly separated, and the rate of hybridization is increasing dramatically as a result. That hybridization has consequences for conservation and management of populations in the modern ocean, for example, as aquaculture species escape into the field.

Functional Organ­ization of Pelagic and Benthic Life Ecosystem pro­cesses are the outcome of what species do. Process-­oriented ecol­ogy therefore requires organ­izing biodiversity by the functions that organisms perform. An individual organism works as an integrated ­whole; but, like a machine, examining its parts helps us understand how it works. ­Those parts include any characteristic of the organism, ­whether morphological, biochemical, physiological, or behavioral, and are referred to as traits. A functional trait is any characteristic of an organism that influences its fitness by affecting growth, reproduction, or survival (Violle et al. 2007). Functional traits include fundamental characteristics common to all organisms, such as body size, fecundity, and rates of metabolism and growth, as well as more specific characteristics—­ability to fix nitrogen, produce defensive chemicals, and so on. We can further distinguish two types of traits: effect traits influence what species do, including their use of energy and resources, growth rates, and interactions with other species, whereas response traits affect how species respond to environmental forcing (Lavorel and Garnier 2002). The functions that organisms perform are of course also ­shaped by the environments where they live. The major dichotomy among marine ecosystems is between the pelagic realm, where organisms are suspended in the open, three-­dimensional ­water column, and benthic systems, where life is concentrated around organisms anchored to the two-­dimensional seabed. The physical conditions of ­these realms f­avor dif­fer­ent functional types of organisms and thus differences in ecosystem pro­ cesses. Below we outline pelagic and benthic functional diversity.

21

22

Ocean Ecology

Functional groups of pelagic life The open ocean is home to a wider range of life forms than anywhere e­ lse on Earth (figures 2.3, 2.4, 2.7). Biological oceanographers have traditionally classified them from a functional perspective, mainly ­because of sampling logistics, since dif­fer­ent size classes of plankton are caught with dif­fer­ent gear. Plankton sampling relies mostly on nets, and the larger eukaryotic algal cells and small invertebrates they retain are often referred to as net plankton, in contrast to the smaller microbes that can be captured only on very fine filters. ­Today plankton are still categorized mainly by body size (­table 2.2). The picoplankton comprise cells 0.2–2 μm in size and include the smallest eukaryotes, but mainly prokaryotes, including the dominant phytoplankton of the open ocean, the cyanobacterium Prochlorococcus. Most ocean Bacteria and Archaea fall within the picoplankton size class, which contains a large part of the ocean’s biomass and phyloge­ne­tic and metabolic diversity. Picoplanktonic eukaryotes are also diverse, phyloge­ne­tically and functionally, comprising many lineages of photosynthetic and heterotrophic protists, primarily bacterivores, that are impor­tant links in the microbial loop, discussed in chapter 10 (Massana 2011). The nanoplankton (2–20 μm) include both phototrophic and heterotrophic eukaryotes, at the smallest limit of resolution by light microscopy. Photosynthetic flagellates in the nanoplankton size range are major components of many plankton communities, particularly ­under stratified, low-­nutrient conditions where their high surface-­to-­volume ratio and the mobility conferred by the flagellum provide an advantage over other phytoplankton types in accessing scarce nutrients. The microplankton (20–200 μm) include many of the familiar taxa of the classical net phytoplankton, such as diatoms and larger dinoflagellates. The mesoplankton (0.2–2 mm) are at the low end of the metazoan size range, notably the small copepods historically considered the major herbivores of the open ocean. The macroplankton (> 2 mm) include larger copepods, krill, and the gelatinous animals, such as medusae, ctenophores, and salps (figure 2.7). Size classes of organisms generally correspond roughly to functional groups ­because body size strongly affects biological rates and modes of life. Functional groups may include phyloge­ne­tically distinct species that perform similar pro­cesses, such as the several unrelated photosynthetic lineages lumped as algae. But functional groups often correspond to distinct phyloge­ne­tic lineages (i.e., higher taxa). Among primary producers, examples include nitrogen-­fixers or diazotrophs (certain lineages of cyanobacteria), silica producers (diatoms), and calcifiers (coccolithophorids). Functional groupings are more subjective than phyloge­ne­tic classifications since species may be grouped quite differently depending on the traits of interest. But most functional schemes group species by some combination of body size and metabolic capacity.

­TABLE 2.2 ​Size-­based classification of plankton Group

Size (ESD)

Major taxa

Macroplankton

> 2 cm

Metazoans: pteropods, chaetognaths; euphausiids, medusae, ctenophores, salps, pyrosomes, amphipods

Mesoplankton

0.2–20 mm

Metazoans: copepods, cladocerans, ostracods, chaetognaths, pteropods, heteropods

Microplankton

20–200 μm

Protists: foraminifera, ciliates; metazoans: rotifers, copepod nauplii

Nanoplankton

2–20 μm

Small protists: diatoms, flagellates, pyrrophytes, chrysophytes, chlorophytes, xanthophytes

Picoplankton

0.2–2 μm

Bacteria, cyanobacteria, Archaea, chrysophytes, smallest protists

Femtoplankton

 5% of the surface value, at least as much production from benthic algae.

Marine ecoregions of the world Longhurst’s main focus was the pelagic zone. Meanwhile, the growing needs of conservation and management in the coastal zone stimulated a new effort to revisit the unique ocean margins. This resulted in the Marine Ecoregions of the World (MEOW) (figure 3.15), a scheme that attempts to

Arctic

Temperate Northern Pacific

Temperate Northern Atlantic

Temperate Northern Pacific

Eastern Indo-Pacific

Arctic

Tropical Eastern Pacific

Tropical Atlantic

Temperate South America

Western Indo-Pacific

Central Indo-Pacific

Temperate Southern Africa Temperate Autralasia Southern Ocean

Figure 3.15. ​Coastal Marine Ecoregions of the World (MEOW), consisting of 12 biogeographic realms of the coastal ocean. The smaller unmarked divisions are the 62 coastal marine provinces (­after Spalding 2007).

Chapter 3 Geography of Marine Life

integrate both evolutionary history and current environment (Spalding et al. 2007). A specific goal of MEOW was to achieve finer spatial resolution than previous schemes, identifying areas biologically homogeneous enough to represent natu­ral management units. At the broadest scale, MEOW divides the coastal ocean into biogeographic realms—­biotically distinct (coastal) areas separated by land masses, wide ocean barriers, and temperature gradients that reduce biotic mixing over long spans of evolutionary time. The 12 realms are large-­scale regions where biotas are “coherent at higher taxonomic levels, as a result of shared and unique evolutionary histories.” Endemism within realms is high, including unique genera and families in some groups. The MEOW realms are broadly equivalent to Longhurst’s biomes but divided more finely, especially in the Southern Hemi­sphere where large ocean expanses and complexity of coastal regions contrast with the continuous circumpolar circulation of the pelagic. Nested within the MEOW realms are 62 provinces defined primarily by distinctive abiotic features and characterized by endemism at the species level. Again, the scale is roughly comparable to that of Longhurst’s provinces. The MEOW provinces are meant to encompass the w ­ hole life histories of organisms that disperse widely as adults or larvae. MEOW goes beyond Longhurst to divide provinces into 232 ecoregions, defined by relatively homogeneous species composition that is distinct from adjacent systems. Despite being based primarily on patterns of species endemism, the MEOW biogeographic units reflect oceanographic conditions closely, confirming that the geographic distributions of coastal marine species are determined by environmental conditions on a relatively fine scale. The distribution of well-­studied coastal bivalves within MEOW provinces ­were predicted with 89%–100% accuracy by a model that included only annual mean and range in temperature, salinity, and productivity (Belanger et al. 2012). This result quantitatively supports Longhurst’s (2007) emphasis on oceanography in shaping pelagic marine biogeography and confirms that it works very well for coastal benthic species as well. Accordingly, the marine ecoregions classification is increasingly used to or­ga­nize biogeo­graph­i­cal approaches in marine conservation (Simpson et al. 2011, Edgar et al. 2014).

Large marine ecosystems Fi­nally, a spatial classification that emerged for practical reasons of international fisheries management divided the world ocean into 64 large marine ecosystems (LMEs), conceived as large regions of the ocean, > 200,000 km2, and characterized by distinct bathymetry, hydrography, productivity, and trophically dependent populations (Sherman 1991). The LME concept aimed to facilitate the ecosystem approach to fisheries, which grew out of a sense among fishery scientists and man­ag­ers that the historical focus on single species populations in fishery management was seriously flawed. As such the LME concept was explic­itly connected to applied science and the geo­graph­i­cal classification of ocean regions into LMEs included both biophysical and po­liti­cal criteria.

The Biogeography of Functional Traits From the perspective of biological oceanography, the main value of biogeography is what it tells us about the distribution of functional types of organisms that influence ecosystem pro­cesses. The distributions of traits, like species, are strongly affected by evolutionary history, especially among the benthos, which have more l­imited dispersal than pelagic organisms. The composition of communities can differ substantially among regions and in ways that have far-­reaching consequences for ecosystem pro­cesses. In many cases, the differences appear due to what are sometimes called accidents of history—­ legacies of past events that are not closely related to current environmental conditions. Large herbivorous patellid limpets are common and impor­tant grazers on rocky shores of the Northeast

63

64

Ocean Ecology

Atlantic but are absent from the other­wise environmentally and biotically similar northwestern Atlantic, which explains in part why grazing is considered a stronger control on intertidal macroalgae in Eu­rope than on the American side of the Atlantic ( Jenkins et al. 2008). The situation is reversed for predators that dig in nearshore sediments: swimming crabs, waterfowl, skates, and rays are more common, abundant, and larger-­bodied on the western side of the Atlantic, and ­these size differences have persisted for three million years (Vermeij et al. 2008). The sediment disturbance and predation mediated by t­hese animals is thought to result in greater sediment disturbance, bioturbation, and top-­down control of infauna on the western side of the Atlantic ( Jenkins et al. 2008). The story of the rise and fall of North Atlantic cod fisheries provides a striking example of the consequences of biogeographic variation in functional traits (box 3.2). But ­there are also clearly deterministic aspects of functional biogeography, involving patterns in traits that vary systematically with latitude. One of the best documented involves life history. Gunnar Thorson first noted that planktonic larvae are rare among marine animals of cold regions compared with ­those from warmer ­waters, a pattern ­later dubbed Thorson’s rule. Subsequent synthesis of life history data from more than 1000 marine invertebrate species confirmed that planktotrophic (feeding) larvae are indeed less common at high latitudes, as well as in the deep sea (Marshall et al. 2012). This is likely b­ ecause productivity of their planktonic food is less reliable in the highly seasonal environments at high latitudes and b­ ecause colder temperatures prolong the risky phase of larval development (figure 3.16A, B). In contrast, most species in cold ­water produce large, rapidly developing larvae or eggs that hatch directly into benthic juveniles. Such direct development may be doubly favored at high latitudes ­because well-­provisioned offspring survive better and ­because oxygen is more soluble in cold ­water, relaxing the constraint on diffusion into tissues (Strathmann 1985). ­These temperature-­related trends in life history may help explain some other­wise puzzling exceptions to the latitudinal diversity gradient. In several marine taxa, species richness does not decline with latitude, and t­ hese tend to be organisms that do not produce planktotrophic larvae. Species richness declines poleward in the classical pattern among marine animals with planktotrophic larvae. In contrast, species richness of ­those with direct development instead increases ­toward the poles (Fernández et  al. 2009). ­These patterns closely follow temperature (figure 3.16C–­F). A pos­si­ble explanation is that the more rapid development of planktonic larvae in warm temperatures shortens dispersal range, likely fostering ge­ne­tic subdivision and thus greater speciation of planktotrophs in the tropics. Conversely, the advantages of direct development in colder w ­ aters may have led to greater ecological diversification or less extinction ­there, explaining the higher richness ­toward the poles in brooding taxa. ­There are also strong biogeographic patterns in functional traits of pelagic organisms (Barton et al. 2013). Among phytoplankton, cells tend to be larger, and nitrogen-­fixers (diazotrophs) rarer, at high latitudes compared with the tiny picoplankton that dominate warmer, oligotrophic w ­ aters. Among zooplankton, polar w ­ aters are dominated by large-­bodied copepod species with generation times of up to three years, whereas tropical ­waters feature small, rapidly growing copepods. ­These patterns characterize communities with very dif­fer­ent dynamics and contributions to ecosystem pro­cesses. For example, the communities of small copepods in the oligotrophic tropics generate less vertical particle flux and weaker trophic transfer through the “classical” food chain to fishes than do large copepods in boreal w ­ aters.

The Biogeography of Species Interactions Geographic patterns in species richness, trait distributions, and environmental forcing suggest that ­there should similarly be geographic variation in species interactions. Like so many enduring themes in ecol­ogy, this idea was first explored systematically by Robert MacArthur (1972), who was intrigued by the idea that competition should be more intense in the diverse communities of the tropics. Multiple lines of evidence support a gradient of stronger interaction strengths t­oward the

Chapter 3 Geography of Marine Life

Box 3.2. ​Functional biogeography and fisheries: A tale of two oceans (A)

Latitude

60°N 30°N 0° 30°S 60°S

> 50 cm 40 to 50 cm 35 to 40 cm 30 to 35 cm < 30 cm 150° 120° 90° W

60°

30°

0°

30°

60°

Longitude

90° 120° 150° E

(B) Change in predator landings (% per year)

The primeval abundance of North Atlantic cod (Gadus morhua) was staggering. When medieval Eu­ro­pe­ans learned to dry and salt this abundant fish, they in­ven­ted one of the first industries for producing and storing high-­protein food, an innovation that arguably launched the age of discovery—­the Eu­ro­pean colonization of the globe—­fundamentally transforming the course of world history (Kurlansky 2001). Perhaps as early as the 1300s, the already depleted fisheries of Eu­ro­pean seas impelled intrepid fishermen and Viking raiders to brave the unknown w ­ aters of “Ocean,” as the North Atlantic was called when Eu­rope’s known world comprised the fringes of the Mediterranean. The discovery of huge quantities of cod along the far ocean coasts of Iceland, Greenland, and Atlantic Canada nourished and motivated the Basques, Vikings, and subsequent Eu­ro­pean explorers as they struck out across the Atlantic in a wave of expansion. Cod fueled the new American colonists’ prosperity, and they named the barrier of their Mas­sa­chu­setts Bay colony for it in gratitude. Dried and salted cod from Canadian w ­ aters also fueled the less savory triangle trade that exchanged African p ­ eople as slaves—­for sugar in the West Indies, for rum to colonial North Amer­i­ca. To this day “saltfish”—­salted and dried cod from faraway Canada packed in wooden boxes—is, paradoxically, common in the diet of many Jamaicans. Cod fishing built and continued to sustain much of North Atlantic civilization from the sixteenth ­century on u ­ ntil, suddenly in the 1980s, cod collapsed across the region. The reasons for the collapse of cod are complex, involving synergism between long-­term climate oscillations, sustained overfishing, and nonlinear ecosystem dynamics that f­avor alternative ecosystem states (Hilborn and Litzinger 2009, Pershing et al. 2015). But the rise and fall of cod also seems to have involved a peculiar biogeographic pattern in fish body size. The collapse of cod that ruined the economy of eastern Canada in the 1980s prob­ably would not have happened in the Pacific Ocean. Among more than 12,000 marine species, a striking concentration of the world’s largest fish species live in the North Atlantic (Fisher et al. 2010) (figure B3.2.1). For poorly understood reasons, fishes of the North Pacific tend to be smaller and faster-­growing, and fisheries ­there have been more stable than in the North Atlantic. Why? Large body size not only makes fish attractive targets for fishers but is also associated with a host of life history traits, including slow growth and late maturity. As a result of their generally larger maximum body sizes and slower growth rates, the species targeted by North Atlantic fisheries have been much more vulnerable to overfishing and collapse than t­ hose in the North Pacific. The mighty cod fisheries—­once thought to be inexhaustible—­ represent the poster child of this phenomenon. The vulnerability of large-­bodied cod is a specific instance of a very general macroecological pattern: large-­bodied species are negatively affected by exploitation among fishes (Jennings et al. 1999, Dulvy et al. 2003), modern marine mammals (Davidson et al. 2012), and Pleistocene terrestrial mammals (Sandom et al. 2014).

10 5 0 –5 –10 –15 20

30 40 50 60 Geometric mean species length (cm) in each large marine ecosystem

70

(C)

Figure B3.2.1. Biogeography of fish body size and its consequences for fisheries. (A) Mean maximum length of fish species within 56 large marine ecosystems (LMEs). (B) Declines in landings of predatory fish ­were greater in regions with large fish body size across the ecosystems (­after Fisher et al. 2010). (C) Abundant catch of large Atlantic cod during the good old days.

65

Ocean Ecology

(C)

3,000

Species richness

2,000

1,000

0

40

15

30

10 5 0

0

20 30 40 50 60 70 80 Latitude

(B)

(D)

Species richness

21

14

7

0 0.2

5

10

15 SST

20

25

2.2

3.2

Chl a (mg m–3)

4.2

5.2

10

30

0

5

10

15

20

25

30

20

25

30

SST (F)

Crustaceans

Crustaceans

80

80

60

60

40 20 0

1.2

20

0 0

28

Molluscs

20

Species richness

Offspring size (υm)

4,000

(E)

Molluscs

Species richness

(A)

Tempterature (°C)

66

40 20 0

0

5

10

15 SST

20

25

30

0

5

10

15 SST

Figure 3.16. ​Biogeography of larval development. (A) Latitudinal increase in average offspring size among nonplanktonic marine invertebrates. (B) Proportion of invertebrate species with nonplanktonic larval development (red = high; blue = low) as a function of sea surface temperature and pelagic productivity (chlorophyll a concentration). Species richness of direct developing (C, D) and planktotrophic (E, F) invertebrates as a function of sea surface temperature (SST, degrees C°) in the Southeast Pacific ([A, B] ­ after Marshall et al. 2012; [C–­F] ­after Fernández et al. 2009).

equator (Schemske et al. 2009). First, defenses against predators are generally stronger and more prevalent at low latitudes, suggesting that predation is more intense in the tropics. For example, shells of marine snails are stronger, on average, in the tropics relative to their temperate relatives (Vermeij 1978, Palmer 1979) (figure 3.17A, B). Snail shells also are stronger in the Pacific and Indian Oceans than in the Atlantic, on average, potentially reflecting a longer evolutionary arms race between predators and prey in the former, older oceans (Vermeij 1976). Seaweeds tend to be more strongly defended against consumers at low latitudes, often featuring more calcified species, smaller-­statured plants (Gaines and Lubchenco 1982), and better-­developed chemical defenses (Bolser and Hay 1996), all of which are consistent with an evolutionary history of stronger grazing pressure at low latitudes. Second, experiments confirm that consumer pressure is generally stronger in tropical than in colder ­waters across a range of organisms and systems, although t­ here is considerable local variation. Experiments across latitudes first corroborated more intense predation on snails in the tropics (Bertness et al. 1981) and that predation and herbivory generally are stronger on tropical rocky shores (Menge and Lubchenco 1981) and fouling communities (Freestone et al. 2011) than in temperate areas. A series of experiments exploiting the wide latitudinal range of the salt marsh cordgrass Spar-

Chapter 3 Geography of Marine Life

(B) 100 None Weak Strong

80 60

30°N

40

0°

20 30°S 0°

1 0

10°

20°

30° 40° Latitude

50°

60°

(E)

(D)

Snails Crabs Insects All Others

1.0

–1 –2 –3 –4

1.0 Predation intensity

0

(C)

Effect strength

60°N

Predation intensity

Percent of species with strong shells

(A)

0.8 0.6 0.4 0.2

20°

30°

40° Latitude

50°

0.8 0.6 0.4 0.2

35

40

45 50 55 60 Latitude (°N)

65

5

10 15 20 Annual mean SST (°C)

Figure 3.17. ​Latitudinal trends in biotic interactions in the sea. (A) Gastropod shell strength increases t­ oward the equator along the Pacific coast of North and Central Amer­i­ca. (B) The proportion of fish species capable of crushing gastropod shells (dark portion of pie) is greater in the tropics. (C) Negative impacts of ectothermic herbivores on biomass of coastal vegetation are stronger on average at low latitudes. (D, E) Predation on amphipods declines with (D) latitude, but more steeply along western (blue symbols and lines) than eastern ocean margins (green symbols and lines), reflecting basin-­scale differences in (E) latitudinal sea surface temperature (SST) trends ([A, B] a­ fter Palmer 1979; [C] a­ fter He and Silliman 2016; [D, E] ­after Reynolds et al. 2018).

tina alterniflora showed that herbivores inflict more damage on cordgrass in the warm-­temperate than in the cool-­temperate USA (Pennings et al. 2009) and that cordgrass from the warmer region is accordingly more resistant to herbivory (Pennings et al. 2001). A synthesis of experiments on coastal vegetation suggests this result is general: impacts of ectothermic (but not endothermic) herbivores ­were stronger on average at low latitudes (He and Silliman 2016) (figure 3.17C). Such geographic variation in interactions can be split into two components: direct effects of the environment on interactions and indirect effects of variation in the kinds of species pre­sent in dif­fer­ ent regions. In the first category of direct environmental ­drivers, the most obvious candidate to explain latitudinal patterns in interaction strength is the higher temperatures in the tropics. For example, metabolic theory predicts that herbivore impacts on plants should generally increase with temperature based on fundamental cellular pro­cesses: the rate of respiration is generally more sensitive to temperature than the rate of photosynthesis (O’Connor et al. 2011). Therefore, herbivore control of plant biomass should be stronger ­under warmer conditions. Experiments have confirmed this prediction for both benthic (O’Connor 2009) and pelagic systems (O’Connor et al. 2009), demonstrating that moderate warming can dramatically shift food web dynamics and predictably change

67

68

Ocean Ecology

food web structure and productivity. Similarly, comparative mea­sure­ments confirm that predation rate rises with temperature across broad geographic expanses (Reynolds et al. 2018; figure 3.17D, E) and that consumer impacts on prey are stronger ­under warmer conditions for benthic herbivores (Duffy et al. 2015) and the original keystone predator Pisaster ochraceous (Sanford 1999). The second, biotic component of geographic variation in interaction strength is hypothesized to result from the greater diversity of tropical communities (MacArthur 1972). Higher diversity is predicted to strengthen consumer pressure both ­because diverse communities are more likely to contain a particularly effective consumer species and b­ ecause a more diverse consumer community can generally consume a wider variety of prey (Duffy 2002). Both herbivorous fishes and shell-­crushing fishes are more common in the tropics (figure 3.17A, B), contributing to stronger average consumer pressure at low latitudes. But how impacts of predation vary with changing diversity is less clear, in part ­because increasing predator diversity can have opposite effects of increasing prey diversity. In rocky shore communities, predators had less impact on the total biomass of prey communities where prey diversity was higher (Edwards et al. 2010).

Biogeography of the Anthropocene Ocean Biogeography has often been thought of as a static template on which ecological pro­cesses play out. But rapid climate change and widespread human-­mediated transport of species are creating a global biogeographic reor­ga­ni­za­tion that has not been seen at least since the Arctic was last ice-­free more than two million years ago, long before the evolution of modern ­humans (Knies et al. 2014). Many marine taxa are expanding poleward, resulting in tropicalization of temperate ecosystems and fishery stocks, cool-­ water species are penetrating into deeper ­waters, and vari­ous species have begun moving between the North Atlantic and North Pacific Oceans. Th ­ ere is growing concern about the poleward spread of harmful algal blooms (Hallegraeff 2010). ­These changes ­will continue as global temperatures continue to warm and trade globalizes, with strong consequences for the functioning of marine ecosystems.

Climate warming and re­distribution of global marine fauna Pro­cesses that set species range bound­aries constrain which species are found in a local community. Species range bound­aries generally appear to be set by some combination of physiological temperature tolerance at a key life stage and dispersal limitation across unfavorable habitat (Sanford 2014). ­These effects are related in that warmer temperatures increase larval development rate and thus decrease planktonic larval duration (O’Connor et al. 2007). Poleward range expansion with rising temperatures may happen ­either by reducing planktonic larval duration or by increasing the probability of establishment during temporary reversals of mean equatorward flow (Byers and Pringle 2006). The biogeographic shifts of marine organisms with climate warming are now unmistakable, echoing the signal of ocean warming in the temperature and physics of the w ­ ater. A synthesis of more than 1700 changes in the distribution of marine species over recent de­cades found that over 80% supported predicted impacts of climate change, with rates of change tracking shifts in sea surface temperature (Poloczanska et al. 2013). The climate-­related changes involved timing of life history events (phenology), community composition, abundance, and demography. Rates of change in marine systems ­were generally faster than t­hose of terrestrial systems. Th ­ ese biogeographic shifts are most evident among the small organisms low in pelagic food chains, whose drifting habits and short generation times foster rapid responses to climate-­mediated changes in environment. In the eastern North Atlantic and Eu­ro­pean seas, pelagic copepods have shifted strongly poleward, by more than 10° latitude among warm-­water species, and colder-­water species have declined coincident with increasing Northern Hemi­sphere temperatures and the North Atlantic Oscillation (Beaugrand et al.

Chapter 3 Geography of Marine Life

(A)

(B)

Figure 3.18. ​Predicted geography of extinction and invasion in the Anthropocene ocean. Hotspots of (A) local extinction and (B) invasion intensity of 20% or more expected by 2050, averaged across three models u­ nder carbon emissions scenario RCP 8.5 (red = high; blue = low). Hatching shows areas with moderate to high agreement among three models (­after Jones and Cheung 2015).

2002). But changes in distribution and phenology have also been documented in a range of other organisms, from benthic invertebrates to fishes to seabirds (Poloczanska et al. 2013). To expand t­ hese observed results into a predictive global framework, two approaches have been used—­prospective and retrospective. Prospective efforts rely on species distribution models (Pearson and Dawson 2003), also called environmental niche models—­empirical summaries of species niches based on the environmental conditions in their current ranges. Projections of ­future distribution assume that species ­will follow ­these conditions as they shift over time. Applying this approach to over 800 exploited marine species projected average poleward shifts of 15–25 km per de­cade for low-­and high-­carbon emissions scenarios, respectively ( Jones and Cheung 2015). The predicted changes ­were largely consistent among dif­fer­ent models and included extensive invasion at high latitudes and local extinctions concentrated near the equator (figure 3.18).

69

70

Ocean Ecology

The retrospective approach to predicting climate impacts compares ongoing changes to ­those seen during climate change in the past, focusing on the mid-­Pliocene and Last Glacial Maximum (LGM) as periods of high and low global mean temperatures, respectively. Projecting current species distribution models onto past temperature profiles agrees closely with fossil data, validating this approach (Beaugrand et al. 2015). ­These projections support predictions that climate-­related species loss ­w ill be greatest in permanently stratified regions of the ocean and that invasions w ­ ill outpace local extinctions in temperate and polar biomes. The moderate mean global increase of 1.4°C expected by 2065 would t­riple the changes in species distributions observed over the last 50 years. Soberingly, current carbon emissions are tracking the more extreme scenario of a 2.0°C increase by 2065 (according to the “representative concentration pathway,” or RCP 8.5), which is projected to affect marine pelagic diversity and distributions more strongly than the temperature changes of the LGM or the mid-­Pliocene, over an area of 50%–70% of the global ocean (Beaugrand et al. 2015).

Tropicalization Recent climate change has been moving warm-­water marine species into temperate seas, often re­ ecause warm-­water species often have smaller body sizes and “faster” life ferred to as tropicalization. B histories (rapid maturation, higher fecundity), tropicalization has impor­tant implications for marine ecosystems and fisheries (Stergiou 2002). For example, in the Northeast Atlantic, plankton community composition has shifted in recent de­cades to smaller-­bodied copepod species. This community-­ wide shift is expected to weaken the biological pump that transports carbon out of the upper ocean, potentially affecting cod stocks in shelf seas that depend indirectly on the transport of production from overlying surface ­waters to the bottom (Beaugrand et al. 2010). Tropicalization of fishery catch is also increasing. This can be quantified by an index of mean temperature of the catch (MTC), calculated from the average temperature preference of fished species weighted by their catch (Cheung et al. 2013). Across most of the world’s coastal and shelf areas, changes in MTC closely tracked regional changes in sea surface temperature, meaning that more of the catch consisted of warm-­water species. Between 1970 and 2006, MTC increased by 0.19°C per de­cade and even faster outside the tropics. Thus, ocean warming is already affecting global fisheries throughout the world, with likely consequences for economies and food security (Cheung et  al. 2013). Moreover, tropicalization appears to be exacerbated by fishing: fish communities within reserves in Tasmania w ­ ere more resistant to invasion by subtropical species than nearby fished areas, and the communities within reserves showed weaker increases in the average temperature affinity of the community (Bates et al. 2014). The ecosystem consequences of tropicalization are dramatically evident as poleward-­flowing boundary currents have created ocean warming hotspots worldwide, expanding ranges of tropical herbivores. Herbivores often mediate phase shifts between dif­fer­ent states in benthic systems, such as dominance by macroalgae versus coralline algal crusts (chapters 11, 12), and t­ hese invading herbivorous fishes are having strong rippling impacts via a previously unknown phenomenon: in Japan and the Mediterranean Sea, herbivorous fishes are expanding poleward and shifting temperate reefs away from macroalgal dominance and ­toward coral dominance (Vergés et al. 2014) (figure 3.19).

The Arctic opening Perhaps the most dramatic biogeographic change expected with ongoing climate change is a coming trans-­Arctic invasion of the North Atlantic by North Pacific marine organisms. The north polar region has evidently been covered continuously with ice since the mid-­Miocene (Krylov et al. 2008) but is now melting for the first time in 14 My. This Arctic warming is similar to a period of warmer,

Chapter 3 Geography of Marine Life

(A)

(B)

(C)

(D)

Figure 3.19. ​Tropicalization and changing top-­down control of nearshore ecosystems. Underwater photo­ graphs from Tosa Bay, Southern Japan, showing (A) kelp (Ecklonia cava) bed in the early 1990s; (B) E. cava bed in 1997, overgrazed by immigrating herbivorous fishes; (C) barren area in 2000; (D) coral communities growing on the formerly barren areas in 2013 (­after Vergés et al. 2014).

more productive Arctic conditions during the Pliocene when Pacific lineages expanded across the Arctic and into the Atlantic Ocean. Signs of an accelerating trans-­Arctic exchange are already vis­i­ble. The diatom Neodenticula seminae is a dominant primary producer in the North Pacific, and bloomed in 1999 in the Atlantic for the first time in 800,000 years ­after a year of exceptionally ice-­free w ­ ater in the Canadian Arctic (Reid et al. 2007). Marine mammals and seabirds, both highly mobile, are also early bellwethers of the coming invasions. Pacific gray w ­ hales have been spotted in the South Atlantic and Mediterranean beginning in 2010, the first sightings in the Atlantic for over 200 years, and seabirds formerly restricted to ­either Pacific or Atlantic have recently been showing up in their respective opposite locations (McKeon et al. 2015). The melting Arctic is influencing biogeography indirectly as well. Commercial ships are a major mechanism for the transport and introduction of nonnative marine species, which are unintentionally transported in ballast w ­ ater and on ship hulls. Transport of species as a result of trade has substantially increased introductions of nonnative marine species over recent de­cades, and the expansion and growth of shipping trade routes into the Arctic region is likely to accelerate marine invasions among ocean basins (Miller and Ruiz 2014). What are the ecological consequences of ­these trans-­Arctic exchanges? A pos­si­ble example comes from well-­studied mollusks. North Pacific mollusks are much more diverse, and generally larger-­bodied, than their Atlantic counter­parts, and more likely to move through the Arctic into the Atlantic than in the opposite direction; this influx of stronger competitors may substantially change the character of the North Atlantic fauna (Vermeij and Roopnarine 2008). Killer ­whales have recently expanded through ice-­free areas of the Arctic into Hudson Bay, where they have been recorded

71

72

Ocean Ecology

feeding on belugas, bowhead ­whales, narwhals, and at least four species of seals, potentially leading to trophic cascades with far-­reaching effects on the food webs of the region (McKeon et al. 2015).

The sixth mass extinction? Rigorous analyses conclude that land (and sea) use change, harvesting, and pollution are leading to rates of global extinction about 1000 times higher than background rates of extinction over past geologic eras (Pimm et al. 2014), and some have suggested we are facing the sixth mass extinction in Earth’s history (Barnosky et al. 2011). The reasons are multifarious, but we can see several commonalities with pro­cesses discussed in this chapter. First, the well-­documented relationship between area and species richness allows some predictions about ­future trajectories of biodiversity. Modification and transformation of land-­and seascapes ­toward ­human uses continues to fragment and strongly reduce the area of natu­ral habitats, resulting in patchier, less connected collections of small habitat areas. Empirical species-­area relationships predict that diversity declines in such situations and suggest that large, poorly dispersing, and specialist species w ­ ill be most affected. But h­ umans are also increasing diversity in many areas via nonnative species introductions (Elahi et al. 2015), and species from lower latitudes are invading temperate and polar regions. Classical island biogeographic theory predicts that diversity increases strongly with island area and decreases with geographic isolation. In the Anthropocene, geographic isolation is becoming less impor­tant; the more significant driver of colonization is economic isolation from h­ uman populations (Helmus et al. 2015). Large, con­spic­u­ ous marine animals are especially vulnerable to ­human impacts, mainly through overharvesting, and have declined precipitously over the last few hundred years—­and the last few de­cades in par­tic­u­lar, with the advent of industrial fishing and globalized trade ( Jackson et al. 2001, McCauley et al. 2015). ­These declines often cause far-­reaching impacts in ecosystems b­ ecause many such animals play keystone roles disproportionate to their abundance (chapters 8, 11, 12).

­Future Directions Biogeography remains among the most fertile areas for ecological research b­ ecause the time and space scales on which its patterns play out are large and the pro­cesses are thus difficult to constrain. This challenge is well illustrated by the latitudinal diversity gradient: although recognized for over a ­century as the most robust pattern in ecol­ogy (Pianka 1966), debate about its c­ auses continues. But ­there are also new, pressing questions in biogeography. What is the relative importance of current environmental forcing versus history and dispersal limitation to the distribution of species and structure of modern marine ecosystems? Finding the answer is an increasingly pressing issue for understanding global change and its impacts on h­ uman economies. Much of the taxonomic and functional variation among marine regions is historical in that it arose as lineages took dif­fer­ent evolutionary paths in dif­fer­ent regions. ­These differences among regions are maintained by barriers to dispersal among regions, which are falling fast in the Anthropocene. ­Humans have inadvertently set up a ­grand global experiment that may help reveal what proportion of variation in communities among regions is maintained by environmental forcing versus what proportion is due to dispersal limitation (which, when breached, should homogenize the formerly distinct biotas). If the character of modern communities is determined mainly by environmental forcing, the trait distribution and functioning of ecosystems should change l­ ittle a­ fter invasions. If, alternatively, communities are strongly influenced by accidents of history, then invasions and extinctions w ­ ill strongly change the distribution of traits, with potentially pervasive impacts on ecosystems. In practice, ­these alternatives may be difficult to distinguish since biotic and environmental change often go hand in hand. We frequently return to the theme of how to parse the importance of abiotic forcing versus the characteristics of living organisms in shaping ecosystem functioning. We’ve seen, for example, that

Chapter 3 Geography of Marine Life

large-­scale patterns in extinction appear to be more strongly mediated by traits of organisms than by major abiotic forcing f­actors. Functional diversity among organisms strongly affects ecological pro­ cesses and is often comparable in strength to major environmental d­ rivers (Hooper et al. 2012, Duffy et al. 2016). In many trait-­based analyses, “taxonomic group” remains a significant predictor of ecological patterns even ­after environment and major life history traits have been accounted for (Davidson et al. 2012, Finnegan et al. 2015). Statistically, this tells us that additional traits influence ecological patterns but they remain unidentified. Identifying ­these taxonomic group effects and ­whether they can be related to more general theory ­will strengthen the mechanistic basis of functional ecol­ ogy. It also has impor­tant implications for how the biosphere responds to global environmental change. Range bound­aries of species are likely set by tolerances of par­tic­ul­ar life events, such as reproduction and larval recruitment, or during par­tic­ul­ar seasons (Helmuth 2009), suggesting that studies of the physiological tolerances of the key organisms w ­ ill be central. Can we predict winners and losers? Can we intervene? Should we?

Summary The distribution of species in the ocean, as on land, is ultimately determined by climate. But it also bears a per­sis­tent imprint of history in that the geographic distributions of lineages w ­ ere s­ haped by the historical distributions of Earth’s land masses interacting with climate. Long-­term interactions between climate, currents, and land mass configurations help explain the existence of biodiversity hotspots and centers of species origin, notably within the Indo–­West Pacific Coral Triangle. The winds and ocean currents forced by solar heating of a rotating earth maintain a per­sis­tent pattern of relatively discrete provinces in the ocean with distinct physical and chemical properties and accordingly distinct communities of pelagic species. This biogeo­graph­i­cal provincialism is more marked among the benthos, which are more sedentary and thus more strongly influenced by both current environment and history than are pelagic species. The most robust and general feature of biogeography—­Earth’s “first-­order biodiversity pattern”—is the latitudinal diversity gradient: most organisms are most species-­rich in the tropics and least so near the poles. Fossil and molecular ge­ne­tic data suggest that the latitudinal diversity gradient is driven by a combination of high speciation rates in the warm tropics, likely stemming from higher metabolic rates and more intense biological interactions, as well as greater habitat area and greater stability through time in the tropics, which depresses extinction rates. The climate is now changing faster than at any time in the past two million years, and the distributions of species are increasingly following suit. Many marine species are expanding their ranges poleward, into deeper ­waters, and among ocean basins, resulting in tropicalization of temperate biotas and projected hotspots of invasion in polar seas. Smaller average body sizes of tropicalizing plankton ­will likely reduce the strength of the biological pump, and expansion of herbivorous fishes into temperate seas is already changing the character of coastal benthic ecosystems. Movement of organisms as a by-­product of h­ uman commerce is also changing distributions of species and their impacts on marine ecosystems. Th ­ ese several h­ uman impacts have initiated a global experiment in the reor­ga­ni­za­tion of biodiversity that ­will ultimately shed light on the relative roles of environmental forcing and intrinsic biological traits in the structure and function of ecosystems. It ­will also have far-­reaching, and still largely unpredictable, consequences for humanity’s dependence on ocean ecosystem ser­vices.

73

4

Introduction to the Anthropocene Ocean Environmental change on earth is as old as the planet itself, about 4 billion years. Our genus, Homo, has altered earthly environments throughout our ­career, about 4 million years. But ­there has never been anything like the twentieth c­ entury. J. R. McNeill (2001)

T

hree millennia ago, at the dawn of civilization, an already jaded King Solomon proclaimed that ­there is nothing new u­ nder the sun. In an impor­tant sense he was wrong—in l­ ittle more than the last ­century, surging ­human population and rapidly evolving technology have profoundly transformed this unique planet, changing even the climate itself, prob­ably irreversibly (McNeill 2001). This unpre­ce­dented boom, which raised h­ uman living standards immeasurably, has been powered largely by a one-­time inheritance from Earth’s early vegetation, buried in colossal quantities before herbivores evolved the ability to eat it, and metamorphosed over eons into coal and petroleum. The costs of humanity’s dependence on fossil fuel—­externalities in the lingo of economists—­ are now soberingly clear. Our numbers, and especially our appetites, continue to grow as the developing world naturally strives to become, in Thomas Friedman’s words, “carbon copies” of Americans in the sense of reaching the high standards of living that fossil fuels have brought to the industrialized world. Happily, ­those ­human development goals have been spectacularly successful (figure 4.1). But sustaining them at even our current rate of per capita energy consumption would require the natu­ral resources of 1.7 earths (Lin et al. 2018). We only have one. Over the long course of earth history, several major evolutionary events fundamentally transformed the biosphere and the earth ecosystem (Szathmary and Maynard Smith 1995): the evolution of eukaryotic cells, the origin of photosynthesis and consequent oxygenation of the atmosphere and ocean, the evolution of multicellularity, the evolution of animal social life, and the rise of Homo sapiens. In this chapter we focus on the last of ­these, a transformation that emerged largely within the span of recorded history and is very much in pro­gress: the age of ­humans, now known as the Anthropocene epoch. Across ­those four billion years, the Anthropocene is the only geological period named for a single species. The reason for this dubious distinction is that within the two centuries since the Industrial Revolution—­the blink of an eye in geologic time—­our species has so pervasively altered the atmosphere, hydrosphere, and biosphere that humanity can be considered the major force of nature. In this chapter we review ­these events, their manifestations in the modern ocean, and some consequences for management and conservation.

First, the Good News

74

We live in a golden age. It may not seem so from the disquieting news that bombards us on a daily basis from all directions, but the well-­being of the average person on Earth is indescribably better than it was a few centuries ago when most ­people woke each morning looking in dread over their shoulder for the four ­horse­men of the apocalypse—­death, famine, war, and pestilence. During the twentieth ­century, average life expectancy more than doubled worldwide, demo­cratic governments steadily replaced autocratic ones, undernourishment in the developing world had declined by roughly half, and despite two horrific world wars and many smaller ones, war had actually declined over

Chapter 4 Introduction to the Anthropocene Ocean

(A)

(B)

70 60

Americas Europe World Asia Africa

4 Democracy score

Average life expectancy

80

50 40 30 20 1760

0 –2 –4 –6

1800

1840

1880

1920

1960

–8 1800

2000

(C)

1840

1880

1920

1960

2000

1600

1700

1800

1900

2000

(D) 100 Major wars (percent of years)

35 Percent undernourishment in developing world

2

30 25 20 15 10 5 0 1970

Developing world Latin America Southeast Asia South Asia East Asia Sub-Saharan Africa 1980

1990

2000

2010

75

50

25

0 1500

Figure 4.1. ​On average, ­people everywhere are better off now than ever before (­after Pinker 2018).

recent centuries (see figure 4.1). All of ­these statistics have bounced up and down, but the long-­term trends are unmistakable and well documented (Pinker 2018). We have dodged the bullet so far, and risen above many predicted catastrophes. But this dizzying pace of advancement unfortunately has not been matched in the nonhuman world, and that is a dangerous blind spot for modern civilization. ­There is a hard limit, defined by the laws of thermodynamics, that we cannot breach; and we are getting closer to it.

The ­Great Acceleration ­ umans have modified Earth’s habitats and environments from surprisingly early in our history, even H at low population densities with primitive technology. The first major increase in global population began approximately 10,000 years ago with the domestication of plants and animals during the Neolithic Revolution and subsequent spread of agriculture, accompanied by widespread clearance of forests, cultivation, and steady innovation in technology (figure 4.2). Even at this early stage, increasing atmospheric concentrations of the green­house gases CO2 and CH4 ­were anomalously higher than during comparable interglacial periods (Ruddiman 2013) (figure 4.3). But the Anthropocene is not simply about strong or novel ­human impact. Like other geological epochs, the designation is based on a signature of widespread, rapid change in the earth system as a w ­ hole (figure 4.4). The Anthropocene was first proposed as beginning with the Industrial Revolution in the late eigh­teenth ­century when the first fossil fuel, coal, was mobilized as a source of power on an industrial scale (Crutzen

75

76

Ocean Ecology

2002). Interdisciplinary analy­sis now recognizes its start in the mid-­twentieth ­century (­Waters et al. 2016), marked unambigu6 ously in the geologic rec­ord by the advent of the nuclear era, beginning with the first atomic bomb test at Alamagordo in 5 New Mexico, USA, in 1945 and continuing detonations over Steam engine: 300 the next de­cade that dispersed a global, permanent stratigraphic Windmill: 1200 4 Gunpowder: 1200 signature of radioactivity. Other distinct markers of the AnthroWater mills: 2300 Roads: 2500 pocene also accumulated rapidly in the geologic rec­ord, includMetal plows: 2500 3 ing fossil fuel combustion products (black carbon), plastics, Bronze: 5000 Irrigation: 5000 2 concrete, and widespread bones of the domestic chicken. Any Cities: 5700 Wheel: 6000 such discrete boundary in time is somewhat arbitrary, and recBow & arrow: 10000 Animal domestication: 10000 1 ognition of the Anthropocene’s dawn in the mid-­twentieth ­century glosses over major h­ uman impacts over tens of thousands of 0 years. Th ­ ese include not only the Industrial Revolution of the 10000 8000 6000 4000 2000 0 eighteenth–­nineteenth c­ entury but also the mass extinctions Years before present of large animals on all continents in the wake of ­human colonization, which began as long as 40,000 years ago. The latter offer Figure 4.2. ​Trends in global h­ uman population and innovation a cautionary tale in the context of modern fishing, to which we over the last 10,000 years. Population and the major innovations ­will return. that helped expand ­human carry­ing capacity are divided into The abrupt change in the earth system recognized as the Anthree time periods (­after Nekola et al. 2013). thropocene stems from a series of technological advances that gathered steam with the Industrial Revolution and crossed a threshold around 1950, sparking a chain reaction in rising global economic activity, resource consumption, and population growth dubbed the ­Great Acceleration (Steffen et al. 2007). This transition is clear in a wide range of indicators of both the earth system and ­human socio­economics, from atmospheric chemistry to foreign investment (see figure 4.4). A central result of t­ hese developments is 7

Global population (billions)

Green revolution: 40 years ago Computer: 60 Antibiotics: 70 Plastics: 90 Airplane: 100 Internal combustion engine: 150 Vaccine: 200

(A)

Early Holocene

Mid-Holocene

(B)

Late Holocene 200

Greenland ice core NO3- (ppb)

Anthropocene concrete (100 Tg)

100

plastic (Tg)

200 0

0

Temperature anomaly (°C)

400

1.2 0

(C)

–1.2

280

Carbon dioxide (ppm)

300 200 1600 1100

Methane (ppb)

CO2 (ppm)

400

260

240

600

12

10

8 6 Age (years x 103)

4

2

0

10,000

5,000 Age (years)

0

Figure 4.3. ​Major markers of ­human impacts on the earth system since the end of the Pleistocene ice ages. (A) Trends in atmospheric temperature and green­house gas concentrations and (B) in the last 50 years, with cumulative production of concrete and annual plastic production (­after ­Waters et al. 2016). Note the dif­fer­ent vertical scales. (C) Observed concentrations of carbon dioxide (green symbols) compared with modeled expectations in the absence of ­human activities (blue symbols and lines) (­after Ruddiman 2013).

Chapter 4 Introduction to the Anthropocene Ocean

8.0

Ocean acidification

60 Trillion US dollars

Hydrogen ion, nmol kg–1

70

7.5

7.0

Real GDP

50 40 30 20 10

6.5 1750

1800

1850

1900

1950

0 1750

2000

1800

1850

1900

1950

2010

2010 4 Nitrogen to coastal zone

Urban population 3

60

Billion

Human N flux Mtons yr–1

100 80

40

2

1

20 0 1750

1800

1850

1900

1950

0 1750

2000

1800

1850

1900

1950

4

600 Shrimp aquaculture

500 Exajoule (EJ)

Million tonnes

2000 2010

2010

3

2000

2

Primary energy use

400 300 200

1 100 0 1750

1800

1850

1900

1950

2000

0 1750

1800

1850

1900

2010

Figure 4.4. ​Trends in selected indicators of earth system status (blue) and h­ uman socioeconomic development (orange), 1750 to 2010. The dashed vertical lines mark the proposed beginning of the Anthropocene epoch in 1950 (­after Steffen et al. 2015).

the accumulation of over 7 billion ­people sharing the earth ­today, more than double the number 50 years ago, with 9–11 billion projected by c­ entury’s end. And ­those numbers tell only part of the story: the ­Great Acceleration delivered a quantum leap in humanity’s average standard of living as well as our global impact, colloquially known as our ecological footprint. The metabolism of this booming industrial civilization, powered by fossil carbon since the beginning, has profound consequences for all life on the earth and in the ocean.

1950

2000 2010

77

78

Ocean Ecology

Coal and climate change

Proxy global temerature (ºC)

Antarctic composite (σ)

CO2 (p.p.m.v.)

Carbon flux (Gt year–1)

Several key developments catalyzed the ­Great Acceleration. The most fundamental was the shift from animal (largely h­ uman) power to fossil carbon as a source of energy for ­human industry (Steffen et al. 2015). The symbiotic nature of technological innovation and fossil fuel use is illustrated by the initial development of steam engines in the seventeenth and eigh­teenth centuries, which was motivated by the need to drain groundwater out of deepening coal mines. Rapid technological improvement soon spread coal-­fueled steam power to multifarious applications, launching the Industrial Revolution. Most of the disparate impacts of ­human society on nature ultimately stem from this adoption of fossil fuel. The surge in combustion of coal, and l­ater petroleum oil and natu­ral gas, took large reservoirs of carbon previously sequestered in the ground, put them to work, and dumped the waste carbon in the atmosphere, destabilizing the rough steady state of the global carbon cycle that emerged over the previous ten millennia durEmissions from fossil fuel 10 Net release from land-use change ing which ­human culture evolved. The CO2 released by fossil Unidentified sink 8 Atmospheric accumulation fuel combustion accumulated in the atmosphere faster than it 6 Oceanic uptake could be removed by photosynthesis, diffusion into the ocean, 4 or geochemical pro­cessing (figure 4.5). And b­ ecause CO2 Releases 2 gas absorbs heat, its rising concentration increased the heat-­ 0 trapping capacity of the atmosphere, a phenomenon known as –2 Accumulations the green­house effect. Chemical analy­sis of air trapped in Antarc–4 tic ice cores shows that atmospheric CO2 concentration and –6 temperature have covaried closely for at least 800,000 years, and –8 that rising CO2 generally preceded rising temperature, consis–10 tent with theory and laboratory evidence that the green­house 1850 1865 1880 1895 1910 1925 1940 1955 1970 1985 2000 effect of CO2 gas increases atmospheric temperature (Lüthi et al. 2008, Shakun et al. 2012) (figure 4.6). Atmospheric CO2 is now Figure 4.5. ​A summary of Earth’s carbon bud­get, and the over 400 ppm, likely higher than at any time in the last 20 million ocean’s role in it, over the last 150 years (­after Normile 2009). years, and global temperatures are following suit. The Intergovernmental Panel on Climate Change (IPCC) reckons that, in the Northern Hemi­sphere, the period 1983–2012 was the warmest 30-­year period in the last 1400 years (Stocker et al. 2013). Much 0 260 of that heat is absorbed by the ocean, a point to which we return 1 below. As a result, over the last c­ entury or so the mean global 240 –1 atmospheric and sea surface temperatures have increased by 0 220 –2 0.85°C and 0.6°C, respectively (figure 4.7). And t­here is ­little reason to expect they ­will slow down in coming de­cades. 200 –3 –1 Over the same period, warming has reduced Arctic sea ice by 40%. An open passage through the frozen Canadian Arctic Holocene 180 –4 appeared in summer 2008 for the first time in prob­ably at least 22 20 18 16 14 12 10 8 6000 years, and has opened each summer since. A ­ fter five cenAge (years × 103) turies of imperial intrigue, naval ­battles, and rueful ends to frostbitten voyages, the elusive Northwest Passage appears to Figure 4.6. ​Green­house gas forcing of climate in ancient and be open for business as a regular summer feature. The melting modern times. Rising CO2 concentration, estimated from of polar ice caps and glaciers, together with associated thermal Antarctic ice cores (red), precedes global climate warming (blue) expansion of ocean ­water, have increased sea level by ~14 cm through the late Pleistocene deglaciation and the Holocene. The climate curve (blue) is based on global proxy temperature over the twentieth c­ entury, a rate that is “extremely likely” to indicators as deviations from the early Holocene mean. The CO2 be faster than in any of the preceding 27 centuries (Kopp et al. rec­ord (red) is based on Antarctic ice-­core composite 2016). ­These effects vary considerably among regions, howtemperature rec­ord. Yellow dots show atmospheric CO2 concentrations (­after Shakun et al. 2012). ever (figure 4.8).

Chapter 4 Introduction to the Anthropocene Ocean

1.0

Temperature anomaly (ºC)

Nitrogen: Detonator of the population explosion

0.5

79

Air temperature

0.0 –0.5

Extent (105 km2)

Sea level anomaly (mm)

Temperature anomaly (ºC)

Temperature anomaly (ºC)

A second key trigger of the Anthropocene, arguably of compa–1.0 rable significance to the mobilization of fossil fuel, came in the 0.4 Sea surface temperature early twentieth ­century when the German chemists Fritz 0.2 Haber and Carl Bosch in­ven­ted a pro­cess to fix elemental N2 0.0 gas into ammonia and thereby changed the world. For most of –0.2 –0.4 ­human history to that point, production of crops (and of all –0.6 other plants) had been ­limited by ­water, the quality of the soil, 0.4 Marine air temperature and availability of what­ever nutrient source—­dung, compost, 0.2 0.0 seaweed, fish—­could be scrounged locally. ­After the Indus–0.2 trial Revolution, ­these organic fertilizers ­were supplemented –0.4 by mined potassium nitrate (known as saltpeter), but this –0.6 100 nonrenewable source dwindled rapidly. Suddenly, with the inSea level 50 dustrialization of the Haber-­Bosch pro­cess, humanity gained 0 –50 access to the essentially unlimited tap of nitrogen that makes –100 up 79% of Earth’s atmosphere. As a result, the efficiency of –150 –200 crop production qua­dru­pled, also benefiting from the inven12 tion of organic pesticides, and both food production and 10 ­human population growth followed suit. The Haber-­Bosch 8 pro­cess has been called the detonator of the population explo6 sion: “Of all the c­ entury’s technological marvels, the Haber–­ Sea ice extent (arctic summer) 4 Bosch pro­cess has made the most difference to our survival” 1850 1900 1950 2000 (Smil 1999). Smil estimated that nearly 40% of ­people alive ­today owe our existence to the Haber-­Bosch pro­cess. Figure 4.7. ​Climate change and the ocean. Multiple The impact of industrial nitrogen fertilization on the earth complementary indicators of a changing global climate, from system is mainly mediated indirectly, by the sharp increase in the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Colors represent dif­fer­ent data sets (­after ­human numbers it made pos­si­ble and all the downstream conStocker et al. 2013). sequences of that population explosion for resource demand, habitat modification, and waste production. Notably, N-­fueled ­human population growth drove a massive increase in conversion of natu­ral ecosystems to managed crop and grazing lands. The result has been a rise in ­human appropriation of global net primary production from 13% to 25% (Krausmann et al. 2013). A more direct impact of the revved-up nitrogen cycle stems from the flood of nitrogen into ecosystems where production was formerly l­imited by this ele­ ment, including much of the world’s fresh and marine ­waters. ­Today h­ uman activity adds as much or more nitrogen to the land as do all natu­ral sources combined (Vitousek et al. 1997). Much of the nitrogen added as fertilizer misses its target crops and runs off the land into streams and rivers, the total nitrogen concentrations of which are strongly correlated with inputs of both fertilizer and atmospheric nitrogen from fossil fuel combustion in the watershed (figure 4.9). This loading of new nitrogen has greatly increased biological productivity of coastal ­waters—­a pro­cess known as cultural eutrophication—­often by species that ­humans ­don’t appreciate, such as toxic algae, and thereby transformed the ecol­ogy of estuaries and coastal oceans. We consider in more detail the general ecological importance of nitrogen, and the consequences of anthropogenic nitrogen addition, in chapter 11. The miracle in agricultural productivity launched by the Haber-­Bosch pro­cess also has a dark side, requiring substantial costs in fossil fuel input, soil degradation, ecosystem conversion to managed and human-­built landscapes, and associated loss of biodiversity. Indeed, two centuries a­ fter the Industrial Revolution and one ­century ­after the detonation of the population explosion, the accelerating impacts of humanity on the earth system are so pervasive that some experts believe we

80

Ocean Ecology

–0.8

–0.5

–0.4

–0.3

–0.2

–0.1

0.1

0.3

0.5

0.7

1.0

4.8

Figure 4.8. ​Hot ­water. Distribution of sea surface temperature anomalies in 2001–2005, relative to historical baseline (1870–1900) (­after Hansen et al. 2006).

are approaching planetary bound­aries—­thresholds in the stocks of resources and rates of change beyond which h­ uman impacts are likely to irreversibly destabilize the earth system. They argue that bound­ aries for biodiversity loss and nitrogen inputs may have already been crossed (Rockström et al. 2009).

Total nitrogen flux (kg N km–2 yr–1)

(A)

Total nitrogen flux (kg N km–2 yr–1)

(B)

The limits to growth

8,000 6,000 4,000 2,000 0

0

5,000 10,000 15,000 Fertilizer inputs (kg N km–2 yr–1)

20,000

2,000

1,000

0

0

200

400

600

800

1,000

NOy deposition (kg N km–2 yr–1)

Figure 4.9. ​Synthetic nitrogen, detonator of the population bomb. The panels show flux of total nitrogen to the ocean via rivers as a function of (A) fertilizer inputs to the watersheds upstream and (B) atmospheric deposition, largely from fossil fuel combustion. Upper and lower curves in B are for the US and Sweden, respectively (­after Howarth et al. 2012).

In 1968 a group of 30 individuals calling themselves the Club of Rome commissioned an analy­sis of “the predicament of mankind.” The challenge was taken up by a group of scientists in the new field of systems dynamics at the Mas­sa­chu­setts Institute of Technology, and the result, published in 1972, was The Limits to Growth (Meadows et  al.). The book channeled the era’s growing environmental consciousness—­and dread—­and was pioneering in several ways (Turner 2008). Based on a computer model known as World3, the analy­sis was the first application of system dynamics modeling and scenario analy­ sis to environmental issues, approaches that are now routine in exploring policy options. It was also the first model to integrate the global economy with the biophysical environment that supports it, including feedbacks between systems and lags between forcing and response. The proj­ect thereby initiated the field of ecological economics. The World3 model simulated interactions among five components of the global system: ­human population, food production, industrial production, pollution, and consumption of nonrenewable resources. It examined three scenarios: the standard run (business as usual) continued then current trends; the comprehensive technology

Chapter 4 Introduction to the Anthropocene Ocean

scenario simulated a world with high recycling rates, low pollution, a doubled agricultural yield, and universal availability of birth control; and the stabilized world scenario added social policies intended to achieve long-­term sustainability, notably universal birth control and two-­child families, as well as preferences for ser­v ices over material goods and more vigorous maintenance of agricultural land. The central result of this analy­sis was that continuation of current trends in economic growth—­ the business-­as-­usual scenario—­would exceed the limits of the global environment during the twenty-­first ­century, likely causing a collapse of the world population and the economic system. The business-­as-­usual simulations for the early twenty-­first ­century showed resources and food diminishing, economic activity and wealth declining, and pollution increasing (figure 4.10). But this result

Population

1.0

0.9

0.9

0.8

0.8

0.7

0.7

0.6 0.5 0.4 0.3

0.6 0.5 0.4 0.3

0.2

0.2

0.1

0.1

0.0 1900

1940

1980

(C)

2020

2060

0.0 1900

2100

0.9

0.9

0.8

0.8

0.7

0.7

0.6 0.5 0.4 0.3

0.3

0.1 2060

2100

2100

0.4

0.1 2020

2060

0.5

0.2

1980

2020

0.6

0.2

1940

1980

Global pollution 1.0

0.0 1900

1940

(D)

Nonrenewable resources 1.0

Normalized value

Normalized value

Food per capita

(B)

1.0

Normalized value

Normalized value

(A)

0.0 1900

1940

1980

2020

2060

2100

Stabilized world Comprehensive tech’y Standard run LtG history Observed data

Figure 4.10. ​Comparison between predictions of the 1972 Limits to Growth earth system model and observed trends in the global economic and environmental indicators. The colored curves represent scenarios produced by models with dif­fer­ent policy assumptions about technology investment, environmental protection, and population growth (LtG history = period prior to 1972 publication of Limits to Growth) (­after Turner 2008).

81

82

Ocean Ecology

was not inevitable: the other model scenarios suggested that collapse could be delayed and even prevented if changes in technology, policy, and ­human be­hav­ior ­were implemented soon. The Limits to Growth report was translated into 30 languages and widely discussed. Not surprisingly, it was highly controversial. It was also widely (and often intentionally) misinterpreted, previewing a strategy that would prove increasingly common and, unfortunately, effective in public discourse about challenges to the status quo, including health effects of tobacco, evolution, and most recently climate change (Oreskes and Conway 2011). Was this groundbreaking analy­sis wrong, as widely reported? Four de­cades ­after publication we have enough data to evaluate several of the World3 model’s predictions (Turner 2008). The most straightforward test is for world population, which is mea­ sured regularly with considerable effort and good precision. Since 1970 both global birth and death rates have fallen steeply, with the net result of a still increasing trend in global population that closely matches the business-­as-­usual and comprehensive technology scenarios but is well above that envisioned u­ nder a stabilized world (see figure 4.10). World agricultural production is also mea­sured reasonably accurately, so it can be compared with the predicted trends in food availability. The data show that per capita food supply (total energy per person) has continued to increase at a low rate, following predictions of the business-­as-­usual scenario, and well below the food production output envisioned by the comprehensive technology and stabilized world scenarios (see figure 4.10), both of which allocate more resources to pollution control and enhanced agricultural productivity. Trends in nonrenewable resources have been evaluated for fossil fuels, which are judged to provide global limits on industrial productivity. Evaluating trends in this energy base is less straightforward than for food production b­ ecause it depends on assumptions about ­whether and how currently inaccessible energy sources, such as nonconventional oil and methane hydrates, are eventually developed. The analy­sis brackets this uncertainty by estimating upper and lower bounds for likely fossil fuel development, the upper bound assuming that even technologically challenging and remote resources are exploited. Trends in nonrenewable (fuel) resources ­under t­ hese assumptions bound the range of scenarios modeled by World3 and suggest that fuel extraction w ­ ill not divert capital from agriculture and industry ­until at least 2030. In such a scenario where energy remains effectively superabundant, “the prob­lem of the acquisition of energy is replaced by the prob­lem of its dissipation” (Hardin 1968)—in other words, pollution. This leaves the predictions for per­sis­tent pollution. Pollutants are diverse, including heavy metals, organic compounds, radioactive materials, and more, most of which lack sufficient data to estimate global inventories. The analy­sis therefore focuses on atmospheric CO2 as a per­sis­tent substance with detailed global data. The trend in atmospheric CO2 since 1970 follows the business-­as-­usual scenario closely and is well above that of the other scenarios (see figure 4.10). The bottom line of the Limits to Growth analy­sis is that the world has proceeded with business as usual since 1970, as the data since then generally match predictions of the standard run model much more closely than scenarios assuming major technological advances or implementation of progressive social policies (Turner 2008). The latter, more optimistic scenarios overestimated food, ser­v ices, and provision of material goods to the world population. And the level of pollution projected by World3’s business-­as-­usual model is roughly consistent with the IPCC’s projections for green­house gas emissions at the mid-­t wenty-­first ­century. It should be emphasized that the World3 model used highly aggregated global data and was intended more to explore the interactions among sectors of the world system than to make precise predictions. Nevertheless, the concordance between the business-­as-­usual projections and the historical data accumulated since then suggests that the model got the general dynamics right. It emphasizes that avoiding the predicted overshoot and collapse that emerges from current trends requires more than cosmetic changes to business as usual.

Chapter 4 Introduction to the Anthropocene Ocean

The Natu­ral and Cultural History of Homo Sapiens To understand how we have come to this extraordinary juncture, some history—­both natu­ral and conventional—is in order. All ecol­ogy begins with natu­ral history, the detailed scrutiny of the habitats and interactions of organisms in their environment. Understanding a planet dominated by one species naturally entails a detailed focus on the natu­ral history of that species. A few hundred thousand years ago, Homo sapiens would have seemed unremarkable: just one of the handful of large, omnivorous mammal species, specifically primates, roaming the forests and savannas of Africa. We shared the world at that time with prob­ably several other kinds of h­ umans, meaning contemporaneous species in the genus Homo, a fact that has been widely recognized only recently (Schwartz and Tattersall 2010, Caspari and Wolpoff 2013). Like our close relatives, early Homo sapiens used s­ imple stone tools, which ­were evidently the limit of our technological repertoire. By modern standards, early Homo sapiens had a subsistence economy, gathering just enough resources to survive and reproduce. Available evidence suggests that h­ umans lived in small bands, consisting mainly of extended families, hunting and gathering a varied diet, and the global population was relatively small. Analy­sis of gene coalescents suggests that at two points in our history, about 50,000 and 20,000 years ago, each of the main lineages of the Homo sapiens population in Africa, Eu­rope, and Asia declined to effective population sizes of only a few thousand individuals (Li and Durbin 2011), which might have brought us perilously close to global extinction. It’s an in­ter­est­ing exercise to contemplate the quite dif­fer­ent course that earth history nearly took at that fork in the road. But it d­ idn’t. Instead, the ­human population pulled through and grew at near exponential rates as ­humans migrated out of our African homeland and spread first into Eu­rope and Asia, then Australia, and fi­nally the Amer­i­cas and the remotest islands of Oceania. The ­human population now numbers over seven billion and covers the Earth. Why? Although we have many features of biology and be­hav­ior in common with our close relatives, like chimpanzees (with whom we share fully 96% of our genes), the h­ uman species is obviously unique. Modern h­ umans have achieved o­ rders of magnitude greater abundance and interaction strength than expected from a mammal of comparable size, essentially replacing the biomass of other large mammals, most of which we ­either ate or outcompeted for habitat (Barnosky 2008). Our geographic range and realized niche cover virtually ­every conceivable habitat on e­ very continent. Much of this c­ areer stems from an intimate two-­way interaction with technology—an extreme example of the extended phenotype (Dawkins 1982)—­ that has no parallel elsewhere in the animal kingdom. From stone tools and fire, we progressed at an accelerating rate to metallurgy, architecture, steam power, electricity, the internal combustion engine, antibiotics, the internet, and artificial intelligence. Understanding our place in the world requires considering both the evolutionary heritage we share with other animals and the unique developments of ­human natu­ral history. We are first a product of our evolutionary history, like all organisms, and our explosion to dominate the earth can be better understood in that evolutionary context. Population growth in ­humans, as in other organisms, results from the general tendency to increase exponentially ­until the population approaches environmental limits, combined with our Darwinian tendency to adapt to the environment to push back ­those limits and keep growing (Nekola et al. 2013). Where Homo sapiens departs from other organisms is in the central role of cultural mechanisms in that pro­cess of adaptation, which has vastly sped up adaptation compared with what would be pos­si­ble by ge­ne­tic evolution alone. Culture, encompassing the combination of our social norms and technology, has driven our exponential expansion and dominance of Earth via a positive feedback between population growth and innovation. Thus, Earth’s multifaceted transformation ­under ­human influence and its feedback to quality of life and economic activity illustrate that the biosphere, with humanity embedded within it, constitutes

83

84

Ocean Ecology

a complex adaptive system. The growth of ­human populations and our coalescence into larger, more complex social groups fostered broad sharing of information and resources that stimulated technological evolution, which in turn allowed ­humans to continually push back the ecological limits faced by a naked ape without technology. We have already discussed how fossil fuel and industrial nitrogen fixation pushed back our ecological limits by o­ rders of magnitude. The corresponding increase in collective brainpower of the large and connected h­ uman population further accelerated innovation in a positive feedback loop. This feedback is strikingly evident in the “superlinear” scaling (power law with β > 1) of numerous indices of social interaction and innovation—­including numbers of inventors, patents, and employment in private research and development—­w ith population size in cities (Bettencourt et al. 2007) (see figure 4.4). The result of this coupled technological innovation and population growth is our modern technologically based, globally interconnected civilization. In the context of ocean ecol­ogy, the major innovations include deep-­ocean mining, large petroleum-­fueled ships, refrigeration, sophisticated electronic fish-­finding, communication and trade networks, and of course the fossil fuel and nitrogen booms that are transforming ocean biogeochemistry. Although ­humans are unique and our impacts far exceed t­ hose of any other species, we nevertheless remain subject to the same laws of physics and biology and are ultimately l­imited by energy and other resources (Brown et al. 2011), key facts that have often been overlooked or willfully ignored during the heady boom of the late twentieth ­century. And we conform to many macroecological patterns similar to t­ hose of the rest of the living world. Most fundamentally, economic activity increases with energy use, mostly from fossil fuels, just as production and population growth increase with energy use among other species (Brown et al. 2011). This is evident in the close correlation between per capita energy use and per capita gross domestic product (GDP), both among countries and within countries through time (figure 4.11), and similar relationships between energy and many other indicators of socioeconomic status and ecological impact (Brown et al. 2011). ­These patterns illustrate that Homo sapiens has evolved a massive per capita interaction strength via constant technological innovation. In synergy with growing population density, this has created the unpre­ce­dented impact of our species on the earth system, which defines the Anthropocene. The pervasive, intertwined connections between the h­ uman economy and the biophysical world emphasize that the ­human sciences—­psy­chol­ogy, sociology, economics—­are now essential components of ecol­ogy. Integrating them requires more than simply recognizing h­ uman activity as an external driver of ecological state. It requires integrating h­ umans and our complex be­hav­ior and culture as key players in ecological networks, connected by multiple direct and indirect paths to other species and environmental f­ actors.

Ecol­ogy for the Anthropocene Culture and the evolution of ­human society The most defining feature of Homo sapiens natu­ral history is our ultrasociality, the tendency to form large, complex social groups that extend well beyond ge­ne­tic kin (Ellis 2015). This is unknown among other species and central to nearly every­thing unique about modern ­humans. Sociality is key to the genesis of culture and industry, an evolutionary phenomenon arising from the mutation and exchange of ideas rather than genes, and proceeding at far faster rates than ge­ne­tic evolution. Sociology and, more specifically, politics comprise the science and practice of ­human ultrasocial be­hav­ior. Politics might be considered the extension of what in other species is called behavioral ecol­ogy to the level of the large nonkin groups that are uniquely ­human. Its focus is how ­people associate into groups, behave, and interact with one another. The daily news is full of examples of how politics, from village to international scales, profoundly influences our interactions and impacts on the natu­ral

Chapter 4 Introduction to the Anthropocene Ocean

(A)

(B)

(C)

Infant mortality per 100k

2.5

3.5

10,000

Calories

Mean per capita total power consumption (watts)

3.6

1,000

3.4 3.3 3.2

2.0 1.5 1.0 0.5

100 3.1 100

1,000

0.0

10,000

Mean per capita GDP (constant US$, 2000) (D)

Log per capita energy consumption (E)

Log per capita energy consumption (F) 3.0

1

1.6

1.2

0.8

2.8

0

Waste (kg)

Nobel Prizes per mil

National poverty (%)

2.0

–1

–2

Log per capita energy consumption

2.4 2.2

–3

0.4

2.6

2.0

Log per capita energy consumption

Log per capita energy consumption

Figure 4.11. ​The metabolic ecol­ogy of ­human civilization. (A) ­Human economic activity (per capita gross domestic product [GDP]) is closely related to per capita energy use across countries (­after Brown et al. 2014). [B–­F] Socioeconomic status and indicators of standard of living are strongly correlated with energy use (­after Brown et al. 2011).

world. It is thus a key facet of the complex adaptive nature of the global ecosystem with ­humans as a central player—­what academic jargon calls the coupled human-­natural system. The rise to global dominance by h­ umans has also involved several phenomena familiar from general ecol­ogy (Ellis 2015). ­These include ecosystem engineering, by which ­humans modify the environment intentionally and unintentionally in ways that last for generations. Industrial agriculture, domestication of waterways, and especially the growth of cities are examples. Th ­ ese modifications can have long-­term evolutionary consequences for our own fitness and that of associated organisms. That is, they involve niche construction, in which choices of habitats and resources define an organism’s living conditions such that its activities modify its own selective regime and influence subsequent evolution (Odling-­Smee et al. 1996). For example, molar size declined sharply in the lineage leading to Homo sapiens, evidently as an evolutionary response to softer foods produced by cooking and other food pro­cessing ­adopted at least 2.6 million years ago (Organ et al. 2011). More recently, population ge­ne­tics show that ­human lactose tolerance is clustered in the same geographic regions where Neolithic Eu­ro­pe­ans established c­ attle farms over 5000 years ago (Beja-­Pereira et al. 2003). The history of ­human accomplishment depended intimately on the increased capacities of social cooperation, and this sociocultural niche construction became the central pro­cess by which ­humans transformed the earth system (Ellis 2015).

85

86

Ocean Ecology

Energetics and economics of Homo sapiens A mechanistic ecol­ogy of the Anthropocene starts with a bottom-up focus on resources. The central resource for ­human society, serving as a proxy for all o­ thers and as an ecological currency, is energy. We’ve seen that economic indicators are tightly related to energy use (see figure 4.11), which implies that the broad scope of h­ uman activity and achievement are l­ imited by the hard biophysical constraint of resource availability. This link also emphasizes that society’s feverish rate of energy consumption is mismatched to the finite stock of fossil fuel that powers it. The earth system has been likened to a chemical battery, charged over more than three billion years by conversion of solar energy into plant biomass via photosynthesis (Schramski et al. 2015). Over that period much of the plant biomass was sequestered as a vast reserve of fossil fuel. In the short centuries since the Industrial Revolution, ­humans have aggressively mined that reserve to build our modern society of more than seven billion ­people, in the pro­cess radically transforming the biogeochemistry and biodiversity of the earth. Both the long, gradual charge of the battery and its much more rapid discharge follow the universal—­and inviolable—­laws of thermodynamics. In princi­ple, solar energy provides a renewable source many times larger than the fossil reservoir we are rapidly draining, but we have not learned how to tap more than a fraction of it. U ­ ntil we do, fossil fuel and the wastes (including atmospheric CO2) produced by exploiting it set hard physical limits on h­ uman society’s long-­term dynamics and well-­being, as envisioned in the Limits to Growth analy­sis and corroborated by data (see figure 4.10). The uniqueness of ­humans takes us a bit outside the box of what is normally considered ecol­ogy. Given the prominence of culture in the rapid rise of Homo sapiens, the scientific natu­ral history of modern ­humans must include the field of economics, which aspires to understand the interactions of individuals with one another and the rest of the material world using the currency of money, a uniquely ­human proxy for energy and resources that can be converted into fitness and influence. Like the population dynamics of other species, the development of ­human society proceeds in a decentralized way, emerging from the interactions among individuals and groups in the marketplace of commerce, ideas, and culture. To the extent that such development involves planning, the guiding princi­ples come from the science of economics. Conventional economics has a potentially dangerous blind spot, however, in largely ignoring the biophysical constraints and feedbacks between h­ uman society, biosphere, and the earth system that ultimately constrain the economy. As we saw above, the finitude of resources is no longer an abstract academic concept but a real­ity approaching steadily on the horizon (Costanza 1989, Turner 2008). Paradoxically, the intense re­sis­tance in some quarters to evidence of climate change supports this view in revealing climate change as a threat to the foundations of the economic worldview at the heart of modern society. As the UK’s review on the economics of climate change led by Nicholas Stern concluded, “Climate change pre­sents a unique challenge for economics: it is the greatest and widest-­ranging market failure ever seen” (Stern 2007). Integrating the finite nature of Earth’s resources into the field of economics is an urgent frontier in h­ uman affairs.

The tragedy and triumph of the commons In 1968 Garrett Hardin captured arguably the central challenge of Anthropocene ecol­ogy: in­de­pen­ dent users of a common-­pool resource—­one that cannot be monopolized—­have an economic incentive to use it as fast as they can ­because, if they ­don’t, ­others ­will. The logical result would seem to be depletion, a tragedy in the philosophical sense of “resid[ing] in the solemnity of the remorseless working of ­things” (Hardin 1968). In the modern, globalized world, however, the stakes are higher than ever b­ ecause the arena of competition has expanded from livestock grazing on the village commons to the finite store of clean air, ­water, minerals, and other resources of the planet as a ­whole. ­Human beings, however, are more complex than the actors envisioned by s­ imple economic game theory. Our strong tendency t­ oward sociality, including development of long-­term reciprocal interac-

Chapter 4 Introduction to the Anthropocene Ocean

tions and relationships among individuals, can modify the remorseless working of ­things. Studies of how ­humans interact with one another and their resources in the context of social-­ecological systems have shown that individuals can self-­organize into groups that behave more sustainably than envisioned by the tragedy of the commons and that the tendency t­oward such sustainable structures can be influenced by policy (Ostrom 2009)—­a field of inquiry that earned Elinor Ostrom and Oliver Williamson the Nobel Prize in Economics in 2009. Historically, sustainable systems of resource use have emerged on local scales where communities have an incentive to maintain resources to which they have some exclusive access. The accelerated modern global economy challenges this situation ­because markets and the reach of ­those who exploit the resources are increasingly global and new export markets (e.g., for novel seafood species) can develop and be exploited to depletion faster than institutions can respond (Berkes et al. 2006). This is a classic example of the tragedy of the commons.

The Anthropocene ocean Despite being terrestrial mammals, technology has now made ­humans the top predator or, more accurately, the top omnivore in the world ocean. Humanity is also a major agent of bottom-up forcing, injecting massive quantities of carbon, nitrogen, and other materials into the ocean and reducing light penetration via sediment runoff in many coastal regions. Indeed, ­humans are unique in being strong interactors in nearly all Earth’s ecosystems, affecting many other keystone species across habitats and thereby amplifying impacts through a complex web of strong indirect interactions (Worm and Paine 2016). On land, we have s­ haped ecosystems to our own ends for millennia. Major changes to ocean ecosystems are more recent but have caught up rapidly during the ­Great Acceleration of the ­later twentieth ­century (figures 4.12, 4.13). We have modified the ocean by using it as a source of resources, primarily of animal protein, as well as a sink for wastes, notably for nitrogen fertilizer and the carbon exhaled by our industrial metabolism. Th ­ ese impacts remain largely invisible above w ­ ater, so understanding them is correspondingly more challenging than on land. As recently as 1986 a prominent review concluded that “­human use of marine productivity is relatively small . . . ​[and] ­human exploitation of marine resources therefore seems insufficient by itself to alter on a large scale any but the target populations and ­those of other species interacting closely with the target populations” (Vitousek et al. 1986). ­Those populations and species of course include a large part of the biosphere, and evidence to the contrary has mounted. ­Here we introduce the major pathways through which ­humans transform the ocean, which we take up in detail in ­later chapters. Assessing impacts requires a baseline for comparison, analogous to an experimental control: What would the ocean look like in the absence of p­ eople, or u­ nder pre­industrial conditions? Answering this question is challenging b­ ecause h­ umans have affected parts of the ocean for centuries, even at low population densities and with primitive technology. We also face the basic cognitive blind spot of the shifting baseline syndrome, first proposed in the context of fisheries but more widely applicable: Each generation of fisheries scientists accepts as a baseline the stock size and species composition that occurred at the beginning of their c­ areers, and uses this to evaluate changes. When the next generation starts its c­ areer, the stocks have further declined, but it is the stocks at that time that serve as a new baseline. The result obviously is a gradual shift of the baseline, a gradual accommodation of the creeping disappearance of resource species, and inappropriate reference points for evaluating economic losses resulting from overfishing, or for identifying targets for rehabilitation mea­sures. (Pauly 1995)

Reconstructing a baseline can be approached in one of three ways: empirically from comparisons across space, empirically using comparisons through time, or mechanistically from first princi­ples of

87

Opportunity to rebuild marine life Efforts to slow down pressures Sharp increase in pressures on and decline in marine life Rebuilding marine life

1860

Debate on whether industrialized fishing could lead to permanent exhaustion of fish stocks (International Fisheries Exhibition, London, 1883) Electric and gas street lights reduced hunting 1890 for animal oil 1920

International Convention for the Prevention of Pollution from Ships (MARPOL, 1973)

Geneva Convention on the Law of the Sea (1958) 1950

UN Conference on Environment and Development (1992) UNCBD (1993)

CITES (1975) IWC Whaling Moratorium (1982) UNCLOS (1982)

1980

IPCC established (1988)

UN SDGs (2015) and Paris Agreement of UNFCC (2016) adopted 2020 Maximum Pressure scale

Hunting Fishing Deforestation Habitat loss Fertilizer, plastic and synthetic chemicals Climate change

1860

Pressures on marine life

1890

1920

1950

Petroleum industry develops Onset of large-scale industrial production of fertilizer, plastic and synthetic chemicals

Steam engines applied to fishing boats

High Medium

Low 1980

2020

Climate dynamics can no longer be reproduced without invoking human perturbation

Warming-induced global coral-bleaching events

Figure 4.12. ​Global pressures on marine life. H ­ uman impacts on ocean life and ecosystems began centuries ago, even millennia ago in some regions. But some pressures appear to have peaked, including fishing and habitat loss, which ­were maximal in the 1980s. Climate change and pollution continue to increase. Together t­ hese led to steep declines in many marine organisms during the ­Great Acceleration of the late twentieth c­ entury (­after Duarte et al. 2020).

A

B

D

E

C

F

Figure 4.13. ​Poster ­children of the Anthropocene ocean. (A) The g ­ reat auk (Pinguinus impennis), extinct from overhunting, 1844. (B) The 130 m deep-­sea trawler MV Margiris, tool of the world’s top marine predator, Homo sapiens. (C) The Pacific diatom Neodenticula seminae, recorded blooming in the Atlantic for the first time in 800,000 years ­after apparently traversing a warm Arctic Ocean. (D) The king crab Paralomis birsteini, a shell-­crushing predator newly arrived on the warming Antarctic shelf to feed on a diverse benthic fauna evolved largely in the absence of predators. (E) A crustose coralline alga, impor­tant substratum for coral recruitment, declining in many regions due to ocean acidification. (F) The elkhorn coral Acropora palmata, formerly dominant coral of energetic shallow Ca­rib­bean reefs, now endangered by multiple stressors, including disease, climate warming, and algal overgrowth.

Chapter 4 Introduction to the Anthropocene Ocean

89

Ocean Warming As ­we’ve seen, a principal driver of Earth’s transformation over the late twentieth ­century was a steep rise in fossil fuel use. Since the beginning of the industrial era in the mid-­eighteenth ­century, a ­little over 400 billion metric tons of carbon have been released to the atmosphere by ­human activities,

Lis FFS ian sk Ni i ho Ga a rd ne r P & H

Ka

M uai ol ok a Oa i hu Ha w aii K a Ma ho ui ol aw e Ku Ne re ck er M ar La o ys an

Biomass (t ha–1)

5.0 macroecol­ogy. In the first case, spatial comparisons seek out Northwestern Apex predators 4.5 Hawaiian Islands places far from h­ uman influence as controls and are l­imited to Other secondary consumers (mostly unpopulated) 4.0 Herbivores systems where such lightly impacted sites still exist. Coral reefs 3.5 around uninhabited islands in the remote Pacific, for example, Main Hawaiian Islands 3.0 (heavily populated) support far higher biomass of fish and especially of large carni2.5 vores than reefs near populated islands (figure 4.14) (Fried2.0 lander and DeMartini 2002, Sandin et al. 2008), suggesting 1.5 that such conditions prevailed elsewhere in the ocean prior to 1.0 intensive ­human exploitation. Spatial comparisons are more 0.5 difficult for other ecosystem types where pristine exemplars 0.0 are gone, such as coastal plain estuaries, which have attracted dense ­human settlement since our species’ earliest days. Second, baselines can be estimated from temporal comparisons, using paleontology and historical sources to reconFigure 4.14. ​Differences in fish biomass and trophic structure struct likely ecosystem states before substantial ­human impacts. among reefs in the Hawaiian Islands along a gradient of h­ uman Historical sources illustrate how profoundly p­ eople have influence. Locations are ranked from left to right, lowest to changed marine ecosystems in the last few centuries. A dihighest, by mean biomass. The main Hawaiian Islands (left of dotted line) are populated by p ­ eople and the Northwestern verse range of historical information from the fossil rec­ord, Hawaiian Islands (right of dotted line) are remote and largely archaeology, fifteenth-­century nautical charts, and the like tounfished (­after Friedlander and DeMartini 2002). gether reveal sustained and major declines in large animals throughout the ocean that have followed ­humans wherever we have gone, even in small populations with primitive technology ( Jackson et al. 2001). Ships’ logs show that sea turtles ­were widespread in past centuries and ­were far more abundant than they are ­today (“Why, a­ fter all, are so many hundreds of sites around the Ca­rib­bean, such as the Dry Tortugas, named ­after turtles that almost no living person has ever seen?” ( Jackson 1997)). Local reef and bottom fishes dominated Hawaiian restaurant menus in the 1940s but have largely dis­appeared from modern ones, replaced by aquaculture species and pelagics taken far offshore (Van Houtan et  al. 2013). The mean size of trophy fish displayed in photos from Florida Keys charter boats declined by nearly an order of magnitude between the late 1950s and 2007 (McClenachan 2009). Such individual historical comparisons are complicated by changes over the same time periods in fishing gear, management, and market globalization. But, taken together, ­these disparate sources of information paint an unmistakable picture: large marine animals have declined precipitously in size and abundance with intensifying ­human influence, especially since 1950 ( Jackson et  al. 2001, Payne et  al. 2016). ­These historical analyses have helped overcome the shifting baseline phenomenon. Fi­nally, baselines may be estimated from first princi­ples by modeling—­asking how much fish biomass could be supported by the primary productivity characteristic of a region given empirically documented conversion efficiency and predator-­prey interactions (Pauly and Christensen 1995, Jennings et al. 2008). Bioenergetic modeling has been used, for example, to confirm that Chinese officials systematically overreported marine fishery catches for years. ­Because China is one of the world’s major fishing nations, this creative accounting inflated estimates of marine fishery production for the world as a ­whole and obscured for a de­cade the evidence that production of marine capture fisheries peaked in the 1980s (Watson and Pauly 2001). The world appears to have reached the limit of ocean fishery production, while appetites continue to grow.

90

Ocean Ecology

Onset of anaerobiosis

CO2 Hypoxia

T (°C)

Onset of denaturation

Competition, food web interactions, phenologies Coexistence ranges

Spawners Growing adults Juveniles Eggs, early larvae

Aerobic thermal window

Aerobic performance and productivity

Onset of loss of performance and abundance

Thermal window widths across life stages (fishes)

Sequence of life stages

Scope for aerobic performance

Thermal windows for animals

Temperate eurytherm Polar stenotherm Warm adapted

Spring warming cue T ranges in climate zones, T ranges throughout seasons

Figure 4.15. ​Physiologically mediated impacts of climate warming in the ocean relative to thermal win­dows (i.e., temperature ranges in which specific life pro­cesses can happen safely) (­after Pörtner and Farrell 2008).

with half of ­those emissions since the late 1980s (Boden et al. 2017). Although the contribution of fossil fuel burning to climate warming is widely known, it is less appreciated that more than 90% of the heat produced thereby has been absorbed by the ocean (Levitus et al. 2012). As global mean temperature increased by 0.4°C since the 1950s, the ocean’s heat content also increased substantially, reducing potential heating of the atmosphere. Indeed, the detection of warming in the deep ocean is considered one of the clearest signals of global climate change. Climate warming varies geo­graph­i­ cally, raising temperatures over continents more than over the oceans. This can strengthen wind fields along ocean margins and increase upwelling, particularly in eastern boundary currents, raising productivity and paradoxically lowering coastal ocean temperatures in ­these regions. Outside of upwelling regions, warming of the surface ocean more often strengthens stratification and nutrient limitation, and has reduced productivity of surface ­waters over large regions of the ocean in recent de­cades (Behrenfeld et al. 2006, Boyce et al. 2010). As we learn in chapter 5, temperature exerts a fundamental control on biological and chemical pro­cesses at all levels, from enzyme kinetics to ecosystem fluxes. Climate warming accordingly has a wide range of effects on organisms, their interactions, and therefore the structure and functioning of ecosystems (figure 4.15). ­These include physiological effects on individual per­for­mance, effects on community interactions, and biogeographic shifts (the last considered in chapter 3). Temperature affects rates of individual growth and development, often differently. Thus, warmer temperatures tend to accelerate maturation more strongly than growth, resulting in smaller adults ­under warmer conditions, a phenomenon well documented among copepods (Forster et al. 2011). B ­ ecause body size influences a range of ecological pro­cesses (chapter 5), ­these temperature-­mediated changes can have far-­reaching consequences for communities and ecosystems. Warmer temperatures also generally accelerate larval development rates, reducing larval development times and thus potentially the degree of connectivity among populations (O’Connor et al. 2007).

Warming effects on communities Changing temperature often affects species differently, which can alter the rates and outcomes of their interactions (Gilbert et al. 2014). We consider ­here three general consequences for communities. First, phenological mismatch occurs when a temperature change alters the timing (phenology) of life history in interacting species differently, such that key events no longer overlap and their interaction is disrupted. In the ocean, many animals have evolved to produce pelagic larvae around the time of the spring phytoplankton bloom when food for their larvae is generally abundant (see box 6.4).

Chapter 4 Introduction to the Anthropocene Ocean

The steadily warming climate has brought ­earlier springs and longer summers over recent de­cades, disrupting the coevolved timing of many such consumer-­prey interactions (Edwards and Richardson 2004). Survival through the feeding larval stage often strongly affects adult abundance, and mismatches between consumer and prey life history timing due to small changes in temperature can cause large, nonlinear effects on population dynamics and interactions (Gilbert et al. 2014). In the UK over the three de­cades ending in 2005, warming-­related life history changes w ­ ere slower for secondary consumers than for primary producers and herbivores across marine, freshwater, and terrestrial systems, suggesting that interactions that depend on synchronization of life history events are being disrupted (Thackeray et al. 2010). A second general effect of warming is alteration of plant-­herbivore interactions. Warming temperatures increase respiration, and thus animal activity, faster than they increase photosynthesis, on average. This mismatch is expected to affect plant populations more strongly u­ nder warmer temperatures as a consequence of stronger herbivory, a prediction confirmed by experiments in both benthos and plankton (O’Connor 2009, O’Connor et al. 2009). Third, climate change can affect ecological interaction by bringing formerly separated species into contact. A striking example involves the recent warming-­related invasion of the Antarctic shelf by the king crab Paralomis birsteini (see figure 4.13D), bringing a shell-­crushing predator into a rich and diverse ecosystem that has evolved without them for tens of millions of years (Aronson et al. 2015). An example that combines both changing species distributions and intensifying herbivory is the decline of kelp beds in southeastern Australia, as warmer ­waters have allowed invasion of herbivorous fishes (Vergés et al. 2016). Large-­scale reor­ga­ni­za­tion of marine ecosystems is expected to redistribute the map of fishery productivity, with substantial economic impacts—­good for some but likely bad for many (Cheung et al. 2009, Lam et al. 2016). The sensitivity of ecological interactions to changes in temperature means that species with key functions in a community can become leverage points (or leverage species) for climate change effects, amplifying small changes in temperature into disproportionately large responses at the community level. The classic example is the original keystone species, Pisaster ochraceous, whose feeding rate on mussels declined substantially when upwelling reduced w ­ ater temperatures along the US west coast (Sanford 1999), potentially allowing mussels to dominate the shore at t­ hese cooler sites. Other species that potentially occupy such climate leverage points in marine communities include diseases and other pests (Kordas et al. 2011). Modest warming has often favored the spread or severity of disease in marine organisms, including reef corals, oysters, and even the original keystone sea star (Burge et al. 2014).

Sea level rise In coastal regions, climate warming has a second major impact: sea level rise. The sea level mea­sured at a given point on the coast is determined by the vertical height (level) of a reference point on the land relative to the height of the sea. Heights of both land and sea change through time in response to several ­drivers, so sea level is dynamic and depends on the rates and directions at which they move relative to one another. Climate warming is raising the global average sea level both by melting the polar ice caps, which adds ­water to the ocean, and through thermal expansion, which increases the volume of existing w ­ ater. At a given point on the coast, sea level also depends on the rate of vertical movement of the land, which responds to tectonic activity, continuing rebound from deglaciation, and sediment compaction. The high-­latitude regions of the northern continents, for example, w ­ ere covered with hundreds of meters of ice during the Pleistocene, and the deglaciation at the end of that epoch relieved them of this burden. As a result, the continental crust began to rebound, springing gradually back up, a pro­cess called glacial isostatic adjustment. In regions where this rebound is ongoing, sea level has risen only slowly and in some places is even falling b­ ecause the land is rising faster than the sea. Conversely, in places where the land is sinking, sea level is rising faster than the global average.

91

Ocean Ecology

Global average sea level has been rising steadily since around 1860 (see figure 4.7), averaging 1.4 ± 0.2 mm per year during the twentieth ­century, according to synthesis of regional sea level reconstructions and tide gauges (Kopp et al. 2016). A sea level model calibrated with ­these data estimates that, without anthropogenic carbon emissions, global sea level rise over the twentieth c­ entury would have been between −3 and +7 cm, rather than the observed 14 cm. Thus, at least half of the sea level rise observed over the last ­century is likely attributable to ­human influence. Applying that model to the IPCC’s “business-­as-­usual” scenario for expected carbon emissions (RCP 8.5), and associated climate warming, proj­ects a sea level rise of 52–131 cm over the current ­century (Kopp et al. 2016). Rising sea level has several consequences for p­ eople and communities in coastal regions, including vulnerability of lives, property, and infrastructure to flooding and storm surge, and salt intrusion into groundwater. It’s estimated, for example, that a sea level rise of 1.5 m over the coming 150 years ­will displace 17 million ­people in Bangladesh alone. From a purely ecological perspective, major consequences of sea level rise are greatest for shallow and intertidal habitats, such as coastal wetlands that occupy specific elevations relative to sea level. Such habitats ­will have to migrate inland as sea level rises, and in many cases the areas immediately inland are already occupied by ­human habitat, generating conflicts. 8.38

(A) 400

CO2

375

Mauna Loa CO2 record

8.33 8.28

350

8.23

325

8.18 8.13

300 275 1950

Ocean Acidification pH

92

8.08 1960

1970

1980

1990

2000

2010

8.03 2020

1990

2000

2010

2020

1990

2000

2010

2020

(B) Aragonite saturation state

Ωaragonite

4.50 4.00 3.50 3.00 1950

1960

1970

1980

(C) Calcite saturation state

Ωcalcite

6.50 6.00 5.50 5.00 1950

1960

1970

1980

Figure 4.16. ​Changes since 1960 in (A) concentration of atmospheric CO2 (parts per million volume) at Mauna Loa, Hawaii, far from industrial sources (red), surface ocean pH (blue), and surface ocean pCO2 (μatm, tan) at Ocean Station ALOHA in the subtropical North Pacific Ocean; (B) aragonite saturation state; and (C) calcite saturation at Station ALOHA (­after Doney et al. 2009).

A subtler but potentially more far-­reaching impact of fossil carbon combustion is ocean acidification, which results from gradual diffusion of CO2 into the ocean, where it reacts with ­water molecules to form dissolved bicarbonate (HCO3−) and a proton (H+), increasing the acidity (decreasing the pH) of seawater. Ocean acidification represents a fundamental change in the chemistry of seawater. Seawater has a high buffering capacity, meaning that its pH changes only slowly with addition of acid; this is due to its high concentration of dissolved bicarbonate and other components of the carbonate system. But the buffering capacity has limits. The absorption of anthropogenic CO2 has already reduced global ocean pH by 0.1 unit, from 8.21 to 8.1 on average, since pre­industrial times (figure 4.16), and is projected to decline by an additional 0.3–0.4 units by the end of this ­century (Orr et al. 2005, Doney et al. 2009). Ocean acidification has potentially profound consequences for ocean life, by changing the biogeochemistry of calcium carbonate, organic carbon, nitrogen, phosphorus, and trace metals (Doney et  al. 2009). Ocean acidification depends mainly on two variables: the CO2 concentration of the atmosphere, which increases its dissolution into the surface ocean, and sea surface temperature, which decreases solubility of CO2 as the ­water warms. Thus, ocean warming and acidification are expected to synergistically affect marine life.

Effects of acidification on organisms Ocean acidification has a wide range of physiological effects that compromise the per­for­mance of individual organisms and can ­ripple through interaction webs to affect communities and

Chapter 4 Introduction to the Anthropocene Ocean

ecosystems. A major consequence, and the first to be widely recognized, is that ocean acidification lowers the calcium carbonate saturation states of ocean w ­ ater (see figure 4.16), affecting organisms that form shells and skeletal ele­ments out of CaCO3. ­These include many key players in ocean ecosystems, such as planktonic foraminifera and coccolithophorids, and benthic coralline algae, mollusks, echinoderms, and corals (see figure 4.13). Effects of ocean acidification are most evident among calcifying organisms, in which laboratory studies show that calcification and growth rates are often reduced ­under high CO2 (Kopp et al. 2016). In reef-­building corals, calcification rates are reduced by both warming and acidification, delivering a double blow in the modern ocean, where t­ hese stressors tend to co-­occur. Acidification also threatens other calcareous foundation species, such as oysters and mussels. Experimental acidification affects metabolism, productivity, consumption, calcification, and sensory capabilities in a wide range of marine organisms. Meta-­analysis of some 600 experiments identified several broad trends in response related to organismal traits (Nagelkerken and Connell 2015). First, b­ ecause inorganic carbon is required for photosynthesis, increasing CO2 could boost plant productivity. Experiments with single species confirm that, on average, acidification raises primary production among noncalcifying plankton from temperate regions, but the opposite is true for tropical plankton. Experimental acidification reduces secondary production on average in both calcifying and noncalcifying species. Such changes could intensify top-­down control since metabolic and foraging costs of carnivores typically increase with acidification as well as with temperature (Nagelkerken and Connell 2015). In addition to its direct physiological effects on organism survival, growth, and metabolism, acidification can affect sensory physiology, with potential downstream consequences for key demographic pro­cesses and ecological interactions. ­These physiological effects generally can be classified into ­those resulting from increased metabolic load; “info disruption,” or compromised ability to gather and pro­cess sensory information; and avoidance of acidified locations (Briffa et al. 2012).

Effects of acidification on communities A sobering theme emerging from global change research is that results of single-­species experiments often do not hold, and may even be reversed, in natu­ral communities of multiple interacting species. Extracting generalizations about global change effects on communities is a frontier for ocean acidification research, as for global change research generally. This requires moving beyond studies of physiological effects of stressors on individual species and pairwise interactions to address how effects might ­ripple through community interaction webs. This prob­lem has been approached through controlled experiments, comparisons in space between naturally acidified CO2 vents and ambient areas, and comparisons through time at sites where long-­term pH rec­ords are available. Early acidification experiments focused on coral reefs b­ ecause calcification of both corals and the crustose coralline algae that often serve as specific settlement substrata for their larvae are sensitive to changing pH. Crustose coralline algae are especially vulnerable to acidification and illustrate how effects of ocean acidification ­ripple through interaction webs. Th ­ ese strange organisms constitute a key functional group in many benthic communities. Heavi­ly calcified and resembling paint blotches on rock surfaces (see figure 4.13E), crustose corallines are major space occupiers and producers of carbonate sediments on reefs, as well as being settlement habitats for recruiting coral larvae. In experiments that raised CO2 concentration on Hawaiian reefs, the cover of crustose corallines declined by more than 90%, yielding to overgrowth by filamentous and fleshy algae that recruiting coral larvae avoid (Kuffner et al. 2007). Such declines of crustose coralline algae have far-­reaching implications for reefs b­ ecause coral recruits often respond to highly specific chemical cues. Perversely, the coralline algae species most attractive to recruiting coral larvae are sometimes the species most vulnerable to acidification. Settlement by larvae of the coral Acropora millepora fell by nearly half ­under experimentally

93

94

Ocean Ecology

acidified conditions on the ­Great Barrier Reef, mainly b­ ecause t­ hose corallines preferred by larvae (Titanoderma spp., Hydrolithon spp.) declined most. Perhaps most surprisingly, ­under acidified conditions coral larvae avoided the crustose coralline species they preferred ­under normal conditions, instead selecting other substrata that support poor survival and growth (Doropoulos et al. 2012). If ­those effects are common in nature, ocean acidification could deliver a one-­two punch to coral reefs, slowing calcification and growth of adults while also taking away their nursery. The most integrative studies of ocean acidification effects on communities come from exploiting naturally acidified ocean ­waters around shallow volcanic vents. At such sites, ­water takes on high concentrations of CO2 in contact with deep hot rock and then streams out through fissures, creating local oases of altered ocean chemistry. Experiments and surveys at t­ hese CO2 vents corroborate lab findings of reduced per­for­mance by calcareous organisms at lower pH but also illuminate broader community consequences. At vents in the Adriatic Sea, CO2 altered the competitive hierarchy among benthic algae, as crustose coralline algae ­were quickly overgrown by fleshy seaweeds. This appeared to result from both reduced growth of the calcareous crusts and reduced grazing by sea urchins in the acidified w ­ ater (Kroeker et al. 2012). Th ­ ese results hint that shifts t­ oward dominance by fleshy seaweeds, already common in many marine ecosystems, may be exacerbated in an acidifying ocean, and illustrate that changes in the physiology of species can ­ripple through interaction webs to produce large effects. By inhibiting crustose coralline algae and grazing, acidification can also ­favor the spread of invasive fleshy algae (Hall-­Spencer et al. 2008). ­Because fishes often play impor­tant roles in marine communities, behavioral shifts caused by acidification can cascade to change communities. Comparisons of CO2 vent communities with nearby unacidified areas in both the Adriatic Sea and New Zealand corroborated lab findings that fishes are less able to detect and avoid predators, but also found that predatory fish appear more sensitive than their prey (Nagelkerken et al. 2015). Together with increased availability of crustacean prey around CO2 vents, t­ hese comparisons suggest that acidification alters the structure and functioning of benthic communities through both direct physiological effects and indirect effects of changed interactions, tipping them from top-­down ­toward bottom-up control. It is striking that ­these studies of natu­ral vent communities found a result opposite the intensified top-­down control expected from single-­species experiments (Nagelkerken and Connell 2015). An impor­tant path by which ocean acidification can impact ­whole communities is through its influence on foundation species like corals. Studies on coral reefs illustrate that interactions within communities can complicate predictions based on experiments with individual species in isolation. On Indo-­Pacific reefs, coral communities normally consist of a structurally diverse mix of branching, plating, and boulder corals. But around vents in Papua New Guinea, chronic exposure to CO2 shifted coral communities to dominance by boulder corals, and structurally complex branching species dropped out. As a result, the acidified reefs harbored much lower densities and diversity of invertebrates that normally shelter among the branches. But sea urchins and some other invertebrates that associate with boulder corals w ­ ere more abundant at the high CO2 sites. Thus, when studied in natu­ ral communities in the field, several groups showed opposite responses to acidification than expected based on their physiological tolerances in laboratory experiments (Fabricius et al. 2014). Similar declines in corals occur along gradients in seawater pH around vents in the Adriatic Sea (Hall-­Spencer et al. 2008). And parallel changes are also seen in time series of a temperate intertidal community where seawater pH has been declining steadily. In the Northeast Pacific, coralline algae and other calcified organisms declined in years with more acidified seawater, and model projections based on well-­studied community interactions suggest they ­will gradually transition from dominance by calcified benthos to fleshy algae as pH continues to decline (Wootton et al. 2008). Results of ­these case studies are general: meta-­analysis confirms that acidification reduces average species diversity and abundances in both tropical and temperate marine communities and tends to shift them ­toward dominance by uncalcified species and microbes (Nagelkerken and Connell

Chapter 4 Introduction to the Anthropocene Ocean

2015). In summary, disparate lines of evidence suggest that the carbonization of the Anthropocene atmosphere and ocean portends simplified marine communities and altered trophic interactions with still uncertain implications. A sobering feature of ocean acidification is that, like climate warming, it is caused by changes in the atmosphere that are essentially global in scale. ­Because the atmosphere is vast and relatively well mixed, it responds slowly to forcing, with the consequence that the green­house gases already added to the atmosphere w ­ ill continue to warm it and acidify the ocean for de­cades before reaching equilibrium, even if we ­were able to reduce additional loading to zero tomorrow. Acidity ­will be a major feature of the Anthropocene ocean for a long time to come.

Homo Sapiens: Top Predator of the Ocean The history and extent of fishing Climate change and ocean acidification portend a profound shake-up of the global ocean ecosystem in coming de­cades. But the major h­ uman impact on the ocean t­oday and since the beginning of ­human history has been fishing. P ­ eople have fished almost e­ very conceivable kind of marine animal and plant, from intertidal shellfish and seaweeds to the g­ reat w ­ hales. Fisheries range from artisanal efforts of individuals fishing for subsistence with hook and line, still common throughout the world, to massive industrial operations by the largest ships on the ocean (see figure 4.13B), working with state-­of-­the art electronics and removing gigantic quantities of fish. In many coastal regions, the primary production required to support fish harvests is about a third of the total fixed by the ecosystem (Swartz et al. 2010). Not surprisingly, this level of sustained, intense harvest—­often targeting top predators—­ has fundamentally transformed the community structure, ecosystem fluxes, and biogeochemistry of much of the world ocean ( Jennings and Kaiser 1998, Jackson et al. 2001, Estes et al. 2011). Marine fisheries are a major source of food and income to humanity. World trade in fish and fish products (including aquaculture) reached $143 billion in 2016, providing 17% of global animal protein intake, the primary animal protein source for one billion p­ eople, and 6.7% of all protein consumed by h­ umans worldwide (FAO 2016). B ­ ecause of this high value and b­ ecause most fish stocks are common property resources that are difficult to monopolize, most fisheries are subject to scramble competition, with a strong incentive to overharvest. The far-­reaching consequences of fishing for both the economy and the environment naturally have stimulated a sophisticated science of fishery ecol­ogy and management, which in turn has been a major stimulus to scientific understanding of marine populations and ecosystems. Indeed, the earliest research recognizable as marine ecol­ogy was motivated explic­itly by the need to understand and manage fisheries (Petersen 1918, Hardy 1924). ­People have fished since the dawn of h­ uman history, producing impacts apparent even in primitive socie­ties at low population densities. Localized effects are clearly vis­i­ble in archaeological rec­ords from hunter-­gatherer socie­ties around 1000  years BP on some Ca­rib­bean islands (Wing and Wing 2001) and the east coast of Africa (McClanahan and Omukotu 2011). At t­ hese sites, harvested fishes declined in average body size and trophic level over a few centuries of occupation, as documented by fish vertebrae in kitchen middens (figure 4.17). Historical rec­ords reveal severe resource depletion as early as Roman times (2500 years BP) in the Mediterranean and during medieval times (1000  years BP) in the Wadden and Baltic Seas (Lotze et  al. 2006). Industrial-­scale fishing took off shortly ­after World War II, directly ­shaped by the new technology and infrastructure of the war effort and market opportunities left in its wake (Holm 2012), and developed in concert with other components of the G ­ reat Acceleration (see figure 4.4). Fishing subsequently expanded rapidly in intensity and reach, both geo­graph­i­cally and into the deep, with the advent of motorized vessels, inexpensive oil, refrigeration, increasingly global commodity markets, and heavy government subsidies to increase fleets (Pauly et al. 2002) (figure 4.18).

95

Average reef fish weight (g)

1200 1000 800 600 400

(B)

(C) 3.8

100 Percent reef fishes

St. Thomas Saba St. Martin Nevis Puerto Rico

Trophic level index

(A)

3.7 3.6 3.5

200 0 2000

1500 1000 500 Years before present

0

3.4 2000

1500 1000 500 Years before present

0

80 60 40 20 0 2000

1500 1000 500 Years before present

0

Figure 4.17. ​Evidence for fishing impacts of prehistoric ­human communities. Archaeological data from five Ca­rib­bean islands show that harvested fishes declined steeply through time in (A) body size and (B) trophic level, and (C) ­were taken farther from shore, even by small h­ uman populations with primitive technology (­after Wing and Wing 2001).

(A)

(C) 90

25

80

20

60

Catch

15

50 40

10

Nominal effort

30 20

100

5

10

Effective effort

0 (B)

140

160

180

200

1.0 0.8

80

0.6

60

0.4

40

0.2

20

0.0

0

1950

1960

1970

1980

1990

2000

220 Percentage of global oceans fished (yellow)

Relative catch per unit of effective effort (green)

1.2

120

Mean depth of catch (m)

70

Effort (gigawatt)

Catch (million tonnes)

80

240

260 1950

1975

2000

Figure 4.18. ​Indicators of accelerating h­ uman impacts on marine fisheries during the Anthropocene. (A) Global fishing effort has increased tenfold, while (B) catch per unit effort (green symbols) has halved since 1950 (­after Watson et al. 2013). (C) Fisheries have moved ever deeper over the last half c­ entury (pelagic fishes (green); bottom-­dwellers (blue); in all fisheries (open) versus high-­seas fisheries only (solid)) (­after Morato et al. 2006).

Chapter 4 Introduction to the Anthropocene Ocean

97

Marine fisheries now use 24%–35% of primary production on continental shelves and in major upwelling areas (Pauly and Christensen 1995), including bycatch (marine animals caught in the nets inadvertently)—­a figure similar to the roughly one-­fourth of the land’s potential net primary production appropriated by ­humans.

The current state of marine fisheries

Percent of maximum

An enterprise this large and wide-­ranging requires management. Fish is a renewable resource insofar as fishes are living organisms that reproduce and can replace individuals harvested by h­ umans. ­W hether replacement keeps pace with harvesting depends on how fast fish are taken out of the population (or stock, in fisheries jargon) relative to how fast they reproduce and grow to maturity. In managed fisheries, the goal, explicit or implicit, has historically been to harvest the stock down to a number (1) that is producing new fish at the same rate they are being removed (i.e., yield is sustainable over time) and (2) where this sustained yield is maximal. Implementing this strategy takes advantage of the relationship between abundance (or biomass) and productivity (yield), which is generally hump-­shaped, such that maximum sustainable yield (MSY) occurs at an intermediate population size (figure 4.19). A smaller population has too few breeders to produce a high yield, whereas a very large and dense population is often less productive ­because competition depresses growth and per capita productivity. An area where fishing has depressed a stock below its MSY is considered overfished. (Longhurst [2010] notes that the seemingly innocuous term “overfishing” hides a value judgment in that it implies ­there is a level of sustainable fishing, known in advance or at least knowable, which he argues is rarely the case.) In order to implement MSY as a management target in practice, we ideally need to know the population size, fishery-­induced mortality, and Rebuilding Overfishing productivity of the stock and that ­these population pro­cesses are MMSY 100 deterministic and minimally affected by other species or changes in the abiotic environment. Th ­ ese assumptions seem suspect on first princi­ples (Larkin 1977, A. Longhurst 2007), and the data re80 quirement for making such estimates is daunting. Nevertheless, we have to start somewhere. 60 Key to fishery management is understanding the ecol­ogy of the predator, ­humans, in the fishery ecosystem. ­Human cultural evolution has proceeded much faster than the ge­ne­tic evolution of 40 other predators or prey defenses. The technological sophistication of both commercial and recreational fishing, along with cheap oil, 20 has eliminated the energetic costs associated with prey search, pursuit, and capture that normally limit the efficiency of other predators, and the subsidy of the ­human diet from land-­based agricul0 0.0 0.2 0.4 0.6 0.8 1.0 ture has eliminated the feedbacks from a declining prey resource Exploitation rate that depress population growth of wild predators in classical Total catch Mean Lmax predator-­prey models. Perversely, low abundance of prey can acTotal biomass Collapsed species tually drive stronger exploitation ­because rare resources are often especially prized in the free-­for-­all competition typical of an inadFigure 4.19. ​Expected effects of increasing exploitation rate equately regulated fishery (Darimont et al. 2015). A striking on a fish community. Exploitation rate is defined as the example is the mighty Pacific bluefin tuna, which is considered proportion of fish biomass caught each year. Body size of threatened by the International Union for Conservation of Nature fishes in the community (Lmax) and total biomass decline steadily with exploitation, while the number of collapsed (IUCN). The first bluefin sold in Japan’s Tsukiji ­wholesale market stocks (with biomass less than 10% of unfished biomass) each year famously fetches astronomical prices (the rec­ord is increases. Total catch across all species peaks at intermediate $1.76 million for a single fish in 2013). But even everyday prices exploitation rate. MMSY refers to the multispecies maximum are startling: in early 2017 fresh bluefin at the Tsukiji market was sustainable yield (­after Worm et al. 2009).

98

Ocean Ecology

g­ oing for $120/kg, or about $27,000 for a prize 500 kg fish. For ­these several reasons, the median rate of fisheries exploitation is estimated to be three times higher than the median rate of predation by the most effective nonhuman marine predator of the same prey (Darimont et al. 2015). The status of marine fisheries can be evaluated ­either at the fine-­grained level of individual stocks or at the aggregate global level. The first approach relies on detailed data, which generally are available only from managed fisheries of the developed world. The United Nations’ Food and Agriculture Organ­ization (FAO) undertakes a comprehensive assessment of the status of global fisheries ­every four years, based on such data provided by national governments. This database necessarily excludes much fishing that is illegal, unreported, and un­regu­la­ted (IUU in fisheries jargon), which is especially common in developing countries with poor governance. In West Africa, for example, total fisheries catches are estimated to be 40% higher than reported catches (Agnew et al. 2009). Nevertheless, the FAO numbers provide a start. They estimate that the proportion of stocks that are overfished increased worldwide from 10% in 1974 to 31% in 2013 (FAO 2016). Slightly more than half the stocks are considered fully exploited, meaning that current catches are at their maximum sustainable production. This combined percentage of overfished and fully exploited marine fish stocks is the highest in history. The FAO reported in 2009 that “the maximum wild capture fisheries potential from the world’s oceans has prob­ably been reached” (FAO 2009). The second, global perspective approaches the state of fisheries from the bottom up, estimating the proportion of local primary productivity needed to sustain fishing. Primary production in dif­fer­ ent regions of the ocean is estimated from satellite mea­sure­ments of ocean color and physical oceanography and is used to model data on fish catches from each region. This approach reveals that, averaged globally, marine fisheries appropriate 17%–112% more of the available primary production than is sustainable over time (Chassot et al. 2010). This estimate has wide uncertainty but implies that, in some areas at least, current levels of harvest cannot continue. Of equal concern and less widely appreciated is that overfishing results not only in degradation of fish stocks and associated ecosystems but also in overfishing debt, a substantial loss of revenue and food security, particularly in nations that can least afford it. ­Because overfishing exceeds MSY by definition, it reduces the long-­term yield of protein to fishers as the stock is depressed below the level at which it is maximally productive. Gross revenue lost to fishing beyond sustainable levels is estimated at 6%–35% of the landed value of the fisheries, averaged globally, resulting in undernourishment of 20 million p­ eople worldwide (Srinivasan et al. 2010).

Ecosystem impacts of fishing The removal of such large quantities of fish has transformed the world’s oceans. Most obvious are the direct effects on target species. Many stocks are fished well below the target abundance of roughly 50% of unfished biomass, and many ­others are not regulated at all. This affects more than abundance. Fishing typically removes larger and older individuals from the population, chronically altering size structure and abundance, which in turn can alter the population’s productivity and destabilize its dynamics, making fish populations more susceptible to collapse u­ nder environmental and other stressors (Hsieh et al. 2006) and even driving evolutionary change (Audzijonyte et al. 2016). Removing large numbers of fish from marine ecosystems also has far-­reaching indirect effects as ­those removals ­ripple through the food web to change prey populations and even primary producer abundance (chapter 8) and alter habitat structure and biogeochemistry (chapter 9). Fishing effects can flow up the food chain in cases where fishing targets the prey of piscivorous fishes, birds, and mammals, or when discarding large quantities of bycatch subsidizes the growth of scavenger populations ( Jennings and Kaiser 1998). A distinct class of fishing impacts results from effects of fishing gear on seafloor habitat. Bottom-­ dwelling fishes and shellfish are caught with mobile dredges and trawls that scrape the bottom, dis-

Chapter 4 Introduction to the Anthropocene Ocean

turbing benthic habitats, homogenizing and resuspending sediments, and flattening three-­dimensional habitat essential to many fishes and other animals. Th ­ ese mobile gears are extremely destructive to epifauna—­the sponges, corals, and other sessile invertebrates that provide habitat structure essential to many fishes and mobile invertebrates (Watling and Norse 1998). Such biogenic structure is impor­ tant as nursery habitat, offering both food and refuge for juvenile fishes, which suffer lower recruitment in degraded habitat. This was one motivation for the mandate to protect essential fish habitat in the US Magnuson-­Stevens Fishery Conservation and Management Act. Fi­nally, a central challenge in fisheries management stems from the tendency, already encountered in the context of ocean acidification, for populations to respond to stressors in unexpected ways as a result of the complex interactions with other components of their ecosystems. Th ­ ere is growing evidence that complex interactions can produce nonlinear responses to stressors and sudden shifts into new stable states that resist attempts to restore them to the original state (Daskalov et al. 2007, Petersen et al. 2008). We explore ­these in more detail in chapter 9.

The ­future of fisheries The pervasive impact of fishing on marine ecosystems raises the question of what ­future we can expect u­ nder the demand from a growing and more affluent global population. Boris Worm and colleagues (2006) dropped a depth charge into the already turbulent ­waters of fishery issues with the extrapolation in 2006 (appearing in the press release but not the publication itself) that, if current trends continue, all seafood species could be collapsed by 2048. The somewhat hyperbolic statement created a firestorm. Critics pointed out, correctly, that the fisheries catch data used in t­ hese extrapolations do not necessarily reflect fish abundance in the ­water ­because of changing gear types, markets, and management regulations. Some of the statistical assumptions required to work with such heterogeneous data ­were necessarily simplistic, so it is easy to find counterexamples to the projected global decline. ­After a de­cade of reflection and new analyses, we can predict with confidence that fisheries ­will still exist at that date. But, to use a terrestrial analogy, this argument may miss the forest for the trees. While many f­actors can obscure population trends for individual stocks, no compelling evidence has been offered that globally averaged catch data significantly misrepresent trends in global fish abundance. On the contrary, as noted e­ arlier, we appear to have reached a global ceiling in what can be captured sustainably from the oceans, and the large proportion of stocks that are fully or overexploited continues to grow steadily. The 2048 controversy stimulated an effort to convene opposing parties to search for consensus on the ­future of fisheries. Academic ecologists and fisheries scientists conducted a detailed analy­sis of available data and concluded, with cautious optimism, that we now understand how to manage fisheries sustainably and can point to several good examples (Worm et al. 2009). This is good news. But the more pressing question is not ­whether we know how to manage fisheries sustainably but ­whether we are likely to actually do so on the scale necessary in the foreseeable ­future. ­Here the outlook is not as bright. Most evidence for sustainable fisheries comes from carefully managed stocks within the exclusive economic zones of rich, stable, well-­governed western democracies that collect rigorous data, and even within ­these countries the model fisheries come primarily from regions (e.g., Alaska, New Zealand) with low h­ uman population density (Worm et al. 2009) and therefore less local pressure on fisheries than elsewhere. This situation is uncommon—­only 7% of coastal nations conduct rigorous scientific assessments of their fisheries to inform such management and a mere 1% have effective mechanisms for enforcing compliance (Mora et al. 2009). It should not be surprising, then, that a recent analy­sis of more than 4700 fisheries based on integrated catch, life history, and stock assessment data reached a conclusion similar to the original projection by Worm et al. (2006) a de­cade ­earlier: global fish stocks are in poor shape and declining on average, with 68% below the target biomass that would support maximum sustainable yield (Costello et al. 2016).

99

100

Ocean Ecology

The empirical evidence thus indicates decidedly mixed pro­gress in sustainable management of fisheries. But this is only part of the story. The holy grail of management for sustainability may be elusive not only in practice but in princi­ple. This is the crux of Longhurst’s blunt analy­sis of fisheries (mis)management, which comes to a rather dif­fer­ent conclusion than that of the Worm et al. (2009) convocation. Mismanagement of Marine Fisheries (Longhurst 2010) is a lively and power­ful critique of the history, science, and politics of global fishery management. It should be required reading not only for marine biologists but for all interested in the tantalizing goal of sustainability. In a nutshell, Longhurst’s thesis is that fish population dynamics are inherently variable and unpredictable, reflecting complex, nonlinear environmental forcing, and this makes deterministic modeling and management impossible not only in practice but in princi­ple. The outlook for marine fisheries in 2048 and beyond is thus a cause for concern. ­There seems ­little reason to expect that the steady increase in overexploitation w ­ ill reverse. Following the current trajectory, many fisheries are likely to be overharvested, especially in the developing world, and poor management ­will continue to cost the world’s poorest ­people valuable revenue and essential animal protein supply. Ray Hilborn, hardly a critic of the fishing industry, captured the prob­lem succinctly (Hilborn and Hilborn 2012): “We now have the technology to overfish almost e­ very imaginable marine resource. The question is, do we have the po­liti­cal ­will and the social and cultural institutions to restrain ourselves?” How do we solve t­hese prob­lems? Effective conservation and management of fisheries ­will likely involve several parallel approaches, including the following. Ecosystem-­based fisheries management (EBFM), also known as the ecosystem approach to fisheries (EAF), seeks to incorporate into management the complex interactions among fish populations, the broader ecosystems that support them, and the central role of ­people, mediated through fishing and other activities in the system (Levin et al. 2009). More specifically, EBFM draws its conceptual basis and operational approaches from both conventional fisheries management and ecosystem management.  Both approaches involve maintaining single-­species exploitation rates lower than would produce MSY and avoiding bycatch of nontarget species. In this re­spect, they depart from the traditional aim in fisheries management of maximizing productivity of an individual target stock in isolation (see figure 4.19). ­Because complex indirect interactions are the rule in ecosystem dynamics, and are often unpredictable, EBFM ascribes par­tic­ul­ ar importance to precautionary mea­sures that aim to avoid depleting stocks. The concept of EBFM has been controversial in part ­because it adds complexity to already difficult management situations (and Longhurst [2010] argues that EBFM has been embraced by some with a faith-­based zeal similar to that surrounding MSY). But many of the criticisms have been addressed (Patrick and Link 2015), and EBFM is u­ nder way in many systems. It has been mandated by the ­Great Barrier Reef Marine Park Act of 1981  in Australia, the Magnuson-­ Stevens Fishery Conservation and Management Act (1996, reauthorized 2006) in the USA, and the International Convention on the Conservation of Antarctic Marine Living Resources. Marine spatial planning focuses on integrating management of the multiple, often conflicting, ­human activities in the sea (Douvere 2008, Foley et al. 2010). Many of the challenges facing fisheries specifically and the ocean more generally derive in part from un­co­or­di­nated governance. In US territorial ­waters, fishing, mining, oil and gas extraction, marine mammal conservation, shipping, and other activities are regulated by more than 20 separate agencies, largely in isolation from one another. One approach to resolving this situation is a form of zoning, which has been used routinely on land for many years. Marine spatial planning aims to strategically site compatible activities together and separate incompatible ones in order to accommodate potentially conflicting uses, such as recreation, fishing, and energy generation. Improved incentive systems  for sustainable fishing can include dedicated access, such as catch-­ share programs. Managing ­people that fish can involve both restrictions, such as t­hose historically used in management, as well as incentives for making fishing sustainable.  One such approach involves ­limited access privilege programs (LAPPs), or “catch shares,” in which a secure share of fish is

Chapter 4 Introduction to the Anthropocene Ocean

allocated to an individual fisher, community, or association. The rationale is to short-­circuit the tragedy of the commons: ­because the shares are allocated before the season begins, fishers know how much fish they are allowed to harvest that year, so ­there is incentive to do it efficiently rather than in the counterproductive and expensive race to get the most fish pos­si­ble that results u­ nder the often complicated historical regulations. Catch-­share programs have shown strong promise to reduce fishery collapses (Costello et al. 2008). It’s worth noting, though, that the successes come mainly from developed nations with strong scientific capacity, so their transferability remains uncertain. Catch shares also limit access by definition and therefore entail difficult decisions about how shares are allocated. Fi­nally, any account of the ­future of fish must consider the rapidly growing global aquaculture industry, which equaled wild capture fisheries in total global production for the first time in 2014 (FAO 2016). Although aquaculture is often suggested as a solution to the environmental impacts of fishing, the situation is complicated for several reasons (Diana 2009). First, many farmed fish proposed for large-­scale ocean ranching—­notably salmon and tuna—­are carnivorous. Carnivore farming operations can cause even greater harm than wild capture fisheries ­because such predators require large quantities of food, which comes from forage fish that must be harvested from the ocean. Second, aquaculture operations often produce prodigious waste and pose risks of disease and parasitism to wild relatives. Third, some types of aquaculture, notably for shrimp, involve extensive conversion of sensitive wetlands. If aquaculture is to be part of a solution for sustainable fisheries rather than part of the prob­lem, it ­will need to focus on species low in the food chain, such as catfish and tilapia, avoid transmitting diseases and ge­ne­tic defects to wild fish, and reduce waste and habitat destruction. Making such changes could advance an impor­tant solution to h­ uman food security: models predict that, if designed properly, aquaculture could produce protein equivalent to current wild capture fisheries using less than 1% of the global ocean, with some production in nearly ­every coastal nation (Gentry et al. 2017).

Marine Biodiversity in the Anthropocene Species decline and extinction Widespread and ongoing declines of species during the Holocene and Anthropocene have raised the concern that Earth may be facing a sixth mass extinction, traditionally defined using fossil data as a period in which more than 75% of species dis­appear within an interval of less than two million years, sometimes much less (Barnosky et al. 2011). Comparing modern extinctions with estimates of ancient ones is fraught with challenges, including limitations of the fossil rec­ord and the biased fraction of modern taxa (mostly mammals) for which accurate estimates are pos­si­ble. Nevertheless, even highly conservative calculations suggest that average extinction rates during the ­human domination of Earth are well above preanthropogenic averages (Barnosky et al. 2011, Pimm et al. 2014). When species considered “critically endangered” by IUCN Red List criteria are assumed to go extinct within the next thousand years (a conservative assumption), Anthropocene extinction rates “would far exceed any reasonable estimation of the upper boundary for variation” in past extinction rates. On land, the princi­ple driver of changing biodiversity in modern times has been extensive habitat conversion and loss, whereas in the ocean it has so far been overexploitation; climate change w ­ ill clearly grow into a major driver of biodiversity change in coming de­cades on both land and sea (Pereira et al. 2010). Very few marine species are known to have gone globally extinct in modern times—­James Carlton and colleagues (1999) found unequivocal evidence for only eight species of birds and mammals. ­These ­were generally slow-­moving, large vertebrates like Steller’s sea cow and the g­ reat auk (see figure 4.13A). Other known or suspected extinctions involve highly specialized, small-­range species, like the sea slug Phyllaplysia smaragda, which was apparently restricted to seagrass epiphytes in a small region of Florida and has not been seen since 1980. Analy­sis of marine vertebrates

101

102

Ocean Ecology

and mollusks confirms that modern extinctions have fallen hardest on mobile, large-­bodied species, reflecting ­human exploita80 tion (figure 4.20), a pattern of selectivity not seen in previous extinctions over the last 60 million years (G. Vermeij 2004, 60 Payne et al. 2016). Large species are vulnerable partly ­because they are valuable commodities but also ­because their popula40 tions tend to grow slowly and so do not rebound well from harvesting (chapter 6). At the extreme end of this slow life history 20 spectrum are sharks and rays, which typically produce only one or a few young each year. The elevated vulnerability of large, mo0 7 36 478 82 88 530 1061 bile species also holds for many marine animals that are still with Number of species us but threatened to vari­ous degrees with extinction, including several w ­ hales, all sea turtles, bluefin tuna, and many sharks. Analyses of fishery data suggest that many large marine fishes Figure 4.20. ​Extinction vulnerability of marine species, have been reduced to a small fraction of their pristine abundance estimated according to the categories of the International Union ( J.  K. Baum et  al. 2003, Myers and Worm 2003, Baum and for the Conservation of Nature’s (IUCN) Red List. Animals that contact land during some portion of their life are distinguished Myers 2004). Such inferences from fisheries catch data are from ­those that do not. From left to right ­these are sea turtles; complicated by changing gear and markets and have been highly pinnipeds and otters; seabirds and shorebirds; sea snakes and controversial (Hilborn 2006). But t­hese are not the only marine lizards; cetaceans and sirenians; diadromous and indicators—­fisheries-­independent evidence leaves ­little doubt brackish ray-­finned fishes; cartilaginous fishes; marine ray-­ finned fishes; and marine invertebrates. The estimated number that large marine animals, and top predators in par­tic­u­lar, have of species in each group is shown below the graph (DD = data been strongly reduced throughout the ocean, with many rendeficient) (­after McCauley et al. 2015). dered ecologically extinct over the course of de­cades to a few centuries ( Jackson et al. 2001, McCauley et al. 2015). Although documented modern extinctions in the ocean are few, this is no cause for cele­bration. Estimating extinction threats for most marine animals is nearly impossible for lack of data (see figure 4.20). One exception is reef corals, which are con­spic­uo­ us, sessile, and widely studied. Among 704 coral species with enough data to assign a conservation status, 33% ­were judged to be at elevated risk of extinction by IUCN criteria, mainly as a result of bleaching and diseases driven by ocean warming (Carpenter et al. 2008). And the fraction of coral species that is threatened has risen rapidly over recent de­cades, even exceeding that of most terrestrial groups. A disproportionate number of the highly threatened coral species are from the Ca­rib­bean, paradoxically including even the two widespread species of Acropora (see figure 4.13F) that dominated reefs throughout the Ca­rib­bean for hundreds of thousands of years (Jackson 1992). Like the famous passenger pigeon and American chestnut, staghorn and elkhorn corals are a reminder that even superabundant community dominants can dis­appear. Terrestrial contact

Exclusively aquatic Percent unreviewed Percent DD Percent extinct Percent endangered

~1 5, 05 0 ~1 51 ,1 50

Percent of species

100

Functional consequence of declining biodiversity What are the consequences of t­ hese marine species declines? We return throughout the book to the question of how species loss and invasion affect ecosystems. A key theme is that the species declining fastest often play strong roles as key interactors in marine systems. ­These are of two kinds. First are large species high in the food chain, which often act as keystone species, with impacts large and disproportionate to their low abundance that ­ripple through ecosystems in multifarious ways (chapter 7). ­These include especially sharks, ­whales, sea otters, and large fishes. Ecological extinction of ­these large predators and megaherbivores often has pervasive cascading effects on the ecosystems they are part of (Estes et al. 2011, 2016). Second, foundation species anchor ecosystems by creating habitat for many other species. They include corals, seagrasses, and mangroves, all of which are in decline worldwide (Carpenter et  al. 2008, Waycott et al. 2009, Polidoro et al. 2010). Large, perennial organisms, such as reef-­building

Chapter 4 Introduction to the Anthropocene Ocean

corals, mangroves, seagrasses, kelps, and shellfish, are especially vulnerable to h­ uman disturbance and, ­under repeated impacts, often yield dominance to faster-­growing, opportunistic species, such as fleshy and filamentous algae, microbial mats, and certain clonal invertebrates. This is b­ ecause the foundation species generally have slower demographic rates and require better ­water quality than the weedy species that replace them. A common result is transformation of communities from structurally complex perennial forests to more homogeneous, unstructured bottoms that support lower diversity and often lower productivity. ­These disturbed conditions also often f­ avor invasive species. We explore ­these trends in more detail in chapters 11 and 12. Some have suggested that the functional consequences of such biodiversity losses have already passed a “planetary boundary” or tipping point that w ­ ill shift the earth out of the safe operating space for humanity, where loss of species seriously compromises key ecosystem pro­cesses and the resilience of the earth system to change (Rockström et al. 2009, Pimm et al. 2014). The idea of a planetary boundary for biodiversity loss is based on the impor­tant roles of diverse interacting species in mediating ecosystem pro­cesses, including production, nutrient and biogeochemical cycling, and climate regulation. ­There is good evidence that we are indeed committed to significant disruption of Earth’s ecosystem pro­cesses as a result of the extinction debt already incurred. But evaluating ­whether we are near a planetary boundary for biodiversity is challenging for several reasons (Mace et al. 2014). The rate of species extinction is a poor metric for assessing the approach to an ecosystem transition ­because t­ here are too few data to estimate it for most taxa and ­because it does not account for changes in abundance and community composition, which are key determinants of ecosystem functioning. More promising candidates for assessing the state of biodiversity in practice include indicators of ge­ ne­tic diversity, functional diversity, and the condition and extent of biomes and habitats.

Marine globalization Globalization—­the increasing connectedness of commerce, culture, knowledge, and demography— is now a central theme of ­human affairs. It is also an ecological theme of the Anthropocene. Ocean life is becoming more homogeneous across space as a combined result of climate change and human-­ mediated dispersal. Commerce is a strong driver of this pattern. Invasions of coastal marine systems by nonindigenous species have increased exponentially over the last two centuries, mainly as a result of commercial shipping but also fisheries-­related introductions (Ruiz et al. 2000) (figure 4.21). On land, urbanization and agriculture drive biotic homogenization by creating habitats that are similar across broad geographic regions and f­avor the spread of species adaptable to human-­built environments (McKinney 2006). The proliferation of coastal infrastructure can be expected to similarly accelerate homogenization of coastal marine communities (Firth et al. 2016). Climate change also contributes to the ocean’s biotic homogenization b­ ecause ongoing and expected warming is expanding ranges of most species, whereas so far only a few species are threatened by climate change (chapter 3). Climate-­mediated range expansion often synergizes with ­human introductions: for example, long-­term ocean warming favored increases in nonnative ascidians at the expense of native ascidians in fouling communities of the northeastern USA (Stachowicz, Terwin, et al. 2002). But range expansions have rarely led to extinctions of the resident species, at least so far. Models based on the expected velocity of climate change proj­ect that, by the end of this c­ entury, species range expansions ­will increase local species richness in most places while increasing the similarity of communities (reducing beta diversity) across space; but as warming continues into the next ­century, extinctions ­will begin to increase (García Molinos et al. 2016). What are the functional consequences of biotic homogenization? Species that are declining worldwide (“losers”) often have a distinct suite of traits compared with ­those that are increasing (“winners”). Across ecosystems, winners tend to have generalist resource requirements, often vegetative reproduction, and life history traits that promote rapid population growth—in short, they are weeds (McKinney

103

104

Ocean Ecology

and Lockwood 1999, Kolar and Lodge 2002). Marine examples include certain fleshy seaweeds and sea squirts that grow rapidly, attach to a variety of substrata, and reproduce clonally as well as sexually. Marine invaders also tend to be lower in the food chain, with some notable exceptions like the lionfish.

(A)

Number of invasions

105

70

Evolution in the domesticated ocean 35

Cr us ta M cea ol lu An sca ne lid a Al ga In e se c Cn ta id a Ch ria or da Br ta yo z Pr oa ot oz o Po a He rife lm ra in th es

0

(B)

Number of invasions

225

150

75

le

m

ul

ca

tip

na

ls

rc h

re s

na or

bi

ea

en m

on oc

he fis

ta

tro

s rie

pi ip sh

l

l

ng

0

(C)

Number of invasions

150

100

50

19

70

–1

99 9

9

40

–1

96

9 –1

93 19

10

90 9 –1

80 18

19

87 9

18

50

–1

9 84 –1

20 18

17

90

–1

81 9

0

Figure 4.21. ​Patterns of coastal marine species invasions in North Amer­ic­ a. (A) Taxonomic repre­sen­ta­tion of nonindigenous species. (B) Vectors of invasion. (C) Time course of invasions as estimated from first rec­ords of occurrence (­after Ruiz et al. 2000).

Divisions of geologic time are generally defined by the new forms of life that arise and spread to dominance such that they provide practical markers in the rock for the transition to a new world order. The ubiquity of domestic chicken bones as a marker of the dawning Anthropocene has already been mentioned. Although ­human activities have not (yet) caused evolution of entirely new species, we are driving pervasive genet­ically based changes—­that is, evolution—­throughout the biosphere, including marine life (Palumbi 2001). Th ­ ese changes result from what might be called domestication of the ocean. Domestication entails taming or cultivating an organism for ­human use. Domestication of animals and plants generally imposes strong, intentional se­lection for rapid growth and other traits favorable to ­humans, and throughout history intensive h­ uman management has driven substantial evolutionary change in managed species, largely through selective breeding, as well as in other species affected by ­human activity. In the ocean, the most consistent evolutionary pressure imposed by ­humans comes from fishing, which has initiated a colossal evolutionary experiment. H ­ umans take a larger proportion of prey populations than other predators do and preferentially target larger, reproductive adults (Darimont et  al. 2009), imposing widespread se­lection on life histories and phenotypes (chapter 6). Many exploited fish stocks are now smaller-­ bodied, mature at smaller sizes, and often faster-­growing than unexploited stocks. ­Because fishes generally have flexible growth and maturation schedules, partitioning ­these changes into ge­ne­ tic versus phenotypic components is challenging. But part of it surely involves directional ge­ne­tic change, that is, human-­induced evolution (Law 2000, Hutchings and Fraser 2008). Aquaculture, too, ­causes evolutionary change. For example, in farmed salmon stocks, genet­ically determined egg size declined by ~30% over less than two de­cades, and models indicate that females producing ­these eggs would have 24% lower fitness if returned to the wild (Heath et al. 2003). The larger concern for aquaculture is disease, which is the single most impor­tant threat to aquacultural production worldwide ( Jennings et al. 2016). Evolution is even faster in microbes than in fishes, and ­there are signs that aquaculture is driving pathogenicity in microbes in a manner parallel to the rapid rise and spread of antibiotic re­sis­tance that threatens ­human medicine.

Chapter 4 Introduction to the Anthropocene Ocean

Take the case of Flavobacterium columnare, which infects juvenile salmon. At a fish farm in Finland, this disease spread steadily over 23 years of salmon culture, evidently due to an evolutionary increase in bacterial virulence (Pulkkinen et al. 2009). This and other instances of disease evolution in cultured populations are facilitated by the practice of growing single species, often in genet­ically homogeneous populations, at high density, with superabundant food, in homogeneous, stable environments. In short, it’s an easy life, selecting for rapid growth and maturation in the host fish. But ­these are also ideal conditions for transmission of parasites and pathogens, which are also selected for rapid growth and reproduction and experience ­little fitness loss for killing their hosts, a combination that manifests as increasing virulence.

Novel ecosystems An overarching result of the many interacting h­ uman impacts on marine ecosystems—­climate warming, ocean acidification, intensive industrial fishing, spreading aquaculture, eutrophication, invasions, rapid evolution—is the emergence of novel ecosystems. Novel ecosystems (also called no-­analog ecosystems) are defined by combinations of species that do not other­wise occur together and thus have no evolutionary history of interaction, in systems strongly affected by ­human activity (Hobbs et al. 2006). Although they result from h­ uman influence, they do not depend on continued h­ uman intervention and comprise an intermediate category between wild ecosystems and t­ hose intensively managed by h­ umans. Often t­ hese are dominated by nonnative species that have established u­ nder abiotic conditions altered by ­human activities. A classic marine example is the San Francisco Bay estuary, which is heavi­ly dominated by nonnative species and bears ­little resemblance to the historical ecosystem (Cohen and Carlton 1998). The bay’s plankton community was completely restructured by the massive population growth of an introduced, suspension-­feeding clam and invasions of zooplankton species (chapter 11). Together with altered freshwater inflow and nutrient loading in the densely urbanized estuary, ­these biological changes caused a collapse of fish populations that depend on the zooplankton as food (Cloern and Jassby 2012). Novel ecosystems can also or­ga­nize around novel habitats created by h­ uman activities. In the ocean, the most obvious examples are physical infrastructure, from small docks to harbor jetties that increase availability of hard substrata in areas other­wise lacking them. This can change both benthic and pelagic components of the ecosystem. For example, one intriguing hypothesis for the global increase in jellyfish over recent de­cades involves worldwide growth in coastal infrastructure that provides substratum for the microscopic polyp phase of the jellyfish life history, which lives attached to hard bottoms (Richardson et al. 2009). Interactions among species may also differ systematically between natu­ral and artificial substrata: canopy-­forming macroalgae are less abundant on artificial substrata along shores of the Adriatic Sea despite good growth conditions ­because herbivore pressure is much stronger on the artificial structures (Ferrario et al. 2016). Artificial structures also f­ avor nonindigenous over native species, fostering their spread at regional scales (Airoldi et al. 2015). Such no-­analog ecosystems are expected to grow in importance as we also experience increasing no-­analog climate conditions (Williams and Jackson 2007).

­Future Directions Science for solutions Ecol­ogy has always harbored a tension between open inquiry in the quest for discovery—­“ blue skies” research—­and the application to solving practical prob­lems. ­These approaches are synergistic and most would argue that both are impor­tant for a robust scientific enterprise. But the need to mobilize scientific talent ­toward solutions in the Anthropocene is more pressing than ever before. ­Doing so

105

106

Ocean Ecology

involves the challenge of integrating economics, politics, and h­ uman natu­ral history more broadly with traditional ecol­ogy focused on the nonhuman components. This goal is reflected in the growing importance of research on social-­ecological systems (Ostrom 2009), which shows, for example, that effective marine conservation depends critically on strong social-­cultural institutions and local engagement in management (Cinner et al. 2016) as well as local ecol­ogy. In the realm of traditional natu­ral science, a central prob­lem of the Anthropocene is incorporating complex ecological (and sociocultural) interactions into workable models of how communities and ecosystems respond to environmental change. The challenge is finding the golden ­middle between two extremes. On one hand are relatively s­ imple, single-­species models common in environmental niche model projections of responses to climate change and the single-­species stock models that remain the work­horses of fisheries management. At the other extreme is the broad precautionary princi­ple that advises reduced h­ uman harvest or intervention based on generic and usually nonmechanistic guidelines. But models of intermediate complexity are pos­si­ble and can even be used in data-­poor situations, as evidenced by the growing application of ecosystem-­based fisheries management (Patrick and Link 2015). Success involves assessment to identify the key pro­cesses in a par­tic­u­ lar situation and incorporating them into the model (Levin et al. 2009). At the simplest conceptual level, this involves keeping fishing mortality low enough to prevent ecosystem-­wide overfishing, reducing bycatch, and avoiding destructive fishing methods that damage habitat, all of which we know how to do in princi­ple (Worm et al. 2009, Hilborn 2011). Questions for f­ uture research include what circumstances and environments require additional model components and which interactions are most sensitive (box 4.1). Another ave­nue ­toward improving models of global change is identifying generalizations about how changing temperature and other conditions influence interactions among species. Promising approaches to this prob­lem focus on how functional traits influence species interactions and on impor­tant demographic transitions (e.g., larval settlement, age at maturity) that have large effects on population pro­cesses, interactions among species, and the spatial scales over which t­hese play out (Harley et al. 2006, Bruno et al. 2015). Such approaches could transcend the idiosyncrasies of individual species, but it remains unclear ­whether such a general functional ecol­ogy of environmental change is achievable (Kordas et al. 2011). As human-­built structures grow to constitute an ever larger fraction of the marine environment, another frontier is adapting ecol­ogy to incorporate emerging new anthropogenic biomes. The proliferation of artificial structures in coastal areas has been likened to the “urban sprawl” familiar in our terrestrial surroundings. Understanding and managing the ecol­ogy of t­ hese systems increasingly requires both classical ecological and engineering princi­ples, and more purposeful planning, design, and operations (Dafforn et al. 2015). How specialized versus generalized are marine organisms in habitat requirements? What types of development and artificial structures are most hospitable to a broad range of native marine life and to harmonizing nature conservation with ­human needs? How can marine spatial planning minimize injurious impacts and maximize benefits? One promising example involves integrated aquaculture, which ­couples cultures of functionally complementary species to maximize recycling and productivity. Common applications involve culturing fish or shrimp and capturing nutrients in their waste stream with terrestrial vegetables, algae, or shellfish (feeding on algae) that convert them to new biomass (Neori et al. 2004). When designed properly, such integrated polyculture systems can achieve high biomass production with low waste output.

Policy for solutions ­ uture directions in Anthropocene ecol­ogy involve questions of philosophy, sociology, and politics F as much as natu­ral science ­because ­these are the domains where decision making ­will ultimately determine the ­future of the ocean, the earth, and ­human well-­being. Most fundamentally, what do we as

Chapter 4 Introduction to the Anthropocene Ocean

107

Box 4.1. ​Can ocean fertilization save us from a climate disaster? As impacts of anthropogenic climate change grow broader and more sobering, attention has turned to the pressing question of what to do about it. Modern civilization’s fundamental dependence on fossil fuel, and the earth system’s slow response to CO2 forcing, has already committed us to de­cades of rising CO2 concentrations and unpre­ce­dented climate warming by the end of the twenty-­first ­century. Major climate-­mediated reor­ga­ni­za­tion of the earth system cannot be prevented by incremental improvements in renewable energy and recycling alone. Aggressive global-­scale interventions—­geoengineering—­ are now being seriously considered (Vaughan and Lenton 2011). A highly publicized example involves proposals to fertilize the ocean on a g ­ rand scale in order to draw down CO2 and, ideally, boost fish production in the bargain. The premise for such plans came from the once controversial suggestion that productivity throughout much of the ocean is l­imited by the availability of dissolved iron. Iron is needed in trace quantities by phytoplankton for photosynthesis, and its low solubility in seawater keeps it vanishingly rare in large areas of the ocean away from continental sources (chapter 10). This suggested to the oceanographer John Martin that iron scarcity might explain the paradox of so-­called high-­nitrogen low-­chlorophyll (HNLC) regions in the open ocean. Martin developed and began testing this iron hypothesis. In 1988 he famously told an audience “Give me a half tanker of iron, and I w ­ ill give you an ice age.” The idea ignited a long-­running debate, but the iron hypothesis was eventually vindicated: fertilization with iron of a swath of ocean off the Galapágos indeed caused phytoplankton to bloom, as did subsequent experiments in several other regions. Martin’s bold hypothesis and provocative statements soon attracted the attention of a public increasingly concerned about climate change. The 1997 Kyoto Protocol required countries to substantially reduce their carbon emissions and proposed tradable carbon credits for ­doing so. Several entrepreneurs and companies saw ocean iron fertilization as a way to achieve this and make some money in the pro­cess. ­These proposals ­were met with widespread alarm among oceanographers who emphasized, rightly, that large-­scale fertilization of the ocean could produce a host of unintended consequences. If we have learned anything about ecol­ogy in the last c­ entury, it is that ecosystems are complex webs of interactions that often respond unpredictably to disturbance. In fact, l­ater ocean fertilization experiments stimulated blooms of toxic algae rather than fish food (Trick et al. 2010). Algal blooms often decompose before they sink to the depths required to sequester carbon, and fertilization

may actually exacerbate climate change as it can also generate methane, a far more potent green­house gas than CO2. The ocean fertilization controversy came to a head in 2012 when a group representing the Haida First Nation of Canada dumped 100 tons of iron sulfate into the Pacific Ocean off British Columbia in an effort to boost salmon production and collect carbon credits in the pro­cess (Tollefson 2012). The action was widely condemned for proceeding before having scientific evidence that it was safe and for being on shaky l­egal and economic ground. It remains unclear w ­ hether this par­tic­u­lar experiment had the desired outcome of boosting local salmon production. The intensive scientific study of the iron hypothesis provided a unique opportunity to explore the complex issues involved in science and governance of proposed geoengineering schemes (Boyd and Bressac 2016). The several ocean fertilization experiments conducted to date have shown quite variable results, likely depending on the composition of the phytoplankton community prior to fertilization, depth of the mixed layer, abundance and composition of grazing zooplankton, and duration of contact between the fertilized w ­ ater and the atmosphere. But even using the most optimistic estimates of carbon export and atmospheric uptake efficiencies, research suggests that long-­term, large-­scale iron fertilization could only offset less than 10% of projected emissions (Williamson et al. 2012). Scientific consensus has thus emerged that ocean fertilization for the purposes of carbon drawdown is a bad idea: it would likely be in­effec­tive and entail a range of unintended consequences (Strong et al. 2009). As a result, the UN Convention on Biological Diversity passed a resolution in 2008 requesting member states to restrict ocean iron fertilization to basic science experiments. Similar misgivings apply to other ocean fertilization schemes, including a proposal to fertilize the ocean around the Philippines with urea (Glibert et al. 2008). In the end, the best science indicates that large-­scale ocean fertilization is unlikely to substantially reduce carbon in the atmosphere, and its unintended consequences are worrisome and uncertain. But this story emphasizes a broader existential prob­lem: h ­ umans now have the power to control climate on a planetary scale; indeed, we are already influencing it unconsciously. If not ocean fertilization, what? Inaction is also a choice, for business as usual, which also has undesired ends and unanticipated consequences. What specific policy actions ­will we accept to make a real difference in reducing the serious and growing threats of climate warming and sea level rise?

a global society want from the oceans? Are we content to treat them as essentially mines, highways, and planetary factory farms for inexpensive fish fillets? Or do we want something more—­a biodiverse, resilient ecosystem that provides a range of ecosystem ser­vices? Even more impor­tant, what are we willing to give up to realize our choice? The ­future of marine fisheries is a microcosm of the ­future of h­ uman society generally. The basic questions that need attention are about our values: How

108

Ocean Ecology

impor­tant is abundant and inexpensive food, and the jobs associated with providing it, relative to the many other benefits we receive from healthy ecosystems? It is a s­ imple fact that t­ hese benefits cannot all be achieved si­mul­ta­neously. How impor­tant is the availability of abundant fish now relative to long-­term maintenance of a functioning ecosystem that can provide fish to our grandchildren and beyond? Answering ­these questions is not easy for two reasons: (1) they involve personal and cultural values, many of which are beyond the scope of scientific analy­sis, and often only dimly understood by ­those who hold them; and (2) ­these values differ, sometimes extremely, among dif­fer­ent stakeholders. A related challenge is that modern systems of governance are not well suited to addressing prob­ lems at global scales and long time spans that affect primarily ­future generations (Reid et al. 2010). Nevertheless, science can be applied to solve prob­lems of h­ uman social be­hav­ior, as it has done in ­every other realm of inquiry. A frontier for research and policy in the Anthropocene is understanding how ­human motivations and be­hav­ior influence such common-­pool resources and how that understanding can be leveraged by policy to foster sustainable practices in our globally connected world (Miller et al. 2014). Even in the realm of natu­ral science, fishery management neatly encapsulates a central challenge for the Anthropocene: prudent management of a fishery requires knowledge at e­ very level of organ­ ization of nature, from functional biology of individual organisms, through populations, to the communities and ecosystems (including h­ umans) in which they are embedded. Even sophisticated, data-­ rich models of t­hese populations are challenged in practice by the complex and often nonlinear responses of the organisms to an environment, notably climate, that is changing on multiple scales of space and time. And ­because we rely heavi­ly on fish for food and commerce, the stakes are high for getting it right. Some have argued that recognizing fisheries as complex social-­ecological systems requires radically dif­fer­ent approaches to management that emphasize self-­organization, learning, and adaptation (Mahon et al. 2008). How do we best evaluate and implement such recommendations?

Reasons for cautious optimism The state of the world ocean is sobering—­but ­there is hope. We know how, in princi­ple, to solve many of the challenges described in this chapter, and some have been successfully implemented in practice. The area of the global ocean u­ nder some form of protection has surged in the last two de­ cades, as has restoration of degraded coastal habitats, and many charismatic megafauna have begun to recover from dangerously low populations (figure 4.22). ­Those success stories are growing, and they offer valuable insights into what works and what d­ oesn’t in conservation and management. Achieving a sustainable ocean ecosystem that supports humanity into the long term ­will continue to require research in traditional marine ecol­ogy, but the major frontier lies especially in ecol­ogy of the ocean’s keystone species Homo sapiens. This w ­ ill involve basic research both in h­ uman behavioral and social ecol­ogy as well as in applied forms of decision theory, ecological economics, and politics. If humanity can rally to mobilize the knowledge we already have, we could make substantial pro­gress ­toward rebuilding many degraded ocean ecosystems, and the critical ser­vices they provide, by the ­middle of the pre­sent ­century (Duarte et al. 2020). As is true for stabilizing carbon emissions, we have sufficient technology and understanding of good governance to achieve sustainable fisheries—in princi­ple (Worm et al. 2009, Levin et al. 2009, Hilborn 2011, Patrick and Link 2015, Costello et al. 2016). At the most basic level, sustainable fisheries involve taking fewer fish from the ocean. But beyond that it quickly gets complicated. How many fewer? How does changing climate, both the long-­term trend and the background of decadal regime shifts, influence sustainable harvest targets? What management tools and governance structures are most effective in protecting the ecosystems that produce fish? ­These questions are at a dynamic interface of natu­ral and social sciences that offers some hope for the ­future.

Chapter 4 Introduction to the Anthropocene Ocean

(A)

109

(B) Restoration sites

MPAs Before 1990

1990–2000

2000–2010

2010–2018

Coral reefs Oyster reefs

Before 1990 Before 1990

1990–2000 1990–2000

2000–2010 2000–2010

2010–2018 2010–2018

1,768 Oyster reefs restoration efforts across US East coast

(C) Number of individuals (n) or area (ha)

10,000 8,000 6,000 4,000 2,000 0 1950

1960

1970

1980

Humpback whale (n), East Australia; 400 Sea otter (n), West Canada; 70 Northern elephant seal (births), West USA; 13 Guadelupe fur seal (n), Mexico; 51 Green turtles (n × 10), Japan; 1,190

1990

2000

2010

Leatherback turtle (n × 10), Virgin Islands; 190 Seagrass area (ha × 100), Connecticut USA; 0.1 Baltic Sea cormorants (n pairs × 20); 3,304 Baltic Sea grey seals (n × 3); 3,645

Figure 4.22. ​Some good news for marine life. Efforts to safeguard and restore marine biodiversity have increased dramatically since the mid-­twentieth c­ entury, including (A) the global extent of marine protected areas (MPAs), (B) the number of proj­ects to actively restore coral and oyster reefs, and (C) successful efforts to rebuild populations of threatened megafauna (­after Duarte et al. 2020).

Like the prob­lem of stabilizing climate (Pacala and Socolow 2004), marine conservation ­will require a portfolio of approaches deployed si­mul­ta­neously and customized to local conditions. Marine protected areas are one well-­studied piece of the puzzle. Protection from fishing has proved effective in stabilizing and rehabilitating many ecosystems (Lester et  al. 2009, Sala and Giakoumi 2018), and the area of ocean protected has risen steeply in recent years, although this is not a panacea (Fox et al. 2014). Economic analy­sis estimates that a global network of protected areas conserving 20%–30% of the world ocean would cost between $5 billion and $19 billion annually and could create a million jobs, among many other less easily monetized benefits. This price tag is lower than, and could be offset by, the perverse subsidies that governments already pay to their industrial fisheries (Balmford et al. 2004). ­These subsidies—­government payments to industrial fishing that are unintentionally detrimental to fisheries and economies—­have been quantified and their impacts publicized (Sumaila et al. 2010). Moreover, restricting access to insurance can shut down vessels involved

110

Ocean Ecology

in illegal fishing (Miller et al. 2016). Improved governance and regulation of fisheries can provide additional solutions. Generally, ­these solutions involve changing incentives, both economic and social norms, to align conservation goals with economic benefits (Lubchenco et al. 2016). Solutions also are facilitated by better access to data. For example, by making satellite data publicly available, programs such as Global Fishing Watch have crowdsourced apprehension of illegal fishing vessels. A promising general approach to solving complex challenges in social-­ecological systems is to identify “bright spots”—­systems that are in better shape than expected based on conditions. A search for bright spots among the world’s coral reefs reveals that t­ hose in better than expected condition are in communities with strong sociocultural institutions, such as traditional taboos and rights of access, strong local engagement in management, and significant dependence on local marine resources (Cinner et al. 2016). From a broader perspective, t­ here is reason for hope in the recognition that many attitudes and social mores are changing nearly as rapidly as technological innovation. We have overcome seemingly insoluble prob­lems before: slavery, once the mainstay of economies throughout the world, has essentially dis­appeared. ­Human attitudes ­toward gender and race equality, the welfare of ­children and animals, warfare and vio­lence generally have changed fundamentally in the last few centuries (Pinker 2011). Popu­lar awakening of environmental consciousness in the 1970s led to “dolphin-­safe” regulations on tuna fishing. More recently, nearly 100 countries have banned shark fishing or sale of fins and many airlines refuse to transport them. In 2015 more than two million square miles of ocean was put u­ nder official protection, more than in any year in history. This includes the first formal agreement among nations to protect an area of the high seas beyond national jurisdictions, which in 2016 established the largest protected area in the world, 1.5 million square kilo­meters in the Ross Sea of the Southern Ocean, most of which ­will remain ­free of fishing and mineral mining.

Summary ­ umans have modified Earth’s landscapes, flora, and fauna for millennia, but impacts increased exH ponentially in intensity and scope during the G ­ reat Acceleration of global population and economic growth that began in the mid-­twentieth ­century. The current age is now recognized as a distinct geological epoch, the Anthropocene, defined by global domination of earth system pro­cesses by a single species, Homo sapiens. The Anthropocene formally commenced with the G ­ reat Acceleration, marked in the rock rec­ord by fallout from the first nuclear weapons tests in the 1950s. The rise to global dominance by h­ umans has been attributed to our large brains, high behavioral adaptability, and the propensity—­unique among animals—to form large, complex social groups with nonkin. Th ­ ese traits fostered sociocultural niche construction as a central theme of ­human pro­gress and drove cultural and technological innovation at rates ­orders of magnitude faster than ge­ne­tic evolution. Over the last ­century especially, the exponentially increasing growth of population and economic activity powered by fossil fuel has transformed the earth system. Globally increasing CO2 concentration resulting from fossil fuel combustion drives continuing climate warming and ocean acidification, and ­these changes interact with massive increases in biologically usable nitrogen, overharvesting of large animals, and global homogenization of biotas through trade and transport. Ocean warming, in concert with other stressors, is changing the distributions of species, altering the strength and nature of key community interactions, and creating novel ecosystems. Ocean acidification threatens ecosystems dominated by calcifying organisms, including coral reefs and certain pelagic systems. Industrial nitrogen fixation has strongly increased biological production in estuaries and the coastal ocean, leading to replacement of perennial foundation species adapted to oligotrophic conditions with fast-­growing algae, much of which remains ungrazed and decays to form low-­oxygen dead zones. Whereas many of t­ hese climate-­related impacts are just emerging, ­human harvesting of fish has been ­under way for millennia and has already transformed the ocean; 80% of harvested marine fish stocks are ­either fully exploited

Chapter 4 Introduction to the Anthropocene Ocean

or overfished, and the rate of capture from the world ocean appears to have reached maximum capacity in the 1990s. Intense, sustained fishing has shifted fish populations to lower densities, smaller body sizes, and e­ arlier maturity, in some cases through evolutionary change, and has shifted communities to dominance by smaller species and weaker top-­down control. Projections of an early earth system model in 1972 have generally been upheld by trends over the subsequent 40 years of increasing population growth and nonrenewable resource depletion and decreasing food availability, adding weight to the model’s prediction that the world economic system could collapse during the twenty-­ first ­century ­unless major changes in technological efficiency and social policy are implemented. The central challenges for h­ uman society in the Anthropocene are stabilization of global population, transition from fossil fuels to renewables as our primary energy source, and integration of biophysical bound­aries into economic theory and policy.

111

5 T

112

Organisms

he individual organism is the fundamental unit of biology and ecol­ogy. Individual plants transform light and inorganic materials into life, other organisms consume that living ­matter and energy and transform it into their own forms of life, and still ­others close the loop by decomposing it back into inorganic materials. The activities of organisms link the dif­fer­ent levels and pro­cesses of ecol­ogy, from environment, to population, to community and ecosystem, interacting with one another, creating habitats, and altering the composition of seawater and atmosphere (figure 5.1). In this chapter we review the basic features of how individual organisms work, which provides a foundation for understanding their interactions in populations, communities, and ecosystems. We consider the organism from two complementary perspectives: functionally, as a chemical reactor; and historically, as the product of an evolutionary pro­cess. First, like the rest of the universe, living organisms are composed of m ­ atter, differentiated into chemical ele­ments and powered by energy. We review how energy and materials are captured, stored, and used by organisms, how the rates of ecological pro­cesses are affected by the fundamental environmental and biological variables of temperature and body size, and the special challenges and solutions of autotrophs (plants) and heterotrophs. We then turn to the unique feature that distinguishes life from nonliving m ­ atter: the capacity of living organisms to adapt to the environment through changes in their ge­ne­tic code or genome—­a set of editable instructions that guide the organism’s self-­construction, the way it captures energy and materials from the environment, and its other affairs. That mutable code allows regular updates to the population’s operating system, so to speak, adapting it to changing environments. ­These adaptations define the organism’s niche and the suite of traits that mediate how it functions and interacts in communities and ecosystems. Like a snowflake, but much more complex, each individual organism’s ge­ne­tic code is unique. That code interacts with the environmental conditions the organism experiences to direct its development via a complex web of biochemical pro­cesses. Each organism is also one of a dispersed collection of individuals, similar to one another as a result of common ancestry, that make their living in much the same way and, in most cases, share and mix their genes through sexual reproduction. This collection of individuals is called a population (chapter 6). The continuous ge­ne­tic adaptation of organisms to variation in Earth’s environment over billions of years has produced the diversity of life forms that exists t­ oday. The unique functional features of an organism, its traits, guide its interactions with the environment and other organisms to produce the structure of its community—­the abundances and interactions of species in an area (chapters 7, 8). Over larger spatial scales and ranges of environments, the unique traits of organisms become less impor­tant in shaping the distribution of life, whereas the basic physical and chemical constraints common to all life become more so; ­those general relationships are the subject of macroecol­ogy, which we consider at the end of the chapter.

Chapter 5 Organisms

Carnivores Biomass Productivity Community composition Genetic composition Behavior

Herbivores

Detritivores

Biomass Productivity Community composition Genetic composition Behavior

Biomass Productivity Community composition Genetic composition Behavior

Plants

Detritus

Biomass Productivity Community composition Genetic composition

Supply rate Quality

Decomposers Biomass Productivity Community composition Genetic composition

Abiotic resources Supply rate Quality

Figure 5.1. ​Major pathways of energy and influence among producers and consumers in ecosystems. Within each functional category are characteristics that influence the flows in both bottom-up (black arrows) and top-­down (blue arrows) directions. Green and orange boxes show, respectively, the grazing pathway based on living plants and the detrital pathway leading from dead plants.

Building Blocks of Life Despite their uniqueness, living organisms are composed of the same chemical ele­ments as nonliving ­matter and are powered by the same sources of energy. Thus, we can use an engineering perspective to understand how organisms work, uniting the disparate pro­cesses of ecol­ogy from molecular to ecosystem levels. The organism’s engine can be thought of as a complex and diffuse chemical reactor, a carefully regulated web of chemical reactions that require materials from the environment as building blocks and energy to construct and maintain its body. ­Those materials and energy are captured from the environment and allocated among many competing demands of the organism’s life pro­ cesses. As it goes about its activities, it burns fuel and transforms the materials, releasing (recycling) some of them back to the environment, generally in modified form (see figure 5.1).

Ecological stoichiometry Linking biological pro­cesses among dif­fer­ent organisms and the nonliving environment requires a common currency. The ele­ment carbon is the most widely used currency in ecol­ogy for two related reasons. First, it is the backbone of all biochemicals and thus a major component of all organic m ­ atter,

113

Ocean Ecology

(A)

(B) 100

Abundance in humans (% atoms)

1

P

B

10

–6

10–8

Mn

Ca Na K

Fe

S Mg

Zn Cu Al Pb Ti Cr Ce Sn SeMn Ge Co Ni Hg Ag Cs As Sb La Nb Sc Y Zr Ga Ta Tl Be In Th Bi V Au U Sm W

10–10

Cl

F

Rb

B

Li

H

O

N Si

10–2 10–4

100

H C

Essential for all mammals Essential for some mammals Not regarded as essential

10–12

Abundance in bacteria (% atoms)

114

O

C

1 P

N Na

10–2 Mn Fe

10–4

Cu Co Cr Cd

10

–6

10–8

S

Mg

Al

Cl

Zn B As

Se

10–10 10–12

10–12

10–10

10–8

10–6

10–4

10–2

Abundance in seawater (% atoms)

1

100

10–12

10–10

10–8 10–6 10–4 10–2 Abundance in seawater (% atoms)

1

100

Figure 5.2. ​Elemental composition of life. Abundances of dif­fer­ent ele­ments in (A) h­ umans and (B) bacteria versus ­those in seawater. Note in both cases the strong correlations between environmental and organismal concentrations, and the elevated concentrations of nitrogen, phosphorus, and many metals in organismal tissue (­after Chopra and Lineweaver 2008).

living or dead. The relative proportions or stoichiometry of dif­fer­ent ele­ments in an organism’s body is constrained within relatively narrow limits, although it can differ dramatically among dif­fer­ent kinds of organisms. As a result, organic carbon content is a more accurate mea­sure of living biomass than is wet or dry mass ­because organisms often differ greatly in content of ­water and inorganic materials used in shell and bone (for example, a stony coral is mostly rock, whereas a jellyfish is mostly w ­ ater). The second reason why carbon is a universal currency is that carbon-­carbon bonds in biochemicals are the principal means of storing energy in the tissues of living organisms, mainly as carbon-­rich fats and oils. Thus, carbon content can be used as a common mea­sure of biologically usable energy content of both living organisms and dead organic materials in the environment. Organisms consist of more than carbon (figure 5.2), but of roughly 90 naturally occurring ele­ ments, only 4 (C, H, O, N) form the major building blocks of living tissues, together comprising about 99% of biomass. Another 7 (Ca, P, K, S, Na, Cl, and Mg) are essential in all living organisms but occur in much smaller quantities (Sterner and Elser 2002). Perhaps 20 other ele­ments, mostly metals, are required in very small (“trace”) amounts as cofactors in enzymes and other biological molecules. Among all of ­these, nitrogen is of comparable importance to carbon. Nitrogen is a major component of protein, which is the building block of the enzymes that catalyze and direct biochemical traffic within all organisms, and of muscle in animals. Despite the fact that Earth’s atmosphere is nearly 80% nitrogen, the elemental N2 in air is not reactive and therefore not usable by most organisms, which must obtain their nitrogen in other forms. As a result, usable nitrogen has been in short supply for much of Earth’s history, at least during the last two billion years of oxygenated ocean and atmosphere. Thus, while carbon serves as a good mea­sure of state variables like biomass and energy content, nitrogen is often a better predictor of fluxes of energy and materials through organisms and ecosystems ­because growth rates depend on protein synthesis, which is often ­limited by availability of nitrogen. Other ele­ments may limit growth in other situations—­phosphorus commonly limits plant growth in freshwater, and iron limits phytoplankton growth in large areas of the ocean far from continental sources of this micronutrient (De Baar et al. 1995).

Chapter 5 Organisms

Herbivory rate (% production)

In the early twentieth ­century Alfred Redfield (1934) first recognized a strong similarity between the relative abundances of ele­ments in seawater and in living plankton across wide areas of the ocean. The chemical composition of organisms has been a lens for understanding ocean ecosystem pro­cesses ever since. More recently, this approach has been elaborated and generalized as the field of ecological stoichiometry (Sterner and Elser 2002), which seeks to understand biological pro­cesses from molecular to ecosystem scales in terms of the relative abundances of biologically impor­tant ele­ ments and how ­these ratios among ele­ments influence ecological pro­cesses. The most central princi­ ple of ecological stoichiometry involves the dif­fer­ent concentrations of key ele­ments in the tissues of plants and animals, which have far-­reaching consequences for ecological pro­cesses. The differences in stoichiometry between plants and animals stem from the dif­fer­ent biochemical building blocks of their tissues. The physical structures of plants, both their mechanical support and vascular systems, are composed of carbon-­rich materials, notably cellulose and, in seed plants, also lignin. In contrast, animals are largely made of nitrogen-­rich protein. Thus, primary producer tissues tend to be significantly lower in nitrogen than t­ hose of the herbivores that eat them, generating a stoichiometric mismatch between consumer and prey. As a result, fitness and growth of herbivores and detritivores are often l­ imited by nitrogen, and their food (A) se­lection is tuned ­toward maximizing nitrogen intake. 60 Such stoichiometric mismatches between plants and herbivores Phytoplankton differ strongly among ecosystems dominated by dif­fer­ent types of plants, resulting in big differences in energy and materials fluxes among 40 ecosystems (figure 5.3) and even across regions of the pelagic ocean (Cebrián et al. 1998, Martiny et al. 2013). Primary producers of high Marshes nutritional quality (high ratios of N:C and P:C in their tissues) are Seagrasses 20 more heavi­ly grazed, more rapidly decomposed, produce less ungrazed detritus, and bury proportionally less carbon in sediments than do less Mangroves nutritious producers. The latter include the flowering plants that dominate coastal marshes, seagrass beds, mangrove forests, and most terres0 0 2 4 trial ecosystems (Cebrián 1999, Cebrián et al. 2009) (figure 5.4; chapPlant N content (%) ter  11). The primary reason for ­these ecosystem-­level differences in rates and fates of production involves the quality of the plant biomass as (B) food for animals. 0.1

115

The first step in essentially all ecological pro­cesses is fixation of organic compounds by algae and plants via photosynthesis. Th ­ ese organic compounds are used by organisms in two ways: they store energy for d­ oing the organism’s work, and they provide the structural building blocks of biomolecules. But a host of other ele­ments, mentioned above, are also required to build and sustain functional cells. The principal two are nitrogen, a major component of proteins that constitutes around 7% of cellular biomass, and phosphorus, which is a component of nucleic acids (DNA and RNA) and ATP, the currency of energy transmission in cells. In seawater, nitrogen is found primarily in four forms: nitrate (NO3), nitrite (NO2), ammonium (NH4), and urea. The first two of t­ hese are oxidized forms, which are more stable than the other forms of nitrogen in seawater. Nitrate is the most stable of all, and therefore accumulates in deep ­water as organic ­matter decomposes into inorganic components

Phytoplankton Decomposition rate (day–1)

Nutrient uptake and use

6

10–2 Seagrasses Marshes Mangroves 10–3

0

2

4

6

Plant N content (%)

Figure 5.3. ​Ecosystem rate pro­cesses depend on primary producer stoichiometry. Figures show percentage of net primary productivity (A) consumed by herbivores or (B) decomposed as a function of autotroph nitrogen content across several major types of ecosystems (­after Cebrián et al. 1998).

116

Ocean Ecology

(A)

(B) Mangroves Herbivory (52%)

Marsh plants

*

Seagrasses

27.8

***

Macroalgae

***

Coral reef algae

**

Benthic microalgae Coastal phytoplankton Oceanic phytoplankton

Decomposition 19 (36%)

*

Total NPP 53

Redistribution (10%)

* 0.65 Stored (1.2%)

** *

0 10 20 0 Net primary production (gC m–2 d–1) Mangroves

10 20 Decomposition (% NPP)

0.15 Buried (0.25%)

*

Marsh plants

*

Seagrasses Macroalgae

** *

Coral reef algae Benthic microalgae Coastal phytoplankton Oceanic phytoplankton 0

10 20 Herbivory (% NPP)

0

10

20

Storage (% NPP)

Figure 5.4. ​Magnitudes and fates of net primary production in the ocean: (A) among dif­fer­ent types of marine systems and producers and (B) summary for the world ocean in Pg C yr−1 (­after Duarte and Cebrián 1996). NPP = net primary production.

in the absence of photosynthesis. Nitrate and nitrite must be chemically reduced (by the addition of a proton, or equivalently a H+ ion) to be assimilated into the cell, which requires energy. Hence ammonium and urea, which are produced by the excretion of heterotrophs, require less energy to assimilate and are therefore preferentially taken up from the environment by phytoplankton. In the open ocean, NO3 comes primarily from upwelling and mixing of deep w ­ ater (in coastal regions, much NO3 also enters from terrestrial sources), whereas NH4 and urea come from excretion by heterotrophs, including animals, protozoan micrograzers, and bacteria in the immediate vicinity of the phytoplankton. ­Because most NH4 and urea originate in plant tissue that has been grazed locally and excreted by herbivores, production fueled by t­ hese reduced forms is often called regenerated production, whereas production fueled by inputs of NO3 is termed new production (Dugdale and Goering 1967). Availability of usable nitrogen is therefore an impor­tant predictor of rates of production and transfer of materials through marine (and other) food webs, and ­these fluxes have historically been ­limited by nitrogen availability—­I say “historically” b­ ecause h­ uman activities have massively increased the input of usable nitrogen into the biosphere, with wide-­ranging environmental consequences (chapter 4). Particularly at lower levels of the food web, materials bud­gets of individual organisms and of ecosystems are better characterized by the availability and fluxes of nitrogen than of carbon.

Chapter 5 Organisms

Iron The chemical nutrients that limit marine primary production vary regionally. Throughout much of the history of oceanography, a vexing puzzle was the existence of vast regions of the ocean with apparently abundant nitrogen and phosphorus but anomalously low phytoplankton abundance—­so-­ called high-­nitrogen low-­chlorophyll (HNLC) regions. The solution to this puzzle began to emerge with one of the major discoveries of latter twentieth-­century oceanography—­that trace nutrients, particularly iron (Fe), limit primary production across substantial areas of the open ocean. Iron is a component in many parts of the cellular machinery, notably the electron transport chains of respiration basic to all cells, but is especially impor­tant for photosynthesis, which requires iron for chlorophyll synthesis and 22 Fe atoms for each photosystem complex. The microbes active in the nitrogen cycle, including NO3 and NO2 reduction and N2 fixation, are especially hungry for iron, meaning that the key ecosystem pro­cess of nitrogen fixation is especially prone to iron limitation (Twining and Baines 2013). In the 1980s John Martin and colleagues proposed that, b­ ecause of t­ hese requirements, phytoplankton need much more iron than is available in open-­ocean w ­ aters and that their anomalously low biomass in HNLC regions is explained by iron deficiency (Martin and Gordon 1988). Subsequent work confirmed that the iron requirements of phytoplankton contrast markedly with its availability in the ocean. At the pH of average seawater, iron exists primarily as Fe3+, which has very low solubility and tends to complex with particles and sink rapidly out of the euphotic zone. Unlike dissolved NO3, the primary form of nitrogen in seawater, iron species are not easily returned to the surface ocean by upwelling. Instead, the major source of iron in the ocean ­water column is the fine dust of soil blown off the continents, which leaves regions of the ocean distant from continents, such as the Southern Ocean and the North Pacific, with very l­ ittle of this essential micronutrient. We consider the ecol­ogy of ­these HNLC regions in more detail in chapter 10.

Powering Life Living is work, and work requires energy. Ultimately, the energy of life comes from the sun’s radiation, which is captured by autotrophs (i.e., algae and plants) in energy-­rich biochemicals that are then available to herbivores and detritivores. The pro­cessing of energy and materials by organisms is called metabolism, and, like all reactions, it follows the basic princi­ples of physics and chemistry. ­These princi­ples include, most fundamentally, the first law of thermodynamics, which states that neither ­matter nor energy is created nor destroyed, only transformed among states. Mass balance is a way of accounting for ­these transformations, recognizing that what goes into a system, ­whether an individual organism or an ecosystem, must come out somewhere in some form. Metabolism provides a set of princi­ples that unite phenomena at scales of organ­ization ranging from molecules to ecosystems—­ metabolic ecol­ogy (Brown et al. 2004). Considering organisms in a physical-­chemical framework helps integrate biological pro­cesses across levels of organ­ization, from molecules to the earth system, ­because the terms in an individual organism’s energy bud­get are similar to ­those of an ecosystem and are mea­sured in the same units; so organismal pro­cesses can be straightforwardly plugged into ecosystem models. For example, a major motivation for research in marine ecol­ogy has always been food—­managing fisheries and the food webs that support them. In this context, the central phenomenon of interest is fish biomass production, or yield in fisheries lingo. Primary production ultimately constrains the productivity of the ecosystem as a ­whole, and at each step in a food chain production of prey constrains the productivity of the predators that eat them. Constructing energy and ele­ment bud­gets for ­these pro­cesses is a basic tool in fishery management, as we learn in chapter 9.

117

118

Ocean Ecology

Autotrophy For plants (and algae), metabolism begins with the capture of energy from the environment in the form of light, which they then use to build organic m ­ atter—­their tissues—­from organic molecules. Primary production is so called b­ ecause it creates living from nonliving m ­ atter. But plants can also be considered consumers in that they require light and nutrients to grow (see figure 5.1). In the sea, as on land, the limiting ­factors for growth are primarily light from above and inorganic nutrients from below. By far the most impor­tant type of primary production is photosynthesis, the formation of living organic ­matter from inorganic components using the energy of light. Plants and algae capture light energy via biomolecules called pigments, principally chlorophyll a, situated within the chloroplasts of eukaryotic cells or associated with folded internal membranes of prokaryotic cells. Th ­ ese pigments convert the captured light energy into chemical form and use it to convert (“fix”) inorganic carbon into ­simple organic molecules that store energy and build living biomass. In most photosynthetic organisms, that energy is used to combine the protons (H+) in the ­water molecule with inorganic carbon, producing organic carbon and releasing oxygen as a by-­product. Basic oxygenic photosynthesis consists of three steps. The first two, the light reactions, capture light energy in the pigments and convert the energy into chemical forms—in the energy storage molecule, ATP, and the molecule that stores protons (i.e., reducing capacity), NADH. Th ­ ese energy stores are used to do the cell’s work. The third step, the dark reaction, can proceed without light and uses the energy stored in ATP and the reducing power stored in NADH to convert inorganic carbon to organic compounds. The quantum yield of photosynthesis is the number of carbon atoms fixed per unit light energy absorbed (often mea­sured alternatively as the number of oxygen molecules produced per unit light energy). Photosynthesis runs by harvesting solar energy in the vis­i­ble range of the electromagnetic spectrum, roughly 300–720 nm, which is referred to as photosynthetically active radiation, or PAR. Chlorophyll a is the principal light-­capturing pigment involved in photosynthesis, but vari­ous kinds of algae and bacteria use other light-­sensitive accessory pigments to capture light of dif­fer­ent wavelengths and transfer it to chlorophyll a. Thus, the light spectrum represents a major axis of niche space for photosynthetic organisms, and each organism’s combination of pigments potentially determines its ability to grow u­ nder dif­fer­ent conditions of light quality and intensity. The types of light-­harvesting pigments are impor­tant functional traits of algae, and pigment diversity within a phytoplankton assemblage influences both its productivity and the ability of dif­fer­ent phytoplankton species to coexist (box 5.1). The magnitude and efficiency of photosynthesis are key controls on ecosystem pro­cesses. They are often expressed in terms of gross primary production, which is the total fixation of light energy into biochemical energy (organic carbon), and net primary production, which is gross primary production minus the fixed energy respired to maintain the plant’s basic metabolic machinery, since the pro­cess of photosynthesis itself requires cellular work. Primary productivity is defined as the rate of net photosynthetic fixation of carbon per unit time.

Heterotrophy An organism’s energy bud­get can be visualized as a flowchart among components and analyzed using the princi­ple of mass balance (figure 5.5). For animals and other heterotrophs, metabolism begins with consumption, the ingestion of energy-­containing material originally produced by autotrophs. Some portion of this remains unused and is egested as feces, along with excretion of excess nitrogenous compounds. The remaining portion is assimilated, that is, broken down into simpler chemical components and rearranged into energy-­dense storage compounds, such as fats, oils, and glycogen—­the organism’s battery or fuel depot. This fuel is then consumed as needed to convert biochemical building blocks into growth of living tissues, and to do the organism’s work via biochemical pro­cesses that convert the chemical energy into the mechanical work of locomotion, feeding, behaving, mating, and so on.

Chapter 5 Organisms

119

Box 5.1. ​Diversity and light harvesting in phytoplankton communities A fundamental question in biological oceanography is, What controls algal primary production? Historically, approaches to this question have focused on control by the physical environment, particularly ocean mixing regime, light, and nutrient availability. But the diverse algal lineages that make up the assemblages also differ markedly in the biochemistry and cellular apparatus of light harvesting. This physiological diversity strongly affects ecosystem-­level productivity. Comparative studies of phytoplankton productivity among 550 locations in northern Eu­rope found that more diverse phytoplankton assemblages used resources more efficiently, ­whether efficiency was mea­sured as mass of chlorophyll a or mass of carbon produced per unit phosphorus, the nutrient often limiting productivity in ­these ecosystems (Ptacnik et al. 2008). Unexpectedly, phytoplankton diversity predicted resource use efficiency better than did major abiotic forcing ­factors, including total phosphorus, pH, temperature, salinity, or lake morphometry (chapter 9). (A)

(B) 24

12 8

0

1 2 3 Ln taxon richness

Phytoplankton population density

4

15 Time (days)

0.04

4

30

6

4

0

15 Time (days)

4.0

2.0

0.0

0.0

30

0.5 1.0 1.5 2.0 Ln taxon richness

2.5

White (green + red) light

(F)

8

0

6.0

–2.0

8 10 12 14 16 18 20 Pigment richness Green light

(E)

8

0

0.08

0.00

4

Red light

(D)

0.16

Phytoplankton population density

4

Specific net primary production (μg C mg C–1 h–1)

Mean light absorbance in vitro

Pigment richness

16

0

(C) 0.16

20

Phytoplankton population density

What explains t­hese patterns? Some insights come from experiments with freshwater algae (Striebel et al. 2009), which found that more diverse phytoplankton communities had higher pigment diversity as expected but also higher rates of light absorbance, primary production, and biomass accumulation (see figure B5.1.1). The greater productivity of diverse phytoplankton in t­hese experiments was explained by dif­fer­ent co-­occurring algal species using dif­fer­ent spectral components of the available photosynthetically active radiation (PAR), and this niche complementarity among species was best explained by the variety of light-­harvesting pigments represented in the community. Laboratory experiments also showed that phytoplankton species coexist by using dif­fer­ent portions of the light spectrum, captured by their dif­fer­ent pigments, a classic example of niche partitioning (Stomp et al. 2004).

8

4

0

0

15

30 45 Time (days)

60

Figure B5.1.1. Phytoplankton pigments mediate community diversity and ecosystem-­level primary production. (A) The number of photosynthetically active pigments increases with phytoplankton richness in both experimental (blue symbols) and natu­ral (green symbols) algal communities. (B) Pigment richness increases total light absorbance, leading to (C) higher net primary productivity (NPP) in more diverse assemblages. In experiments, pigments mediate coexistence of two marine phytoplankton (Synechococcus) species: (D) one species (green symbols) dominates ­under red light, (E) another (red symbols) dominates ­under green light, and (F) the two species coexist ­under white light ([A–­C] ­after Striebel et al. 2009; [D–­F] ­after Stomp et al. 2004).

120

Ocean Ecology

Consumption

Assimilation

Reproduction

Molting

Digestion, transport, and use of metabolized energy

Growth

Feces

Deamination and other processes

Activity metabolism

Excretion of metabolized compounds

Metabolizable energy Maintenance metabolism

Defecation

Assimilated energy

Urine

Egestion

Respiration

Net production

Figure 5.5. ​The organism as a bioreactor. Generalized components of the energy and materials bud­get of an individual animal (­after Valiela 1995).

The work of living burns a lot of energy. Respiration is the portion of the assimilated energy that the organism uses to power its life pro­cesses, and it is typically the largest term in the energy bud­get (see figure 5.5). Respiration is considered the fundamental rate of metabolism, the rate at which the potential (i.e., chemical) energy stored in assimilated food is transformed into the mechanical and chemical work that an organism does. When an organism is inactive or resting, its respiration mea­sures the basal metabolic rate (its overhead in economic terms)—­the expenditure of energy required to keep the body’s complex structure and functioning systems intact and prevent it from succumbing to the dissolution of entropy. Active metabolic rate is higher than the basal rate ­because it includes the energy requirements of hunting, feeding, fleeing from enemies, mating, and other activities. Averaged over time, active metabolic rate tends to be two to three times basal metabolic rate in most organisms (Brown et al. 2004). Respiration therefore often serves as a practical mea­sure of overall metabolism. In aerobic organisms, it is mea­sured as the rate of oxygen consumption since energy production from metabolized biochemicals requires a specific quantity of O2. If we can make reasonable assumptions about what biochemical compounds an organism burns to generate energy, we can convert oxygen consumption into yield of energy (calories) using the oxycalorific coefficient, which varies depending on chemical composition of food but ranges around 3.2–3.4 kcal per mg of O2 consumed, depending mainly on dietary protein content (Elliott and Davison 1975). Mea­sured respiration can then be extrapolated to estimate production and carbon use at both organismal and ecosystem scales. Among heterotrophic organisms, most active time is spent acquiring and digesting food, which thus often limit rates of growth and production. I say “active time” b­ ecause, as the pioneering ecologist Charles Elton (1927) observed, “all cold-­blooded animals spend an unexpectedly large proportion of their time ­doing nothing at all, or at least nothing in par­tic­ul­ar.” In general, individual growth rate—­and thus, ultimately, production of the population—is ­limited e­ ither by acquisition of energy to fuel life pro­cesses or by usable forms of nitrogen for building the tissues that make up their bodies, as discussed above.

Kinetics of Life Temperature fundamentally influences the rates of all biological pro­cesses, from cellular biochemistry to ecosystem fluxes. The rates of chemical reactions, including t­ hose that power life, vary with temperature. At the theoretical limit of absolute zero, no reaction can happen, whereas above the boiling point of ­water (and well below it for most organisms) the proteins on which life depends decompose. Therefore, all organisms are constrained to live within a ­limited range of temperatures. For any par­tic­u­lar species, this range reflects an evolved coordination of molecular structure and rates of biochemical, cellular, and organismal pro­cesses. Rates of biochemical reactions typically increase exponentially with temperature over the organism’s normal range. An organism’s per­for­mance is quantified by its metabolic

Chapter 5 Organisms

scope, that is, its oxygen consumption above resting level, which is highest over a small range of optimal temperatures and declines in e­ ither direction (Pörtner and Farrell 2008). But metabolic scope is plastic: it can be depressed by vari­ous stressors, narrowing the v­ iable temperature range, and the range can also vary among life stages. For example, it is narrower for females carry­ing embryos with a high oxygen demand (see figure 4.15). Some species, termed stenothermal, have a narrow range of temperature tolerance, whereas eurythermal species have a broader range. For example, reef-­building corals with their symbiotic algal cells tend to be stenothermal and to live near the upper limit of their thermal tolerances. As a result, small increases in temperature can push corals over the edge to death or morbidity through bleaching—­the expulsion of their algal symbionts to reduce coral respiration (chapter 12). The temperature tolerances of species shape their geographic distributions and therefore community composition, and climate shifts can alter community structure, particularly as they influence species’ propensity to grow or decline near the edges of their temperature ranges (Stuart-­Smith et al. 2015). Temperature also influences interactions among species in communities. On average, metabolic rates are higher in warmer regions and during warmer times of the year, and t­hese can translate to higher rates of interactions among species in warm places and times. Dif­fer­ent species commonly respond differently to changing temperature, which can alter their interactions, for example, when warming increases rates of predation more than rates of prey population growth. In general, consumer-­ prey interactions are a function of three pro­cesses: biomass accumulation of the resource (prey), the rate at which predators consume the resource, and consumer mortality (Gilbert et al. 2014). Understanding how ­these pro­cesses change with temperature for dif­fer­ent taxa and ecosystems provides a key component of a mechanistic theory of community interactions in a changing world.

Dimensions of Life Marine life varies in size over nine ­orders of magnitude, from viruses (which straddle the line between molecules and organisms) to ­whales (see figure 2.8). Individual body size strongly shapes all biological pro­cesses. Organismal structure, physiology, and ecological interactions all change predictably with body size, making it the closest ­thing we have to a master trait in ecol­ogy: the size of an organism predicts, often within an order of magnitude, its metabolic rate, feeding rate, mobility and dispersal capacity, and even the mode and range of its sensory abilities (Andersen et al. 2016). The influences of body size on functional ecol­ogy are summarized mathematically as allometry, originally defined as the regular change in body shape with size but now used more broadly to include scaling of any physiological or metabolic pro­cess with size. Th ­ ese regular relationships are the basis of a fundamental law of biology: metabolic rate (B) scales positively and allometrically with body mass (M). Metabolic rate provides a unifying theme in ecol­ogy ­because the amount of energy consumed by an organism is related in predictable ways to characteristics of both organism and environment, notably body size, temperature, and stoichiometry (Brown et  al. 2004). Julian Huxley (1932) first recognized that many organismal characteristics, including metabolic rate, are related to body size by a power function, meaning that the relationship is linear on a log-­log scale: Y = Y0Mb, where Y is a dependent variable (say, basal metabolic rate), Y0 is an empirically estimated constant that is specific to par­tic­u­lar taxa or pro­cesses, M is body mass, and b is a scaling exponent that quantifies how metabolic rate (for example) changes with body size. The mathematical relationships between body size and shape or function define their allometry. The power functions that describe them are often called allometric equations, and b is the allometric exponent. It has been known since the pioneering work of Max Kleiber (1932) that metabolic rate scales with body mass as a power function. A wealth of empirical data from diverse taxa and systems supports this power law scaling of metabolism (figure 5.6) and shows that the scaling coefficient b falls

121

122

Ocean Ecology

(A)

In (development rate M1/4)

30

–4 –5

–3

–6 –5 40 41 42 38 Inverse of temperature (1/kT)

40

20

Plant Mammal Bird Bat Zooplanton Insect Fish

10

–30

–20

–10 0 In (mass)

10

(D)

20

In (development rate e E/kT)

T-corrected maximal biomass production (In [PeE/kT], g individual-1 year-1)

40

Field data for fish y = –0.68x + 24.15

Zooplankton (C) y = –0.73x + 24.81 –1

(B) y = 0.76x + 25.22

42

(E) y = –0.27x + 20.37

26 25

25

24

24

23

23 –18

–14

–10

y = –0.22x + 22.49

26

–10

–8

–6

–4

In (mass)

Figure 5.6. ​Metabolic scaling as a function of body size and environmental temperature. (A) Maximal rates of biomass production are tightly linked to body mass across a range of organisms from unicellular phytoplankton to trees and mammals (­after Morgan Ernest et al. 2003). The allometric exponent (slope) is close to the predicted value of ¾. (B–­E) Temperature and mass dependence of developmental rates for (B, D) zooplankton eggs in the laboratory and (C, E) fish in the field (­after Brown et al. 2004).

within a relatively narrow range of values. Much debate surrounds the question of how uniform the scaling coefficient is (i.e., ­whether and how it differs among taxa in nature) and what values are expected from theory. Indeed, ­there is still disagreement on the basic question of ­whether the metabolic theory that predicts such values is actually mechanistic (i.e., emerging from first princi­ples of physics) versus phenomenological (simply presenting a mathematical description of correlated patterns). (For a flavor of the vari­ous viewpoints, see the forum in Ecol­ogy 85 [7].) In any case, most empirical studies have estimated that the scaling exponent relating metabolism to body size approximates ¾, or ~0.75: B ∝ M3/4. Why this (relative) constancy? Two main types of explanations have been offered (Savage, Gillooly, Woodruff, et  al. 2004, Vandermeer 2006). Early theories suggested that metabolic scaling might be driven by the dif­fer­ent allometry of an organism’s surface area versus its volume. Metabolism occurs throughout the volume of the body (which is proportional to length3), whereas fluxes of nutrients, gases, and wastes must pass through its surface (which is proportional to length2), so we might expect metabolic rates to scale to the ⅔ power of body mass. Yet, as data accumulated, they more frequently corresponded to an exponent of ¾, puzzling early physiologists. The second and currently most widely accepted model predicts this ¾ scaling of metabolism on body mass based on geometric constraints of the architecture of internal transport systems rather than external surface area (West et al. 1997). Specifically, the latter model has two key components. First, it proposes that transport of material through the organism’s volume is optimized by natu­ral se­lection via a fractal network of transport structures, for example, the circulatory system of vertebrates or the vascular system of land plants, which branches from major vessels to ever smaller ones as they approach the tissues they serve. Second, the model assumes that the smallest ele­ments of the network are constant in

Chapter 5 Organisms

size and structure regardless of the organism’s body size, as is true of the mitochondria and chloroplasts that power respiration and photosynthesis in eukaryotes. When ­these two conditions are met, the theory predicts that metabolic rate w ­ ill scale with the ¾ power of body mass (West et al. 1997). This scaling means that, all ­else being equal, 10 g of small invertebrates (10,000 individuals weighing 1 mg each) w ­ ill be 5.6-­fold more productive than 10 g of large invertebrates (10 individuals weighing 1 g each), even though the two samples are equal in total biomass. Numerous aspects of metabolism and life history follow this ¾ power scaling, from trunk dia­meter and mass in trees, to zooplankton egg development rate, to population growth rates and density of phytoplankton (Belgrano et  al. 2002, Brown et al. 2004, Savage, Gillooly, Woodruff, et al. 2004) (figures 5.6, 5.7). The dependence of biological rates on temperature and body mass can be combined into a power­ful predictive framework (Gillooly et al. 2002, Brown et al. 2004). Syntheses of metabolic data across organisms spanning the complete 20 ­orders of magnitude in body mass from viruses to ­whales show that basal metabolic rate scales consistently with body mass and temperature (see figure 5.6). The allometric scaling laws describing such relationships are grounded in fundamental pro­cesses of organismal metabolism and therefore place clear constraints on our expectations for many patterns (A)

(B)

19

Ln fish mortality

Ln fish mortality

3

1

–1

17

15 β = –0.24

β = –0.45 39

40 41 Inverse temperature (1/kT)

0

5

(C)

15

(D) Unicellular eukaryotes Multicellular eukaryotes Vertebrates

26

In rmax

1

In rmax

10

In (mass)

–1

–3

β = –0.68 37

39

41

Inverse temperature (1/kT)

43

22

18

–10

β = –0.23 0

10

In (mass)

Figure 5.7. ​Population growth and mortality rates depend on body mass and temperature. (A, C) Temperature and (B, D) mass dependence of (top) fish mortality rates in the field (from Pauly 1980) and (bottom) maximal rates of population growth for a variety of organisms (from Savage et al. 2004). Note that temperature decreases from left to right (­after Brown et al. 2004).

20

123

124

Ocean Ecology

and pro­cesses at higher levels of organ­ization, including population density, community body size distribution, and ecosystem pro­cess rates (chapters 6–9). Allometric scaling thus helps link ecological pro­cesses from molecular to ecosystem scales. Across organisms from bacteria to w ­ hales, body size is tightly correlated with rates of metabolism, development, biomass production, population growth, and even population density (see figures 5.6, 5.7). And the theoretical constraints identified by metabolic ecol­ogy are consistent with the data in some surprising ways. For example, rates of ecological interactions, including competition, predation, parasitism, and disease, scale with temperature (Brown et al. 2004), and even rates of speciation and extinction appear to increase with temperature in the fossil rec­ord (Mayhew et al. 2012). Physical pro­cesses, such as fluid dynamics, change substantially over the 20 o­ rders of magnitude in size spanned by life in the ocean (Andersen et al. 2016). ­These changing physical regimes strongly influence how organisms function, particularly the effectiveness of vision, mechanical sensing, modes of locomotion, and modes of resource capture (figure 2.8). Small organisms experience the world in fundamentally dif­fer­ent ways than larger ones. A bacterium interacts with the world at a molecular scale, experiencing its fluid medium as something like a viscous gel, sensing individual biomolecules in its surroundings, and exchanging materials with the outside world and within its body via molecular diffusion and active transport across its cell membrane and wall. At the other end of the spectrum, a blue w ­ hale experiences the world at a scale that nearly encompasses the planet, traversing ­whole ocean basins in migration using sophisticated ­mental maps built with sonar and a language of vocalizations that proj­ect over vast distances underwater to communicate with its dispersed companions. To exchange the same materials that the bacterium pumps across a cell membrane, the ­whale requires a complex fractal circulatory system powered by a heart the size of a small automobile to ser­vice its 190-­metric-­ton body, among other systems that have no parallel in unicells. Ken Andersen and colleagues (2016) have traced the changes in physical pro­cesses by body size and correlated them with changes in the way organisms function, specifically the sizes where scaling relationships change or break down. They identified three transitions in body size where physics renders one sense more efficient than another in obtaining resources: (1) below a body length of 100 μm organisms can effectively use chemical diffusion gradients to locate resources; (2) between 1 mm and 1 cm the optimal sensory mode for detecting prey shifts from hydromechanical to visual; and (3) at about 1 m in size, predators are capable of seeing as far as w ­ ater visibility allows—­roughly 80 m in the clear ­water. Above this size, predators can extend their sensory range only by following chemical trails, as in sharks, or by echolocation, as in marine mammals. Body morphology is also influenced by such size-­based physical transitions, specifically between viscous and inertial regimes around a body length of 7 cm. Below this threshold organisms have a wide range of body shapes, whereas above it organisms (primarily fishes and mammals but also squids) are streamlined and converge on one another in form.

Mechanics of Life Just as organisms are composed of the same components as other ­matter, they are subject to the same laws of physics. How organisms live is influenced by both what they are made of and the physical medium in which they live. ­Water is a fundamentally dif­fer­ent medium than air, with far-­reaching implications for the structure and ecol­ogy of organisms that live in it. The main difference is the greater density of w ­ ater (Strathmann 1990), which has several impor­tant consequences. First, since living organisms are about the same density as seawater, they tend to float and are thus able to maintain a position in the ­water column with ­little effort. Many marine organisms can regulate their buoyancy by adjusting the chemical composition of their tissues. Fishes do so by changing the gas composition of their swim bladders. Phytoplankton cells do so by adjusting the content of lipid, which is relatively light, or of dissolved osmolytes.

125

Chapter 5 Organisms

Box 5.2. ​Metabolic theory and species interactions Herbivory, the consumption of plants by animals, is the first link in moving primary production up through the food chain. Metabolic theory predicts that herbivore impacts on plants w ­ ill generally be greater in warmer climates and seasons b ­ ecause herbivore metabolism (respiration) increases faster with temperature than does plant production (photosynthesis) (figure B5.2.1A). Experiments have confirmed this prediction for both benthic and pelagic systems. First, the brown seaweed Sargassum filipendula grew faster in warm w ­ ater, consistent with the common increase in general metabolic rates at higher temperatures, but feeding on the seaweed by an herbivorous crustacean increased with temperature more steeply than did plant growth. Therefore, seaweed biomass tended to decrease with rising temperatures in the presence of the herbivore (O’Connor 2009). Similarly, in a plankton community (figure B5.2.1B, C), top-­down control of primary production by grazing copepods was stronger at higher temperatures, but only when resources ­were abundant. In contrast, at lower resource levels, food web production was constrained at all temperatures

(A)

(B)

–25 Respiration –30

0

30 Temperature

2.0 Phytoplankton (mg C/L)

Photosynthesis

–35

(C) 1.5

Herbivore/algal biomass

–20 Resting metabolic rate

(O’Connor et al. 2009). T ­ hese results demonstrate that moderate warming can dramatically shift food web dynamics, causing predictable changes in food web structure and productivity, and that temperature interacts with nutrient loading. ­These fundamental differences between plant and herbivore metabolism appear to translate up to regional scales as well. Field experiments and natu­ral history show that predation and herbivory are generally stronger in tropical regions than at higher latitudes. For example, along the east coast of North Amer­i­ca, salt marshes at lower latitudes support an order of magnitude higher herbivore density, 2–10 times higher herbivore damage to leaves, and two o ­ rders of magnitude higher grazing damage to standardized plants deployed in the field than do marshes of the same plant species at higher latitudes (Pennings et al. 2009). ­These results have potentially impor­tant implications for food web pro­cesses generally, the fate of primary production, and the structure of communities and vegetated ecosystems in a warming ocean.

1.0

0.5

0.0 20

22

24

26

28

1.5 1.0 0.5 0.0 20

22

Temperature

24

26

28

Temperature

Figure B5.2.1. Metabolic effects on algae-­herbivore interactions in plankton. (A) Herbivore (and plant) respiration increases faster than photosynthesis with warming, raising the impact of herbivory with temperature. (B) In experimental plankton communities, the ratio of herbivore biomass to algal biomass changes with increasing temperature and with nutrient status (green = nutrient addition, blue = control). (C) Total standing phytoplankton biomass declines with temperature u­ nder nutrient addition (green) but is ­little changed ­under ambient, low-­nutrient conditions (blue) (­after O’Connor et al. 2009).

Single-­celled phytoplankton are the dominant primary producers of the open ocean. Their buoyancy in the fluid medium of seawater keeps them near the surface source of sunlight and bathes them in dissolved nutrients. As a result, phytoplankton do not need costly structural materials, like the stems and trunks of land plants, to lift them ­toward the light, nor roots to find and extract nutrients from the soil, nor bark to protect them from desiccation. The low requirements of phytoplankton for structural materials, such as wood (which is mostly carbon), means that marine algae are generally richer in nitrogen than land plants and thus highly nutritious to herbivores (box 5.2). Phytoplankton are therefore tiny and grow fast. Consequently, marine ecosystems have higher grazing rates than ter-

126

Ocean Ecology

restrial systems, as well as higher production-­to-­biomass ratios, a larger fraction of primary production grazed, and more efficient conversion of production to herbivore biomass (Steele 1991, Cebrián 1999, Shurin et al. 2005). Both the microscopic size of marine primary producers and the fluid connections among marine habitats and regions lead to much closer and more rapid coupling between physical ­drivers and biological pro­cesses in marine (pelagic) systems than ­those on land (Steele 1985). For example, nutrient loading to the coastal ocean can produce responses of primary producers within days or even hours in the sea, whereas major changes may require months to de­cades on land. Thus, many marine communities, especially pelagic communities, tend to be more sensitive to disturbances and to rebound more rapidly ­after disturbances than terrestrial ones.

Coding Life So far we have considered organisms as machines, chemical reactors built from the same stuff as the rest of the universe, powered by energy and chemical reactions and working according to physical laws. This perspective explains how organisms interact with the environment and drive biogeochemical pro­cesses. But organisms are obviously more than machines. The critical feature that distinguishes life from nonliving m ­ atter is the ability to replicate itself with modification and thus to adapt to change in a way that improves its per­for­mance in its environment. And the key to that ability to adapt is a set of editable instructions for building and maintaining the organism, its ge­ne­tic code, written in the letters of DNA bases. An individual organism is the culmination of a long pro­cess of adaptation to the environment by its ancestors. Unlike a carbon atom, ­every organism is unique.

Natu­ral se­lection and adaptation Adaptation has two meanings in biology. First, it is the pro­cess by which natu­ral se­lection adjusts the functional fit of an organism to its environment. Second, the product of that pro­cess is also referred to as an adaptation. For example, the streamlined body form that helps a pelagic fish swim fast or the slow metabolism that allows a deep-­sea animal to survive with very l­ ittle food are adaptations to their environments. Adaptation occurs via natu­ral se­lection, which is a consequence of three features of living organisms: (1) variation among individuals, (2) some degree of heritability (i.e., ge­ne­tic basis) of that variation, and (3) differences in fitness among the variants. Natu­ral se­lection is strongest among individual organisms within a population, but the pro­cess can operate at any level of organ­ization, from gene to social group (Dawkins 1976). Individuals of the same species vary in their ge­ne­tic constitution and therefore in morphology, be­hav­ior, and biochemistry. Although we often make the simplifying assumption that members of a species are uniform and interchangeable, we know that they are not (figure 5.8). The phenotype of a species is ­really just an average across that variation in individual phenotypes. And the variation is impor­tant. When two individuals of the same species compete for food or other resources, the winner is generally the one that is e­ ither better at monopolizing the resource (interference competition; chapter 7) or better at surviving and performing ­under depleted levels of the resource (exploitation competition). ­Those individuals that obtain more resources in turn convert them into more offspring. If that competitive ability is at least partly heritable, the consequence is that the traits of better competitors become more frequent in the next generation. This is the essence of natu­ral se­lection. The a­ ctual pro­cess of adaptation is a good deal more complex than this s­ imple summary suggests (Larson 2009), but more than 150 years since Darwin first articulated it, the prodigious accumulated evidence shows that, in general outline and in many specifics, he got it right. The spectacular diversity among living t­ hings has been produced largely by natu­ral se­lection as organisms have solved, in their vari­ous ways, the challenge of acquiring materials and energy to survive and reproduce. ­Those challenges come not only from the physical environment but also from

Chapter 5 Organisms

Figure 5.8. ​Phenotypic variation within a species: Individuals of the Philippine marine snail Vittina waigiensis vary like snowflakes, no two reputedly the same.

other organisms that compete with it for resources, provide its resources, or try to eat it. The characteristics of organisms (i.e., traits) are sculpted by ­these challenges. Ecol­ogy and evolution are thus inseparable.

Genotype and phenotype The genotype is the ge­ne­tic code of an individual organism, written in its DNA sequence. The phenotype is every­thing built from that code—­the totality of the organism’s structural and functional properties. The phenotype emerges from the genotype as it interacts with the environment, starting with the developmental environment of the egg cytoplasm within its m ­ other and continuing as the newborn individual, larva, and adult interacts with the external environment throughout life. The amount of ge­ne­tic variation among individuals within a population determines the population’s evolutionary potential, that is, its ability to adapt to environmental change. Some phenotypic traits are ­under ­simple ge­ne­tic control by a single gene, like the color and texture of Mendel’s famous peas. But most traits of ecological interest (size, weight, morphology, life history) are polygenic, meaning that they are controlled by multiple genes, and usually also by the environment. Such traits vary continuously rather than as discrete states. An organism’s fitness, its contribution of gene copies to the next generation, is strongly influenced by life history traits, including survivorship, growth rate, and reproductive rate. ­Because of their influence on fitness, life history traits tend to be ­under strong se­lection, and we expect se­lection to have pushed such traits to values close to what is optimal for the environment in which the organism lives. Nevertheless, life history traits often show significant heritability, that is, ge­ne­tic variation that is passed from parent to offspring. For example, heat tolerance varies among populations of the Australian reef coral Acropora millepora, and genomic and experimental studies of coral larvae show that this tolerance is heritable. Coral larvae from parents living at warmer latitudes survive up to 10 times better u­ nder heat stress than t­ hose from parents at cooler locations, the larvae show heritable

127

128

Ocean Ecology

differences in cellular stress pro­cesses, and experimental crosses of parents from dif­fer­ent areas show that coral thermal tolerance has a strong ge­ne­tic basis (Dixon et al. 2015). A review of over 1000 studies found that median heritability for life history traits is ~25%, somewhat lower than for morphological traits but nevertheless substantial (Mousseau and Roff 1987). This means that ­there is much potential for evolutionary adaptation to changing conditions in most wild populations. Despite the strong influence of genotype on an organism’s traits, much phenotypic variation is not genet­ically determined, and this variation is critical to many ecological and evolutionary pro­cesses. The ability of a given genotype to produce dif­fer­ent phenotypes u­ nder dif­fer­ent environmental conditions is called phenotypic plasticity, and generally results from environmental influences during development. This relationship is often summarized as the reaction norm, that is, the distribution of phenotypes produced by a single genotype across a range of environmental conditions. Phenotypic plasticity is most pronounced in plants and sessile animals that cannot move as adults (West-­Eberhard 1989) and must adjust their phenotype to prevailing conditions, but it also occurs in some mobile animals for whom environmental challenges can be predicted reliably from certain cues. We consider the mechanisms and consequences of evolutionary change and adaptation in more detail in chapter 6.

Functional Ecol­ogy and the Niche Functional traits may be thought of as a bridge connecting organisms to ecosystems for a c­ ouple of impor­tant reasons. First, an organism’s functional traits necessarily affect ecosystem pro­cesses ­because the activities ­those traits help to perform consume resources from the environment, affect transformations of energy and materials, and affect other organisms in the ecosystem (figure 5.9). Second, ­because functional traits can be defined by mea­sur­able characteristics, such as size and chemical composition, they transcend the taxonomy of par­tic­u­lar groups and allow more direct comparisons among even unrelated species and systems. Traits tend to vary together. No organism can si­mul­ta­neously grow fast, reproduce constantly, and live forever. This is ­because ­there are competing demands for resources that support ­these activities, referred to as trade-­offs. As a result of trade-­offs, some traits are negatively correlated across related species (e.g., egg size vs. number) whereas ­others are positively correlated (e.g., growth rate and maturation rate). Trade-­offs in organismal energy bud­gets and life history drive the functioning of ­whole organisms and thus the coordinated evolution of their interacting traits. An impor­tant consequence of ­these relationships among traits is that species often cluster into functional groups that share sets of traits. Such functional groups can be summarized using multivariate methods that separate species into a small number of groups that share similar trait combinations, or even along a single axis through multivariate trait space (Madin, Hoogenboom, et al. 2016). Be­hav­ior is another impor­tant way that organisms match their current environment to their desired situation. Animals, and to some degree plants, seek out environments at both landscape and microenvironment scales where conditions are favorable for them. So the match between organism and environment is achieved not only through morphological and physiological adaptation but also through evolution of be­hav­ior, which amounts to a form of phenotypic plasticity.

Historical and modern concepts of the niche An organism is an integrated complex of traits. That complex is influenced strongly by its environment, including both physical forces and the other organisms with which it interacts; in a word, its niche. The pioneering ecologist Evelyn Hutchinson (1957) characterized the niche, in a famous bit of ecological jargon, as an “n-­dimensional hypervolume”; in other words, the set of many environmental conditions in which an organism can live, with ­those dimensions defined not by the three dimen-

Chapter 5 Organisms

Fisheries yield

Atmosphere composition

Eukaryotic consumers (Micro) (Macro)

• Body size • Growth rate • Diet breadth • Mobility

• Metabolic rate • Stoichiometry • Defenses

Primary producers

Heterotrophic prokaryotes

• O2 metabolism • C metabolism • N metabolism

• Cell size • Growth rate • Sinking rate

• Stoichiometry • Edibility • Coloniality

DOM

Export flux (C storage)

Figure 5.9. ​Some organismal traits that influence pelagic ecosystem pro­cesses (Duffy and Stachowicz 2006).

sions of ­actual space but by a “hypervolume” with a potentially infinite number (n) of variables—­ physical characteristics, ­water quality, resource levels, predation and disease pressure, and so on. The fundamental niche is usually defined as the complete set of abiotic environmental conditions in which a species is capable of living. The realized niche is the subset of conditions that a species actually occupies in the field. The difference results from interactions with other species that share part of the fundamental niche and reduce its fitness, thereby restricting the species to a narrower range of conditions (box 5.3).

­Toward a trait-­based ecol­ogy of marine organisms In the case of phytoplankton, the traits influencing fitness or competitive ability include requirements for light and nutrients (see boxes 5.1, 5.3) as well as cell size, defenses against herbivores, and so on. Focusing on such functional traits helps illuminate the mechanisms by which organisms interact and distribute in nature. Taking an example from Sommer’s experiment (see box 5.3), the dominance of Stephanopyxis palmeriana ­under high Si:N ratios tells us only about that species, whereas knowing that this species is a diatom with a silica shell helps us understand that it dominated ­because ­there was plenty of silicon in this treatment, and suggests that other diatoms w ­ ill also do well u­ nder ­these conditions, which was in fact true (Sommer 1994). In other words, functional traits allow us to move beyond the idiosyncrasies of species’ taxonomic names to understand the general mechanisms by which species make a living.

129

130

Ocean Ecology

Box 5.3. ​Defining the niches of phytoplankton An organism’s niche encompasses the conditions u ­ nder which it is capable of living and sustaining a population. Where species compete for resources, as most do, differences between their niches may reduce competition and allow them to coexist. But despite the power­ful intuitive concept of the niche, it is difficult to define in practice. Hutchinson (1961) highlighted this difficulty of defining niches in his famous “paradox of the plankton” (chapter 7). Phytoplankton need relatively few basic resources—­light, inorganic nutrients, and carbon—to grow, so their niches should be more straightforward to mea­sure than t­ hose of many other life forms. A logical way to begin defining an organism’s niche is to mea­sure its per­for­mance in obtaining resources u ­ nder dif­fer­ent combinations of ­these variables. Ulrich Sommer (1994) used this approach to characterize the niches of Baltic Sea phytoplankton. He raised 11 species of phytoplankton together in lab cultures, allowing the populations to grow and compete through multiple generations. To find the optimum conditions for each species, the cultures w ­ ere grown ­under ­every combination of four light intensities, three daylengths, and six ratios of the nutrients nitrogen and silicon—­resulting in 72 separate treatments. By recording which species came

(C)

L:D = 16:8

S

10

12:12

8:16 R1 R2

D

1

R3 Diat R 0.1

0

1

Prym

Oth 2

3

4

Time (weeks)

5

1 R4

Daily light dose (mol quanta m2 s –1)

(B)

Vi/Vtot

Cell volume (106 μm3 ml–1)

(A)

to dominate u ­ nder each set of conditions, the realized niche of each species could be defined in the form of a three-­or four-­dimensional hypervolume of light intensity, daylength, and N:Si ratio. Dif­fer­ent phytoplankton species dominated ­under dif­fer­ent resource conditions, showing that they had partly complementary niches: diatoms fared best u ­ nder high ratios of Si to N, whereas flagellates did better u ­ nder other conditions, and species within t­ hese groups separated along the light gradient (figure B5.3.1). This suggests that adaptation to dif­fer­ent conditions may help phytoplankton species coexist u ­ nder the range of environmental conditions in nature. This experiment illustrates well that the phytoplankton niche is multidimensional. We can think of the 72 experimental treatments as 72 potential niches. Yet ­these encompassed only three environmental gradients. In nature, the niche is even more complex ­because t­ here are more competing species and more environmental gradients as well as grazers and microbes that can e ­ ither facilitate or infect the phytoplankton, all of which can vary seasonally (the inclusion of daylength as one niche axis in the experiment shows how temporal variation can alter per­for­mance).

10 D S 1 10 C L

1 10 R

N

1

0 0

50 100 150 0

50 100 150 0

Si:N ration (mol/mol)

50 100 150

10

50 10010 50 100 Si:N ratio (mol/mol)

Figure B5.3.1. Niche differentiation in phytoplankton. (A) Biomass trajectories of several phytoplankton species (lines) and their summed biomass (unshaded area) in multispecies culture. (B) Biomass of one diatom, Lauderia borealis, ­under dif­fer­ent daylengths (columns, L:D = light:dark), irradiances (rows), and Si:N ratios (x axis). (C) The multidimensional niches of six phytoplankton species based on their growth rates in multispecies culture (­after Sommer 1994). Darker green = higher abundance.

Biological oceanography has been based on traits since its inception. Phytoplankton, microzooplankton, and nekton are categories that each include phyloge­ne­tically diverse species but are defined by shared traits, specifically mode of nutrition (auto-­versus heterotrophic), body size, and locomotory ability. Similarly, food web ecol­ogy is structured around the trait (complex) of trophic level: ­whether an organism obtains its energy from inorganic sources, from plants, from animals, or from dead organic detritus. A general frontier in ecol­ogy is identifying the traits that are most impor­

Chapter 5 Organisms

tant to key ecological pro­cesses, such as growth, competitive ability, vulnerability to predators, and geographic distribution. Such trait-­based frameworks have been developed with varying degrees of success for macroalgae (Littler and Littler 1980, Steneck and Dethier 1994), herbivores (Steneck and Watling 1982), benthic invertebrates (Bremner et al. 2006), and fishes (Potter et al. 2015, Albouy et al. 2015, Froese and Pauly 2019). One of the best developed concerns corals (Madin, Anderson, et al. 2016; chapter 12). Throughout the remainder of the book, we focus on traits of organisms and how they help illuminate their roles in species interactions and ecosystem pro­cesses.

Functional Ecol­ogy of Marine Primary Producers Plants and algae vary widely in morphology and physiology, but a few key traits help define their interactions and have far-­reaching effects on the structure and functioning of ecosystems. One of the most central traits of primary producers is their nutrient requirements, reflected in the nutrient content of their tissues, which is closely correlated with life form and habitat. A plant’s life form is in turn strongly conserved phyloge­ne­tically: the major traits influencing a plant’s vulnerability to herbivory and decomposition diverged millions of years ago among major lineages. For example, single-­celled algae, like phytoplankton, have high nutrient content compared with flowering plants, such as seagrasses, ­because seagrasses produce a lot of structure out of cellulose, which lacks nitrogen and thus results in nutritionally poor seagrass tissues. Grazing by herbivores is highest on nutrient-­rich microalgae, both pelagic phytoplankton and benthic microalgae, averaging more than 40% of net primary productivity in systems dominated by t­ hese forms (figures 5.3, 5.4). Conversely, grazing is lowest on coastal flowering plants, such as seagrasses, marsh grasses, and particularly mangroves. As a result, marine ecosystems dominated by microalgae pro­cess nearly all production within the system and export or store very ­little carbon, whereas coastal systems dominated by higher plants tend to export a substantial fraction of their production and store 10%–17% of net primary production in sediments (Duarte and Cebrián 1996). Across 600 field experiments in benthic systems, impacts of herbivory on plant biomass ­were better explained by plant functional group, which is closely related to taxonomic group, than by habitat type, temperature, or nutrient levels (Poore et al. 2012). Focusing on individual functional traits helps us understand how organisms work. But it is impor­tant to keep in mind that an organism is an integrated w ­ hole and its traits function together, incorporating trade-­offs among activities such as resource acquisition, competition, and herbivore deterrence. Often functional form corresponds relatively closely to a higher taxonomic rank (Poore et al. 2012), showing that suites of interrelated functional traits are remarkably conserved over evolutionary time. Thus, despite the ­great diversity of marine primary producers, we can deduce some general patterns in their functional roles based on their traits. The major traits of marine algae and plants that affect biomass and distribution are body or cell size, resource use, physiological responses to temperature, and re­sis­tance to grazers (chapters 10, 11).

Functional groups of phytoplankton Phytoplankton cells live suspended in the nutrient broth of seawater, which is very dilute in the open ocean far from terrestrial nutrient sources. ­Because ocean ­water is a buoyant medium that contains all the substances phytoplankton need to grow, they need invest nothing in the complex and often expensive structural tissues that land plants and some marine macroalgae use to reach for the sun from their rooted positions. Phytoplankters are essentially stripped-­down photosynthetic machines (see figure 2.3A), adapted to maximize capture of light and nutrients from the medium. Most species live as single cells; their small size maximizes the area of cell surface through which they can absorb inorganic nutrients from the ­water, and reduces self-­shading of their photosynthetic apparatus. As a consequence, most phytoplankton can grow rapidly and are highly efficient at capturing nutrients. In

131

132

Ocean Ecology

(A)

Photosynthesis (pg C cell–1 h–1)

10

4

102

100

Diatoms Dinoflagellates Coccolithophores Cyanobacteria Chlorophytes Others RMA slope = 0.87

10–2

10–4 –2 10

100

102

104

106

102 104 3 Cell size (μm )

106

Cell size (μm ) 3

102

Respiration (pmol O2 cell–1 h–1)

(B)

100

10–2

10–4

10–6 –2 10

(C)

100

C-specific photosynthesis (h–1)

0.25

0.20

0.15

0.10

0.05

0.00 –2 10

100

102

104

106

Cell size (μm ) 3

Figure 5.10. ​Size scaling of phytoplankton metabolism and growth. Plotted against cell size are (A) photosynthesis per cell, (B) dark respiration per cell, and (C) carbon-­specific photosynthesis in 22 species growing ­under the same conditions (­after Marañón 2015).

turn, the lack of nitrogen-­poor structural materials means that most phytoplankton are highly nutritious and vulnerable to grazing, compared with benthic macrophytes (see figure 5.3). The trade-­off between nutrient capture and defense against grazing is an impor­tant axis distinguishing functional groups of phytoplankton. The most fundamental trait of phytoplankton is cell size, which varies over several ­orders of magnitude and strongly affects metabolic rates, light absorption, nutrient uptake, sinking rate, interactions with grazers, and thus biogeochemical pro­ cesses (figure 5.10) (Marañón 2015, Litchman et al. 2015). As a result, cell size is a major axis of functional differentiation for phytoplankton. Both body size and stoichiometry of marine producers are response traits, predicting how species respond to environmental forcing, as well as effect traits, which influence growth rates, food web interactions, and biogeochemical cycles (Finkel et al. 2010, Litchman et al. 2015). As cell size increases, the ratio of surface to volume declines, reducing the rate of nutrient diffusion into the cell per unit volume. As a result, larger cells generally require higher nutrient concentrations to grow (Marañón 2015). Self-­shading of the pigment molecules that capture light energy also increases with cell size, further reducing efficiency of resource capture and growth. For ­these reasons, small-­celled phytoplankton are favored u­ nder low-­nutrient and low-­light conditions. Against ­these advantages, small cells are constrained by reduced biosynthetic ability. The reason for this is that the basic cellular machinery, such as the nucleus required by all cells, takes up a certain amount of cell volume and resources regardless of size; small cells thus have less remaining volume to use for nutrient acquisition and biosynthesis (Marañón 2015). In pelagic systems, nutrient availability is lower in stratified than in well-­mixed ­water columns ­because phytoplankton in surface ­waters deplete nutrients, which cannot be readily replenished due to ­limited exchange with deeper ­waters. ­Under such stratified conditions, surface ­waters are typically dominated by very small cells whose high surface area per volume fosters efficient nutrient uptake (Marañón 2015). The extreme case involves the chronically oligotrophic, permanently stratified surface w ­ aters of central ocean gyres, where the picoplanktonic cyanobacterium Prochlorococcus dominates the phytoplankton (chapter 10). ­These tiny cells experience seawater as a viscous medium, so they sink very slowly into the deep ocean. They are grazed heavi­ly by protistan microzooplankton, which can grow very rapidly and respond quickly to phytoplankton production pulses. Thus, picoplanktonic production passes mainly through the microbial loop (chapter 10), cycling between production, rapid grazing, and regeneration of nutrients by micrograzers in the same ­water parcel.

Chapter 5 Organisms

In contrast, the w ­ ater column in more nutrient-­rich coastal systems is generally dominated by larger phytoplankton cells, which in turn are grazed primarily by metazoan zooplankton like copepods. Since t­ hese animals are impor­tant food for young fishes, more production enters the classical food chain leading up to harvestable fish in ­these more nutrient-­rich w ­ aters. Thus, the modal phytoplankton cell size depends on the abiotic conditions of water-­column stratification and nutrient concentration, and is in turn a major controller of both biogeochemical and food web pro­cesses.

Functional groups of benthic macrophytes Approaches to understanding the functional ecol­ogy of marine benthic autotrophs have focused on functional form—­a suite of related traits that vary together between major plant and algal groups. An influential early model examined how suites of traits affect productivity and grazing across a range of marine macroalgae and proposed that traits ­were integrated into functional forms, that is, suites of co-­ occurring traits that correspond to specific ecological strategies, among macroalgae (Littler and Littler 1980). For example, opportunistic species (sometimes called ephemeral algae) devote ­little energy to structural materials or defenses against herbivores, therefore have high net productivity, tend to colonize and grow rapidly, and dominate early in successional sequences. Conversely, the macroalgal species typical of “mature, climax” communities (also called perennial algae) have lower productivity and slower growth, but produce more structural tissues that raise them up to compete for light and make them more resistant to herbivory (see figure 2.11). A corresponding model for marine herbivores took a similar approach, focusing on how algal functional forms interact with herbivore functional morphology to influence vulnerability to grazing (Steneck and Watling 1982). Subsequent research has found many exceptions to the generalizations proposed in ­these functional group approaches (Padilla and Allen 2000), but on hard substrata across a wide geographic area the dominant macroalgal types correlate reasonably well with environmental productivity and herbivore pressure (figure 5.11) (Steneck and Dethier 1994). Again, many of the traits considered in such functional form models—­ morphology, growth rate, tissue nutrient content, vulnerability to herbivores—­are substantially conserved phyloge­ne­tically, meaning that functional forms are often (though certainly not always) roughly concordant with evolutionary lineages (Poore et al. 2012).

Macroecol­ogy We have approached ecol­ogy from two distinct directions, reflecting the field’s historical development. At one end of the spectrum, in this chapter, w ­ e’ve focused on the traits of individual organisms and how they influence population dynamics and interactions among species in local communities. At the other end, w ­ e’ve explored how Earth’s rotation and distribution of land masses shape the long-­ term evolution of global and regional patterns of species richness (chapter 3). ­These two approaches must interact and influence one another, yet they have generally been studied by dif­fer­ent scientists using dif­fer­ent methods at dif­fer­ent scales (Ricklefs 1987, Brown 1995). The challenge is that, while organisms operate at scales dictated by their own body dimensions, many of the major ecological patterns we seek to understand emerge at the scales of coasts, ocean basins, and continents—­requiring a long-­term evolutionary perspective that includes drifting continents and diverging lineages (Witman and Roy 2009). How can we bridge ­these micro and macro perspectives? And do the bridges reveal something that is not evident in ­either approach alone? For example, can patterns in organismal traits move us beyond taxonomic idiosyncrasies to reveal general princi­ples of how ecosystems work? Macroecol­ogy focuses on the patterns and relationships that emerge when considering large numbers of species and individuals across broad spatial scales, linking levels of organ­ization from gene through biochemical reaction to landscape. The bricks in the bridge between scales, and between subfields of ecol­ogy, are functional traits, which mediate resource use, interactions with competitors

133

Ocean Ecology

al ote nti

Mid

1,000

Low

vi t

Algal functional groups Leathery macrophytes Articulated calcified Corticated macrophytes Foliose algae Filamentous algae Microalgae Crustose algae

V. low

(D) General model

High

High

tia ity

ity

uc Low

Pro d

uc

Pro d

Low Low

tiv

tiv

100 0

po

po

200

tia

Mid Mid Mid

l

l

ND

ten

(C) St. Croix, USVI

0

ten

0

500

Pro du c ti

Low Low

1,000

yp

yp

2,000

High High Mid Mid Low Low V. low

vi t

3,000

300

1,500

al

High

ND

Pro du c ti

4,000

(B) Washington, USA

ote nti

(A) Maine, USA

Biomass

134

Disturbance potential

Figure 5.11. ​Distribution of seaweed functional groups on hard substrata as a function of environmental productivity potential and herbivore pressure (disturbance potential). (A–­C) Empirical data from three regions and (D) predictions of a general model (­after Steneck and Dethier 1994).

and enemies, population growth, and the resultant biogeochemical fluxes. Although t­ here are many such traits, a surprisingly small subset explains a good deal of the variance in composition and functioning of ecosystems. Chief among t­ hese, as w ­ e’ve seen, are the traits that underlie metabolism and the stoichiometry of living organisms. Macroecol­ogy focuses on two general classes of questions. First, do organismal traits and population characteristics vary systematically with one another across space and environmental gradients? Second, what can we deduce about ecological pro­cesses from ­those patterns? By necessity, macroecol­ ogy addresses t­ hese questions from a primarily statistical, observational a­ ngle. It seeks to understand relationships among the organismal traits of body size and trophic level, the population characteristics of abundance and range size, and community characteristics of species richness and total biomass. Macroecol­ogy was born from empirical patterns, but it is increasingly grounded in theory, notably the metabolic theory of ecol­ogy (Brown et al. 2004), which links ­these fundamental properties of life via the under­lying pro­cesses of metabolism, life history (West et al. 1997, Enquist et al. 1999, Savage, Gillooly, Brown, et al. 2004), and regularities in trait distribution that emerge among interacting species (Loeuille and Loreau 2005, Harfoot et al. 2014). Macroecol­ogy is in a sense an extension of the allometry of individual organisms to the allometry of ecosystems, aiming to unite t­ hese scales via common princi­ples based on the biophysical constraints on life.

135

Chapter 5 Organisms

Organisms in the Anthropocene

3.6

4.0

Trophic level 4.4

4.8

5.2

1 Biomass (log10)

From a practical perspective, macroecol­ogy offers promise in reducing complex ecosystems to more manageable dimensions (Brown 1995) and, for some management prob­lems, provides alternatives to labor-­intensive approaches based on species or functional groups, requiring fewer par­ameters. Examples include modeling global fish production ( Jennings et al. 2008) and using regularities in the shape of biomass distributions among trophic levels as null distributions for mea­sur­ing fishing impacts ( Jennings and Blanchard 2004, Graham et al. 2005) (figure 5.12; chapter 9).

0

–1

1982 2001

–2

0

1

2

3

4

5

Body mass (log10)

Thinking of an organism as part of a continuing ge­ne­tic lineage—­a timeline—­emphasizes that it is more than a particle responding Figure 5.12. ​Estimation of fishing impacts from size spectra. passively to environmental forcing. Species are capable of adaptThe dotted line shows the predicted size spectrum of an ing to change, within limits. This evolutionary adaptation is conunexploited fish community, assuming a typical trophic transfer efficiency (0.125). The dashed and solid lines show size spectra stant and can be surprisingly rapid. Evolution has implications for the exploited North Sea fish community in 1982 and 2001, for how marine ecosystems respond to the rapidly changing enrespectively, with filled dots representing mea­sured biomass in vironment of the Anthropocene, driven by h­uman impacts each body mass interval in 2001. Note that the observed size spectrum follows the predicted one closely in smaller size (Palumbi 2001). Evolutionary se­lection is part of the explanaclasses but declines sharply in larger size classes, a signal that tion for why heavi­ly harvested fish stocks often fail to rebound larger size classes are well below steady-­state biomass due to to large body sizes even well ­after fishing has ceased (Neubauer harvesting (­after Jennings and Blanchard 2004). et  al. 2013), why hatchery-­raised juveniles released into the wild reduce the fitness of wild salmon populations (Heath et al. 2003), and why most antibiotics ­don’t work like they used to (Williams and Nesse 1991, Gluckman et al. 2011). A trait-­based focus helps us understand how marine biodiversity is changing in the Anthropocene (chapter 4), both by favoring or disfavoring organisms with par­tic­u­lar types of traits and by evolutionary adaptation, which is faster in species with certain traits (small size, short generation times) than in ­others. The two primary variables in metabolic scaling models, temperature and body size, are respectively a major component of global change and the central trait mediating biological responses to this and other ­human pressures on ecosystems. With re­spect to climate, metabolic models predict that, all ­else being equal, warming ocean temperatures w ­ ill raise individual metabolic rates, potentially reducing average population sizes. Warming can also reduce average body sizes within species since maturation rates respond more strongly to warming than do growth rates (Angilletta et al. 2004). The increasing acidity of seawater resulting from absorption of anthropogenic carbon influences organisms in several ways. In most marine animals, physiological per­for­mance declines as PCO2 increases (Wittmann and Pörtner 2013). Th ­ ese individual effects can have far-­reaching ecosystem impacts when they affect ecologically impor­tant species. One of the best-­documented effects of ocean acidification involves the dissolution of calcium carbonate. Many marine organisms incorporate CaCO3 as a structural component in their bodies, including coccolithophorid phytoplankton, pteropod mollusks among zooplankton (Orr et al. 2005), benthic bivalve mollusks, reef corals, and echinoderms. Calcifying organisms, including most echinoderms and mollusks, are especially sensitive to acidified seawater ­because their calcareous shells or skele­tons are vulnerable (Kroeker et al. 2010). Since bivalve mollusks can filter large volumes of w ­ ater and sea urchins are impor­tant benthic herbivores, the vulnerability of ­these animals may knock on to affect algal populations and the efficiency of trophic transfer up the food chain. Indeed, sea urchins are less abundant around naturally acidified shallow-­water vents in the Mediterranean Sea, and instead ­these vent areas support luxurious growth

136

Ocean Ecology

of benthic algae, including invasive species (Hall-­Spencer et al. 2008). Similarly, calcareous intertidal organisms have fared poorly during intermittent episodes of low pH in the Northeast Pacific (Wootton et al. 2008). Biological effects of ocean carbon absorption are often not easily predictable. For example, calcareous coccolithophorid phytoplankton have increased, not decreased, in North Atlantic w ­ aters over the last half c­ entury (Rivero-­Calle et  al. 2015), prob­ably due to release from carbon-­limited growth. The resulting shift in phytoplankton community composition has consequences for the nature of export to the deep ocean: data from sediment traps show that carbonate flux has increased relative to silicate flux. Ocean acidification also interferes with sensory abilities of many organisms, influencing their interactions. For example, among predators, acidification altered the be­hav­ior of mud crabs and therefore decreased their predation rate on eastern oysters (Dodd et al. 2015), and also reduced the ability of hermit crabs to locate prey via chemical cues (de la Haye et al. 2012, but see Clark et al. 2020).

­Future Directions A frontier to which we return repeatedly is the goal of identifying traits of organisms that help predict their interactions and responses to environmental change. Focusing ecol­ogy around functional traits can make the science more mechanistic and more generalizable, transcending the idiosyncrasies of geography and taxonomy. Applications include (1) describing interactions and community structure (chapters 7, 8) in terms that are inherently related to function; (2) predicting what types of organisms ­will be most affected by environmental change (chapter 4); (3) predicting what types of organisms have strong effects on par­tic­u­lar ecosystem pro­cesses and ser­vices (chapter 9); and (4) connecting t­hese approaches to develop a predictive framework for how environmental change ­will affect ecosystem pro­cesses, mediated by functional changes in the biological community. In this chapter we have concentrated on two of the most fundamental functional traits: body size and organismal stoichiometry. The evidence is abundant and clear that both strongly affect the structure of communities and ecosystems as well as the fluxes of materials and energy through them. The metabolic theory of ecol­ogy provides a foundation for the importance of body mass and temperature in ecosystems. Further development of this theory, and challenging it with empirical data, should prove useful in building a more nuanced, mechanistic basis for global change ecol­ogy and applications. A promising example involves marine plant-­herbivore interactions since it is well documented that plant stoichiometry strongly influences the fate of primary production across a range of ecosystem types (figures 5.3, 5.4), and herbivore impacts also change predictably with temperature (chapter 7). Body size influences many aspects of demography and trophic interactions in predictable ways (chapters 6, 7). An impor­tant step for applied ecol­ogy ­will be refining pa­ram­e­terization of body size effects for dif­fer­ent taxa and systems, ideally tied to general princi­ples based on traits. For example, a global analy­sis of marine mammals that incorporated birth rate, social group size, foraging area, and range size, in addition to body size, identified 15 species at risk of extinction that ­were not on the IUCN Red List (Davidson et al. 2012). Similar searches for sets of functional traits that allow useful predictions in ecol­ogy w ­ ill bear much fruit. Despite the benefits of trait-­based approaches to functional ecol­ogy, ­there are several challenges to implementing them more broadly (Violle et al. 2007, Luck et al. 2012). First, the traits most useful in mechanistic ecol­ogy are ­those that are directly linked to ecological pro­cesses, which are often challenging to mea­sure or derive from the lit­er­a­ture. Traits most useful in functional ecol­ogy should also be more rigorously comparable among studies. This argues for a systematic approach to selecting and justifying traits. Trait-­based approaches are best developed for terrestrial plants, for which a global archive of standardized, quality-­checked plant trait data has been assembled from some 69,000 species (Kattge et al. 2011). Analyses of this database have revealed inherent trade-­offs among dif­fer­ent traits,

Chapter 5 Organisms

synthesized the numerous traits into syndromes of functional variation, and linked t­ hese to demographic and ecosystem pro­cesses. For example, the “leaf economics spectrum” synthesizes the correlated traits of leaf mass per area, nitrogen content, and life span into an essentially one-­dimensional axis, from species with rapid growth and nutrient utilization to ­those with slower rates (Wright et al. 2004). Trait-­based approaches are less well developed for animals, but corals are not far b­ ehind plants (Darling et al. 2012, Madin, Anderson, et al. 2016; Madin, Hoogenboom, et al. 2016). Trait lists have been developed for freshwater benthic invertebrates (Poff et al. 2006), marine invertebrates (Bremner et al. 2006), and vertebrates generally (Luck et al. 2012). Among animals, trait-­based approaches are also well developed for fishes, where they have been used successfully to predict the characteristics promoting nonnative invasion (Kolar and Lodge 2002) and extinction (Olden et al. 2008) in freshwater, and susceptibility to fishing in marine fish assemblages ( Jennings et al. 1998). For example, the steady increase in fishing in the North Sea since 1925 produced community-­wide changes in average life history traits, resulting in a community with lower average body size, lower age and length at maturity, and faster mean growth rate—in other words, an increase in “weedy” species. Systematic frameworks for identifying and quantifying functional traits and applying them to field data are advancing rapidly (McGill et al. 2006, Villéger, Mason, et al. 2008, Luck et al. 2012, Reecht et al. 2013) and show promise for making ecol­ogy more mechanistic and general.

Summary The individual organism is a unique product of a ge­ne­tic code, s­ haped adaptively by many generations of interaction with the environment and by development during individual ontogeny. From a functional perspective, organisms can be analogized to chemical reactors that extract energy and materials from their living and nonliving environment and pro­cess the materials to build their own bodies and do vari­ous kinds of work. The rates at which they do so are governed by princi­ples of physics and chemistry and mediated by functional traits, that is, phenotypic characteristics that influence their fitness. Carbon is the backbone of all biochemicals, both structural materials of organisms and energy reserves; thus carbon is a commonly used currency in energy bud­gets of organisms as well as ecosystems. Nitrogen and phosphorus are key components of proteins and nucleic acids, and their availability often limits organismal growth in per­for­mance. In the open ocean far from land, trace ele­ ments, notably iron, can limit growth. Differences among organisms in their requirements for, and ability to obtain, ­these nutrients are key dimensions of their fundamental niches. Body size is the most fundamental characteristic of an organism influencing its function, and temperature is the most fundamental such aspect of the environment. The metabolic theory of ecol­ogy predicts that many biological pro­cesses scale with quarter powers of body mass (exponent with absolute value of ¼ or ¾) when corrected for temperature, prob­ably driven ultimately by relationships between body size and the structure of biological transport networks. Substantial evidence supports t­ hese predictions for metabolism, growth, maturation, and mortality, although debate continues about both the evidence and proposed mechanisms. Applying metabolic princi­ples to data on body size and environmental temperature yields quantitative generalizations—­laws of allometric scaling—­that link ecological pro­cesses from molecular to ecosystem levels. Similarly, at a finer scale, we can approach an organism’s activities in its ecosystem as a function of its traits, which define its niche—­the set of abiotic conditions and interactions with other species within which the species is capable of living. Th ­ ese functional traits provide a basis for community and ecosystem ecol­ogy that is mechanistic and transcends taxonomic idiosyncrasy.

137

6 I

138

Populations

t is a curious fact that variation in living nature is not continuous but is instead divided more or less discretely into groups of organisms that are similar to one another and distinct from o­ thers. ­These groups are species. With some impor­tant exceptions (notably among life stages, discussed below) members of a species are usually similar enough to one another in appearance and in what they do that they are generally considered interchangeable in ecol­ogy. A population is a group of such individuals of the same species that live within some defined area and that are isolated to some degree from other such populations. From an evolutionary perspective, a population is more narrowly defined, at least in sexually reproducing species, as t­ hose conspecific individuals living close enough to one another to have some probability of interbreeding. The structure, growth, and dynamics of populations are central to both basic and applied ecol­ogy. The adaptation that molds individual organisms to their environment is a product of changes in gene frequencies that happen within their populations. The mathematical ele­ments that describe population growth are mostly the same ones that describe the relative growth rates of ge­ne­tic lineages that produce evolutionary change. At the community level, population dynamics quantify how species respond to environmental forcing, resources, competitors, predators, and ­human activities. At the ecosystem level, population dynamics influence biomass production and the rates of organism-­mediated energy and materials fluxes; fishery management, for example, is largely a science of quantitative population dynamics. Defining a population is easier in theory than in practice among marine organisms for at least two reasons. First, ­because the ocean is continuous around the globe, t­ here are fewer clear bound­ aries to dispersal of organisms than on land, and therefore marine populations are often less clearly defined in space. Second, populations of marine organisms tend to be less discrete ­because many produce long-­lived larvae capable of dispersing over large distances. Indeed, in the ocean, habitats are often patchier and more discrete than the populations of organisms that inhabit them. As a result, many marine organisms exist as metapopulations, that is, “populations of populations” (Levins 1970), or groups of spatially separated populations occupying relatively discrete habitat patches, which interact by exchanging individuals via dispersal and/or migration. The connectivity among such populations—­the frequency and magnitude of individual movement between them—­has impor­tant consequences for the dynamics and stability of the species, its response to perturbations (like fishing), and hence the effectiveness of management. For example, a population wiped out by a disturbance can be rapidly restored if abundant larvae flow in from elsewhere, whereas even a thriving population may die out quickly if its larvae disperse widely and it depends on distant larval sources for recruitment. Thus, larval dispersal is a central theme in both basic and applied marine ecol­ogy. We begin this chapter by reviewing the development and life history of marine organisms, then turn to the conceptual framework that connects organismal traits to population growth and thus facilitates predictions of population dynamic responses to environmental change. The traits most directly affecting population dynamics are the components of life history, so we then turn to a

Chapter 6 Populations

139

discussion of how variation in life history affects the vital rates that influence population dynamics. Differences in t­ hese vital rates among individuals can result in differential repre­sen­ta­tion of their ge­ne­tic lineages in subsequent generations, that is, evolutionary change in the population. With this general foundation, we then consider examples of how population dynamic models are applied in conservation and management to explore responses of marine species to threats or exploitation, including both ecological effects on population growth and evolutionary effects on the ge­ne­tic constitution of populations.

Development and Life History Most organisms begin life as a single cell. For microbes, which typically live as individual cells or loose aggregations, development is a straightforward pro­cess of growth and division into two copies of the parent. But for sexually reproducing multicellular plants and animals, the fusion of gametes into a single-­celled zygote is the beginning of a long journey though drastically dif­fer­ent conditions of life and often dif­fer­ent environments. This is especially true for the many marine algae and animals with complex life histories. A “typical” marine benthic invertebrate might begin as a fertilized egg; hatch into a planktonic larva radically dif­fer­ent in size, form, and ecol­ogy from the adult; drift for some time as a temporary member of the plankton (meroplankton) in the open w ­ ater column, feeding (or not) on even smaller suspended materials; ­settle to the bottom; undergo a radical metamorphosis; and begin growing into an adult (figure 6.1A). Many marine algae similarly pass through parts of the life history that differ not only in size and ecol­ogy but in ploidy level (i.e., the number of replicate sets of chromosomes); ­these organisms undergo so-­called alternation of generations (figure 6.1B).

(A)

(B)

Sporophyte

Six naupliar stages

Larval release

Early stage nauplius

Spores

Gametophyte (female)

Late stage nauplius Moult to cypris

Diploid (2n)

Haploid (1n)

Cypris

Egg

Sperm Settlement and metamorphosis

Zygote

Figure 6.1. ​Complex life histories are typical of marine organisms, exemplified by (A) an acorn barnacle and (B) a kelp. The barnacle passes through two larval stages, eventually s­ ettles to the bottom, and metamorphoses into a sessile, plated adult attached to the bottom and suspension-­feeding from the ­water column. The kelp displays alternation of haploid and diploid generations that differ strongly in size and ecol­ogy (http://­www​.­asnailsodyssey​.­com​/­IMAGES​/­BARNACLE​/­PechenikEtAl1998Fig3​.­gif).

Gametophyte (male)

140

Ocean Ecology

The life history of an organism has far-­reaching consequences for population size, distribution, response to environmental forcing, and biogeography. Obviously, organisms in their early life stages—­ differing in body size, habitat, and so on—­experience quite dif­fer­ent conditions and needs than adults of the same species, so it’s not surprising that they also differ in their traits—­morphology, be­ hav­ior, diet, and habitat use. A mighty bluefin tuna, for example, can reach 450 kg as an adult but must pass through a single-­celled stage like every­one e­ lse, meaning that a young bluefin larva is perhaps seven ­orders of magnitude smaller in body mass than the same individual as an adult. So the bluefin’s role in the pelagic ecosystem differs depending on its life stage. Whereas an adult bluefin is a top predator of the open ocean, a bluefin larva might be eaten by many of the same fish species that the adult eats, or by a jellyfish, or even a shrimp. ­These transitions among life stages and their ecological consequences are molded by basic constraints on life in the ocean, both intrinsic to organisms and extrinsic in the environment. Intrinsic constraints on an organism’s development stem from the finite nature of resources and consequent trade-­offs among competing demands for t­ hose resources. Natu­ral se­lection acts to allocate resources among growth, reproduction, and other demands so as to approach maximal lifetime reproductive success. Such constraints include trade-­offs between growth and reproduction, current and ­future reproduction, and so on. One of the major intrinsic trade-­offs, which defines the broad categories of marine life histories, is between number and quality (size) of offspring. Given finite resources, an organism can produce many small offspring or few large ones. The “many-­small-­eggs” end of the spectrum is known as planktotrophy: planktotrophic larvae must feed on other, even smaller plankton to complete development. They hatch from small eggs, spend some weeks or even months feeding in the plankton, and may be dispersed over hundreds of kilo­meters before settling. Examples include many familiar fishes, some of which can produce millions of eggs during their lifetimes. At the other end of the spectrum, the “few-­large-­eggs” situation is lecithotrophy (“yolk-­feeding”): eggs are provisioned with energy (yolk) to give them a head start in development, which in some cases can carry the larvae all the way through settling without the need for food. An extreme form of lecithotrophy is direct development (brooding): eggs hatch into juveniles that are miniature adults, bypassing the planktonic larval stage entirely. ­These brooded eggs require enough energy to carry the offspring through emergence as a miniature adult. Sharks and rays, for example, brood their eggs and give birth to live young—­some species produce only a single offspring each year. Some deep-­sea amphipods produce fewer than 10 eggs each year. What explains the wide variation in life history among marine organisms? This question has a long history (Thorson 1950, Strathmann 1985). Planktotrophy offers high potential returns in the number of offspring produced but also involves a greater risk of recruitment failure ­because the small offspring are minimally provisioned by their ­mother and are therefore more vulnerable to starvation and predation in the plankton. For ­those reasons, planktotrophy is predicted to prevail in food-­rich environments. Several lines of evidence support this hypothesis. First, studies of a marine copepod show that, within the species, life history is adjusted in response to food availability, illustrating the trade-­off between offspring size and number (Guisande and Harris 1995, Guisande et al. 1996). During a bloom of its phytoplankton food, the copepod Euterpina acutifrons produced large numbers of small eggs, presumably b­ ecause the abundance of food would allow many small larvae to survive. But as phytoplankton concentration decreased, the copepods produced smaller numbers of larger, better provisioned eggs that could carry the larvae through lean times (figure 6.2). Thus, t­ here was a trade-­off between egg size and number, which responded to a change in environmental favorability. Lab experiments mirrored ­these field data and confirmed that the trade-­off was driven by food availability. Experiments with another copepod showed why: larger eggs are advantageous u­ nder poor food conditions b­ ecause newly hatched larvae from big eggs indeed survived better than ­those from small eggs (see figure 6.2D). Comparisons across a range of pelagic marine species support this general relationship between investment in eggs and survival (figure 6.3):

Chapter 6 Populations

(B) Field

17

4

40

16

3

15

2

14

30

30 20

4

25

20

3

2

0

0

15

1

50

69

69

67

350

67

67

65

30 65 20 10 0

ND J F MAM J J A S OND J

1994

1995 Month

65

63

63

61

61

1996

Naupliar body length (μm)

12

Eggs size (μm)

10

Eggs size (μm)

1

Eggs size (μm)

13

40

Eggs per sac

(D) Lab

Naupliar survival (days)

50

(C) Lab

Eggs per sac

5

Eggs per sac

18

Fluorescence

Temperature (°C)

(A) Field

63

61

0

1

2

3

4

Fluorescence

5

6

59

141

0

10,0000 20,0000 30,0000

328 306 284 262 240 140

Food concentration (cells/ml)

150

160

170

180

Eggs size (μm)

Figure 6.2. ​Trade-­offs between egg size and egg number in the marine copepod Euterpina acutifrons. (A) Field time series show that high phytoplankton biomass (fluorescence; green) tracks declining temperatures (black), and is in turn tracked by increasing clutch size (egg number; open symbols, lower panel) and decreasing egg size (filled symbols, lower panel). (B) Egg number increases and egg size decreases with higher phytoplankton biomass in the field. (C) Egg number increases and egg size decreases with higher phytoplankton biomass in a laboratory experiment. (D) Larvae from large eggs survived starvation better in a lab, suggesting that large eggs ­were adaptive ­under low food conditions (­after Guisande and Harris 1995, Guisande et al. 1996).

–1

Natural mortality (d )

species with small eggs suffer greater mortality in the plankton (McGurk 1986). Thus, among species and in some cases also within species, egg size and number are adjusted to maximize reproductive success u­ nder prevailing conditions. Although planktotrophic species can produce large numFish eggs bers of offspring, they experience greater risk of recruitment 1 Fish larvae failure due to starvation and predation in the plankton. This Juvenile/adult fish Whales risk is illustrated by comparisons of recruitment variability –1 10 Pelagic invertebrates among 86 commercial fish stocks, where recruitment variability was strongly related to the change in body length of larvae, 10–2 that is, duration of larval life (which tends to last longest in species with small eggs that develop into planktotrophic lar10–3 vae) (Pepin and Myers 1991). In other words, well-­provisioned larvae from large eggs develop rapidly and have more consis10–4 tently successful recruitment, whereas ­those from small eggs 10–6 10–4 10–2 1 102 104 106 stay longer in the plankton and suffer greater average mortality Dry mass (g) and greater variance in recruitment. This risk of recruitment failure is greater in more variable environments, where larvae Figure 6.3. ​Large, well-­provisioned larvae develop rapidly and are more likely to encounter poor food conditions. And that survive better. Estimated instantaneous daily mortality declines leads to a second prediction—­planktotrophy should be more with body size across a broad range of marine organisms, common in more constant environments, whereas brooding including larvae (McGurk 1986).

142

Ocean Ecology

of eggs should prevail in highly variable or food-­poor environments. This prediction is supported by global variation in the distribution of marine life histories: brooding and direct development are common among animals in the deep sea, where food is very scarce and, at high latitudes, where food availability is highly variable seasonally and from year to year ( Jablonski and Lutz 1983).

The Prob­lem of Larval Dispersal Our discussion of life history strategies has so far glossed over a more fundamental question: Why should an animal that lives on the seabed send its young out into open ­water far from home in the first place? A benthic organism that successfully reproduces is living at a site that, by definition, is favorable for its survival and reproduction. What could be the advantage of scattering its offspring to the four winds where their prospects may be poor? Yet most marine organisms do exactly that. The adaptive significance of larval dispersal—or lack thereof—is an enduring prob­lem in marine ecol­ogy (Thorson 1950, Strathmann 1985). Two general types of hypotheses have been offered to explain larval dispersal, based on w ­ hether it provides a benefit in itself or is simply a by-­product of another adaptation. The first and historically dominant hypothesis posits that planktonic dispersal provides a direct fitness benefit by boosting the chances of per­sis­tence in unpredictable environments. In other words, spreading larvae is a form of bet-­hedging, reducing variance in survival in a heterogeneous environment. Bet-­hedging amounts to trading off some cost in reduced average (or maximum) per­for­mance in exchange for the benefit of reduced variance in per­for­mance. Field evidence from marine invertebrates provides some support for this hypothesis: among 570 time series from 170 species of benthic invertebrates, variation through time in adult abundance was indeed lower in planktonically dispersing species than in ­those with direct development, prob­ably ­because planktonic dispersal averaged recruitment success over space (Eckert 2003). Interestingly, this is opposite to the conclusion from the study of variability in fish recruitment (Pepin and Myers 1991) mentioned above. The second class of hypotheses suggests that larval dispersal is not an adaptation at all but rather an incidental consequence of se­lection for high fecundity. That is, given a fixed allocation of energy for reproduction, producing more eggs requires making them smaller on average and, as w ­ e’ve seen, small eggs hatch into small larvae that must feed and grow substantially before settling. Key to evaluating this hypothesis are the very dif­fer­ent conditions of life experienced by dif­fer­ent life stages of the same organism. A large adult necessarily passes through small stages to get ­there. If small stages have very dif­fer­ent environmental needs (food, predation refuge) than ­those for adults, and if ­those needs are better met in the w ­ ater column, then dispersal into the ­water column could be adaptive. What evidence exists for this hypothesis? One of the key differences between the pelagic and near-­bottom environments is the generally much higher density of animals, including predators, on and around the bottom. Very small animals are especially vulnerable to predators, and thus one hypothesis for larval dispersal is that release of larvae into the plankton reduces their exposure to predators during vulnerable early life stages (Strathmann 1985). William Hamner and colleagues (1988) coined the term “wall of mouths” to describe the high density of planktivorous fishes that occupy reef edges, scouring zooplankton from the incoming oceanic ­water. More generally, benthic and demersal animals may represent a “carpet of mouths” that poses high risk of predation for small animals, like larvae near the bottom, and therefore selects for dispersal into the w ­ ater column, where predators are fewer. Some support for this hypothesis comes from a clever experiment with postlarvae of spiny lobsters (Panulirus argus). ­These crustaceans support valuable fisheries in the Florida Keys, USA, as their relatives do throughout the world tropics. Spiny lobsters produce large, transparent larvae that drift for months in the coastal ocean. When the time comes to return to shallow reefs and ­settle down to benthic life, they must pass through the gauntlet of dense predatory and planktivorous fishes—­the wall of mouths. They do so only at night, in surface ­water, on flood tides, during the new moon (figure 6.4). Why? To find out, Charles Acosta and Mark Butler (1999) tethered lobster postlarvae to floats,

Chapter 6 Populations

(A)

(B)

Florida

Atlantic Ocean

Gulf of Mexico

Caribbean Sea Florida Bay

Reef N Reef

(D)

(C) 100

100

Surface Bottom

80 Percent mortality

Percent mortality

80 60 40 20 0

New moon Full moon

60 40 20

Reef

Lagoon Habitat

Bay

0

Bay - surface Reef - deep Habitat/depth

Figure 6.4. ​Pelagic larvae avoid predation by benthic animals. (A) Trajectory of recruiting spiny lobster larvae over the Florida Keys reef tract, USA. (B) Large, transparent phyllosoma larvae of the spiny lobster. (C) Predation on drifting postlarvae tethered near ­water surface was less than that of postlarvae tethered near the bottom, and declined ­after passing from the ocean over the reef into the lagoon. (D) Predation on postlarvae was much lower during the new moon (dark) but only near the ­water surface (­after Acosta and Butler 1999).

drifted them across the reef in the Florida Keys at night u­ nder dif­fer­ent conditions, and mea­sured how many remained on the tether at the end. The experiment tested how predation on larvae varied by depth in the w ­ ater column (near bottom versus near surface), offshore versus reef versus inshore lagoon (bay), and bright full moon versus dark new moon. Consistent with the carpet-­of-­mouths hypothesis, they found that predation on larvae was highest near the bottom, where predators ­were densest, and during the full moon, when predators could see. ­These results support the idea that releasing larvae into the ­water column can reduce their exposure to the abundant predators on the bottom. Protecting larvae from benthic predators may also explain one of the most spectacular biological phenomena in the sea. On Australia’s G ­ reat Barrier Reef, numerous species of corals spawn en masse during a few nights ­after the full moon in spring (Harrison et al. 1984). Although many species spawn on the same night, time of gamete release generally differs among species. On one hand it makes sense that colonies within a species should spawn in synchrony to maximize fertilization. But why should multiple species do it on the same nights, raising the risk of producing inviable hybrid zygotes? The most plausible hypothesis is that semisynchronous mass spawning is an adaptation to overwhelm predators, that is, a predator satiation strategy. Thus, the offset spawning times on the

143

144

Ocean Ecology

(A)

(B) 16

20

15

18 N (0/00)

14

15

15

N (0/00)

14 16

13 12 11

12 10

10 4

6

8 10 12 Log2 maximum total mass

14

16

9

0

10

5

15

Log2 body mass

Figure 6.5. ​Trophic level is strongly related to individual body size rather than to maximum size of the species in North Sea fishes. (A) Estimated mean trophic level (indexed by the stable isotope signature, del 15N) plotted against maximum body size by species. (B) Estimated mean trophic level plotted against mea­sured body size of individual fish (Jennings et al. 2001).

same night could represent a compromise—­colonies within a species spawn synchronously to maximize fertilization, but spawning time is offset somewhat from other species to minimize hybridization. Since many species spawn on the same night, the massive density of larvae washing over the reef means that predators can only eat a tiny fraction of them, ensuring that any individual larva has a low probability of being eaten. One impor­tant consequence of complex life histories is that traits often change radically through the life history, meaning that the species name masks a potentially wide range of ecologically significant variation among individuals and, through time, even in the same individual. Fortunately, some of this variation follows predictable rules. The most universal such rule brings us back to body size. First-­feeding larvae of fishes are generally very small, and ­there is ­little that can support their rapid growth other than even smaller zooplankton, which is what most first-­feeding fishes eat. As a result, first-­feeding juvenile fishes from many families have far more in common with one another functionally than any one of them has to an adult of its own species. A striking demonstration comes from North Sea groundfishes, in which trophic level was estimated using the naturally occurring stable isotope 15N ( Jennings et al. 2001). In a large sample of fishes, t­here was l­ittle correlation between trophic level and the maximum body size of the species but a strong correlation between trophic level with the body size of individual fish (figure 6.5). In this community at least, trophic level is a characteristic of individuals rather than a species-­specific trait, and it changes consistently through the life history: individual fish rise higher in the food chain as they grow in size. As a result, community predator-­prey interactions are structured by body size rather than by species.

Population Growth: A Brief Review Understanding the distribution and abundance of organisms means understanding what limits the size and growth rate of their populations. Impor­tant ­factors include both intrinsic aspects of biology, such as body size and life history, and extrinsic aspects of the environment, such as habitat, resource availability, and temperature. To approach ­these population pro­cesses systematically, we need a quantitative framework for population growth. For unicellular organisms, like bacteria and phytoplankton, population growth is straightforward—­grow, divide, repeat. This can be modeled very simply with the logistic growth equation familiar from introductory ecol­ogy. But for higher organisms, the story

Chapter 6 Populations

145

(A) Life table is more complex as we have seen. Multicellular organisms grow through several life history stages that differ, sometimes radiFi = Pi = i l(x) b(x) x l(i)/I(i–1) b(i)Pi cally, in ecol­ogy. Juveniles ­don’t reproduce and reproductive 0 1.0 0 output of adults often differs with age or size. Calculating pop1 1 0.8 2 × 0.80 = 1.60 ulation growth in ­these species therefore requires considering 2 2 0.4 3 0.50 1.50 the population’s age structure. Fortunately, despite the diver3 3 0.1 1 0.25 0.25 sity of marine life histories, we can express virtually any envi4 4 0 0 0.00 0.00 ronmental influence on a population in terms of its effect on ­either (1) survivorship, (2) age at maturity, or (3) fecundity. (B) Resulting Leslie matrix ­These basic ele­ments combine to determine the rate of increase or decrease of the population, and they provide a ­simple 1.6 1.5 0.25 0 F1 F2 F3 F4 Fecundity (contribution 0.8 0 0 0 P1 0 0 0 framework for quantifying population pro­cesses. We assume A= A= to age class 1) 0 0.5 0 0 0 P2 0 0 ­here some familiarity with the basic equations of population 0 0 0.25 0 0 0 P3 0 growth; for a clear and concise treatment, see Nicholas Gotelli’s (2008) Primer of Ecol­ogy. Survivorship At the most basic level, population growth represents the (contribution to next age class) difference between rates of births and deaths of individuals. In unicellular organisms, such as phytoplankton and bacteria, birth (b) and death (d) are simply rates of cell division and Figure 6.6. ​(A) Life t­ able and (B) Leslie matrix for estimating death, respectively, and t­hese organisms approximate logistic population growth. Variables are described in the text (­after Gotelli 2008). population growth well in lab culture. In multicellular organisms, estimating population growth is more complicated ­because organisms may reproduce multiple times, and ­because survival and reproduction vary with an individual’s age. Calculating the growth rate of a population with age structure therefore requires dividing the individuals into age classes and, for each class, estimating per capita birth rate (i.e., fecundity, bx) and death rate (or its inverse, survivorship, lx), then integrating ­these over an average individual’s lifetime. ­These estimates are made using life ­tables of survivorship and fecundity by age (figure 6.6). A population’s mortality schedule, or age-­specific pattern of survivorship, and fecundity schedule, or age-­specific pattern of reproductive output, are obtained by estimating the following terms for each age class (x):

nx = the number of individuals alive at the beginning of the age interval (for age class 0 = n0), lx = the proportion of individuals in the original cohort surviving at the beginning of the age interval (nx/n0), bx = age-­specific fecundity (i.e., the number of female offspring per female during that age interval). Traditionally, ­these calculations focus on females b­ ecause it is relatively straightforward to mea­sure their reproductive output as the number of eggs produced, whereas this is much more difficult for males. The number of offspring produced by each age class (lxbx) is then summed over all age classes to get the total lifetime reproductive output of the average female (R0), that is, the number of surviving ­daughters expected per female in an average lifetime (or generation): n

R0 = ∑lxbx . x = 0 

Since R0 describes the reproductive output of an average female, it is also equivalent to the mean population reproductive output, or the per generation rate of multiplication of the population (Nt/ N0). It can be interpreted as the ratio between population size in the offspring generation (Nt) compared with the parental (N0) generation:

146

Ocean Ecology

R0 = Nt /N0. Lifetime reproductive output is used to define the more general finite rate of population growth, λ, as follows: Nt+1 = (1 + R0)Nt = λ Nt ,

λ = Nt+1/Nt = the finite rate of increase of the population. The net reproductive rate R0 is scaled to the organism’s generation time (lifetime). In contrast, λ (lambda) is a more general expression b­ ecause it can be defined for any time unit of interest. λ is an impor­tant quantity in ecol­ogy, evolution, and conservation. It is called the finite rate of population growth ­because it is calculated over a discrete (finite) time period, typically a year. This distinguishes λ from r, the instantaneous rate of population growth familiar from introductory ecol­ogy. Thus, λ is the ratio of population size one time unit in the f­ uture relative to its size now. For example, if λ = 1.2, then the population next year ­will be 20% larger than now. If lambda = 1.0, t­ here ­will be exactly the same number of individuals next year (i.e., the population is stable). It’s easy to see that a population ­will be stable (no change) when λ = 1, growing when λ > 1, and declining when λ  0.1 indicated in bold.

of loggerhead sea turtles is highly sensitive to survival of the large juveniles most vulnerable to bycatch in shrimp trawls, whereas protecting eggs and hatchlings on nesting beaches is expected to have ­little effect on populations (box 6.2). Extension of the matrix modeling approach to loggerhead turtles worldwide showed that fishing impacts differ regionally depending on w ­ hether fishing overlaps the habitats used by vulnerable life history stages of turtles and on the gear types used (Wallace et al. 2008). For example, turtles caught in longline fisheries tend to be smaller juveniles with low reproductive value (and low influence on population growth), whereas trawls take large juveniles with high reproductive value and elasticity for population growth. This analy­sis confirmed that the most effective conservation strategy for loggerheads throughout the world ocean would be management of turtle mortality from trawling. The challenge to exploiting demographic modeling more widely in conservation and management is that elasticity analy­sis requires detailed data on population structure and age-­specific vital rates, which are difficult to collect and therefore lacking for most wild organisms. Selina Heppell and colleagues (2000) found a way around this prob­lem. They showed, using detailed life ­table data from well-­studied mammals, that useful estimates of population growth can be calculated without full life ­table data, using only data on average adult survival, age at maturity, and an estimate of population growth rate, which are available for most exploited species. ­These three quantities in turn allow calculation of (1) fertility elasticity, the effect of a change in reproductive output across all adult age classes; (2) juvenile survival elasticity, the effect of a proportional change in all survival rates from age one to the period just prior to maturation; and (3) adult survival elasticity, the effect of a proportional change in all survival rates for mature individuals. Elasticity analyses across a range of marine taxa calculated using this approach showed that protecting adults (i.e., reduced mortality) has the

Chapter 6 Populations

153

Box 6.2. ​Matrix population models in sea turtle conservation The world’s seven species of sea turtles are in trou­ble. Adults and larger juveniles are killed as bycatch by shrimp trawlers and longline fishers, and the need to lay their eggs on beaches makes the young stages vulnerable to harvesting and coastal development. Population models can be used to explore how management actions can change survival or fecundity of dif­fer­ent life stages, thereby influencing the population’s growth rate and viability. Biologists long disagreed about ­whether sea turtle conservation should focus on saving eggs on beaches, large adults on the high seas, or somewhere in between. For example, efforts in the southeastern US have focused on protection of eggs on nesting beaches, yet a ­ fter de­cades of protection t­here was ­little sign that this increased the abundance of nesting turtles. Since resources for conservation are ­limited, the practical question arises, What is the most effective way to help turtles? To explore dif­fer­ent conservation scenarios, Deborah Crouse and colleagues (1987) built a matrix model using data from a 20-­year study of loggerhead turtles at ­Little Cumberland Island, Georgia, USA, and used it to proj­ect ­future population trends u ­ nder dif­fer­ent policy scenarios. Using previously calculated growth curves, they showed that population growth was most strongly affected by survival of large juveniles at sea, whereas the number of eggs produced and survival of small juveniles had relatively small effects (figure B6.2.1). T ­ hese findings have clear management implications. First, they imply that protection of eggs and juveniles cannot halt decline in loggerhead populations ­unless large, seagoing juveniles are also protected. Second, t­hose large juveniles are, perversely, of the size and location (coastal w ­ aters) most vulnerable to shrimp trawling, which kills more sea turtles in (A)

(B) 0.06 Intrinsic rate of increase (r)

0.1

0

–0.1

–0.2

= –10% mortality = –50% mortality = –90% mortality

0.00

Stage class

Eg

g/

ha t Sm chli ng all ju v en La rg ile ej uv en ile Su ba du lt Ad ul t Fe cu nd ity

–0.05

Eg Fe gs cu /h nd Sm atc ity all hlin La juv gs rg en e j ile uv s e No Su nile vic bad s 1s e b ul t-y re ts r e M rem der s at i ur gra eb n t re s ed er s

Change in rate of increase (r)

US w ­ aters than all other ­human activities combined. Fi­nally, the model helped pinpoint where additional data would be most useful: simulation showed that an acceleration of three years in age at first reproduction almost halted the projected population decline. Since maturation age is poorly known for most sea turtle populations, this emphasizes the value of obtaining such data. Happily, this study led to concrete conservation mea­sures, and the story has a hopeful ending. In the 1970s and 1980s, widespread deaths of turtles in fishing trawls inspired the development of the turtle excluder device (TED), a door in the trawl that allows trapped turtles to escape. Where used properly, TEDs virtually eliminated death of turtles in trawls and showed significant promise to turn around the decline of sea turtle populations. Therefore, TEDs ­were required on all US shrimp trawlers in 1987. Field evidence suggests that required use of TEDs indeed reduced sea turtle deaths. Comparison of turtle strandings (deaths) in South Carolina, USA, during periods when TEDs ­were required versus not used suggests that TEDs reduced strandings by 37%, and that the loggerhead turtle population grew at a rate similar to that estimated from the matrix model (Crowder et al. 1994). Striking evidence for the value of TEDs comes from an unpre­ce­dented spike in turtle strandings in South Carolina following expiration of TED requirements ­there in the fall of 1990, showing that the devices had been achieving their goal. A similar spike in strandings in the Gulf of Mexico in 2010 has been attributed to trawlers taking advantage of reduced enforcement of TED regulations in the wake of the Deepwater Horizon oil spill. Turtle excluder devices work ­because they target the life history stage with the strongest effect on long-­term population viability.

Stage class

Figure B6.2.1. Effects of changing fecundity and survival on projected changes in growth rate of a loggerhead sea turtle population. Variation in population growth is estimated from simulated changes in rates of individual life history stages, using a matrix population model. (A) The dashed line represents the intrinsic rate of population growth, r, estimated from the initial matrix. Simulations (bars) represent a 50% increase in fecundity or an increase in survivorship to 1.0 (Crouse et al. 1987). (B) Effects of three dif­fer­ent levels of reduced mortality at par­tic­u­lar life stages on loggerhead turtle population growth (Crowder et al. 1994). In both cases, large juveniles show the strongest response to simulated management actions that would affect survival or fecundity.

Ocean Ecology

Elasticity of adult mortality

3.5 3.0 2.5 2.0 1.5 1.0 0.5

de rin

g

Or ca alb at Le ro ss op ar d sh ar Gr k ay w ha Lo le gg Se iw er he ha ad le se Ke at m ur p’s tle rid ley tu rtl Ha e r bo No r rth se al er n So s ea ut he lio n rn se ao tte r Sa lm Re on d se au rc hi n Lit to rin S a No t r i p e d rth b as Se s ah ad do ck

0

W an

154

Figure 6.9. ​Modeled effects of protection in marine reserves on marine species spanning a range in life histories. Bars show elasticity of population growth in response to a 20% change in adult mortality (­after Gerber and Heppell 2004).

largest positive impacts on population growth in invertebrates and fishes, whereas it has ­little effect on marine animals with naturally low adult mortality, such as w ­ hales and oceanic seabirds (figure 6.9) (Gerber and Heppell 2004).

Life history and the effectiveness of marine reserves A major thrust of marine conservation in recent de­cades has aimed to establish protected areas where fishing activities that harm target species are controlled. Our discussion so far suggests that protection might have quite dif­fer­ent effects on species depending on their life histories. How can we predict how par­tic­u­lar species and life history stages ­w ill respond so management interventions are designed for maximum benefit? And, importantly, how are t­hese predictions affected by dispersal between protected and unprotected areas? Demographic models have been used not only to evaluate how marine reserves affect animals with dif­fer­ent life histories but also to prioritize habitat types required by ­those life history stages (essential fish habitat). As we saw for loggerhead turtles, elasticities can be calculated to estimate the influence of par­tic­ul­ ar life stages on population growth and thus identify life stages most sensitive to protection. This in turn can guide the choice of habitats most impor­tant to protecting them (Mangel et  al. 2006). ­These issues have been explored with stage-­structured matrix models like ­those discussed above, estimated separately for reserves and unprotected areas, including terms for migration between them (Gerber et al. 2005). To explore how animals with dif­fer­ent life histories would be affected by no-­take reserves, Leah Gerber and colleagues (2005) simulated population dynamics of four marine species—­a grouper, a sea urchin, a sea turtle, and a whale—­whose populations lived both in and outside of the reserve. They compared effects of a 20% reduction in mortality within a reserve to the same reduction spread over the ­whole population, both in and outside the reserve. Creating a no-­take reserve increased population growth rates for all species as expected, but when compared with an equivalent reduction in fishing spread across the region, the value of the reserve depended on the magnitude of dispersal, which differed strongly among species. Thus, the effectiveness of marine protected areas for boosting populations depends strongly on the species’ life history and be­hav­ior.

Chapter 6 Populations

California halibut (Paralichthys californicus) provides an example of the links between life history, habitat, and conservation strategy. Juvenile halibut use a variety of inshore habitats, including exposed coast and coastal embayments. ­After the adults spawn, larvae spend about a month in the coastal ocean before settling into shallow areas along the exposed coast. Some of the juveniles then move into bays and estuaries. Demographic analy­sis showed that both exposed areas and coastal embayments could contribute roughly equal numbers of recruits to the adult stock, but that positive growth of the regional population depended on the subpopulation of juveniles using coastal bays, lagoons, and estuaries (Fodrie, Levin, et al. 2009). In contrast, use of the exposed coast by juveniles actually depressed overall population growth in most study years b­ ecause of high juvenile mortality along exposed coast. The analy­sis illustrates that simply tallying recruitment into a habitat may be a misleading proxy for its importance since the recruits must also survive and grow to adulthood and habitats differ in their favorability.

Organismal Fitness and Adaptation to the Environment Up to this point we have treated individuals within age or stage classes as identical. This approximation is usually reasonable for estimating population growth rates and the sorts of scenario analyses we have considered. But of course individuals are not identical, and matrix models are also useful in exploring how environment and h­ uman impacts f­ avor individuals with certain traits and thus influence evolution of the population. Variation among individuals within a population (discussed in chapter 5) influences who reproduces and who d­ oesn’t and thus the population’s ge­ne­tic composition. The components of life history that influence population growth are therefore the same components necessary to understand evolutionary change. By comparing the fitness of individuals with certain traits, we can predict which w ­ ill be favored by se­lection and increase in frequency. Fitness is usually equated informally with superior survival and/or reproduction, but treating fitness quantitatively requires a more formal definition. The most commonly used quantitative estimate of fitness is the same pa­ram­e­ter used to define population growth, the instantaneous rate of natu­ral increase (r), or its equivalent in discrete time, λ, the geometric rate of increase. In general, we use the same math to describe changes in trait or gene frequencies in evolving populations as we do for describing changes in population density in ecol­ogy. As discussed in chapter 5, natu­ral se­lection arises as an inevitable consequence of three features that are essentially universal among living organisms: phenotypic variation among individuals, some degree of heritability of that variation, and differences in fitness among the variants. Darwin showed that if ­these premises are all met, one of a few outcomes necessarily follows, depending on the match between the population’s current trait distribution and what would work best in the current environment (i.e., the trait optimum). First, if the population’s mean trait value is near the optimum for the current environment, then variations at both extremes of the trait distribution (especially large or especially small offspring, for example) ­will tend to have lower survival and/or eventual reproduction, so they are weeded out. This reduces variance in the trait within the population without changing the mean value and is known as stabilizing se­lection. Stabilizing se­lection is often mediated by trade-­offs in advantages at the two extremes of a trait distribution. In the Ca­rib­bean damselfish Stegastes partitus, for example, large males have the highest reproductive success ­because they are better able to compete for and hold territories. In other words, (intra)sexual se­lection strongly ­favors large male body size. But larger males also mature ­later and are more vulnerable to predation, and t­hese fitness costs push back, generating natu­ral se­lection for somewhat smaller body size. The result of t­ hese opposing se­lection pressures is stabilizing se­lection for intermediate adult body size, which matches well the size distribution of individuals observed in the field (figure 6.10) ( Johnson and Hixon 2011). Stabilizing se­lection has been suggested as one explanation for the surprising consistency of morphology through long periods of evolutionary time

155

156

Ocean Ecology

(A)

(B) Age at first reproduction (days)

Ln (reproductive success)

10 8 6 4 2

1,200 1,000 800 600 400

7.0 7.5 8.0 8.5 9.0 9.5 Asymptotic size (male length, cm) (C)

(D)

(E) 2.5

0.95

15

0.90 0.85 0.80 0.75

Frequency

2.0 Relative fitness

Mean proportional survival

1.00

7.0 7.5 8.0 8.5 9.0 9.5 Asymptotic size (male length, cm)

1.5 1.0

10

5 0.5

0.70 0

2 4 6 8 Mean size (length, cm)

0

6

7 8 9 10 11 Asymptotic size (male length, cm)

0

6

7 8 9 11 10 Asymptotic size (male length, cm)

Figure 6.10. ​Stabilizing se­lection on male aysmptotic body size in the Ca­rib­bean damselfish Stegastes partitus. (A) Sexual se­lection strongly f­ avors larger males, who are better able to compete for mating territories. But larger males suffer fitness costs in terms of (B) l­ater maturation and (C) lower survival at the largest sizes. ­These trade-­offs result in (D) stabilizing se­lection for intermediate body size, which (E) reflects the size distribution observed in the field (­after Johnson and Hixon 2011).

in many organisms, known in paleontology as stasis (Eldredge et al. 2005). The same argument may help explain the high frequency of cryptic species in the ocean—­those that are morphologically very similar but genet­ically distinct (Knowlton 1993). The second pos­si­ble outcome of natu­ral se­lection occurs when the population’s current trait distribution does not match well with what works best in the current environment—­say, if smaller, faster-­maturing individuals in a fish population have an advantage b­ ecause trawl nets more effectively capture large individuals. In this case, the mean trait value (body size) may decrease as large individuals are preferentially captured and killed, leaving smaller mature individuals to contribute most of the population’s reproduction and therefore passing more genes from smaller individuals to the next generation. This is known as directional se­lection since the mean trait value moves away from the former average through time. This is what we typically think of as evolution, a directional change in a trait through time. Some of the most striking examples of directional se­lection in marine systems involve impacts of fisheries, which typically target large individuals. Since both size and growth rate in fishes are heritable, that is, partly determined by ge­ne­tics (Law 2000, Carlson and Seamons 2008), size-­selective harvesting should impose evolutionary se­lection on fish populations. This prediction is supported by laboratory experiments (box 6.3) and is also consistent with data from time series of several harvested fish populations in nature, which show substantial changes in age at maturation not

Chapter 6 Populations

157

Box 6.3. ​Fish evolution caused by fishing: An experiment they ­were harvested by fishers (the scientists) before they could reproduce. In this situation, individuals maturing at smaller sizes ­were the only individuals able to reproduce, and their genes increased in subsequent generations. The opposite was true of the population where small individuals ­were selectively removed: in this case se­lection favored rapid growth to pass through the vulnerable small-­bodied stage of the life history, and the population became dominated by large individuals. This experiment had a second, intriguing result. Population biomass production evolved rapidly in the direction opposite the size-­dependent se­lection. That is, perhaps counterintuitively, selective harvest of large individuals soon resulted in lower, not higher, biomass production (i.e., fishery yield) in each generation than in controls where individuals ­were harvested at random. The reason for this is that when large individuals ­were continuously removed from the population, the fish that remained to produce offspring tended to mature at small size b ­ ecause they grew slowly. This experiment demonstrated that a fishing practice qualitatively similar to that of much commercial fishing can cause dramatic evolutionary change in fish populations and also affect the major traits of value in the fishery, namely, body size and productivity.

(A)

(B) 100

0.15

95

0.10 Growth rate (mm/d)

Mean length (mm)

Fishing typically targets large individuals within a population and often imposes heavy mortality. This is a ­recipe for strong evolutionary se­lection on body size. To test the hypothesized evolutionary impact of fishing, David Conover and Stephan Munch (2002) conducted a laboratory experiment with silversides (Menidia menidia), a small schooling fish of the West Atlantic. Although silversides are not fished, they grow rapidly and are easily kept in the lab. The researchers subjected the fish populations to size-­ selective harvest in the lab for four generations, along with experimental controls (figure B6.3.1). In each generation, the fish faced intense simulated fishing in one of three treatments, all of which harvested 90% of the individuals but in dif­fer­ent ways: (1) selecting the smallest 90% of individuals in the population, (2) selecting the largest 90%, or (3) a control group from which 90% of individuals ­were removed at random. The experimental populations ­were then grown up in a common environment without simulated fishing to examine ­whether and how they had changed genet­ically. A ­ fter only four generations of this se­lection, the dif­fer­ent treatments produced ge­ne­tic evolution: where large individuals ­were harvested, the population became dominated by small-­bodied fishes—­even a ­ fter fishing ceased. In essence, the larger fish had low fitness ­because

90 85 80 75

0.00 –0.05 –0.10

70 65 –80

0.05

–60

–40 –20 0 20 Cumulative selection (mm)

40

60

–0.15

0

1

2 Generation

3

4

Figure B6.3.1. Directional se­lection on body size at maturity in Atlantic silversides, imposed by size-­selective harvesting over four generations in the lab. (A) Heritability of body size (mean length) in Menidia menidia was estimated as the relationship between experimentally imposed se­lection (for large body size) and the length of offspring grown ­under identical conditions. (B) Genet­ically based growth rates decreased over four generations when large individuals w ­ ere selectively harvested (yellow symbols) and increased when small individuals w ­ ere harvested (blue symbols), compared with randomly harvested controls (green symbols). Data come from offspring raised in a common-­garden experiment (Conover and Munch 2002).

easily explained by environmental ­factors ( Jørgensen et al. 2007), although such phenotypic changes are notoriously difficult to assign to ge­ne­tic change with confidence (Browman et al. 2008). ­These examples illustrate that populations are inherently variable and ever changing in response to their changing environments. The three features that foster evolutionary change—­variation, heritability, and fitness differences—­are universal in organisms, and so evolutionary change is a fact

158

Ocean Ecology

Number of taxa

of everyday life. This is impor­tant ­because evolution in real time affects our expectations, which are usually based implicitly on the assumption that populations are genet­ically static or uniform. For example, a warming climate might render a place uninhabitable for a species ­because it ­can’t stand the heat. This is the default assumption in environmental niche models used for predicting responses to climate change. But a warming climate might also select for ge­ne­tic adaptation to higher temperatures, yielding a dif­fer­ent result for predicting how climate change w ­ ill influence species distributions. Just such a result is suggested by a 15-­year time series of phytoplankton distributions off Venezuela, which shows that the realized niches of phytoplankton with re­spect to w ­ ater temperature, light, and nitrate levels tracked long-­term trends in rising ­water temperatures, and that species occupied warmer ­waters over time (Irwin et al. 2015). This trend is unlikely to be explained by nonge­ne­tic physiological acclimation, which tends to be rapid in unicellular phytoplankton and would easily equilibrate within the monthly sampling interval. It therefore appears that species niches are shifting away from environments that are declining, and d­ oing so faster than the expected rate of climate change. Thus, phytoplankton community composition is changing in a way that would be missed by assuming fixed traits in ­these species. It is increasingly recognized in environmental science that evolution is often of similar speed and magnitude to ecological pro­cesses such as habitat change and succession. Indeed, evolution in response to ­human impacts is comparable to—or even stronger than—­the background rates of “natu­ ral” evolution (Palumbi 2001, Darimont et  al. 2009). The rapid evolution of fish in response to ­human harvesting (see box 6.3) is echoed by field evidence that eggs of captive salmon held in a hatchery declined by ~30% over less than two de­cades of domestication and that this difference was genet­ically determined. Soberingly, models indicated that females producing t­ hese eggs would have 24% lower fitness if returned to the wild (Heath et al. 2003). Similar changes in marine species can be expected to be widespread, as domestication of marine species is accelerating rapidly (Duarte et al. 2007). Such human-­induced evolution appears common—­synthesis of more than 40 studies of human-­harvested systems found that resulting phenotypic changes are much faster than responses to natu­ral or other ­human ­drivers (Darimont et al. 2009). 6 4 2 0 4 2 0

Macroalgae

8 6 4 2 0

Sessile filter feeders

Dispersal, Recruitment, and Metapopulations

Marine herbivores

30 20

Terrestrial plants

10 0 6 3 2 0

Herbivorous insects

0.1m

1m

10m 100m 1km 10km 100km 1000km Estimated dispersal scale

Figure 6.11. ​Dispersal distances vary widely among functional types of organisms. Panels show differences in distribution of dispersal distances among groups of marine and terrestrial organisms (­after Kinlan and Gaines 2003).

Dispersal biology is a central ­factor in structuring communities (chapter 8). Dispersal distances vary over five o­ rders of magnitude among marine organisms generally, from certain seaweeds and sessile invertebrates whose propagules manage to move only a few centimeters before settling, to certain large vertebrates whose ambits span the entire global ocean (Kinlan and Gaines 2003) (figure 6.11). For ­these reasons, some widely dispersing species, such as pelagic seabirds and large oceanic predators, can have a single global population, whereas highly sedentary species have genet­ically distinct populations on nearly e­ very reef or rock outcrop (Burton and Feldman 1981, Taylor and Hellberg 2003). Most of this variation stems from life history characteristics, such as the mode of larval development. The commonness of pelagic larvae, with the potential to disperse over long distances, is a key feature of life in the sea, and distinguishes it from that on land, although realized dispersal distances in marine species are often much smaller than larval biology would imply. Maximum dispersal distances of marine

Chapter 6 Populations

Juveniles

Juveniles Juveniles

Larvae

Adults

Larvae

Adults

Adults

Larvae

Adults

Adults

Figure 6.12. ​A conceptual framework for modeling marine metapopulations connected by dispersal of larvae, movement of adults, and/or ontoge­ne­tic shifts in habitat use (­after Gaines et al. 2007).

species with sedentary adults are estimated to be one to two o­ rders of magnitude greater than t­ hose of terrestrial plants, on average (Kinlan and Gaines 2003). That difference has consequences: in contrast to the terrestrial case, local demographic models often poorly predict dynamics of marine populations as larval influx from afar decouples local population size from recruitment. In other words, marine species are typically distributed as metapopulations—­groups of partially isolated populations connected by some degree of dispersal. The essence of a metapopulation is that the spatial structure of the populations and the exchanges among them are just as impor­tant to population growth and stability as local birth and death rates (Hanski 1998). Especially for the rare and ecologically specialized species that dominate highly diverse regions, metapopulation pro­cesses are likely critical to long-­term per­sis­tence. This framework is increasingly relevant to all species as ­human impacts fragment natu­ral habitats. The dynamics of metapopulations and the nature of connectivity among them are thus central themes of marine ecol­ogy with impor­tant implications for management and conservation. Dispersal by larvae and, for ­those species that can swim, by other stages determines the scales at which species interact with their environment and one another, evolve, and respond to environmental change and h­ uman pressures. Therefore, dispersal strongly influences the scales of their local and geographic distribution (figure 6.12). Understanding larval dispersal involves two main questions: “Where do larvae go?” (dispersal) and “Where do larvae come from?” (connectivity) (Levin 2006). Estimating dispersal of marine organisms can be approached in several ways. Larger vertebrates can be fitted with vari­ous types of tags that track their movement. But tagging is resource-­intensive, and usually only practical for a small number of individuals larger than a few kilograms. For microscopic larvae, dispersal trajectories can be simulated using knowledge of life span and swimming ability together with hydrodynamics of the area. In some cases, dispersal distances can be estimated from the rates of observed invasions by a species into a previously unoccupied area. But most of what we know about dispersal and connectivity among marine populations comes from ge­ne­tic and, to a lesser degree, geochemical markers.

Tagging and tracking Many large marine animals undertake voyages of Homeric proportions, suggesting that their populations may be genet­ically mixed across the world ocean. The migrations of gray ­whales along the California coast, for example, is famous and a major attraction for w ­ hale watchers. Gray w ­ hales are the champion migrators among marine mammals. One female western gray ­whale by the name of Varvara was fitted with a satellite-­monitored radio tag off Sakhalin Island in the Rus­sian far east, tracked across the North Pacific to the Aleutians, down the west coast of North Amer­i­ca to the tip of Baja,

Adults

159

160

Ocean Ecology

California, and back—­a journey of 172 days and 22,511 km, the longest migration recorded by any individual mammal on land or sea (Mate et al. 2015). An Arctic tern banded as a chick off the British east coast reached Melbourne, Australia, within three months a­ fter fledging—­a sea journey of over 22,000  km. Jackson, a southern elephant seal fitted with a satellite transmitter, logged nearly 30,000 km over the course of a year—­that works out to 666 marathons, at a rate of roughly 2 per day. How typical are t­ hese journeys? Biologists have followed such animals using acoustic and radio tags to understand not only the detailed, three-­dimensional paths of their movements but a wealth of information on be­hav­ior, feeding habits, and the structure of their pelagic habitats. Indeed, a major motivation for tagging marine animals is to employ them as “animal oceanographers.” During the Census of Marine Life, the Tagging of Pacific Predators (TOPP) proj­ect deployed over 4000 satellite and acoustic tags on 23 species of sharks, tunas, sea turtles, ­whales, and seabirds inhabiting the California Current System and tracked them for periods of up to several months each (Block et al. 2011). The migratory powers of many of t­ hese animals are impressive, but equally remarkable are the highly regular movements of many of the species (figure 6.13). Eastern Pacific white sharks (Carcharodon carcharias) fitted with pop-up transmitting tags off central California, for example, followed a highly predictable migration cycle, departing the coast in winter and migrating more or less directly across 2000–5000 km to a site in the open tropical ocean midway to Hawaii, the white shark “café.” The sharks spent much of the spring and summer at the café, prob­ably both feeding and mating, before returning in autumn to California ( Jorgensen et al. 2009). Thus, despite their wide-­ranging movements across superficially unstructured pelagic habitats, the Pacific white sharks’ distributions w ­ ere quite structured and kept them separate from other populations of white sharks. This separation was supported by ge­ne­tic data showing that eastern Pacific white sharks clustered closely with conspecifics from Australia and New Zealand but formed a monophyletic clade distinct from populations elsewhere in the world ( Jorgensen et al. 2009). Indeed, many widely distributed oceanic fishes show strong population ge­ne­tic structure despite their seeming potential to circumnavigate the globe (Graves and McDowell 1995).

Figure 6.13. ​Long-­distance journeys of large marine vertebrates in the Northeast Pacific. Colored density shows frequency of occupation of each area by electronically tagged, tracked tunas, seals, sharks, seabirds, turtles, and ­whales (Block et al. 2011).

Chapter 6 Populations

Hydrodynamic simulation of larval movement Tracking of dispersing larvae pre­sents more formidable challenges than following adults and requires integration of several approaches. ­Because most marine larvae are planktonic, they move with ­water currents. Therefore, in princi­ple, their movements and distribution through time can be simulated with hydrodynamic models. Larvae are not simply passive particles, however, and models incorporating mortality and larval be­hav­ior show that realized dispersal can be much more local than the potentially long lifetimes of drifting larvae would suggest. Simulated dispersal of reef fish larvae around the island of Barbados showed that even when large numbers of larvae ­were released from a source population, almost none of them reached downstream locations 140 km away—­the distance to the closest island—­under realistic levels of diffusion and larval mortality (Cowen 2000). The model simulated a population of sedentary reef fish whose dispersing larvae spend up to 28 days in the plankton and die at a rate of 20% per day, estimated from field data. A regional hydrodynamic model was then used to estimate how many larvae reach downstream sites u­ nder dif­fer­ent conditions. The impor­tant implication is that s­ imple models of larval movement (advection) can greatly overestimate rates of larval exchange among sites ­because they ignore the very large (up to nine ­orders of magnitude) decline in larval concentrations from diffusion and mortality as the larvae drift. Another f­ actor that tends to retain larvae near their birthplace is the coastal boundary layer, the band of reduced current velocities near the shore. When the reduced currents within the coastal boundary ­were incorporated into a ­simple simulation model, median (but not maximum) larval dispersal distances decreased by up to 59%, and local retention of larvae increased by three o­ rders of magnitude (Nickols et al. 2015). Together t­ hese models suggest that even long-­lived larvae often recruit much better at highly local scales than at distances the larvae are biologically capable of reaching. The upshot is that even populations of marine animals with potentially long-­lived dispersing larvae generally recruit most successfully very close to home. ­These results help explain common findings that populations of reef fishes with long-­lived planktonic larvae show substantial ge­ne­tic structure among islands that are concordant with hydrodynamic model simulations (Taylor and Hellberg 2003, Cowen 2006).

Larval be­hav­ior If most dispersing larvae perish before recruiting downstream, as simulations suggest (Cowen 2000), se­lection should ­favor mechanisms to maximize their local retention. Such mechanisms are indeed common: despite their tiny size and seeming helplessness, larvae of many marine organisms have considerable control over their fate. Around Barbados where Cowen’s (2000) study was conducted, fish larvae tend to descend in the w ­ ater column to a depth of ~35 m, where the w ­ ater has a mean onshore flow, thus avoiding being carried offshore and enhancing local retention. Indeed, larvae of many marine species actively seek certain vertical positions in the ­water column, which exploit directional currents to ­either keep them in place (Kingsford et al. 2002) or bring them to favorable sites for growth (box 6.4). A classic example of this control is the migration of crab larvae between the vertically stratified shoreward and seaward flows in estuaries, which allows them to remain in local habitats (Cronin and Forward 1986, Bilton et al. 2002). As larvae reach the time to ­settle into benthic habitats, they can display sophisticated be­hav­iors that also contribute to putting them in the right place. Marine larvae use a variety of chemical, acoustic, physical, light, and even magnetic cues to guide them in ­doing so (Kingsford et al. 2002). The planula larvae of many reef corals, for example, s­ ettle preferentially and in some cases exclusively on crustose coralline algae (Morse et al. 1996, Heyward and Negri 1999), likely b­ ecause ­these algal crusts are reliable cues to the high-­energy, well-­grazed habitats in which stony corals tend to thrive. The specificity of cues used by marine larvae can be quite remarkable, as illustrated by several coral reef fishes (box 6.5).

161

162

Ocean Ecology

Box 6.4. ​A stormy love affair (A)

NE Cape Hatteras

Temperature (°C) 12 14

16

17

18

19 21 23

NW

0 50

Cape Lookout

F

MS

GSF

am tre lf S

Gu

100 150

Depth (m)

200 250

SW

MSF = Midshelf Front GSF = Gulf Stream Front Meteorological forces

Circulation

Isotherms

Nitrate (μM)

Survivor spawning frequency (individuals per day)

(B) 120 80 40 0 1.2 0.8 0.4 0.0 Wind stress

Marine animals have an ingenious range of strategies for encouraging their planktonic larvae to end up in the right place. Some animals time their reproduction to maximize food availability in remarkable ways. A dramatic example comes from the Atlantic menhaden (Brevoortia tyrannus), a planktivorous fish in the herring f­amily that lives in vast schools and supports a lucrative commercial fishery in the Northwest Atlantic. Menhaden spawn over the eastern North American shelf near Cape Hatteras and spend their early life history stages t­ here, exploiting predictable currents and front development (figure B6.4.1). Weather-­ induced changes in surface w ­ ater flow appear to control both food availability and transport of larvae from the shelf edge back to coastal habitats where juveniles live. A study conducted by David Checkley and colleagues (1988) found that strong northeasterly winds in late January, the Northern Hemi­sphere winter, cause upwelling of relatively warm but nutrient-­rich ­water at the Gulf Stream front and set up a circulation cell with shoreward surface flow. A subsequent cold-­air outbreak then cools the w ­ ater and ­causes it to sink as it approaches the midshelf front. The upwelling results in high nitrate concentrations in this shoreward-­flowing w ­ ater; phytoplankton then take advantage of the nutrients and bloom, also attracting microzooplankton (rotifers, copepod nauplii), the main food of larval menhaden. By mea­sur­ing ages of otoliths dissected from the menhaden larvae, the authors of the study w ­ ere able to back-­calculate the dates of spawning, which w ­ ere concentrated in a burst of activity near the Gulf Stream front during the northeasterly storms of January through May. The storm-­associated spawning of menhaden appears to be an adaptation to position their larvae on a con­vey­or ­belt of ­water that provides them first with a rich food source at the critical first-­feeding stage and then with transport back to inshore w ­ aters where the adults spend their lives. Supporting this interpretation, several other fish species similarly spawn shoreward of warm boundary currents in winter storms.

N W

5 dyn cm–2

E S

15 20 25 30 January

4

9

14 19 24

February

1

6

11 16

March

Figure B6.4.1. Storm-­induced spawning of Atlantic menhaden off North Carolina, USA. (A) Strong NE winds in late January cause upwelling at the Gulf Stream front, which sets up a circulation cell with shoreward surface flow. (B) Menhaden spawning was maximal during periods of strong northeasterly winds that caused upwelling of nutrient-­rich ­water. Subsequent onshore winds pushed the nutrient-­rich w ­ ater and feeding larvae ­toward coastal habitats where juveniles grow into adults (­after Checkley et al. 1988).

Population ge­ne­tic markers of dispersal and connectivity It’s difficult, and often impossible, to track the movement of marine animals through space, so connectivity among populations is often estimated indirectly. The most common such approach for estimating dispersal and connectivity among marine populations uses ge­ne­tic markers. The premise of ge­ne­tic approaches to population structure is that selectively neutral mutations accumulate within lineages at some random but relatively constant mean rate that depends on population size, generation time, and so on. If a population is divided and the parts cease exchanging individuals and interbreeding with one another—­for example, if a barrier to dispersal arises between them—­the accumu-

Chapter 6 Populations

163

Box 6.5. ​Larval sensory biology and global change As drifting larvae approach readiness to s­ ettle, they need cues to appropriate habitat, which in sedentary species is where they ­will spend the rest of their life. Sound is a particularly reliable cue in w ­ ater ­because it travels far with ­little attenuation and is not influenced by currents; thus, many underwater habitats have characteristic sound signatures. Underwater playback experiments show that larvae of both corals (Vermeij et al. 2010) and reef fishes (Simpson et al. 2005) are attracted to reef sounds, mainly produced by the snapping shrimp and fishes that abound on healthy reefs. Chemical cues are also valuable ­because they can be highly specific indicators of par­tic­u­lar conditions. A remarkable example involves the clownfish Amphiprion percula that associates with anemones in shallow ­water beneath overhanging rainforest vegetation in Papua New Guinea. Experiments showed that larvae of A. percula, including t­hose raised naive in the laboratory, ­were attracted to ­water from nearshore areas compared with ­water from offshore, including ­water infused with the scent of leaves from nearby rainforest trees or of anemones (Dixson et al. 2008). Larvae of both corals and fishes are similarly able to detect and orient t­ oward specific chemical cues that indicate healthy coral reef habitat (figure B6.5.1) (Dixson

(B) Time spent in each water type (%)

(A)

100

et al. 2014). Experiments showed that w ­ ater flowing from degraded, algal-­dominated reefs repels recruiting fish larvae, including the herbivores necessary to remove algae that outcompete corals. Follow-up experiments showed that settling larvae of both corals and fishes orient ­toward corals over algae, and t­ oward specific seaweed species characteristic of healthy reefs versus t­ hose from degraded reefs (Dixson et al. 2014). For coral larvae, attraction to corals and avoidance of macroalgae ensure that they are in habitat favorable for growth to adulthood. For fishes, the same cues presumably ensure that recruits ­will be in the three-­dimensional habitat structure necessary to grow and develop in safety from predators, which is provided by living corals. Crucially, t­ hese be­hav­iors may result in a positive feedback loop whereby healthy reefs attract recruitment of the animals (corals and herbivorous fishes) that keep them healthy, whereas degraded algal-­covered reefs repel the same prospective recruits and may thus remain in an alternative stable state. If this mechanism is general, it may provide one component of the complex trigger mediating a shift between alternative stable states (Hughes 1994, Roff and Mumby 2012; chapter 12) on tropical hard bottoms.

Water with Acropora chemical cues (colored bars)

Water with Porites chemical cues (white bars)

Water with Chrlorodesmis chemical cues (colored bars)

Water with Sargassum chemical cues (white bars)

80 60 40 20

% time in cue (mean + 1SE)

n = 60

80 60 40 20

A. tenuis

A. nasuta

A. millepora

80 60 40 20 0

Ch

ry

sip

0

100

Ch ro m te is vir Ch ra b idis io ry sip cel la ta Da tera cy Da sc sc yllu ane yll s a a Po u m s tr rua im nu ac a en s tru cula s r Ch spil tus o Ch a ae eto toce d to do on ps Ac n va raff el an g th ab si u Ct urus nd u en oc trios s he t at egu u s Sig s str a Ch anu tus lor s s p ur us inus Ha so rd lic id hl oe Sc us re ar s u Ap trim s sp . og a on cula an tu gu s sta tu s

MPA Non-MPA

100

Time spent in each water type (%)

0

Figure B6.5.1. Larvae of coral reef animals are strongly attracted to chemical cues from organisms of healthy reefs. (A) In lab choice experiments, coral larvae spend much more time in w ­ ater from a healthy reef within a marine protected area (MPA) than in w ­ ater from a degraded reef (non-­MPA). (B) Similarly, reef fish larvae are attracted to specific corals (top) and macroalgae (bottom) from healthy reefs (colored bars) compared with t­ hose from degraded reefs (white bars) (­after Dixson et al. 2014).

164

Ocean Ecology

lation of mutations then proceeds in­de­pen­dently within each ­daughter population, and their ge­ne­tic compositions begin to diverge as a consequence of this ge­ne­tic drift. The magnitude of this neutral ge­ne­tic differentiation can then be used as an estimate of the degree of demographic connectivity among the populations. Studies of population ge­ne­tic structure have been conducted in many marine organisms, beginning with the use of protein allozyme variation in the 1960s and evolving ­toward increasingly sophisticated, high-­resolution genomic markers. What do ge­ne­tic markers tell us about the connectivity and structure of marine metapopulations? A review of 90 benthic marine species from a wide range of marine algae, flowering plants, invertebrates, and fishes estimated dispersal scales from ge­ne­tic mea­sures of isolation by distance and found that dispersal distance ranged over five ­orders of magnitude, from a few meters to hundreds of kilo­meters (Kinlan and Gaines 2003). Interestingly, estimates of ge­ne­tic dispersal differed among functional groups, with benthic herbivores generally dispersing one to two o­ rders of magnitude farther than their algal prey, and sessile suspension-­feeding invertebrates ranging over similar scales as macroalgae. This variability among taxa and functional groups means that environmental change and management interventions ­will often affect co-­occurring species differently. For example, rising sea temperatures are expected to have quite dif­fer­ent effects on species, leading to reshuffling of communities (Williams and Jackson 2007) and potentially the emergence of novel ecosystems (Hobbs et al. 2009). Given the evidence that larval be­hav­ior and planktonic mortality strongly influence the spatial patterns of larval recruitment, it’s not surprising that population ge­ne­tic structure is only loosely related to estimated planktonic dispersal potential. Synthesis of data from 300 studies of marine organisms found that ge­ne­tic differentiation among populations was poorly correlated with average planktonic larval duration, and in fact was marginally nonsignificant when brooding species (with no planktonic duration) ­were excluded (Weersing and Toonen 2009). In applying findings from population ge­ne­tics to management, it’s impor­tant to emphasize that ge­ne­tic structure is much more sensitive to exchange among populations than are demographic rates. That is, a small number of mi­grants per generation can homogenize populations genet­ically, making them appear like a single population, but this level of larval influx is insufficient to sustain a population or rescue it from overharvesting (Hellberg 2009). This emphasizes the value of applying multiple approaches to estimating connectivity in metapopulations, such as geochemical tracers.

Geochemical tags Another valuable tracer of population structure arises from variation in the composition of trace ele­ ments incorporated into animal hard parts. Such variation can provide detailed information on where an individual has spent its life. Analyses of elemental (geochemical) composition have been especially useful in marine fishes, which continually build layers onto the mineral otoliths that serve as their ears. The layers of the otolith preserve a rec­ord of conditions prevailing at dif­fer­ent periods of the individual’s life, much as tree rings do. This approach was impor­tant in revealing formerly hidden stock structure in weakfish (Cynoscion regalis) along the east coast of the USA. Adult weakfish migrate annually from overwintering grounds south and offshore of Cape Hatteras to spawning sites in estuaries and bays along the east coast. The larvae generally are retained in the estuaries where they ­were born by exploiting tidal currents, as discussed above. ­Because of this larval retention, connectivity among weakfish populations is determined by the be­hav­ior of adult fish, notably their tendency to return to their natal estuary from offshore to spawn.  Estuaries differ naturally in the physical and chemical composition of ­water due to variation in source rock composition in the watershed, ­water residence times, and biological pro­cesses. Th ­ ese differences are incorporated into the fish’s growing otolith, providing a geochemical signature of its location when the otolith layer was laid down. Simon Thorrold and colleagues (2001) analyzed adult fish and larvae from several estuaries along the east coast of the USA and found that geochemical signatures from spawning two-­year-­old weakfish

Chapter 6 Populations

(A)

(B)

GA

Mg Mn Sr Ba C O

PS

CB

PS

0.0 –0.5 –1.0

1.0 0.5

–0.5

GA

1.0 0.5

100

200 km

64

DE 72

CB 60

PS

CB GA

81

–0.5 Mg Mn Sr Ba C O

GA Atlantic Ocean

0.0 –0.5 –1.0

0

Mg Mn Sr Ba C O

NY 65

0.0

–1.0 Mg Mn Sr Ba C O

NY

0.0

–1.0

Residuals

Residuals

0.5

0.5

Birth location CB DE

NY

NY Spawning location

0.0 –0.5 –1.0

1.0

DE

Residuals

0.5

1.0

Residuals

Residuals

1.0

DE

PS

Mg Mn Sr Ba C O

Figure 6.14. ​Geochemical tracers show that weakfish (Cynoscion regalis) return to the estuaries where they ­were born. (A) Variation in six elemental tracers combined to characterize geochemical signatures in otoliths of juvenile weakfish collected in 1996 (lighter bars) and two-­year old spawning adults collected in 1998 (darker bars) from locations along the mid-­Atlantic coast of the USA. (B) Rates of homing in spawning weakfish, illustrated as the percentage of fish from each location that w ­ ere spawning in the estuary of their birth and each of the other four estuaries (­after Thorrold et al. 2001).

matched well with the signatures from otoliths of juveniles collected in the same estuary two years previously (figure 6.14), confirming that 60%–80% of adult weakfish returned to natal locations to spawn. Importantly, ge­ne­tic data from weakfish in the same region had implied a single, well-­mixed population from Florida to Maine. The geochemical tags showed that spatial structure of the weakfish population is much stronger than had been assumed by fisheries man­ag­ers, which means that weakfish—­and other species with strong natal homing—­are significantly more vulnerable to fishing than predicted by stock models that assume a single, well-­mixed population. This study highlights a potential weakness of fisheries management policies based on ge­ne­tic approaches that are sensitive to very low rates of exchange (Thorrold et al. 2001).

Macroecol­ogy of Populations Metabolic scaling and life history The close links between individual body mass, metabolic rate, and demographic rates imply that metabolic scaling should strongly influence life history syndromes (Savage, Gillooly, Brown, et al. 2004), specifically rates of birth, maturation, reproduction, and death. The intrinsic rate of population increase r scales with the −¼ power of body mass (Fenchel 1974, Blueweiss et al. 1978, Savage, Gillooly, Woodruff, et al. 2004), and a large body of data confirms that its several life history components

165

166

Ocean Ecology

similarly scale closely with body mass, including maturation rate, generation time, life span, and fecundity (Blueweiss et al. 1978). Mortality rate also scales with body mass (negatively) and temperature (positively), which explained nearly half the variation in instantaneous mortality rate across 175 stocks of marine fishes in the field (Savage, Gillooly, Brown, et al. 2004) (see figure 5.7). Fi­nally, population abundance (N) scales with the −¾ power of body mass, which implies a similar relationship between M and environmental carry­ing capacity, K. Presumably, the biology under­lying t­ hese patterns is that smaller organisms need fewer resources to reach maturity and thus have shorter generation times. But, by the same token, smaller organisms have fewer reserves when facing food shortage and are vulnerable to a wider range of predators. The metabolic theory of ecol­ogy has often taken the simplified approach of lumping the disparate biochemical and physiological pro­cesses of an organism’s life into the composite variable of metabolism. But the components of metabolism can scale differently with temperature. This variation affects predicted relationships between temperature, body mass, and abundance, and separating the organism’s development into life history components helps understand and or­ga­nize it. For example, growth often responds differently to changing temperature and body size than does development. Specifically, maturation rate increases faster with temperature than does growth rate in many animals, including fishes and several zooplankton taxa (Rec­ord et al. 2012), resulting in smaller size at maturity at higher temperatures. In marine pelagic copepods, which have been especially well studied, development rate is more sensitive to temperature than is growth across all life stages. Models incorporating ­these differences predict that adult body size w ­ ill often be smaller u­ nder warmer conditions, which is borne out by field observations (Forster et al. 2011).

Abundance and the energetic equivalence rule The metabolic rate of an organism sets the scope for its demography and population dynamics. Growth rate and time to first reproduction thus scale closely with body size (Savage, Gillooly, Brown, et al. 2004). Th ­ ese organismal scaling relationships can be used in turn to relate the metabolic allometry of the organism to the demographic par­ameters of the population—­its intrinsic growth rate, r (Gillooly et al. 2002) and steady-­state abundance (carry­ing capacity, K), predicting that, for a given resource input, population abundance at steady state (i.e., at carry­ing capacity) scales with body mass to the −¾ power: N ∝ M−3/4. That is, all ­else being equal, smaller organisms w ­ ill be more abundant than larger ones, and this size dependence is quantitatively consistent. If, as we have discussed, both metabolism and population size scale similarly but with opposite sign with body mass, a somewhat nonintuitive generalization emerges. Multiplying the equation describing metabolic scaling with body size (B ∝ M3/4) by the scaling of abundance with body size (N ∝ M−3/4) predicts that within a guild using the same resources—­ say, all phytoplankton or all herbivorous animals—­the total energy flux through populations in the guild (BT) is roughly equal regardless of their body size, a prediction known as the energetic equivalence hypothesis (or rule). This pattern was first identified by Damuth (1981) based on data from herbivorous land mammals, and was subsequently supported by the −¾ power scaling of abundance with body mass among terrestrial plants (Enquist et al. 1998) and marine phytoplankton (Belgrano et al. 2002). A fundamental challenge to the energetic equivalence hypothesis—as for metabolic ecol­ogy more broadly—is disentangling correlation from causation. The theory explains energetic equivalence as a result of diffuse competition among species for a common source of energy. Yet the predicted scaling of abundance with body mass to the −¾ power was also documented in a community

167

Chapter 6 Populations

of rocky intertidal invertebrates strongly disturbed by harvesting, which seems inconsistent with the theory’s implicit assumption of equilibrium among competitors (Marquet et  al. 1990). Both the community disturbed by size-­selective harvesting, presumably far from equilibrium, and an undisturbed control community included multiple trophic levels that competed for space rather than food (energy), suggesting that the general allometric relationship between size and abundance must have a dif­fer­ent explanation in at least some cases (Marquet et al. 1995). Although the mechanisms ­behind such macroecological patterns remain uncertain, t­here is some evidence that allometry of individual metabolism extends to an allometry of population demography, and that community interactions also scale predictably with body mass (chapters 7, 8). This is presumably b­ ecause metabolism defines the fundamental demand for resources as well as the limits within which t­ hose resources are allocated to survival, growth, and reproduction. Th ­ ese fundamental biophysical limits therefore constrain the flow of energy up the food web.

The macroecol­ogy of range size

Mean species home range area (m2)

We saw in our discussion of biogeography (chapter 3) that geographic range size has impor­tant consequences for the origin and extinction of species over evolutionary time scales. But what determines range size? Metabolic scaling once again offers a theoretical explanation for how organismal traits and environment interact to produce patterns in space use, range size, and even species richness. Body mass influences the size of both home range—­the area that an individual traverses in a lifetime—­and the species’ geographic range size through its influence on individual mobility and resource requirements. Home range size scales as a positive power function of body mass among terrestrial mammals ( Jetz et al. 2004), suggesting that home range size is ­shaped by energy Carnivores Herbivores demands and metabolic rates. In contrast to the quarter power Birds scaling of most other macroecological relationships, however, Mammals Reptiles home range size scales with an exponent closer to one in mamFishes mals. ­There is also considerable variation in the value of the ex108 ponent, as might be expected given the many ecological and evolutionary ­factors influencing organism mobility. Among 569 terrestrial and aquatic vertebrates, residual variation around 106 the allometric scaling of home range on body mass is largely explained by mode of locomotion, scale of foraging, trophic level, and prey size, which together explain 80% of variation when controlling for phylogeny and tracking method (Tamburello 104 et al. 2015) (figure 6.15). An obvious candidate for predicting geographic range size 10 among marine organisms is the duration of larval life in the 2 10 10 plankton: long larval life would be expected to increase the area 10 over which larvae disperse. Many planktonic organisms indeed 10 have broad geographic ranges. Surprisingly, however, t­here is ­little empirical support for this hypothesis: among both marine HRA1kg 10 10 10 10 10 100 benthos and fishes, meta-­analysis showed no relationship be100 102 104 106 tween geographic range size and dispersal ability, estimated Mean species body mass (g) ­either from pelagic larval duration or from ge­ne­tic isolation-­by-­ distance data (Lester et al. 2007). Instead, among 590 species Figure 6.15. ​Individual home ranges of animals scale consistently of tropical reef fishes spanning 47 families, the strongest predicwith their body mass and feeding mode. Plots show allometry of tors of geographic range size ­were adult traits judged to influhome-­range area across 569 terrestrial and aquatic vertebrates. ence ability of colonizing individuals to survive and establish Solid lines indicate carnivores; broken lines indicate herbivores. reproductive populations. ­These ­were large body size, schooling Inset : Raw data for marine fishes (­after Tamburello et al. 2015). 6

4

2

0

0

1

2

3

4

168

Ocean Ecology

be­hav­ior, and nocturnal activity, all of which tend to reduce the vulnerability of new recruits to the intense predation common on reefs (Luiz et al. 2013).

Marine Populations in the Anthropocene Our review of marine population biology offers several insights into the ­future of marine life in the Anthropocene. First, a general theme emerging from multiple lines of evidence is that marine species often show stronger geographic population structure than previously assumed. In such cases, local impacts on environments and populations have stronger and more long-­lived consequences than they do for well-­mixed populations. A second theme comes from application of stage-­structured models, which show that survival of large individuals is often a key choke point in the vulnerability of marine populations, whereas recruitment of youn­gest stages has weaker influence. This is particularly true of the large animals most targeted by fisheries and conservation efforts, as we have seen for corals and sea turtles. The other lesson from such models, however, is that the most impor­tant life stages, and thus targets for conservation and management action, vary greatly as a function of life history. In general, short-­lived, early maturing species with high fecundity (“fast” species) tend to be highly sensitive to recruitment variation and can recover quickly from intensive fishing u­ nder the right conditions. In contrast, “slow” species that are long-­lived with low fecundity are often highly vulnerable even to moderate fishing pressure and tend to recover slowly, if at all, from overfishing. We have also seen that species on both land and sea are evolving, sometimes rapidly, in response to the strong se­lection pressures imposed by h­ uman harvesting and domestication. It has often been said that Earth is in the midst of the sixth mass extinction in its history as a result of h­ uman influence. What is less appreciated is that we have also set in motion a global experiment in coevolution, whereby ­those of Earth’s species that survive the extinction are being forced to adapt—to evolve—to a world where habitat types and distributions, climate, food web interactions, and flows of energy and materials are all profoundly changed and controlled, directly or indirectly, by the global keystone species Homo sapiens. The dominance of ­humans focuses attention on our own population biology and how it has been affected by the industrialization that has transformed the world. Can we understand the emergence of the industrialized Anthropocene earth via the princi­ples of population ecol­ogy? Demographic rates of organisms are fueled by food energy, which is consumed and used to produce new biomass and fuel metabolism. As we saw in chapter 5, the fundamental biophysical relationships that underlie ­these pro­cesses produce very general allometric and macroecological relationships among body size, energy use, demographic rates, and abundance. Modern ­humans have escaped t­hese constraints ­because, uniquely among living organisms, we rely heavi­ly on “extrametabolic” energy sources that transcend what our individual physiology is capable of. This ability to capture large quantities of extrametabolic energy raises h­ uman per capita energy consumption to levels well above t­hose predicted for an animal of our size (Burger et al. 2011). Nevertheless, when extrametabolic energy is included in the allometric equations, demographics and life history traits of industrialized ­human populations still depend on per capita energy use (Burger et al. 2011). For example, ­human birth rate scales with the −⅓ power of energy use even across industrialized populations (Moses and Brown 2003), and juvenile mortality declines steeply with energy use to the −1 power (Burger et al. 2011). Interestingly, the net reproductive rate R0 actually declines with per capita energy use across industrial populations, contrary to expectations from nonhuman animals but consistent with widely recognized trends of lower fertility in developed nations, or social groups, with higher mean energy consumption. This results in part from the generally l­ater age of first reproduction in industrialized socie­ties with high energy consumption. Thus, the widely recognized demographic transition (Bongaarts 2009), in which rich industrialized socie­ties tend to show declines in population growth, is largely consistent with very general princi­ples of metabolic ecol­ogy.

Chapter 6 Populations

­Future Directions The prob­lem of dispersal—­that is, an adaptive explanation for the commonness of pelagic larvae among marine organisms—is still with us. Available data have not resolved the questions of what se­ lection pressures have driven and maintained the widespread occurrence of planktonic larvae in marine animals, ­whether and how planktonic dispersal reduces mean and variance in recruitment, or to what extent intrinsic biophysical constraints on brooding eggs dictate embryo size and ecol­ogy. The ecological hypothesis that dispersal in the w ­ ater column provides refuge from benthic predators is also unresolved, despite l­imited experimental and circumstantial evidence. This area is ripe for targeted research and synthesis. Early life history stages of most marine organisms are highly vulnerable to starvation due to small size and meager energy reserves. In many marine organisms, recruitment of ­these early, vulnerable stages coincides roughly with seasons of warmer temperatures and high planktonic production, particularly in the highly seasonal environments of temperate and high-­latitude systems. This is true of both holoplankton like copepods and krill as well as the planktotrophic larvae of fishes and benthic invertebrates. High-­latitude copepods, for example, typically reproduce or emerge from diapause around the time of the spring phytoplankton bloom, and larvae of many invertebrates and fishes are also released into the ­water column in late spring. Populations of commercial fishes in par­ tic­u­lar are thought to depend strongly on synchrony of recruitment with plankton blooms, often of par­tic­u­lar phytoplankton species. Off Nova Scotia, Canada, for example, timing of the spring bloom explained 89% of the variation in year class strength of haddock, with ­earlier blooms more closely matching haddock spawning times and therefore supporting much higher recruitment (Platt et al. 2003). Such linkages between seasonal plankton productivity and fish recruitment w ­ ere first advanced as major controls on fishery production by Johan Hjört in the early twentieth c­ entury and ­later developed by David Cushing, and became known as the Hjört-­Cushing match-­mismatch hypothesis. The hypothesis suggests that recruitment depends on plankton productivity during the “critical period” when fish larvae make the transition from feeding on endogenous yolk to feeding on organisms in the plankton. Yet a critical review suggests that the evidence in support of the Hjört-­ Cushing match-­mismatch hypothesis is generally weak (Leggett and Deblois 1994). Resolution of this issue has impor­tant implications for basic ecol­ogy, fisheries management, and understanding of global climate change impacts. Quantifying and understanding demographic connectivity among populations is central to many issues in Anthropocene marine ecol­ogy and depends on understanding the biology of dispersing larvae. But much of larval life remains a mystery due to small size, wide-­ranging movements, and complex physical oceanography of nearshore habitats. Growing computational power and sophistication offer promise of more complex hydrodynamic models. Together with new molecular tools and automated sampling approaches, the frontier demands integrated field research on physical oceanography and larval biology to validate and pa­ram­e­terize ­those models, and this ­will require close interaction between biologists and physicists, modelers and field ecologists (Cowen and Sponaugle 2009). Making use of new, more power­ful models ­will in turn require substantive collaboration between scientists and man­ag­ers. Larval dispersal in the deep sea, which occupies the vast majority of Earth’s ecospace, is a largely unexplored frontier. Despite formidable logistical challenges, new approaches for cracking this nut are emerging with more sophisticated molecular markers and geochemical tracers, again in combination with advancing physical oceanographic models (Levin 2006). It is clear that many of the pressures h­ uman activities impose on marine organisms, notably fishing, are strong and selective, and act on traits with high heritability, like body size and growth rate. In other words, h­ umans impose strong evolutionary se­lection on marine populations. Controlled experiments and evidence from aquaculture indicate that this se­lection can cause substantial change in life history and vital rates over surprisingly few generations in vari­ous marine organisms. But teasing

169

170

Ocean Ecology

apart the contributions of ge­ne­tic change and phenotypic plasticity to phenotypic change in wild populations has proved difficult. ­Doing so is impor­tant for understanding and mitigating potentially rapid evolutionary responses of marine populations to h­ uman activities. Moreover, studies of evolutionary responses to ­human activity have concentrated on the effects of fishing, whereas we also need attention to evolutionary consequences of ocean warming, acidification, habitat modification, exotic invasions, and other anthropogenic stressors. ­There is ­great potential to weave together the vari­ous threads considered in this chapter, including physical oceanography, larval biology, and matrix (meta)population modeling ­toward a more integrated and predictive framework for designing and evaluating marine conservation and management mea­sures generally, and protected areas specifically.

Summary A population is a group of organisms of the same species living within a defined area. The demographics, growth, and dynamics of populations are central to basic and applied ecol­ogy, from individual to ecosystem levels. In the ocean, the combination of a relatively continuous pelagic environment and the commonness of planktonic dispersal means that most marine species are distributed as metapopulations—­groups of populations occupying relatively discrete habitat patches and connected by exchange of individuals or larvae. The dynamics of a metapopulation are influenced as strongly by spatial structure and the degree and direction of connectivity among its component populations as by local demographics and interactions. Therefore, larval biology and physical oceanography are key to understanding population dynamics of most marine organisms. Although potentially long larval life in the plankton suggests long-­range dispersal, recruitment in most dispersing species is much more localized as a result of mortality and diffusion in the plankton and active larval be­hav­iors that result in local retention. The commonness of planktotrophic larvae in marine life cycles may thus reflect se­lection for high fecundity, coupled with the need to escape benthic predators, rather than for dispersal per se. Approaches to modeling dynamics of such age-­ structured marine populations typically employ matrix projections based on life t­ able data, that is, survivorship and fecundity at each age or life history stage. Models adapted for data-­poor conservation and management applications can estimate elasticities of survival and fecundity using only data on adult survival, age at maturity, and population growth rate, which are available for many species. ­These are useful for identifying life stages and habitats for which conservation investment is most effective and for determining how organisms with dif­fer­ent life histories w ­ ill respond to environmental change or management interventions. Stage-­structured matrix models are also useful in quantifying and predicting evolutionary change within populations. Home range sizes of mobile animals scale positively with body mass but often with exponents > ¾; ­these relationships are more variable, reflecting differences in mobility, trophic level, and feeding mode. At the community level, some evidence supports the energetic equivalence rule—­that energy flux through populations using similar resources (e.g., herbivores) is roughly equal across populations of dif­fer­ent body size. Pervasive ­human impacts on marine ecosystems in the form of fishing, habitat alteration, and climate change are altering the spatial and demographic structure of populations and the selective regimes they experience with widespread but still poorly understood consequences for the ecol­ogy and evolution of marine organisms. Resolving ­these issues ­will involve better integration of hydrodynamics and larval biology, higher-­resolution tracers of larval origins and movements, quantifying the contributions of ge­ne­tic change versus phenotypic plasticity in population responses to stressors, and field collection of key demographic data to pa­ram­e­terize and compare population models for dif­fer­ent taxa.

7

Species Interactions

W

 hen we think of nature, the images that come to mind are often of organisms interacting with one another—­bees pollinating flowers, predators hunting prey, clouds of colorful fish flowing among corals. All organisms interact with other species in a multitude of ways, but t­ hese interactions can be grouped into three general types: competition for a common resource, trophic interactions in which one species consumes another (herbivory, predation, parasitism, disease), and facilitative interactions in which at least one of the species benefits and neither party is harmed (mutualism, commensalism). ­These categories are somewhat fluid and interactions between two species may shift among them as conditions change, as when salt marsh plants competing for soil resources become mutualists ­under harsh conditions (Bruno et al. 2003). ­Because organisms live in complex assemblages of many species, interactions between two species often affect the distributions, abundances, and be­hav­iors of other species as well. Together with environmental forcing (chapter 3), ­these direct and indirect interactions among organisms, knitted together in complex networks or webs, are what create and shape the structure of communities (chapter 8) and mediate biogeochemical fluxes through ecosystems (chapter 9). Ultimately, ­those pro­cesses are ­shaped by the traits of the interacting organisms. What are the most impor­tant characteristics of organism and environment that influence interaction strength? Can we find generalizations that allow us to cope with the ­great diversity of interactions happening in nature? In this chapter we review the main types of interactions among species and how they link together into networks. We cover their consequences for communities and ecosystems in the following two chapters.

Interactions among Species: General Considerations The interactions among species within a community can be represented graphically as an interaction web (figure 7.1), where the topology, or structure of the web, consists of arrows showing who affects whom, and the thickness of arrows represents interaction strength, that is, the direct effect that an individual of one species has on an individual of another species (Wootton and Emmerson 2005). Interaction strength is defined mathematically as the per capita effect of one species on the dynamics of another species. Interaction strengths can be estimated empirically from experimental removals of consumers or competitors in the field, laboratory feeding assays, or observational approaches integrating gut contents and estimated metabolic requirements. The distributions of interaction strengths within communities are highly uneven, generally with a few very strong interactions, ­whether positive or negative, and many weak ones. This skewed pattern of interaction strengths has been found in rocky intertidal communities (Paine 1992, Wootton 1994), kelp forests (Sala and Graham 2002), estuarine infauna (Emmerson and Raffaelli 2004), and coral reefs (Bascompte et al. 2005) (figure 7.2). Species in a community interact not just in a pairwise fashion with competitors or predators but with a potentially large number of other species. Meta-­analyses comparing the strengths of

171

CONNECTEDNESS WEB Acmea pelta Strongylocentrotus Katharina

drift

benthic diatoms

Tonicella

Acmea mitra

Alaria Hedophyllum Corallina Bossiella Lithothamnion

ENERGY FLOW WEB Acmea pelta Strongylocentrotus Katharina

drift

benthic diatoms

Tonicella

Acmea mitra

Alaria Hedophyllum Corallina Bossiella Lithothamnion

INTERACTION WEB Acmea pelta Strongylocentrotus Katharina

drift

Acmea mitra

Alaria Hedophyllum Corallina Bossiella Lithothamnion

(B)

–0.05

25

–0.04 –0.03

×

20

×

–0.02 –0.01

mean

× ×

0.00

×

×

Frequency

Per capita interaction strength

(A)

benthic diatoms

Tonicella

Figure 7.1. ​Alternative approaches to characterizing connections among species in communities. Arrow thickness represents (top) presence versus absence of a feeding relationship, (­middle) magnitude of energy flow, or (bottom) interaction strength (­after Paine 1980).

10 5

0.01

×

0

×

0.02

15

0.03 1 (C)

2

3

4 5 Rank

6

7

–3.5 –3.0 –2.5 –2.0 Log per capita interaction strength

8 (D)

Frequency

0.15

0.10

0.05

0.00

–0.0025 –0.0725 –0.1125 Interaction strength (dynamic index)

Figure 7.2. ​Most communities have a few strong interactions and many weak ones. The figures show the frequency distributions of interaction strengths within marine communities reconstructed from (A) field manipulations of rocky intertidal herbivores (­after Paine 1992); (B) laboratory mea­sure­ments of diverse herbivores feeding on kelp germlings (­after Sala and Graham 2002); (C) gut contents of coral reef fishes and invertebrates (­after Bascompte et al. 2005); and (D) simulated interactions among dif­fer­ent size classes of four estuarine predators feeding on a common prey species (­after Emmerson and Raffaelli 2004).

Chapter 7 Species Interactions

173

(A) Food chain (B) Omnivory competition and predation found that the two pro­cesses have roughly similar impacts on the interacting species, on average, Predator although they vary considerably among trophic groups and often interact with one another—­that is, predation can change the strength and even the direction of competition between Prey-consumer species (Sih et al. 1985, Gurevitch et al. 2000). A first step ­toward organ­izing thinking about species interactions beyond pairs involves community modules (Holt Resource 1997), small sets of species whose dynamics are linked through direct and indirect interactions. Th ­ ese modules include many (D) A fraction of a food web with (C) Intraguild predation of the types of interactions familiar in community ecol­ogy several interrelated modules (figure 7.3). Common examples are exploitative competition, in which one species negatively influences a competitor by reducing their common resource, and trophic cascades, in which a predator positively influences plants by reducing the abundance of herbivores that eat them. By focusing on how interaction between two species influences o­ thers around them, community modules begin to bridge the gap from pairwise interactions to ­whole interaction webs. A key component in any interaction web larger than two species is indirect interacFigure 7.3. ​Examples of interaction modules found in real food tion, in which one species alters the abundance or be­hav­ior of webs (­after Holt 2009). another species through an intermediate, without interacting directly with that species. Indirect interactions are common and often strongly influence the structure and dynamics of communities. In rocky intertidal communities, for example, compilation of experimental results came up with nine general types of indirect interactions and 83 (!) subtypes (figure  7.4) (Menge 1995). Roughly 40% of the community changes caused by ­these experimental manipulations resulted from indirect interactions. Of ­those, the most common ­were P + P P P keystone predation (35%) and apparent competition (25%). + + + p p p p f f p A good illustration of how indirect interactions and varif – – + – – f ance in interaction strength influence communities involves – – B B B B B – – keystone species, defined as species that produce effects on comc Exploitation Keystone Apparent munites that are both large and disproportionate to their abuncompetition competition predation dance. In extreme cases, a single consumer species can drive the structure of entire ecosystems. The iconic original example + P is the predatory sea star Pisaster ochraceous on rocky shores of P P p – Washington State, USA (Paine 1966, 1980) (box 7.1). Despite p – if relatively low abundance, Pisaster has a power­ful influence on – H B the diversity and species composition of this community. The p + term keystone species is now used broadly to refer to any species – Habitat B facilitation with an effect on the system disproportionate to its abundance or biomass, w ­ hether through predation or some other pro­cess Trophic cascade (Power et  al. 1996). Many such species—­predators, mutualists, pathogens—­are known. This contrasts with a community dominant, which is both functionally impor­tant and abundant. Figure 7.4. ​Some examples of indirect interactions among species, showing that even ­simple community “modules” of a few By this definition, disease organisms might be considered the species can show a diverse range of interactions (c = competition; ultimate keystone species as their biomass is negligible but f = facilitation; p = predation) among predators (P), herbivores (H), they often profoundly influence ecosystems. Striking examples and basal species (B) of algae and sessile invertebrates (­after include epidemic diseases that strike ecologically impor­tant Menge 1995).

174

Ocean Ecology

Box 7.1. ​How biological interactions structure communities on rocky shores ecological interactions beginning in the 1960s. Joseph Connell (1961) conducted a seminal experiment examining competition among barnacles on a Scottish shore, which remains one of the clearest examples of competitive exclusion in nature and also illustrates the flip side of the coin—­that some form of niche differentiation is necessary for long-­term coexistence of competing species. Like many pioneers, Connell started with a striking and well-­known pattern that most workers believed had

How impor­tant are interactions among species to the composition and diversity of communities in nature? The first and most influential demonstrations of competitive exclusion in the field came from rocky intertidal shores, where living space on the rocks limits abundance of the seaweeds and sessile invertebrates growing t­ here (figure B7.1.1). The simplicity of identifying and mea­sur­ing this resource, along with ease of manipulating it, made rocky intertidal shores a productive laboratory for studying

(A)

0.0’

80 60 40

0.0’ in tide pool

–1.0’

–2.5’

Number of Chthamalus stellatus present

20 10 5

Stone lost

2 1

0.0’

80 60 40

0.0’ in tide pool

–2.5’

–1.0’

Balanus removed Unmodified

20 10 5 2 1

J J A S ON D J F MAM J

1954

1956

J J A S ON D J F MAM J

1954

1956

J J A S ON D J F MAM J

1954

1956

J J A S ON D J F MAM J

1954

1956

Species richness

(B) 12

12

10

10

8

8

6

6

4

4

2

2

0

0

100

200

300

0

0

100

200

300

Littorina littorea density (#/m2)

Figure B7.1.1. Early field experiments demonstrated that competitive exclusion and herbivory strongly influence the structure of rocky shore communities. (A) Curves show survivorship of the barnacle Chthamalus stellatus on rocks at several sites from which the competitor Balanus balanoides was removed (blue) versus left in place (green) (­after Connell 1961). (B) Herbivore impacts on diversity of their algal prey depend on the competitive ability of the preferred prey. (Left) In tide pools, the preferred prey is competitively dominant, and moderate grazing increases diversity by reducing its cover and releasing other species from competition. (Right) On emergent substrata, the preferred prey is competitively inferior, and grazing at any level tends to reduce diversity (­after Lubchenco 1978).

Chapter 7 Species Interactions

already been explained, but he probed it in a new way. Two barnacles occupy distinct zones on Scottish shores: Chthamalus stellatus at the high end and Balanus balanoides in the low zone (figure B7.1.1A). In the productive w ­ aters of the North Atlantic, barnacle larvae recruit in large numbers and are densely crowded on the rocks. Connell recognized that this situation held par­tic­u­lar promise for studying the outcome of competition: the community is s­ imple, with two major competitors, the habitat is s­ imple in that the barnacles appear ­limited by availability of space, and space can be freed up experimentally by removing barnacles from the rock. Connell reasoned that, b ­ ecause barnacles are sea creatures, both species should fare better low on the shore where cool seawater and food are more available, and therefore that Chthamalus is abundant high in the intertidal prob­ably ­because that is the only place it can escape competition with the more vigorous Balanus. To test this hypothesis, he collected cobbles on which young barnacles had settled, scraped the individual Balanus recruits off of half of each cobble, then planted them back onto the shore. A ­ fter about two years, Chthamalus continued to flourish on the rocks low on the shore—­but only where Balanus had been removed (see figure B7.1.1A). This s­ imple experiment showed that competitive exclusion can happen in nature: Balanus excluded Chthamalus from its preferred habitat low in the shore. Equally important, the experiment showed that coexistence was made pos­si­ble by niche differences, namely, Chthamalus has higher tolerance for the harsh conditions of the drier upper shore, where it has a refuge from the superior competitor. Subsequent experiments revealed that a variety of other interactions among species shape rocky shore communities. Predation can also determine which species

175

dis­appear versus thrive in communities, and therefore their diversity. Robert Paine’s ­simple experiment on a rocky shore of Washington State, USA, dramatically illustrated this (Paine 1966, 1974). ­Here the ochre sea star Pisaster ochraceous feeds preferentially on the community’s dominant competitor, the mussel Mytilus californianus. Paine removed sea stars from a stretch of shore and found that, when protected from predation by sea stars, mussels increased and crowded out most other sessile species over the course of a few years. Thus, sea stars maintained high diversity of the rocky community by removing the community’s dominant species and releasing other species from competitive suppression by mussels. ­Because the sea star so dramatically altered species presence and abundances, Paine dubbed it the keystone species of the community, by analogy with the piece of masonry that supports the arch in classical stone doorways. ­Whether a predator plays such a keystone role depends on its feeding preferences, specifically for the competitively dominant prey species. Jane Lubchenco (1978) demonstrated this by manipulating the herbivorous snail Littorina littorea on rocky shores in New E ­ ngland, USA. In tide pools, the snail’s preferred food, the green alga Enteromorpha sp., was competitively dominant and snail grazing prevented Enteromorpha from excluding other algae, thus maintaining high diversity (except where snails w ­ ere so abundant that they ate every­thing). In contrast, on exposed rocks, the snails preferred Chondrus crispus, which is competitively inferior, and h ­ ere grazing accelerated competitive exclusion and reduced prey diversity (figure B7.1.1B). Lubchenco’s experiments showed that the composition and diversity of communities can be strongly affected by interactions between the traits of predators (food preference) and prey (competitive ability).

host species, such as the Ca­rib­bean long-­spined sea urchin Diadema antillarum, a major grazer that helped keep algae in check, in turn facilitating the dominance of corals on Ca­rib­bean reefs. In the early 1980s populations of this formerly very dense herbivore w ­ ere wiped out over a period of a few months throughout most of the tropical West Atlantic, evidently by a virus, accelerating the dominance of algae on many Ca­rib­bean reefs (Lessios 1988). In the 1930s the seagrass Zostera marina (eelgrass), which dominated shallow estuarine ­waters throughout the Northern Hemi­sphere, was struck by an epidemic of wasting disease caused by the slime mold Labyrinthula zosterae, and eelgrass all but dis­appeared from much of the North Atlantic for years, with huge ecological and economic impacts (Addy and Aylward 1944, Rasmussen 1977).

Interactions between Competitors All organisms require multiple resources to grow and reproduce, and resources are finite. As a result, competition for resources is essentially universal. Species differ in their ability to compete for resources, resulting in interspecific competition, the reduction in fitness of one species by another as a result of competition for a shared resource. Interspecific competition is central to understanding which species assem­ble and persist to form communities, and their relative abundances (chapter 8).

176

Ocean Ecology

Competition may occur for any limiting resource. Plants typically compete for light and nutrients and, on land, for w ­ ater. Animals may compete for food, territory, even refuges from predation or disease. Competition can have one of three outcomes: exclusion of one competitor, stable coexistence, or unstable coexistence. A good starting point in understanding competition is the competitive exclusion princi­ple, which states that no two species using the same limiting resource can coexist in­def­initely, that is, the superior competitor w ­ ill exclude the other (Hardin 1960). The competitive exclusion princi­ple has a strong inherent logic and was supported by classic early lab experiments (Gause 1934, Park 1948). Some of the first field experiments ­were done on rocky intertidal shores, and illustrated that interspecific competition can indeed result in exclusion of a competitor in the field (see box 7.1). Assessing the importance of competition in nature is often complicated by the practical difficulties of identifying which resources are limiting and by the fact that interactions ­don’t persist in­def­initely—­they are often interrupted by vari­ous disturbances and changes in environmental conditions. Connell’s classic demonstration of competition among barnacles (see box 7.1) has been so influential in part ­because the results ­were clear-­cut, involving an easily mea­sured limiting resource: living space on the rock surface. How do we approach competition in more complicated situations? In many cases, organisms compete for multiple resources, which requires considering more complex dynamics. David Tilman (1982) introduced a general theoretical framework for competition based on phytoplankton (and, ­later, land plants) competing for nutrients and light. Assuming that the single resource in shortest supply limits growth of the competing populations, the species that consumes that limiting resource down to the lowest level wins in exploitative competition. In the parlance of the model, R* (“R star”) is the lowest concentration of resource that can support a species population. Thus, all ­else being equal, the species with the lowest R* wins. At least for freshwater phytoplankton, this resource-­ratio theory is reasonably well supported by lab experiments, showing that species dominance changes with environmental nutrient profiles in ways predictable from each species’ nutrient use patterns in monoculture (Miller et al. 2005) (see also box 5.2). Yet, despite abundant evidence that competition is often strong in nature, communities generally contain many more species than expected if competition were the major force structuring them. ­There are many pos­si­ble, nonexclusive reasons for this, as we learn in chapter 8. One contribution to the explanation is that coexistence of competing species can be ­either stable or unstable, involving two distinct classes of mechanisms (Chesson 2000). Equalizing mechanisms reduce fitness differences among species, thus prolonging the pro­cess of competitive exclusion, allowing it to be more easily disrupted by environmental change, immigration from the other populations, or even evolutionary change. Equalizing mechanisms contribute to unstable coexistence in that they only slow exclusion, albeit potentially for a very long time. In contrast, stabilizing mechanisms reduce interspecific competition relative to intraspecific competition, giving each species an advantage over competitors in some portion of its niche, thus fostering stable coexistence. Stabilizing mechanisms include niche differences among species, such as use of dif­fer­ent resources or microhabitats that reduce competition. In general, stabilizing mechanisms include four general niche dimensions along which species differ: resources, enemies, space, and time (Chesson 2000).

Interactions between Plants and Herbivores All organisms compete for resources. For herbivores, predators, and parasites, t­hose resources are other organisms. Thus, the second major class of interactions between species is consumption: herbivory and predation. Plant production is the starting point for all ecological pro­cesses. Herbivores are consumers of plants, as plants are consumers of inorganic resources. The key distinction between ­these two types of resource-­consumer interactions is that plants are living, evolving organisms. Thus, consumption of plants by herbivores not only influences plant abundance but also has strong and often complex effects on their growth rates, ge­ne­tic composition, and trait evolution. Interactions between plants and herbivores create a more intimate and complex feedback system than that be-

Chapter 7 Species Interactions

177

tween plants and their inorganic resources. Notably, intense grazing has selected for a diverse proliferation of plant defense traits and herbivore counterattack traits involving morphology, chemistry, and life history (Lubchenco and Gaines 1981, Hay et al. 1994, Coley and Barone 1996). To persist and maintain positive growth, plants must escape, deter, or tolerate herbivory. Thus, the impact of herbivory can be understood as the product of three terms: probability of encounter, times probability of being eaten once encountered, times the fitness cost of being eaten (Lubchenco and Gaines 1981). Plants have l­ ittle control over their probability of encounter since they are fixed in place, or in the case of phytoplankton, cannot move of their own accord. Nevertheless, plants can sometimes escape by living in habitats inaccessible to herbivores or by producing vulnerable new tissues during seasons when herbivores are inactive. Such spatial escapes from herbivores are most evident in tropical regions where herbivory is high on average. For example, many edible algae and seagrasses in the tropics live primarily in structurally ­simple habitats away from the shelter of reefs, where they can avoid herbivorous fishes and urchins that are exposed to their own predators and reluctant to visit (Ogden et al. 1973, Hay 1981). ­These interactions are dramatically illustrated by the “halos” of bare sand maintained by tropical herbivores in the immediate vicinity of patch reefs vis­i­ble from space (figure 7.5). (A)

(B)

111 m

2655 m (C)

(D) 100 90

80

80 70

60

60 40

50 40

20

30 20

0 –20

10 0

3

6

9 12 15 18 21 24 Distance from reef edge (m)

Figure 7.5. ​The seascape of fear on the ­Great Barrier Reef. (A, B) Halos of grazed areas, vis­i­ble from space as white patches surrounding reefs, are caused by grazing fishes, who are reluctant to venture farther away from the reefs (C) b ­ ecause of predation risk. Panel (D) shows the decrease in experimental grazing (blue symbols) and increase in standing algal stock (green bars) with distance from the reef (­after Madin et al. 2011).

27

30

0

Canopy height (mm)

Percent of algae consumed

100

178

Ocean Ecology

Within such seascapes of intense grazing, palatable algae often persist at smaller, microhabitat scales in holes and cracks inaccessible to grazers (Menge et al. 1985). Some plants can escape herbivores in time, as illustrated by Halimeda, a ubiquitous green seaweed of coral reefs. Halimeda thrives where herbivory is intense b­ ecause it is protected by a combination of chemical defenses and tough calcium carbonate in its tissues. But Halimeda has an Achilles’ heel: its new growth is uncalcified, rich in nitrogen, and thus highly vulnerable to herbivores. Halimeda minimizes damage to its vulnerable young segments by producing them primarily at night when herbivores are asleep and by waiting to pump valuable nitrogen into t­ hese segments u­ ntil near sunrise when they can begin photosynthesizing (Hay et al. 1988). Despite such tricks, most marine plants w ­ ill eventually be found by herbivores. Where herbivores are abundant and effective, grazing imposes strong se­lection for traits that deter herbivory. ­These plant defenses include a range of morphological and chemical traits, discussed below. Plants may also persist in habitats where herbivory is intense by associating with other organisms that are themselves protected from enemies (Hay 1986, Littler et al. 1986). Such associational defenses entail ele­ments of both escaping detection and deterring herbivores. Fi­nally, some plants can persist and even increase in the face of herbivory if the fitness costs of being grazed are low. On coral reefs, small filamentous algae and cyanobacteria are generally the most abundant and impor­tant, albeit inconspicuous, primary producers. Th ­ ese algae form dense turfs, which are nearly invisible b­ ecause heavy grazing maintains very short stature. Turf algae can persist ­under intense grazing—in some cases, 100% of daily production is grazed (Carpenter 1986)—in part b­ ecause they grow extremely rapidly in the intensely bright light of shallow reefs.

Controls on herbivory: Plant traits Far from being passive responders to enemies, plants and algae produce a wide range of defenses that reduce their vulnerability to herbivores. Prominent among ­these are secondary metabolites—­chemical compounds so named b­ ecause they have no known function in basic (primary) metabolism, but are distasteful or noxious to herbivores. Secondary metabolites are best developed in situations where herbivory is intense, suggesting that the compounds evolved for defense against herbivory (Hay and Fenical 1988). Numerous experiments confirm that many secondary compounds of marine algae reduce herbivory and are impor­tant in mediating their distribution, abundance, and community composition where herbivory is intense, such as on coral reefs (Hay 1996). As on land, marine plants also produce structural defenses in the form of tough tissues. In the tropics especially, many seaweeds incorporate substantial CaCO3 into their tissues, which has the effect of both reducing nutritional quality of the tissue to herbivores and stiffening it. The distribution of plant defenses at several spatial scales corroborates their importance in protecting against enemies. On a local scale, reef areas with intense herbivory tend to be dominated by chemically defended seaweeds and sessile invertebrates, like sponges and soft corals, whereas poorly defended species are more common in sandy or other unstructured areas with lower herbivory (Hay 1981, 1996). On a geographic scale, consumer pressure is more intense at low latitudes, and plants in ­these regions tend to have stronger defenses against consumption. In North American salt marshes, for example, experiments have shown that herbivores generally prefer to eat plants from northern regions over plants of the same species from southern regions, which are tougher and less palatable than their northern conspecifics (Pennings et al. 2001, Siska et al. 2002). Similarly, sea urchins graze more heavi­ly on seaweeds from temperate North Carolina, USA, than on the same or related seaweeds from the subtropical Bahamas (Bolser and Hay 1996). Although plants are not normally considered capable of be­hav­ior, some respond to herbivore attack by producing induced defenses. Several kinds of algae up-­regulate production of defenses—­

Chapter 7 Species Interactions

sometimes quite sophisticated ones—­when damaged by grazers. For example, the bloom-­forming phytoplankter Phaeocystis globosa of polar ­waters can live ­either as single cells or as large colonies. Single cells are easily eaten by ciliate protozoans but not by copepods, whereas the reverse is true for large Phaeocystis colonies. Phaeocystis can chemically sense its neighbors being attacked by ciliates and respond by aggregating into the colony form that is too big for ciliates to eat (Long et al. 2007). Remarkably, when its neighbors are attacked by copepods, colony formation ceases and Phaeocystis reverts to growing as single cells too small to be grazed effectively by copepods. ­These shifts potentially translate to impacts at the ecosystem level since Phaeocystis is a dominant phytoplankter at high latitudes, and the growth form it takes influences energy flow to herbivores, nutrient cycling, and export to depth, thus carbon sequestration. Numerous experiments have documented the induction of chemical defenses in benthic and planktonic marine algae (Toth and Pavia 2007), generally ­after sustained exposure to herbivore damage over multiple days. Induction is more responsive to some kinds of herbivores than to ­others, and appears generally stronger a­ fter attack by small crustaceans and gastropods than by large invertebrates—­perhaps ­because the smaller herbivores use the plants as habitat and are thus more likely to remain predictably associated with it and to be affected by the defenses (Toth and Pavia 2007). Fi­nally, ­there is some evidence that macroalgae can mount defenses against microbial pathogens. Although ­little experimental research has addressed this topic, ­there are strong functional similarities in molecular defense systems between macroalgae and the immune systems of vascular plants and animals. ­These systems show homologies in defense-­activating signals and the enzymes involved, suggesting the intriguing possibility that immune-­like defenses against microbial attack may be pre­sent in a range of marine algae (Weinberger 2007).

Controls on herbivory: Herbivore traits Herbivorous animals differ widely in how they feed, what they consume, and the effects of feeding activities on plant biomass and community composition. The biggest differences are between vertebrate and invertebrate herbivores. Vertebrates generally have higher metabolic rates, especially among endothermic mammals and birds, which translate to higher average feeding rates and activity levels and larger home ranges. Th ­ ese differences can have ecosystem consequences: meta-­analysis of experiments showed that trophic cascades from predators through herbivores to plants ­were strongest in systems where vertebrate predators fed on invertebrate herbivores (Borer et al. 2005). The cascade from sea otters through urchins to kelp is a classic example (box 7.2). Herbivores also sort into groups within higher taxa. Herbivorous fishes can be divided into functional groups of scrapers, excavators, grazers, and browsers (Green and Bellwood 2009). Scrapers, including the small parrotfishes common on coral reefs worldwide, crop filamentous algal turfs and sediments from reef surfaces, whereas larger excavators feed in similar habitats but remove bites of reef substratum, causing bioerosion, and some feed on live corals. Grazers also feed on turf algae and detritus, whereas browsers feed on macroalgae, often highly selectively. As we discuss in chapter 12, the dif­fer­ent functional groups of herbivores play quite dif­fer­ent and often complementary roles in mediating transitions between coral-­dominated and algal-­dominated states. Larger grazers, including fishes and sea urchins, tend to differ in be­hav­ior from mesograzers, that is, small invertebrate herbivores that often associate with algae for habitat as well as food. ­Because of their more intimate relationships with host plants, mesograzer species often have more specialized feeding relationships than macrograzers. For example, although marine herbivores generally have much more generalized diets than terrestrial herbivores (Hay 1991), the ­great majority of marine specialists that do exist are mesograzing mollusks and crustaceans, many of which have specialized associations with chemically defended plants that provide some protection from the mesograzers’ own predators (Hay 1992) (figure 7.6).

179

180

Ocean Ecology

Box 7.2. ​Cascading effects of top predator decline

Otter number (% maximum count)

Plants and algae are autotrophs and therefore need nutrients and sunlight to thrive. Many primary producers also depend on animals, including predators. The complex web of interactions typical of communities in nature, and

100

the consequences of disturbing them, are well illustrated by communities associated with habitat-­forming foundation species, such as corals, seagrasses, and mangroves. T ­ hese organisms provide physical structure that defines a habitat,

Sea otter abundance

80 60 40

Amchitka I. N. Adak I. Kagalaska I. L. Kiska I.

20 0 400

Sea urchin biomass

g 0.25m–2

300 200 100 0 60

Sea urchin grazing intensity

% loss 24 hr–1

50 40 30 20 10 0

Number 0.25m–2

10

Total kelp density

8 6 4 2 0

1972

1985

1989 Year

1993

1997

Figure B7.2.1. Strong trophic interactions can cascade through four trophic levels. In the 1990s killer ­whales (orcas) in the Aleutian Islands of Alaska, USA, began feeding on sea otters, apparently due to low abundance of their normal prey, sea lions. Sea otter abundance declined, coinciding with increases in sea urchin abundance and grazing intensity and a decrease in kelp density (­after Estes et al. 1998).

Chapter 7 Species Interactions

often altering the environment and facilitating establishment of other species. The dominance of par­tic­u­ lar foundation species in turn often depends on consumers that reduce their vulnerability to competitors or grazers. A classic example involves the alternative ecosystem states dominated by kelps and coralline algal crusts on temperate rocky reefs (figure B7.2.1). Kelps are large and highly productive macroalgae that dominate rocky coasts at mid to high latitudes throughout the world, providing complex habitat structure and a rich flux of detritus that greatly increases the density and diversity of species in the habitat and even in adjacent terrestrial systems (Duggins et al. 1989, Steneck et al. 2002, Anthony et al. 2008). Sea otters formerly occupied East Pacific coastlines from Alaska to southern California, feeding on the rich invertebrates supported by kelp forests, seagrass meadows, and productive sediment bottoms. But sea otters w ­ ere hunted intensively in the seventeenth ­century, mostly by Rus­sian fur traders, and ­were eliminated or greatly reduced in much of this range, surviving primarily on a few remote and inaccessible islands of the Aleutian chain in present-­day Alaska. In the early 1970s Jim Estes was censusing sea otters in the region for the US Fish and Wildlife Ser­vice when he noticed an intriguing pattern and

A

181

initiated a series of classic observations and experiments showing that the dominance of g ­ iant kelp depends on a trophic cascade. The first clues came from the surveys of sea otter populations, which showed that islands hunted out centuries ago had radically dif­fer­ent subtidal communities than t­ hose with extant otter populations: “otter islands” supported luxurious communities of g ­ iant kelp (Macrocystis pyrifera) and other macroalgae, whereas the submerged rocks on islands without otters w ­ ere paved by coralline algal crusts and overrun by sea urchins (Estes and Palmisano 1974). Comparative studies and experiments confirmed that t­ hese alternative states w ­ ere mediated by a strong trophic cascade, from predatory otters to herbivorous urchins to kelps (Duggins 1980). Remarkably, l­ater studies added a fourth level to this trophic cascade, showing that a behavioral switch by killer ­whales (orcas) to feeding on sea otters cascaded all the way down to reduce kelp beds (Estes et al. 1998), just as had happened with h ­ uman hunters long ago. Synthesis of the comparative and experimental data suggests that subtidal rock bottoms indeed tend to switch between alternative semistable states of kelp beds and urchin barrens in several parts of the world (Filbee-­Dexter and Scheibling 2014).

B

C

Figure 7.6. ​Specialist mesograzers that feed and live on chemically defended algae, thereby obtaining protection from their own predators: (A) the sea slug Costasiella sp.; (B) the sea slug Cyerce nigra; (C) the portunid crab Caphyra rotundifrons, specialist feeder on the green seaweed Chlorodesmis.

182

Ocean Ecology

Interactions between Prey and Predators Consumption of heterotrophs by other heterotrophs (prey-­predator interactions) has basic similarities with plant-­herbivore interactions, with the main exception that animals move and behave, resulting in a richer range of responses by animals to their enemies. Escaping detection, which is difficult for plants, is a primary means by which animals avoid being eaten, involving crypsis, camouflage, and activity patterns that minimize exposure to predators in time and space.

Controls on predation: Prey traits Many traits of organisms affect their vulnerability to predation, including body size, mobility, be­hav­ ior, crypsis, body composition, and physical and chemical defenses. Body size of prey is prob­ably the single most impor­tant characteristic governing prey se­lection in food webs generally. Indeed, size classes are the classic means of classifying plankton and suspended particles and are central to modeling the flow of energy and materials through food chains. In the plankton, size affects consumer-­prey interactions in two ways (Hansen et al. 1994): larger consumers eat larger prey on average, and within the range of acceptable prey sizes a given individual consumer generally takes larger prey items disproportionately to their abundance (figure 7.7). Generally, the se­lection of larger prey is due to their greater energy content, but it can also result from greater ease of detecting larger items. ­These effects of body size on consumer-­prey interactions turn out to be general across ecosystems. A compilation of more than 5000 studies showed that prey size increases linearly with consumer size on a log-­log scale, and that the relationship is quite similar among dif­fer­ent ecosystems but differs between invertebrate and vertebrate predators (Brose et al. 2006). Nutritional composition of animal prey also influences se­lection by predators. Historically, foraging has been thought to be driven by prey energy (carbon content), since the stoichiometry of animal prey is much more similar to that of the predators’ own tissues than is true for herbivores feeding on plant prey. However, predators are often quite selective among prospective prey and even eat certain prey tissues, particularly fat, in preference to ­others (Machovsky-­Capuska et al. 2016). Thus, in general, food choice by herbivores and detritivores tends to be strongly influenced by the elemental stoichiometry and nitrogen content of available foods, whereas carnivore food choice is more often influenced by the balance of prey lipid and protein content, in addition to prey body size, be­hav­ior, and detectability. Interactions between predators and prey involve behavioral complexity that d­ oesn’t apply to herbivore-­plant interactions. Animals are not passive in the face of predators—­their be­hav­ior is often strongly affected by the risk of predation. The mere presence of a predator can depress prey population growth, distribution, and activity even without killing them, with consequences that ­ripple through the interaction web. Such trait-­mediated (nonconsumptive) effects—­changes in prey be­hav­ ior and physiology in the presence of a predator—­comprise a large part of the effects of predators on prey population dynamics, energy flow through food webs, and the structure of communities (Preisser et al. 2005). For example, the amphipod Ampithoe longimana is an impor­tant generalist herbivore common in seagrass and macroalgal habitats in the southeastern USA and, like most small crustaceans, highly vulnerable to fish predation. In mesocosm experiments, the mere presence of chemical cues from a fish predator steeply reduced the amphipod’s grazing rate, reduced its dispersal and colonization of field patches by > 50%, and as a consequence reduced amphipod population abundance by half ­after two generations, compared with controls ­free of fish cues (Reynolds and Bruno 2013). In a similar experiment, the lower amphipod population growth and grazing nearly doubled algal biomass where amphipods ­were exposed to predator cues (Reynolds and Sotka 2011). ­These strong nonconsumptive effects of predators indeed are the norm: effects of prey intimidation by predators ­were generally at least as strong as effects of direct consumption, averaging 63% of the total effect of predators on prey population (Preisser et al. 2005).

Chapter 7 Species Interactions

(A) Dinoflagellate

Optimal prey ESD (μm)

100 18:1

1:1

(Gyrodinium)

Other flagellates

(Ochromonas, bodonida)

30

Ciliates

(oligotrichs)

3

1

Rotifers

8:1

10

(Brachionus)

50:1

Copepod nauplii Copepodites Cladocerans (limnic filtrators)

3:1

Meroplankton

(bivalve, gastropod, and polychaete larvae)

3

10

30 100 Predator ESD (μm)

300

1000

(B)

(C) 10 8

4

Log body mass ratio

Log10(predator mass)

6 2 0 –2 –4 –6 –8 –10 –10

–8

–6

–4

–2

0

2

Log10 (prey mass)

4

6

8

10

Inv Ect End Marine

Figure 7.7. ​Consumer body mass scales consistently with prey body mass. (A) Relationships between log size (equivalent ­spherical dia­meter, ESD) of predator and prey in marine plankton; numbers show ratios of predator size to optimal prey size for dif­fer­ent taxa (­after Hansen et al. 1994). Relationships between predator and prey size (B) across a wide range of taxa and (C) among invertebrate (Inv), ectothermic vertebrate (Ect), and endothermic vertebrate (End) predators in marine systems (­after Brose et al. 2006).

Since herbivores are central links in food webs and many species are relatively small and vulnerable, such nonconsumptive effects of predators might be expected to have impor­tant indirect effects on communities and ecosystems. “Landscapes of fear” are common and impor­tant in a variety of ecosystems and can have surprisingly strong cascading effects on prey populations (see figure 7.5) and communities, even when predators do not kill prey. For example, on rocky shores of New ­England, USA, predatory green crabs (Carcinus maenas) feed on the herbivorous snail Littorina littorea, which in turn feeds on rockweeds (fucoid algae). When grazing snails ­were experimentally confined on rock surfaces in the field, snails exposed to effluent from a predatory crab ate far less algae, resulting in nearly fivefold greater recruitment of the rockweed they normally grazed (Trussell et al. 2002). This was despite the fact that crabs w ­ ere caged and could not touch the snails. Thus, the predator altered prey be­hav­ior and algal abundance via a nonconsumptive effect. Trophic cascades resulting from direct effects of predation typically attenuate down the food chain, whereas ­these nonconsumptive effects of predators do not attenuate and, surprisingly, nonconsumptive effects of

183

184

Ocean Ecology

predators w ­ ere responsible for nearly all the cascading effect of predators on plants in marine systems (Preisser et al. 2005). Clearly, approaches to understanding trophic interactions focusing only on density of organisms miss a big part of the dynamics. Nonconsumptive effects of predators—­the landscape of fear—­can have significant implications for conservation and management. Elizabeth Madin and colleagues (2010) studied fish be­hav­ior along a gradient in fishing intensity across the central Pacific’s northern Line Islands, spanning a range that included nearly pristine coral reefs. Variation in predation risk along the gradient resulted in strong behavioral shifts in foraging patterns among multiple fish species, including both short-­ term and longer-­term “chronic” responses to perceived risk. Th ­ ese areas of reduced fish grazing, resulting from fear of natu­ral predators, are even vis­i­ble from space in Google Earth images (see figure 7.5) (Madin et al. 2011). Such behavioral responses to predators are impor­tant ­because they can affect entire fish assemblages, thus causing greater impacts than killing of prey by predators, and ­because they can respond very rapidly to the mere presence of a predator.

Controls on predation: Predator traits We’ve seen that larger aquatic consumers tend to eat larger prey disproportionately to their abundance (figure 7.7). This scaling occurs not only among species but even among individuals across the ­great range of sizes spanned by a single individual fish during ontogeny. This is dramatically illustrated by a study that tracked ontoge­ne­tic change in fish trophic level using the chemical tracer 15N, the heavy isotope of nitrogen that tends to accumulate predictably in a predator relative to its prey (Peterson and Fry 1987). Among North Sea groundfishes, the stable isotope signature δ15N bore ­little relation to the maximum body size of the species but was strongly correlated with individual body size. That is, in many fishes, trophic level is a characteristic of individuals rather than a species-­ specific trait ( Jennings et al. 2001, figure 6.5). A predator’s degree of feeding selectivity can strongly influence its impact on prey, particularly ­whether it is a generalist or specialist feeder (see box 7.1) and w ­ hether its feeding preferences are fixed or flexible. Just as prey can act to avoid or escape their enemies, predators also have flexible be­ hav­ior that influences their interactions and their effects on communities. Predators that can switch to another prey species when one prey is depleted tend to stabilize community structure b­ ecause they depress populations of abundant species and relax pressure on species that are rare (Murdoch 1969). The impacts of predators on prey also depend on their hunting mode, for example, ­whether they are active pursuers or rely on an ambush or sit-­and-­wait strategy (Schmitz 2007).

Parasitism and disease Microbes that attack and consume macroscopic organisms are special kinds of predators known as pathogens, and their impacts on prey manifest as disease. Essentially all organisms are exposed to one or another kind of disease or parasite. In the open ocean, for example, metagenomic sequencing has shown that parasites are major players in the primarily microbial food webs of the pelagic zone (see box 2.1). Microbial pathogens are often keystone species in that they have virtually no biomass but can have huge impacts on communities. Diseases have had especially strong and wide-­ranging ecological impacts when introduced into naive communities that have been historically isolated from enemies. Indeed, the most detrimental invasions have invariably been pathogens—­organisms that cause disease. The introduction of the avian malaria parasite Plasmodium relictum drove much of the spectacular endemic bird fauna of the Hawaiian Islands to extinction ­after Eu­ro­pean contact; and a fungus, chestnut blight, transformed the eastern forests of North Amer­i­ca in the early twentieth ­century, where it devastated the community dominant American chestnut. In the sea, a still uncharacterized disease, apparently a virus, wiped out more than 95% of the population of Diadema antilla-

Chapter 7 Species Interactions

rum, the dominant herbivorous sea urchin, throughout its range in the western Atlantic over a few years in the early 1980s, releasing its algal prey to bloom throughout the region (Lessios 2016). And the bacterial white-­band disease of elkhorn and staghorn corals was similarly central in the catastrophic decline of ­these dominant reef-­builders in the Ca­rib­bean since the late 1970s.

Predation and community diversity Herbivores and predators play central roles in shaping the diversity of communities (chapter 8). Consumer impacts on community composition and diversity depend on several features (Holt 2009). First and most obvious are the intensity and selectivity of feeding. A consumer that feeds intensively on a prey species can drive it to extinction. H ­ uman predation drove many large vertebrates to extinction in Pleistocene and Holocene times across all continents (Sandom et al. 2014), and nonnative predators have often invaded isolated, oceanic islands such as Hawaii and decimated their native species (Blackburn et al. 2004, Clavero and García-­Berthou 2005). But a predator’s influence on prey diversity depends on traits of both predator and prey. A predator that feeds nonselectively on all prey species imposes a disturbance similar to that of abiotic stressors, such as a storm that clears space indiscriminately. In contrast, a predator that feeds selectively can e­ ither accelerate or arrest competitive exclusion among prey, depending on which prey it targets. If the predator feeds preferentially on the competitively dominant prey, its removal of the dominant species releases other species from competition and increases diversity of prey in the community. Fi­nally, if the predator instead targets inferior competitors, it accelerates competitive exclusion and reduces prey diversity. Experiments on rocky shores have demonstrated each of ­these kinds of predator effects on diversity (see box 7.1).

Facilitation and Mutualism Community ecol­ogy has traditionally focused on the negative interactions of competition and predation. But positive interactions are also widespread, so much so that we often overlook them—­trees provide habitat for insects and birds, corals shelter fishes and crabs, and so on. Facilitation happens when one species modifies environmental conditions so as to make them more favorable for another. Facilitation tends to be most impor­tant in physically and biologically harsh environments, where one species (e.g., a larger foundation species) may ameliorate the stressful conditions for another species (Bertness and Callaway 1994, Bruno et al. 2003). The most general examples of facilitation involve foundation species, which create habitat and often reduce environmental stress for associated organisms, boosting their abundance and diversity. Indeed, facilitation is responsible for the existence of entire ecosystems that assem­ble around foundation species such as corals, seagrasses, and mangroves (chapters 11, 12). Facilitation also influences communities through countless smaller-­scale interactions. A common theme is that species that other­wise compete for resources may facilitate one another when conditions become biologically or environmentally stressful. For example, in the southeastern USA, omnivorous sparid fishes are abundant and voracious consumers of both small invertebrates and algae, with strong impacts on the structure of benthic communities. By feeding on green and red seaweeds, t­ hese fishes foster dominance by the brown seaweed Sargassum, which they find distasteful (Duffy and Hay 2000). In the absence of ­these grazers, competition theory predicts that the dominant seaweed would reduce the diversity of other algae. But Mark Hay (1986) found the opposite pattern: in the field, seaweeds palatable to fishes w ­ ere found only in the presence of the dominant Sargassum and in fact w ­ ere mostly growing attached to it. Experiments showed that the palatable red seaweed Hypnea grew less in as­ ere excluded, but that living with Sargassum benefited sociation with Sargassum when grazing fish w Hypnea when fish ­were pre­sent ­because fish could not easily pick out the tasty Hypnea when it was

185

186

Ocean Ecology

Hypnea

Sargassum

100 80

Change in wet mass (%)

60 40 20 0 –20 –40 –60 –80

With hypnea

Alone

With hypnea

Alone

On sargassum

Alone

On sargassum

Alone

–100

Figure 7.8. ​Associational defenses against herbivory in seaweeds. In a field experiment, the palatable seaweed Hypnea grew less when associated with the unpalatable seaweed Sargassum in the absence of fish grazing (green bars), but survived better in association with Sargassum when fish ­were pre­sent (blue bars). Sargassum was unaffected by ­either fish or presence of Hypnea (­after Hay 1986).

intertwined with the distasteful Sargassum (figure 7.8). Thus, in this system, consumer pressure turned an interaction between two plants from negative (competition) to positive (facilitation). Facilitation occurs in many kinds of organisms and ecosystems (Bruno et  al. 2003). On tropical reefs, noxious fire corals are local hotspots of algal diversity ­because they protect palatable seaweeds from grazing (Littler et al. 1986). On cobble beaches of New E ­ ngland, USA, herbaceous salt marsh plants can survive only where the dominant cordgrass Spartina alterniflora stabilizes the cobbles; thus, a plant that would be a competitor in more hospitable sediments acts as a facilitator where the bottom consists of shifting rocks (Bruno 2000). Similar facilitation extends to the small animals that associate with and feed on chemically defended seaweeds and sessile invertebrates. Most of the seaweeds, sponges, and soft corals that dominate tropical reefs are protected from intense consumer pressure by distasteful or toxic chemical compounds. Th ­ ese protected organisms often harbor small invertebrates specialized to live on them, including vari­ous crabs and amphipod crustaceans, but especially ascoglossan sea slugs, nearly all of which live in specific associations with chemically defended green algae (Hay 1991) (see figure 7.6). Experiments show that many such mesograzers feed preferentially on the chemically rich host algae and suffer lower predation by living on them, consistent with the hypothesis that their association with chemically defended algae was selected for by intense predation pressure (Hay et al. 1989, 1990).

Ecological Networks

The previous sections show that organisms interact with one another in a wide variety of ways, with an even wider variety of consequences. Trying to understand dynamics of numerous, simultaneous interactions by piecing them together pairwise is a formidable challenge in practice and perhaps impossible even in princi­ple. One way around this prob­lem is to focus on the properties of interaction networks that arise from the large number of direct and indirect interactions that shape community structure. As noted above, an ecological network can be summarized as a graph of the links among species in a community and is characterized by topology, denoting who interacts with whom, and the directions, signs, and strengths of the interactions (see figure 7.1). The most familiar type of ecological network is a food web, a community of species joined by trophic (consumer-­prey) interactions. To understand food web dynamics, we need at least three types of data (May 1973): a list of species; the topology or structure of links among them (i.e., who eats whom); and the relative strengths of the links (i.e., per capita feeding rates). A key question in food web ecol­ogy is how ­these networks of diffuse interactions produce emergent structure, meaning properties that transcend the identity of the individual species involved (Loeuille and Loreau 2005, Petchey et al. 2008). Food webs are also useful in many practical applications, for example, to summarize the fluxes of energy and materials among interacting species. Network-­based ecosystem models have been central to biological oceanography and fisheries science, both of which focus on what controls primary production, how it transits to higher trophic levels, and how trophic levels respond to mortality imposed by ­human harvesting. A food web is a con­ve­nient graphical summary of the paths along which ­those pro­cesses happen. We explore t­ hese applications in more detail in chapters 8 and 9. The earliest marine food webs ­were born from fisheries management, the first published by C. G. J. Petersen (1918), summarizing the identity, biomass, and feeding interactions among species in

187

Chapter 7 Species Interactions

shallow eelgrass (Zostera marina) beds of the Kattegat on the west coast of Sweden. The commercial motivation for this work is clear by the designation of the vast majority of invertebrate biomass in the system as “useless animals” (figure 7.9). Around the same time, Alistair Hardy (1924) produced a depiction of the pelagic food web supporting herring in the North Sea (figure 7.10). A glance at Hardy’s web hints at why so much effort has gone into finding emergent properties of food webs—­there is ­little hope of understanding such complex systems by building them up from individual pairwise interactions.

Kattegat Plaice etc.

Cod etc.

5

Plankton

Herring etc.

6

Predatory crustaceans and gastropods (large)

7

Small fish Starfish

50

10?

25

100?

200?

Useful animals: 1000

50

500

150

Useless animals:

Functional traits as a lens into community organ­ization

5000

Zostera Community structure has traditionally been defined by the 24000 number, identity, and relative abundances of species pre­sent. This nomenclatural approach contains only indirect and imperfect information about the interactions among species needed to understand community assembly and dynamics. Species names provide the link to a ­great deal of natu­ral history informaFigure 7.9. ​An early food web of marine animals supported by eelgrass (Zostera marina) on the Swedish coast. The size of tion about the species, but the link is qualitative, and that knowleach box represents estimated biomass of that component in edge often exists primarily in the heads of specialists who know thousands of tons. Note the much greater biomass of plants the organisms well. In contrast, functional traits are explicit, mea­ than of animals and the clear practical motivation shown by sur­able, and inherently connected to ecological mechanisms. the distinction between “useful” and “useless” animals (­after Petersen 1918). Thus, functional traits have the potential to transcend the taxonomy of par­tic­u­lar communities, link more closely to predictive ecol­ogy (McGill et al. 2006), and bridge from individuals to populations to communities to ecosystem pro­cesses (Keddy 1992, Lavorel and Garnier 2002, McGill et al. 2006). The challenge is realizing this potential. The distribution of trait values within a community offers clues to the pro­cesses structuring it. Perhaps the earliest trait-­based approach to understanding species interactions involved the idea of limiting similarity, which predicts that traits involved in resource acquisition, such as mouth width of fishes or body size generally, ­will be more evenly distributed among co-­occurring species than expected by chance b­ ecause competition is expected to exclude functionally similar species from a community (Hutchinson 1959, MacArthur and Levins 1967). Conversely, traits involved in adaptation to the environment, such as transparent bodies in the open pelagic ocean, are expected to be more similar among co-­occurring species than expected by chance. Thus, we might expect the species in a community to show a mix of divergent and convergent traits depending on their functions. Consider the Northeast Pacific seabass species in the genus Sebastes. The cool coastal ­waters of this region host an impressive evolutionary radiation of Sebastes (figure 7.11), with over 70 species differing widely in morphology, diet, and life history. How do so many species coexist? Travis Ingram and Jonathan Shurin (2009) mea­sured two traits hypothesized to mediate adaptation to the environment and competitive partitioning of diet among species. First, the gill rakers of fishes form a sieve to retain ingested food while allowing ­water to be expelled through the gills. The length and number of gill rakers are related to the size and nature of food a fish can consume. Therefore, if competition for food limits the similarity of species that co-­occur within a community, gill raker traits should differ consistently among them. The second trait, eye size, is related to visual per­for­mance in dim w ­ aters at depth. Dif­fer­ ent seabass species occupy a range of habitats, from shallow to deep ­water, so, in contrast to the expected divergence in gill raker traits, eye size should be more similar among co-­occurring species than

188

Ocean Ecology

Young

7–12mm

12–42mm

Herring

42–130mm

Adult herring

Ammodytes juv. Medusae

Cumacea

Pleurobrachia Tomopteris

Semibalanus cypris larvae

Larval Mollusca

Harpacticoid copepods

Nyctiphanes

Decapod larvae

Mysidacea

Paracalanus

Oikopleura

Limacina

Sagitta

Pseudocalanus

Hyperiid amphipods Apherusa

Temora Acartia

Centropages

Calanus Podon Evadne Foraminifera

Noctiluca

Tintinnopsis

Radiolaria

Prorocentrum

Thalassiosira

Melosira

Peridinium

Paralia

Ceratium

Coscinodiscus

Rhizosolenia

Chaetoceros

Biddulphia

Guinardia

Navicula Nitzschia

Phaeocystis

Figure 7.10. ​Alistair Hardy’s classic North Sea pelagic food web supporting herring. Early food webs w ­ ere motivated by the need to understand environmental controls on marine fishery production, including Hardy’s work (1924) on the food web supporting herring in the North Sea.

A

B

Figure 7.11. ​Selected rockfish (Sebastes) species of the Pacific coast of North Amer­i­ca. Ingram and Shurin (2009) showed that rockfish assemblages exhibit two contrasting patterns of trait-­based assembly: co-­ occurring species are similar in eye size, reflecting habitat filtering according to depth and light level, but less similar than expected in morphology of gill rakers, which are related to feeding, consistent with limiting similarity caused by competition. (A) Sebastes ruberrimus and (B) S. mystinus illustrate the morphology and eye size typical of deeper and shallow w ­ aters, respectively.

expected by chance. The investigators first confirmed the assumption that gill raker morphology and relative eye size are related to trophic position and habitat depth, respectively. Then they tested their distribution among the seabass species. As hypothesized, when they mapped traits onto a phyloge­ne­ tic tree, gill raker morphology indeed was more evenly distributed among species than expected by chance, whereas eye size was more clustered than expected (Ingram and Shurin 2009). Thus, func-

Chapter 7 Species Interactions

tional traits mediating resource and habitat use supported the roles of environmental filtering and food competition in community assembly in this diverse group of fishes. Knowledge of functional traits can also provide insights into community dynamics and responses to environmental forcing. For example, in the Gulf of Mexico, demographic changes of estuarine fish species over a 30-­year time series w ­ ere predicted with 72% accuracy by a combination of four biological traits (spawning season, the maximum and range of salinity occupied, and oocyte size), plus taxonomic order, which may be considered a proxy for phyloge­ne­tically conserved traits that ­were not mea­sured directly (Sirot et al. 2015).

Traits in interaction networks The growing body of well-­resolved food webs and new computationally intensive analyses provide intriguing hints of general patterns in the structure of interaction webs. One system where interactions are well characterized is the rocky intertidal zone, where numerous experiments have explored how competition and predation influence community organ­ization (chapter 8). Compilation of data from many such experiments found that, surprisingly, organismal traits w ­ ere better predictors of interaction strength than ­were environmental ­factors. Neither community diversity, environmental stress (proxied by height in intertidal), time, nor size of experimental plot significantly affected interaction strength among competitors, consumers, and their prey in the studies. In contrast, body mass and trophic group well predicted interaction strength, which was stronger in large organisms and stronger in herbivores and carnivores than in plants or suspension-­feeders (box 7.3) (Wood et al. 2010). Interestingly, ­because strong interactors ­were less abundant than weaker interactors, the population-­level interaction strengths of the dif­fer­ent groups ­were remarkably similar (figure B7.3.1C). ­These results suggest that even in complex communities, key patterns in interaction strength may be predictable based on a small number of organismal traits. Simulation models support the generality of ­these results from field experiments. Among 600 model food webs the principal predictors of interaction strength ­were body mass of the species and the ratio of consumer to prey body mass (Berlow et al. 2009). But applying this model to the effects of predator removal in a rocky shore community also illustrates key differences between dif­fer­ent kinds of interactions. The model accurately predicted the outcomes of trophic interactions, that is, of whelks feeding on mussels. But the model was less successful in predicting nontrophic interactions, such as the finding that structure created by barnacle shells facilitated recruitment of mussels and provided an alternate prey (Berlow et al. 2009). One impor­tant lesson is that real interaction networks differ in impor­tant ways from classical food webs in that nontrophic interactions strongly affect species dynamics. This is particularly true where foundation species provide habitat. In North American salt marshes and African seagrass meadows, the presence of ­these dominant plants increases community species richness across trophic levels, as well as the number of interactions per species (van der Zee et al. 2016). ­These results highlight that many communities are as strongly s­ haped by facilitative interactions as by the negative interactions of competition and predation that ecologists have focused most intently on for de­cades (Bruno et al. 2003).

Emergent properties of ecological networks A key question in conservation and management is how harvesting or loss of a par­tic­u­lar species affects the stability of the ecosystem. One way of addressing this question is by simulating extinctions in model food webs built with topologies and interaction strengths from real data. Real food webs tend to have highly uneven distributions of interaction strengths (see figure 7.2). Model simulations confirm that food webs are highly vulnerable to loss of the well-­connected species (Sole and

189

190

Ocean Ecology

Box 7.3. ​Macroecological scaling of community interactions exponent of 0.59–0.70. Surprisingly, interaction strengths varied l­ittle with tidal height. ­These results illustrate that scaling of biological pro­cesses with body mass extends beyond the metabolism and physiology of individual organisms to the interactions among species that or­ga­nize communities. ­These experiments also revealed a pattern parallel to the energetic equivalence rule (chapter 6) in rocky intertidal systems. Per capita interaction strengths of individual organisms scaled with body mass according to a typical macroecological power function—­linearly on a log-­log plot. In contrast, when interaction strength was calculated at the population level by multiplying per capita interaction strength by density of individuals pre­sent, ­these population-­level interaction strengths ­were very similar among producers, suspension-­feeders, herbivores, and predators across the size range (figure B7.3.1C). Thus, while herbivores and predators had stronger individual effects on other species than did plants and suspension-­ feeders, whole-­population impacts of ­these groups ­were roughly equivalent, prob­ably reflecting that intensity of interactions among species is driven by requirements for resources that scale allometrically with body size (Wood et al. 2010).

The rich communities of seaweeds and invertebrates on intertidal rocky shores have provided a wealth of insights on how interactions among species create the structure of communities (chapter 8). Most of this research has focused on the natu­ral history of the organisms and how they mediate interactions that shape species composition and diversity. This is appropriate and necessary—­but can we draw generalizations about community pro­cesses that transcend taxonomy? Spencer Wood and colleagues (2010) addressed this question by removing and adding species in rocky intertidal communities of North Amer­i­ca and New Zealand. They used parallel experiments across all the sites to provide comparable estimates of interaction strength, defined as the per capita impact of a focal (manipulated) species on the population of a target species. The experiments showed that strengths of vari­ous interactions, including competition, consumption, and facilitation, scaled closely with log individual body mass (figure B7.3.1A), mirroring previous results for individual metabolism (chapter 5). Interaction strength also differed systematically with trophic level, being roughly two o ­ rders of magnitude higher among herbivores and predators than for producers and suspension-­feeders (figure B7.3.1B). In each of t­ hese functional groups, log interaction strength scaled linearly with log individual body mass with an

(B)

Per population interaction strength

–2

–8

–14

vo re rn i

iv or e

Ca

rb He

en

sio

n

er

sp

uc

Su

Ca

rn i

vo re

–20 iv or e

–20 n

1000

rb

10 100 Body mass (g)

He

0

sio

0.1

en

1e–08

–14

er

1e–07

sp

1e–06

uc

1e–05

–8

Su

1e–04

Pr od

0.001

(C) –2

Per capita interaction strength

Per capita interaction strength

0.01

Producer Suspension feeder Herbivore Carnivore

Pr od

(A)

Figure B7.3.1. Community-­wide interaction strengths estimated from experiments in rocky shore communities. Standardized experiments on rocky shores showed that (A, B) per capita interaction strengths increase consistently with body mass and trophic group, and that (C) per population interaction strengths are similar among trophic groups (­after Wood et al. 2010).

Montoya 2001). Similarly, simulated species removals from a highly diverse coral reef food web (see figure 7.2C) showed that the network was robust to the loss of a randomly chosen species but was destabilized by the loss of strongly interacting species, such as sharks and other top predators (Bascompte et al. 2005). ­These simulations support experiments and observations showing that species with strong connections to ­others, namely, keystones and foundation species, have large impacts on

Chapter 7 Species Interactions

communities (Estes et al. 2011), even when the lost species w ­ ere rare (Bracken and Low 2012). This result has impor­tant implications for management ­because it is exactly such large, strongly interacting species that are often most vulnerable to ­human impacts. But weak interactions are also impor­tant, as shown by both model simulations and experiments. Perhaps paradoxically, weak interactions are especially impor­tant in stabilizing food webs against random species removals b­ ecause they help reduce fluctuations in abundance within interaction webs (Sole and Montoya 2001). This result is corroborated by experiments: in a rocky reef community, removal of weakly interacting predators destabilized the community (O’Gorman and Emmerson 2009).

Ecological Interactions in the Anthropocene A frontier in Anthropocene ecol­ogy is understanding and predicting how the vari­ous components of ongoing global change affect interactions among species and the consequences of t­ hose altered interactions. Tremendous effort has gone into modeling how species distributions and abundances w ­ ill respond to climate warming (chapter 4), but ­these are commonly based on inferred physiological and habitat requirements of species, with ­little attention to their interactions with other species. Can the ecol­ogy of interactions offer insights for how ecosystems ­will respond to the ongoing changes of the Anthropocene? H ­ ere we consider two of the clearest and most pervasive consequences of Anthropocene change, involving the effects of warming ocean ­waters and the decline of large animals.

Changing species interactions in a changing climate Interactions among species respond to many environmental d­ rivers, and temperature is among the most fundamental (chapter 5). All organisms have an optimal range of temperature, and b­ ecause interactions ultimately depend on metabolic pro­cesses, interactions with other species also tend to increase in rate and intensity with rising temperatures t­ oward the organism’s optimum and then decrease, forming a hump-­shaped relationship with temperature (Kordas et al. 2011). In general, interactions change if metabolism or population growth of interacting species responds differently to changing temperature. The pervasiveness and complexity of temperature effects on organisms create a nearly infinite range of pos­si­ble effects on species interactions. Nevertheless, several general predictions can be made based on metabolic theory and ­limited empirical data. ­These involve asymmetries in the thermal responses of consumer and prey to temperature (Bruno et al. 2015). First, temperature generally affects metabolism of heterotrophs more strongly than autotrophs. This is expected to strengthen top-­down control of plants by herbivores with rising temperatures, and evidence from both pelagic and benthic marine systems supports this prediction (discussed in chapter 4). For example, global compilations of data show that intensity of herbivory increases with environmental temperature in invertebrate herbivores feeding on salt marsh vegetation (He and Silliman 2016). Second, rising temperatures are predicted to affect predator-­prey interactions through differential scaling of organismal traits involved in resource detection and capture, including relative rates of movement and ­handling time (Dell et al. 2014). A model of ­these pro­cesses found that temperature influences predation primarily by affecting the relative movement rates (i.e., body velocities) of the interacting species. Body velocity influences encounter rate of predator and prey, and therefore consumption rates. Asymmetries between predator and prey in how body velocity scales with temperature can change their interactions not only quantitatively but qualitatively. For example, when the prey is sessile, temperature affects only movement rate of the consumer and is likely to increase encounter rates, whereas predictions are more complex for mobile prey. Predation by the sea star Pisaster ochraceous on mussels fell by about two-­thirds during cool-­water upwelling events on the Oregon coast even though the w ­ ater temperature dropped by only a few degrees (Sanford 1999). Such temperature-­mediated changes in interaction strength are strongest among

191

192

Ocean Ecology

ectothermic predators. Endothermic animals are less affected by environmental temperatures, so rising temperatures are likely to ­favor ectothermic prey over their endothermic predators. In general, when consumers and resources have similar body temperatures and mobilities, temperature change is less likely to alter interactions through thermal asymmetry. Review of thermal response curves in some 300 species from vari­ous ecosystems concluded that asymmetries in thermal responses of consumers and prey are common in nature, and therefore likely to drive altered interactions in a warming world (Dell et al. 2014). Third, species interactions can be disrupted if temperature differentially affects phenology. Warming generally accelerates rates of organismal development that affect the timing of seasonal life history events, but such changes often proceed at dif­fer­ent rates in dif­fer­ent species. Phenological mismatch between consumers and prey can disrupt their interactions, with impor­tant consequences for food web structure and flows of energy and materials through ecosystems. One of the first marine studies of this phenomenon mea­sured the change in seasonal appearance of 66 plankton taxa in the North Sea over the second half of the twentieth c­ entury (Edwards and Richardson 2004). Changes in seasonal life history timing varied widely among taxa. In general, the blooms of diatoms that support higher trophic levels did not shift much, prob­ably ­because they respond to day length rather than temperature. In contrast, the planktonic larvae of fishes and invertebrates, whose life histories are very sensitive to temperature, tended to appear much ­earlier in the warmer ­waters of recent years, averaging 27 days ­earlier in 2002 compared with 1958. ­These patterns have strong implications for pelagic ecosystems since the early hatching larvae miss the peak season of food availability and many starve. In the UK over the three de­cades ending in 2005, warming-­related life history changes w ­ ere slower for secondary consumers than for primary producers and herbivores across marine, freshwater, and terrestrial systems, suggesting that interactions that depend on synchronization of life history events are being widely disrupted (Thackeray et al. 2010). Fi­nally, ­there is abundant evidence that interactions among species often shift from negative outcomes (e.g., competition) in benign environments to more positive, facilitative interactions in stressful environments (Bruno et al. 2003). We saw examples e­ arlier in the chapter of vulnerable algae thriving in association with macrophytes that are defended from herbivores, and with which they would other­wise compete. Similarly, salt marsh plants survive and grow better with neighbors in areas with high desiccation or salt stress. To the extent that warming temperatures increase such stresses, we may expect many interactions to shift from competitive to facilitative with climate warming (Kordas et al. 2011). At a broad global scale, predation is strongly and positively related to environmental temperature in several marine systems, including fishes feeding on seagrass-­associated invertebrates (Reynolds et al. 2018) and sessile fouling communities (Freestone et al. 2011). But over the long term, warming-­induced intensification of consumer pressure may be mediated as much or more by changes in the composition of communities as by physiological changes at the organism level. In a survey across 105° of latitude, consumption of a standard bait by generalist marine consumers showed a hump-­shaped relationship with temperature, rising with temperature through most of the range, but declining again at the highest temperatures; predation intensity was best predicted by the taxonomic composition of a site’s fish fauna, which also varied consistently with temperature (Whalen et  al. 2020). Taxonomic composition of communities is changing as the ocean’s ­waters warm, and the changing distributions and abundances result in altered interactions. For example, fish herbivory has historically been restricted mainly to low latitudes (Longo et al. 2018), but herbivorous fishes are expanding poleward with warming seas, producing striking transitions from kelp forests to coral reefs in the western Pacific (see figure 3.19) (Vergés et al. 2014). Similarly, many marine diseases are increasing in prevalence and severity as climate warms (Harvell et  al. 2002).

Chapter 7 Species Interactions

Food web decapitation and trophic skew

Trophic level 4 3 2 1

Original trophic distribution

Recorded extinctions – 24%

193

Recorded invasions

The single strongest and most consistent impact on marine life – 29% – + + 67% over recent de­cades has been intensive ­human predation (chapter  4). Some estimates suggest that fishing has reduced the abundance of large marine fishes globally by upward of 80% Current distribution Projected distribution (after 5.1% turnover) (after 25% turnover) (Myers and Worm 2003), and fishing has profoundly trans– 14.0% – 65.1% formed marine ecosystems in a variety of ways ( Jackson et al. – 5.4% – 24.6% = 2001). Decline of predators results in trophic skew—­a charac+ 8.6% + 50.0% teristic flattening of the trophic pyramid, with fewer predators – 0.1% – 3.3% and more abundant organisms at lower trophic levels (Duffy 2003)—­and is exacerbated by nonnative invasions, which priFigure 7.12. ​Trophic skew in the Anthropocene ocean. The marily involve organisms lower in the food chain. Trophic skew changing shape of a coastal marine food web in the Wadden Sea, is increasingly a feature of coastal marine systems: 70% of docNetherlands, caused by extinctions of predators and invasions by animals at intermediate trophic levels (­after Byrnes et al. 2007). umented marine extinctions have involved top predators and carnivores, whereas nonnative invaders are primarily planktivores, deposit-­feeders, and detritivores (Byrnes et al. 2007) (figure 7.12). The widespread decline of large animals, particularly top predators, has reduced the strength of top-­down interactions in many ecosystems on both land and sea, with far-­reaching consequences for interactions, communities, and ecosystems (Estes et al. 2011).

­Future Directions The holy grail of community ecol­ogy is a mechanistic theory of species interactions, based on biophysical princi­ples and the effect and response traits of organisms, that can predict how communities assem­ble and change in response to environmental forcing. What are the characteristics of organisms and environments that influence the types and strengths of interaction among species? We have the beginning of a theory of interaction strengths based on organismal traits and environmental conditions. Regarding traits, we have seen that in most communities the distribution of interaction strengths is highly skewed, with one or a few strongly interacting species and many with l­ittle or no impact on the community. The strongest interactors are often keystone or foundation species. Interaction strength can be predicted to some degree by body size: larger organisms generally have larger per capita effects on other species (Woodward et al. 2005), larger fishes occupy higher trophic levels ( Jennings, Pinnegar, et al. 2002), and larger marine animals are more vulnerable to h­ uman predation (Payne et al. 2016). Vertebrates generally have larger per capita impacts in interaction webs than invertebrates, all ­else being equal (Borer et al. 2005). The distinction between vertebrates and invertebrates is not a trait per se, but this difference can presumably be traced to a suite of traits that includes greater sensory and cognitive capacity, visual acuity, and mobility. Homeothermic vertebrates tend to have larger per capita effects than poikilothermic vertebrates. Body velocity—­the capacity for rapid movement—is also generally positively related to interaction strength (Dell et  al. 2014). A frontier is extending such generalizations to other traits and quantifying them, which w ­ ill require careful field studies of individual species and interactions, experiments, modeling, and synthesis. Characteristic patterns in distributions of algal traits among habitats (Steneck and Dethier 1994) provide some guidance on promising places to explore such trait relationships. Environment strongly influences interaction strengths. We have well-­developed theory for the key environmental driver of temperature (Brown et al. 2004, Bruno et al. 2015). Warming temperatures usually accelerate herbivory more steeply than plant production, and experiments confirm that herbivore control of plant biomass tends to strengthen ­under warming in both benthic and pelagic

194

Ocean Ecology

systems (O’Connor 2009, O’Connor et al. 2009). ­There are intriguing hints that metabolic theory can help explain other community interactions, as in the rocky intertidal (see box 7.3). Experiments in a salt marsh showed that total metabolic biomass (mass0.75) of grazing snails, calculated from individual body size and density, predicted their effects on plant biomass, supporting theoretical predictions (Atkins et al. 2015). We also have the beginnings of a mechanistic theory for influence of nutrients on competition and dominance among marine primary producers in both pelagic systems (chapters 5, 10) and benthic systems (chapters 11, 12), which can inform management actions responding to eutrophication and climate-­induced changes in surface ocean nutrient concentrations. In contrast, responses of organisms and their interactions to ocean acidification appear more idiosyncratic based on current evidence (Nagelkerken et al. 2015), defining another impor­tant frontier.

Summary Organisms interact in three general ways: competition for resources in which at least one of the parties suffers reduced fitness, consumption of one species by another, and facilitation in which at least one party gains fitness and neither is harmed. Changing environmental conditions can shift the interaction between a pair of species among t­hese categories. B ­ ecause most species interact with many other species, direct interactions between two species often generate indirect effects on the fitness of ­others. The web of interactions arising from such linked interactions can be characterized by topology (link structure) and interaction strengths. Competition, consumption, and facilitation are all common in nature, they often interact to influence community structure, and they are influenced by both environmental d­ rivers and organismal traits. Foundation species create or modify the environment for other species with often far-­reaching consequences for community composition and ecosystem pro­cesses. Keystone species are defined as species that produce strong impacts that are disproportionate to their abundance. The functional traits of organisms are a bridge from environment to organism to population to community and ecosystem pro­cesses. Functional traits influence both what species do in ecosystems (effect traits) and how they respond to environmental change (response traits). Body size is a key trait—­both effect and response—­influencing interactions among species. Prey size scales roughly as a power function of predator size across a range of taxa, although vertebrate predators tend to take smaller prey for a given predator body size than do invertebrate predators. Vertebrate consumers have stronger impacts per body mass than invertebrate consumers, on average, and homeotherms are generally stronger interactors than poikilotherms. Herbivore feeding and impact on plant communities are strongly affected by plant stoichiometry, with more nutrient-­rich plants supporting stronger herbivory and greater rates of trophic transfer up the food chain and biogeochemical cycling. Consumer impacts are also strongly affected by prey chemical and physical defenses. Animal prey additionally rely on be­hav­ior to avoid predators, and predator-­ induced changes in prey be­hav­ior often have as strong or stronger effects as direct killing by predators on prey distribution and demographic rates. Experiments in rocky shore and salt marsh communities show that per capita interaction strengths, like metabolic rates, scale positively with body mass according to a power law. L ­ imited evidence similarly supports roughly equivalent population-­level interaction strengths across dif­fer­ent functional groups in rocky shore systems. Strengthening the mechanistic trait-­based science of species interactions is a frontier for ecol­ogy.

8

Ecological Communities

R

ocky shores around the world are home to many kinds of seaweeds and invertebrates that eke out a living in the space alternately submerged and exposed by the changing tides. ­These are among the best places to see a diverse range of sea life. Even a casual observer is struck by the more or less discrete horizontal bands of dif­fer­ent life forms on t­ hese rocks (figure 8.1). In the cooler latitudes, the bands often grade from blotches of blackish lichen at the highest reaches wet by sea spray, followed by a white band of barnacles, then a mottled zone of diverse invertebrates and algae that are more regularly submerged, and at the lowest level, a glimpse of the brown seaweed forest in the swirling w ­ aters below. The group of such species that occur together at a place is commonly referred to as a community. A community is described in terms of its structure, traditionally defined by the richness (number), composition (identities), and relative abundances of species found in it and also by its traits and phyloge­ne­tic relationships. The striking zonation, and accessibility, of rocky intertidal habitats have made them productive laboratories for understanding how communities of species assem­ble and interact, beginning with the first descriptions of their characteristic zonation in the early twentieth ­century (Baker 1909, Clapham 1926, Colman 1933). Puttering along the rocks and tide pools, many questions come to mind. What c­ auses the striking pattern of horizontal bands? Why do dif­fer­ent species live in the pools than on the exposed rocks a few centimeters away? Why are barnacles so thick on the upper rocks, but virtually absent farther down where conditions would seem to be more favorable? If y­ ou’ve been privileged to wander shores in other areas of the world, you might ask, Why ­don’t I see the big limpets in New ­England that are common on similar shores in Old ­England? Why do mussels form a black carpet on some shores but only tattered patches on ­others? And perhaps most puzzling of all, how can so many species live in a place where they seem to be competing for the same patch of real estate on the rocks? Such questions are at the heart of understanding ecological communities and the pro­cesses that produce their structure. We explore t­hese interactions in this chapter. Rocky shores illustrate a very general pattern in nature: within any given area live many kinds of organisms, typically hundreds or thousands of species. Even in an Arctic salt marsh, where a quick survey may reveal only a handful of species, careful observations uncover a multitude of worms, insects, meiofauna, protists, algae, bacteria, disease organisms, and parasites, as well as birds and mammals that appear only transiently. Why do so many species live in this place? And why is the number so much greater in tropical climes? Why, in other words, should a par­tic­ul­ ar area be home to 50 species rather than 5 or 5000? The pioneering ecologist Evelyn Hutchinson (1959) raised this question over half a c­ entury ago, thus articulating one of the central, enduring prob­lems in ecol­ogy. The answer is not at all obvious from first princi­ples. Attempts to explain the number and kinds of species living together in communities have generated much of the energy of ecol­ogy ever since (MacArthur 1972, Vellend 2016). More recently, a complementary question has attracted interest: What are all ­these species ­doing in the ecosystem—­that is, how do their interactions mediate biogeochemical

195

196

Ocean Ecology

Figure 8.1. ​Zonation of intertidal communities on a rocky shore: Calvert Island, British Columbia, Canada.

stocks and flows of energy and materials? The two questions—­why t­ here are so many species and what they do in ecosystems—­are obviously closely intertwined. But historically they have been studied by dif­fer­ent ­people in dif­fer­ent ways. This chapter focuses primarily on the first question about community structure, the pro­cesses controlling number and identity of species. We treat the second question about functioning mostly in the next chapter. We begin with the seemingly s­ imple question of what a community is, then turn to the pro­cesses that shape the numbers and kinds of species in communities, and the links between communities and ecosystems, which we continue in the next chapter.

What Is a Community? Understanding how and why certain species occur together starts from the observation that species interact with one another and with their environment in vari­ous ways (chapter 7). All species have certain tolerances that act as a filter on the kinds of places where they live—­fishes must live in w ­ ater and most plants require soil, for example. But their requirements also extend to other organisms. On the rocky shore, sea stars eat mussels, so they live where their prey live. Conversely, fast-­growing seaweeds, such as sea lettuce, are a favorite food of snails and cannot persist where snails are abundant. The early plant ecologist Frederic Clements introduced the concept of an “ecological community” to describe the group of species that typically occur together as a result of such interactions. Clements (1916) believed that most species interact intimately with one another such that a local group of species can be thought of as a “superorganism.” He focused on the succession of a community—­the sequence of species that colonize a place through time—­and compared it with the development of an individual

Chapter 8 Ecological Communities

organism, with the relatively stable climax community that emerges in time representing the superorganism’s adulthood. Not every­one agreed with him. His con­temporary Henry Gleason (1917, 1926) argued instead that a community is “individualistic,” meaning that species distribute along environmental gradients—­such as the gradient of exposure from low to high rocks on the shore—­ based on their own environmental tolerances rather than as coordinated groups of closely interacting species. Community ecol­ogy gained a theoretical foundation beginning with Robert MacArthur (1972) and a series of collaborators. Naturalists had long recognized the clear roles of history and biogeography in determining which species lived together in an area, but MacArthur saw a stronger role for local interactions among species, reasoning that competition and predation proceed more rapidly than regional dispersal and speciation. MacArthur’s ideas helped establish the foundations of modern community ecol­ogy and ­shaped the field for de­cades. But alongside the focus on local interactions that came to dominate ecol­ogy, there has always been a current of research emphasizing history and geography in structuring communities. On rocky shores, for example, the Pleistocene glaciers extended farther south on the western side of the North Atlantic than on the eastern side, and ice-­ free refugia persisted throughout the ice ages in northern Eu­rope, helping explain why the coasts of northern Eu­rope have more marine species than ­those of Atlantic North Amer­i­ca, and why certain functional types are altogether absent in the west, such as large grazing gastropods in the intertidal. Such species may well be capable of living in the Northwest Atlantic, but 12,000 years ­after the ice retreated they still have not returned. In the 1980s a renewed appreciation for insights from paleontology, phyloge­ne­tics, and biogeography led to the conclusion that—­contrary to MacArthur’s view—­these historical influences are often stronger than local interactions in shaping the species richness of communities (Ricklefs 1987, Ricklefs and Schluter 1993). An extreme version of this view suggested that “the distributions of species within a region reveal more about the pro­cesses that generate diversity patterns than does the co-­occurrence of species at any given point. The local community is an epiphenomenon that has relatively ­little explanatory power in ecol­ogy and evolutionary biology” (Ricklefs 2008). Data from a range of systems support Ricklefs’ first point that species diversity within a community is strongly influenced by history and interactions with communities in the surrounding region, as we discuss below. Nevertheless, it’s clear from de­cades of experiments that the par­tic­u­lar species pre­sent in a community and their relative abundances are often strongly influenced by local conditions and interactions between competitors, predators, and prey (Bertness, Bruno, et al. 2014). This local environmental control is reflected in the predictable composition of benthic communities in par­tic­u­lar habitats, often dominated by members of the same ­family or genus in similar habitats of widely distant regions (Petersen 1918, Longhurst 2007). So both evolutionary history and local conditions strongly shape community structure. Indeed, the pro­cesses contributing to local community structure are often so complex that some have despaired at finding useful generalizations in community ecol­ogy (Lawton 1999). From its beginning, then, community ecol­ogy has wrestled with an existential question: Are communities real? In other words, is a community more than a transient group of species that have similar environmental requirements and therefore end up in the same place? Or is community membership dictated by strong, coevolved interactions with other species? The answer is somewhere in the m ­ iddle. Communities are mosaics. Our challenge is to understand which pro­cesses are most impor­tant in par­tic­u­lar communities and environments. T ­ oday the most commonly accepted definition of a community tends ­toward the individualistic concept: a group of species populations that co-­occur in space, potentially interacting with one another. A strict definition of community includes all organisms that co-­occur within some area since any organism might interact with a wide range of other plants, animals, and microbes in its neighborhood. In practice, community research often focuses more narrowly, for example, on fishes or benthic invertebrates or seaweeds alone. Such a taxonomically circumscribed entity is often referred to as a community, but purists prefer the term assemblage. The

197

198

Ocean Ecology

simplest definition of a local community is the set of species whose distributions include a par­tic­u­lar point in space and time (Ricklefs 2008). This concept focuses attention on understanding communities as populations of species interacting on regional scales. Hutchinson (1961) famously articulated another question that remains central to understanding ecological communities in what he termed the paradox of the plankton: “The prob­lem that is presented by the phytoplankton is essentially how it is pos­si­ble for a number of species to coexist in a relatively isotropic or unstructured environment all competing for the same sorts of materials.” In other words, why does one of t­ hose species not eventually exclude the dozens or even hundreds of ­others? Hutchinson proposed two pos­si­ble solutions, suggesting that phytoplankton communities rarely reach competitive exclusion ­because ­either (1) the rate of competitive exclusion is slower than the rate of seasonal environmental change, which disrupts it, or (2) small-­scale variation in ­water properties ­favors dif­fer­ent species in dif­fer­ent microhabitats ­because of subtle niche differences. Hutchinson’s essay implanted the concept of disequilibrium firmly in the soil of community ecol­ogy, which previously had been focused mainly on deterministic pro­cesses, and contrasted it with niche differentiation. We see support for both of Hutchinson’s hypotheses across a spectrum of marine communities. But we also know now that competition and other local interactions among species and environment are only part of the equation that determines species diversity.

Community Dynamics: A Conceptual Framework The species composition of a community is molded by numerous pro­cesses across scales of time and space, which can be distilled into four general classes (Ricklefs 1987, Hubbell 2001, Vellend 2010). Species are added to a community through (1) speciation, the evolutionary origin of new species over many generations, and (2) dispersal, that is, the movement of organisms among sites and regions. ­W hether the species that arrive at a site persist ­there is molded by (3) ecological drift—­random fluctuations in species abundance—­and continuing dispersal, and fi­nally by (4) ecological se­lection among species, that is, environmental tolerances and interactions among species that predictably ­favor certain species ­under par­tic­u­lar conditions (see figure 3.6). It is the last of ­these, ecological se­lection, that has dominated research in ecol­ogy over the years (chapter 7). But a comprehensive understanding of ecological communities requires integrating all four classes of pro­cesses. Returning to our rocky shore in New ­England, USA, the species we see are a subset of ­those that ­were born (speciated) and evolved over millions of years in the North Atlantic, along with some that evolved in the Pacific and dispersed across the Arctic Ocean during a period of milder climate. The shore is missing some species formerly h­ ere but driven extinct during the ice ages. Crouching down to examine a par­tic­u­lar patch of rock, we see patterns produced by more local, deterministic interactions (ecological se­lection): certain seaweeds are absent from tide pools with sea urchins, but flourish in pools that urchins have not reached. Newly settled barnacles are common low on the shore, but few adults are found ­there ­because they get overgrown by mussels. Across the local seascape, some outcrops have abundant sea stars and some very few, the result of random variation in recruitment (ecological drift), which cascades to affect abundances of their prey. Of ­these four classes of pro­cesses structuring communities, we considered speciation and dispersal in chapter  3. ­These two classes form regional biotas over long time spans and large spatial scales and are often referred to as “historical” pro­cesses to distinguish them from the con­temporary interactions acting on scales of one or a few generations within organismal neighborhoods (Ricklefs 1987). Speciation and long-­range dispersal have traditionally been the province of evolutionary biology and biogeography. Th ­ ese long-­term historical pro­cesses are key to understanding community interactions at local scales ­because they create the collection of species in a biogeographic region, say, an ocean basin or large island group, referred to as the regional species pool, which are available to colonize local communities. The group of species that interact with one another at the local scale, the

Chapter 8 Ecological Communities

community, is drawn from that regional pool. Thus, the regional pool sets an upper limit on the number of species found in a local community. Similarly, regional richness is defined as the sum of all species recorded from a region (say, the Gulf of Maine ecoregion of the cold-­temperate Northwest Atlantic province), whereas local richness mea­sures the number of species that co-­occur in a small enough area that they could potentially interact (say, a single rocky outcrop). In this chapter we consider how dispersal, ecological drift, and ecological se­lection act on the regional species pool to mold species composition and abundances in local communities. We focus ­here on general princi­ples, and in ­later chapters explore how they interact to produce structure in par­tic­u­lar communities.

Ecological se­lection Ecological se­lection is a general term for the deterministic interactions between organisms and their environment that influence which species are pre­sent in a community. Most of the questions and pro­cesses that have historically occupied community ecol­ogy, discussed in the last chapter, fall ­under the category of ecological se­lection, which results from the physiological tolerances that define the fundamental niche, and the interactions with other organisms that condense it to the realized niche (figure 8.2). Grouping t­ hese pro­cesses ­under the umbrella of ecological se­lection emphasizes their commonalities: they are deterministic and mediated by traits of organisms, as distinct from the random effects of ecological drift, and the much longer-­term and less predictable effects of speciation and long-­range dispersal. Natu­ral se­lection is familiar as the differential survival and reproduction (i.e., fitness) of genet­ically distinct individuals that results in changing ge­ne­tic composition of a population (chapter 6). Ecological se­lection is an analogous but faster pro­cess that happens in communities via the differential local survival and spread of species as a function of their fitness in that environment, which results in changing species composition of a community. If we understood the action of ecological se­lection pro­cesses in shaping communities well enough, they might be summarized as a set of community assembly rules (Diamond 1975) that predict patterns of species co-­occurrence based on hypothesized interactions with the environment and one another. Much research has worked to discover such assembly rules, initially focusing on competition among animal species (Diamond 1975) and subsequently on the role of abiotic environmental ­drivers in plants (Keddy 1992). Community assembly is often conceptualized as a series of filters (figure 8.3) (Cadotte and Tucker 2017). In the first stage, propagules arrive at a site or not based on

(A)

(B)

Resource enhancement

Competition Realized niche Recruitment limitation

Predation

Realized niche Predation refuge

Recruitment enhancement

Fundamental niche

Fundamental niche

Disease and parasitism Habitat amelioration

Figure 8.2. ​The fundamental and realized niches as they are s­ haped by species interactions. (A) In the face of negative interactions, such as competition and predation, the realized niche (green) is compressed to a subset of the fundamental niche (dashed line). (B) When positive effects of facilitation dominate, they can expand the range of conditions that meet the requirements of the fundamental niche (­after Bruno et al. 2003).

199

200

Ocean Ecology

Dispersal filter

Environmental filter

Interaction filter

Figure 8.3. ​Community assembly visualized as a sequence of filters. The filters are proposed to successively reduce the pool of species in the region to the set that actually occurs at a given site as a result of successful dispersal, tolerance of local environmental conditions, and per­sis­ tence through interactions with other species (­after Cadotte and Tucker 2017).

a dispersal filter, that is, w ­ hether they are capable of reaching the site from a source population. Second, they recruit successfully or not based on their tolerance of local abiotic conditions. For example, a barnacle larva can s­ ettle on a rock outcrop but not on a sand bottom. This stage is often referred to as environmental filtering b­ ecause it sorts species into environments based on the match between their physiological tolerances and local abiotic conditions. Third, once past this abiotic environmental filter, ­those species that successfully recruited then interact with the species already t­ here. This interaction filter may exclude the recruiting species from an other­wise favorable site via competition, predation, disease, and so on. For example, the barnacle may be overgrown by macroalgae and eventually smothered. Conversely, biotic interactions may facilitate establishment of species by making the environment more hospitable, for example, by providing shelter or reducing stress or consumer pressure (Bruno et al. 2003) (see figure 8.2). The outcome of ­these biological interactions in community assembly is often attributed to niche differentiation b­ ecause the ability of species to coexist is thought to be promoted by differences in their use of resources, habitat, and so on—­that is, in their niches. This concept of a sequential “stack of filters” is attractive but somewhat misleading, and difficult to test in practice, b­ ecause abiotic conditions and biological interactions happen si­mul­ta­neously throughout community assembly and it is difficult to deduce their respective roles from observational evidence alone (Kraft et al. 2015). The effects of speciation and long-­range dispersal on a community result from large-­scale pro­cesses mostly extrinsic to the organism, namely, geography and the region’s history of speciation and extinction. In contrast, ecological se­lection is driven by intrinsic characteristics of species—­traits—­related to use of resources, competitive ability, tolerance of stress, vulnerability to predators, and thus to realized demographic rates. In other words, ecological se­lection results from niche differences. Environmental filtering and niche partitioning happen si­mul­ta­neously as species colonize and establish in a community. Below we review some of the main mechanisms of ecological se­lection that assem­ble communities.

Dispersal and metacommunities

All organisms disperse through space at some stage in their life history (chapter 6). Dispersal shapes community structure in two main ways. First, it links local communities to the broader, regional species pool. Second, dispersal from the broader region tends to counteract the drift t­oward extinction of species within a local community as they are affected by weather, disease, and demographic stochasticity. In both ways, dispersal generally increases local community diversity (Hubbell 2001). Lifetimes of most species are mea­sured in millions of years, during which environments change, sea levels rise and fall, and continents shift in position (chapter  3). ­These events foster evolution and movement, and most species come to occupy geographic ranges much larger than the area we associate with a local community. Thus, local communities are connected to larger, regional metapopulations of their component species.

Chapter 8 Ecological Communities

Species also move in and out of the community within the lifetimes of individual organisms, especially among marine organisms with pelagic larvae that disperse over large distances. Dispersal ability influences a species’ population dynamics, distribution, and responses to disturbance, among other pro­cesses (chapter  6). For ­these reasons, dispersal can profoundly affect local community structure (Kinlan and Gaines 2003). Adult demography and recruitment are decoupled in many benthic invertebrates, whose pelagic larvae can mix over substantial areas of the regional ocean (figure 6.12). With re­spect to community diversity, a key consequence of this rain of larvae from distant sites is that competitively inferior species have a new chance to hold on in ­every generation, even when benthic adults compete strongly, such that complete competitive exclusion is rarely achieved. Such links emphasize that local communities are connected via dispersal to other communities within a regional matrix. This connected network of communities is called a metacommunity (Leibold et al. 2004). Dispersal among patches of habitat within a metacommunity affects interactions within local communities, influencing their dynamics. Perhaps most importantly, dispersal among communities often results in a local community containing certain species not ­because they are well adapted to local conditions but ­because they regularly arrive via dispersal from larger populations nearby. The contributing and receiving communities are referred to as sources and sinks, respectively. The importance of ­these source-­sink dynamics for communities depends on dispersal rate. When dispersal among communities is low, its influence is primarily on community assembly, that is, in providing propagules that influence which species are pre­sent in the community (i.e., presence versus absence), with ­little effect on demographic rates or interactions. Conversely, when dispersal rate is high enough to affect species demographics in most generations, dispersal becomes impor­tant in influencing how species interact and in their relative abundances. U ­ nder ­these circumstances, dispersal can prevent extinction of populations that are declining or poorly suited to local conditions, a phenomenon known as the rescue effect (Brown and Kodric-­Brown 1977). For example, Scopoli’s shearwater (Calonectris diomedea) is a seabird that nests in burrows on Mediterranean islands and, like many seabirds, is frequently killed as bycatch in longline fisheries. Intensive monitoring of nests and individuals showed that western Mediterranean populations of this bird, although apparently stable, had low numbers of breeding adults and w ­ ere maintained (rescued) by an influx of immigrants from other populations (Sanz-­Aguilar et al. 2016). This finding has impor­tant implications for conservation in showing that such populations are in jeopardy and in focusing attention on the critical importance of understanding demographics of the source populations. In summary, dispersal can strongly influence local community structure by favoring species whose dispersal biology is well suited to local habitat distribution, and by delivering recruits that rescue declining populations from extinction. The dispersal biology of species in a regional pool interacts with seascape structure—­the kinds and arrangement of habitat patches—to shape the composition and richness of local communities. Where species have very similar niches, variation in dispersal capacity among them can have an overriding influence on local community structure (box 8.1).

Ecological drift Within populations, gene frequencies change more or less continuously due in part to inherent unpredictability in biology (e.g., partly random segregation of chromosomes to gametes) but also to environmental disturbances, such as severe weather, that kill individuals at random. The resultant random changes in gene frequency are known as ge­ne­tic drift and are recognized as a key pro­cess in population ge­ne­tics, especially in small populations where death or reproduction of a given individual has a disproportionately large effect on the population. An analogous pro­cess happens among species in local communities: Stephen Hubbell (2001) coined the term ecological drift to describe the random changes in

201

202

Ocean Ecology

Box 8.1. ​Dispersal and diversity in metacommunities The spectacular diversity of fishes on coral reefs is legendary, and many reef fish communities include numerous closely related species that appear ecologically equivalent—­eating much the same food and living in the same habitats. Butterflyfishes, for example, are beautiful coral-­feeders that are highly diverse on Indo-­Pacific reefs, and careful study has found that most species overlap substantially in diet and microhabitat. At Lizard Island on the G ­ reat Barrier Reef, for example, 11 of 14 butterflyfish species feed primarily on one of two corals, Acropora hyacinthus and Pocillopora damicornis (Pratchett 2005). ­These fishes thus pose a similar challenge to Hutchinson’s paradox of the plankton: How can so many species coexist while ­doing basically the same t­ hing? This puzzle motivated one of the earliest versions of neutral theory in ecol­ogy, the lottery hypothesis (Sale 1977), which proposed that living space is severely ­limited on reefs but that its unpredictability in time and space make interspecific competition for space a game of chance with no consistent winner. Studies on the G ­ reat Barrier Reef now show that differences among species in dispersal ability can allow them to coexist even with identical food and habitat

(A)

requirements. First, theory shows that, even among ecologically equivalent species, differences in dispersal distance can promote coexistence when habitat patches are distributed irregularly across the seascape. Specifically, modeled species that disperse well over short distances do well where habitat patches are dense and close together, whereas species with longer-­ distance dispersal do better with more widely spaced patches (Bode et al. 2011). The model found that species with shorter versus longer dispersal can coexist in a heterogeneous seascape ­because each is favored at a dif­fer­ent interpatch distance. And more species can coexist if dispersal patterns vary through time, as is typical in nature. When dynamics of fish assemblages ­were simulated for multiple generations across a real seascape of the ­Great Barrier Reef, the metacommunity stabilized with 27 of the original 500 fish species remaining. Each species dominated t­ hose clusters of reefs that w ­ ere spaced close to their optimal dispersal distance (figure B8.1.1). This study showed that variation in dispersal capacity alone can maintain coexistence of species even when they have identical food and microhabitat requirements.

(B)

17°S

18°S

146.5°E

147.5°E

Figure B8.1.1. Ecologically equivalent reef fish species can coexist across the seascape if they differ in dispersal distance. (A) Three of many butterflyfish species commonly co-­occurring on the ­Great Barrier Reef. (B) Results of simulation models applied to a patchy seascape on the G ­ reat Barrier Reef, the colors indicating the species that dominated each patch a­ fter 500 years when species ­were ecologically identical except that their dispersal abilities varied as a function of environment and time (­after Bode et al. 2011). (Photo credits, top to bottom: Richard Ling, Leonard Low, Rick Stuart-­Smith)

Chapter 8 Ecological Communities

relative abundances of species that happen in communities through time (Vellend 2010). Where ecological se­lection is relatively weak and species populations are small, ecological drift can be as strong as or stronger than the deterministic changes caused by competition and predation, and ecological drift thus plays an impor­tant role in community composition (Mertz et al. 1976, Siepielski et al. 2010). Indeed, Hubbell’s neutral model showed that, in the context of a regional metacommunity connected by dispersal, such ecologically equivalent species can coexist for hundreds to thousands of generations before one drifts to extinction. Historically, stochastic pro­cesses of ecological drift have received less attention than deterministic pro­cesses of ecological se­lection, such as competition and predation. In one sense, ecological drift should be our starting point for considering what controls species diversity in communities since it provides a null or neutral model against which we can evaluate other pro­cesses (Hubbell 2001). But ecological drift is more intuitive ­after we understand ­those other pro­cesses. Neutral pro­cesses are especially impor­tant in testing models of diversity maintenance in ecological communities.

Synthesis: Diversity in Ecological Communities Diversity is the most striking characteristic of living nature, and attempts to explain it have generated much of the energy and effort in ecol­ogy since the field’s birth (MacArthur 1972, Hubbell 2001, Vellend 2016). Some of the theories and concepts include dynamic equilibrium, the intermediate disturbance hypothesis, predator-­mediated arrest of competitive exclusion, and the list goes on (Vellend 2010). Yet this list, for the most part, does not even include neutral and historical pro­cesses. The latter provide a logical starting point for evaluating the mechanisms that generate and maintain diversity in ecological communities.

Neutral models and their assumptions How do we know which pro­cesses are most impor­tant in influencing community composition and diversity? To identify the candidates, a logical step is to begin with a null model that omits biological differences among species that drive deterministic interactions and to explore how patterns observed in nature deviate from the null expectation. ­These deviations can provide clues to what pro­ cesses are influencing community structure in nature. Null models have a long history in ecol­ogy (Harvey et al. 1983). Most theories of community diversity begin by considering the simplified situation of competitive communities, that is, groups of species that compete for one or more limiting resources, ignoring trophic interactions. A general assumption of such models is zero-­sum dynamics; that is, at a given level of resources, a finite number of individuals can be supported in the community. This is a reasonable assumption for rocky shore organisms or terrestrial plants competing for space. The next simplification is to ignore all differences among species that lead to ecological se­lection, essentially treating all individuals as identical regardless of species. This assumption is clearly false but allows us to focus on how local communities would assem­ble from the regional pool if affected only by dispersal and drift. Two such neutral theories have been especially influential: the theory of island biogeography (MacArthur and Wilson 1967), which in its ­later development also included speciation, and Hubbell’s (2001) unified neutral theory of biodiversity and biogeography, which includes speciation, dispersal, and drift, but not se­lection. ­These models explore how dispersal and stochastic pro­cesses (ecological drift) combine with competition to shape or limit local community diversity. Both Hubbell’s neutral theory and the island biogeography theory include spatial structure implicitly, in the sense that a local community depends on a supply of species from the larger regional pool. Island biogeography theory is more explic­itly spatial in that it incorporates the distance between habitat patches and their sizes.

203

204

Ocean Ecology

The unified neutral theory of biodiversity The most comprehensive neutral model in ecol­ogy is the unified neutral theory of biodiversity (Hubbell 2001), which predicts the diversity and structure of metacommunities by incorporating all the nonselective pro­cesses structuring communities: speciation, dispersal, and ecological drift. Hubbell’s interest grew out of the paradox of hyperdiverse tropical forests, which can contain hundreds of tree species, seemingly coexisting on the same limiting resources of w ­ ater, light, and a small number of mineral nutrients—­the same general prob­lem as Hutchinson’s paradox of the plankton. Hubbell was led to consider the role of random pro­cesses in maintaining diversity a­ fter intensive studies of a tropical forest in Costa Rica produced ­little evidence supporting any of the several theories then circulating to explain tropical forest diversity (Hubbell 1979), and he l­ater asked similar questions about coral reefs as well (Hubbell 1997). Hubbell’s theory is neutral in that it treats all species in the community as identical in their per capita probabilities of reproducing, ­dying, dispersing, and speciating. The habitat patches are also identical in the basic formulation of the model and together comprise a metacommunity. Such a null model is useful in formalizing what is expected in the absence of the niche differences of primary interest, which allows estimation of their importance. Hubbell’s model simulated the dynamics of a large population of individual organisms that speciate, disperse among local communities (i.e., habitat patches), and drift to dominance or extinction over large numbers of generations. Thus, it encompasses all populations of the species throughout their geographic ranges, as also emphasized by Ricklefs (1987). This is accomplished with a small number of model par­ameters: the total population size of individuals summed across all species ( J); the per individual rate of speciation (ν); the migration rate among habitat patches (m); and the rather mystically named “fundamental biodiversity number” (θ) equal to 2Jν. Mathematically, the fundamental biodiversity number determines both species richness and relative species abundances in the metacommunity at equilibrium between speciation and extinction. The fundamental biodiversity number increases with both the total number of individuals in the metacommunity ( J) and the speciation rate (ν). Its effect on the distribution of abundances is less straightforward and depends on the arrangement of habitat patches. Hubbell’s neutral theory builds on the ­earlier theory of island biogeography. Both theories predict that the distributions and abundances of species are determined by a dynamic equilibrium between speciation, dispersal, and extinction. The main difference is that island biogeography takes the regional species pool as given, that is, as external to the model, whereas the unified neutral theory closes the loop by including speciation rate, with the aim of predicting species richness from global through regional to local scales.

Testing the neutral theory in nature A remarkable feature of the neutral theory is that it produced patterns of species richness and abundance strikingly similar to t­ hose found in empirical data on forest trees despite considering all species as ecologically equivalent, which is clearly not true (Hubbell 2001). Hubbell concluded that both spatial variation in species richness and relative abundances of species within local communities can be predicted with a small number of par­ameters that include no niche differences. How can this be? Obviously, species are not identical. They vary wildly in body size, resource requirements, demographic par­ameters, dispersal rates, and even speciation rates. But note that the key assumption of the neutral theory is not that species are generically identical but that they are identical in net demographic rate, that is, the product of survival and reproduction. This is a more subtle statement and less easy to refute, although it clearly does not apply universally. Could the theory be right for the wrong reasons? Hubbell’s results stimulated a vigorous research effort aimed at answering ­these questions. One of the first and most rigorous empirical tests of neutral theory came, predictably, from rocky inter-

Chapter 8 Ecological Communities

tidal communities. Tim Wootton (2005) emphasized an impor­tant challenge to testing the theory, which is that most of its basic par­ameters are difficult or impossible to estimate empirically, including regional population size and rates of speciation, migration, and mortality. To get around this and other constraints, Wootton reformulated the model equations in terms of units of resources rather than units of individuals, and then estimated the par­ameters using field data from communities of sessile invertebrates and seaweeds in the rocky intertidal of Washington State, USA (box 8.2). At the local scale of interacting organisms, the patterns of species composition on the shore and their responses to disturbance ­were strongly deterministic, clearly refuting the neutral theory for this community. Rocky intertidal communities are legendary in ecol­ogy precisely ­because they demonstrate so clearly the importance of deterministic interactions in molding community structure. What about coral reefs, the pinnacle of marine diversity? One of the pioneering concepts of neutral dynamics in ecol­ogy was the lottery hypothesis, which explained the coexistence, but seemingly random dynamics, of diverse assemblages of ecologically similar coral reef fishes (Sale 1977). The pattern is similar for corals. On the G ­ reat Barrier Reef, for example, the coral genus Acropora has dozens of species, many of which can be found on the same reef. Maria Dornelas and colleagues (2006) used an extensive data set of Indo-­Pacific coral assemblages to test predictions of neutral theory. Starting with observed distributions of coral species, they shuffled them to simulate neutrally assembled communities and found that both relative abundances of species and similarities in species composition among observed coral assemblages differed strongly from neutral predictions (figure B8.2.1E, F). The coral assemblages ­were more divergent than predicted by neutral theory, suggesting that, on t­ hese coral reefs, diversity is strongly s­ haped by variance in local environment and interactions. The failure of neutral theory in ­these examples does not invalidate it completely, of course. Many communities contain numerous coexisting cryptic species that are difficult to distinguish ecologically and often taxonomically. In many cases, ­these species apparently are nearly equivalent demographically (Shmida and Wilson 1985, Leibold and McPeek 2006). It is difficult to demonstrate conclusively the absence of impor­tant niche differences among species since niches are complex and ­there is always the possibility that we have missed some impor­tant variable. But even in well-­studied rocky shore communities, not all interactions are as clear and deterministic as ­those of the classical experiments described ­earlier. This is clear from the intensive search for niche differences among two coexisting barnacles on Chilean rocky shores. Experiments ­were designed explic­itly to distinguish between two types of coexistence (Chesson 2000): niche differentiation and ecological equivalence. The key to stable coexistence of two competing species is that intraspecific competition should be stronger than interspecific competition. First, experiments demonstrated that the two species, Jehlius cirratus and Notochthamalus scabrosus, in fact coexisted, showing that both recovered ­after experimental reduction—­neither species could exclude the other (Shinen and Navarrete 2014). But no trade-­offs between intra-­and interspecific effects could be found at any tidal level. In the end, the authors concluded that the distributions of t­ hese two barnacles ­were best described by neutral dynamics of species that are essentially equivalent demographically, consistent with the neutral theory. One impor­tant prediction of neutral theory was that modest dispersal rates among local communities result in regional population sizes so big that such ecologically equivalent species take very long times to drift to extinction. The Chilean barnacles thus may represent an example of such unstable (Chesson 2000) but highly per­sis­tent coexistence. Such findings emphasize that the strong deterministic interactions familiar from classical studies of rocky shores are not the only pro­cesses at work ­there. Understanding t­ hese communities must also incorporate interactions with the regional species pool—­the metacommunity—as well as the longer-­term evolutionary pro­cesses that produce the pool. Indeed, local determinism and neutral dynamics may operate side by side, affecting dif­fer­ent species within the same community (Leibold and McPeek 2006).

205

206

Ocean Ecology

Box 8.2. ​Testing neutral theory in marine communities Rocky shore communities offer a promising system for testing the neutral theory of biodiversity ­because space on the rocks is clearly the limiting resource, it can be easily mea­sured and manipulated, and the rocky outcrops are connected through pelagic larval dispersal as the theory assumes. Tim Wootton (2005) was motivated to test the theory ­because, as in Hubbell’s forests, the rocky shores of Tatoosh Island in Washington State, USA, showed a close correspondence between observed patterns of rank abundance among species in the field and ­those predicted by the neutral theory (figure B8.2.1A). Does this mean the neutral theory is correct? Wootton reasoned that if community dynamics are truly neutral as the theory proposes, this should be evident in the dynamics of individual species in the community, not just their total richness and the rank-­abundance curve. To test ­these more specific predictions, he fitted the neutral theory equations to data from experimental field plots from which the competitively dominant mussel Mytilus californianus had been continuously removed for 7–11 years. In t­ hese field plots, the rank-­abundance curve fit neutral model predictions, like the forests originally used to test the theory. But ­because data ­were also available for individual species on the rocky shore, Wootton was able to compare their predicted and observed trajectories. ­These differed

(C) 30 Frequency (number of sites)

2 1 0 0

5

20 10 0

10 15 20 25 30 Species rank

Recruits

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Neutral theory prediction

100 0

0.5 Community similarity

1.0

0

0.5 Community similarity

1.0

(F) 30

Frequency (number of sites)

0.7 0.6 0.5 0.4 0.3 0.2 0.1 0

200

0

60 70 80 90 Community similarity

(D)

(B)

300

Adults

20 10 0

300 Frequency

Log abundance

3

–1

(E)

Frequency

(A)

Observed proportional abundance

substantially (figure B8.2.1B), refuting the neutral model predictions. This is ­because abundances of species in this system are strongly affected by competition and predation, as documented by de­cades of experiments and the success of deterministic models at reproducing community dynamics and composition (Wootton 2005). ­These results highlight a key theme in ecol­ogy, and science more generally: patterns in nature can often be produced by any number of interacting pro­cesses, so a ­simple correspondence between pattern and prediction is a weak test of theory. Support for deterministic control of rocky shore communities also comes from a survey along 800 km of the coastline of Chile. Andrés Caro and colleagues (Caro et al. 2010) found strong spatial variation in recruitment of barnacles among 15 sites, which neutral theory predicts should lead to corresponding ecological drift in adult abundance among communities. Yet the communities that developed from ­those recruits converged to a more uniform range of compositions, comprising only a small subset of the pos­si­ble communities expected ­under a hypothesis of passive ecological drift from the recruiting populations (figure B8.2.1C, D). T ­ hese results from two distant regions confirm that rocky shore community structure is generally deterministic, bearing a strong imprint of ecological se­lection pro­cesses, including competition, predation, and facilitation.

60 70 80 90 Community similarity

200 100 0

Figure B8.2.1. Tests of neutral theory in marine communities. In a NE Pacific rocky intertidal community, (A) observed species rank-­abundance curves are similar to the lognormal prediction of neutral models, but (B) species responses to disturbance depart strongly from neutral predictions (­after Wootton 2005). (C, D) In rocky shore communities of Chile, adult communities are much more similar in composition than expected from neutral expectations based on variance among sites in recruitment of their larvae (­after Caro et al. 2010). In contrast to the Chilean example, (E) based on simulated random dispersal of propagules, (F) observed coral communities on the ­Great Barrier Reef are much less similar than expected (­after Dornelas et al. 2006).

Chapter 8 Ecological Communities

Disturbance and diversity in communities The world is an unpredictable and dangerous place for many organisms. On top of the partly predictable seasonal changes in weather, resources, and predation, a host of seemingly random disturbances—­ storms, heat waves, volcanic eruptions—­are always on the horizon. Hutchinson (1961) introduced the concept of disequilibrium in community ecol­ogy, that is, the species composition of communities may not be entirely deterministic but may result in part from disturbances or environmental change. An impor­tant consequence of such disturbances is interruption of competitive exclusion, which tends to maintain higher diversity. Any number of events might derail competitive exclusion—­a pulse of new recruits or of resources, mortality caused by a storm, or a change in the environment that ­favors one or another species. All environments change, albeit some more than o­ thers—­weather varies by the hour, seasons come and go, migratory species pass through and move on, lightning strikes, and so on. When such disturbances prevent one species from monopolizing resources, they can maintain diversity in a community. Key questions in evaluating the importance of disturbance to communities are how rapidly competition leads to exclusion, and w ­ hether conditions remain stable long enough for competitive exclusion to play out before being interrupted. The most influential idea for nonequilibrium maintenance of species diversity was Joseph Connell’s (1978) intermediate disturbance hypothesis, which predicted a hump-­shaped relationship between frequency of disturbance and species richness. The argument defines three components of the relationship between disturbance and diversity: (1) in the absence of disturbance, competitive exclusion runs its course, eliminating inferior competitors and reducing diversity; (2) as disturbance frequency or intensity increases, it interrupts competitive exclusion and allows inferior competitors to persist; (3) fi­nally, when disturbances become very frequent, species with slowly growing populations c­ an’t recover and they decline to extinction. Thus, diversity is predicted to be maximal at intermediate disturbance. Long-­term data from coral reefs match the pattern predicted by the intermediate disturbance hypothesis (Connell 1978) (figure 8.4). The dynamic equilibrium model (Huston 1979) formalized and expanded the proposed relationship between competition and disturbance in regulating species diversity, proposing that diversity is ­shaped by two opposing forces: (1) competitive displacement, considered proportional to population growth rate, and (2) mortality-­causing disturbances, which prevent competitive exclusion when growth rate is high and cause local extinction of populations whose growth rate is very slow (see figure 8.4). The intermediate disturbance hypothesis can be thought of as considering a cross-­section of the dynamic equilibrium model at a consistent level of productivity. Application of the dynamic equilibrium model proposed that the high diversity characteristic of the open pelagic ocean and deep-­sea floor is a product of nonequilibrium pro­cesses (harking back to Hutchinson’s explanation for the paradox of the plankton) in low-­productivity environments where competitive exclusion proceeds very slowly (Huston 1994). Although the intermediate disturbance hypothesis is consistent with patterns in both terrestrial and marine benthic communities (Connell 1978) and is supported by lab experiments with freshwater phytoplankton (Sommer 1995), the hypothesis has decidedly mixed support empirically (A. R. Hughes et al. 2007) and is problematic theoretically as well b­ ecause disturbances reduce not only abundances but also the strength of competition needed for exclusion (Fox 2013).

The role of history Populations adapt to the environment via evolutionary (ge­ne­tic) change and form communities as a result of shared environmental tolerances and interactions. All of t­ hese pro­cesses take time. Events that interrupt or deflect them can leave a lasting impact. Such legacies of history can be seen at a wide

207

Ocean Ecology

(C)

(A)

Low Disturbances frequent Soon after a disturbance Disturbance large

Infrequent Long after Small

(B)

Population size

Number of species

High

Number of species

208

24 18 12 6 10

30 50 70 % Cover of live coral

90

Time

Figure 8.4. ​The intermediate disturbance hypothesis. (A) Schematic view of the proposed intermediate disturbance hypothesis and (B) observational evidence from coral reefs offered in its support (­after Connell 1978). (C) Model simulations supporting the dynamic equilibrium model of species richness: (top) in the absence of disturbance, fast-­growing species initially dominate but decline as they are outcompeted; (­middle) at intermediate disturbance frequencies, both fast-­growing species and slower-­growing strong competitors sustain populations through the end of the time series; (bottom) with frequent disturbance, only fast-­growing species can keep up (­after Huston 1979).

range of time and space scales—­from small, local disturbances, such as an anomalous patch of algae where waves have ripped mussels from a rocky shore, through the catastrophic asteroid impact that ended the dinosaurs and radically changed the character of the terrestrial fauna forever. In ecological time spans of a few generations of the dominant organisms, history manifests in two general ways. First, as ­we’ve seen, an event may delay a system’s approach to equilibrium and, if such events happen frequently enough, maintain the system in a dif­fer­ent state. This is the basis of the intermediate disturbance hypothesis. Second, a system may be highly sensitive to starting conditions, in which case an event can deflect it ­toward one of two or more alternative stable states, meaning combinations of species and pro­cesses that are distinct from one another and that resist change despite starting with similar abiotic conditions (chapter  9). This phenomenon has fascinated ecologists for de­cades (Sutherland 1974) and is increasingly impor­tant in the context of multiple h­ uman stressors on ecosystems, but the evidence remains challenging to interpret (Scheffer and Carpenter 2003). Over the longer span of evolutionary time, history manifests as evolution in the species richness and trait composition of a regional biota, from which local communities assem­ble (chapter 3). This regional evolution is partly deterministic, driven by habitat area and abiotic conditions that affect speciation and extinction rates. But it is also affected by legacies of history that can persist for millions of years. For example, coral reef diversity is much lower in the Atlantic than in the Pacific due in part to the smaller area and climatic instability of the Atlantic region during the Pleistocene ice ages, which caused more extinction t­ here. In fact, the species composition of some marine communities

Chapter 8 Ecological Communities

bears strong legacies of tectonic events that happened tens of millions of years ago (S. A. Keith et al. 2013). In ecol­ogy, history usually refers to pro­cesses that influence the evolution of regional biotas via speciation, adaptation, and extinction, such as movements of continents and their effects on ocean circulation and climate, in time scales of millions of years. ­These influences ­were traditionally the purview of paleontology and biogeography and have been underappreciated in community and ecosystem ecol­ogy. But it is increasingly recognized that they are e­ very bit as impor­tant to the structure of communities as are local interactions, if not more so (Ricklefs 1987, Ricklefs and Schluter 1993, Hubbell 2001, Vellend 2016).

Dispersal and species richness Nearly all local communities are connected via dispersal to a larger network of populations. As a result, the dynamics of marine species happen at a much larger scale than what we think of as the local community. The large population sizes and variety of conditions among the connected communities mean that competitive exclusion and extinction are very slow at the regional scale. Thus, regional community dynamics and evolutionary dynamics converge on similar time scales, making speciation an integral component of community dynamics (Hubbell 2001, Ricklefs 2004). To what extent is species diversity in a community determined by local interactions versus dispersal from the larger region around it? Comparisons of local richness across regions offer key evidence to answer this question. The logic is as follows (Ricklefs 1987, Srivastava 1999): If local interactions, such as competition for resources, determine diversity, then the number of species in a community should be maintained below some limit determined by local resources. In this case, similar habitats in dif­fer­ent regions should converge on a similar number of species, regardless of how many species arrive from surrounding regions. In other words, local richness reaches saturation at a point determined by the local environment and is unrelated to richness of the surrounding region (Terborgh and Faaborg 1980, Ricklefs 1987) (figure  8.5). Alternatively, if local richness is more strongly influenced by dispersal from the surrounding region, then local richness should correlate with regional richness, a pattern referred to as regional enrichment. The alternatives are actually not so

(A)

(B)

gio n

al

en

ric

hm

en

t

Increased specialization

Local saturation

Species

Re

Local diversity

Increased niche overlap

Regional diversity

Resource gradient

Figure 8.5. ​Alternative models for the relationship between local and regional species richness. (A) A community is said to be saturated when local species richness is similar in communities with similar environmental conditions regardless of the diversity of the surrounding species pool (regional diversity). In contrast, regional enrichment is recognized where local richness of communities rises with richness of the regional pool within which they are embedded. (B) Local saturation (bottom) implies a fixed niche width for species within a local community, whereas the pattern of regional enrichment (top) might be realized e­ ither by greater niche overlap or by greater specialization among species within richer local communities (­after Ricklefs 1987).

209

210

Ocean Ecology

clear-­cut: local richness can increase linearly with regional richness despite strong local interactions if disturbance or dispersal is strong enough (Caswell and Cohen 1991, Mouquet et al. 2003). Nevertheless, a linear relationship between regional and local richness implies that community composition is not determined solely or even mainly by local interactions. Many empirical studies support a role for regional enrichment in setting the diversity of local communities (Ricklefs and Schluter 1993). The two studies summarized in box 8.3 are especially compelling ­because they test the local saturation hypothesis u­ nder the conditions most likely to support it—­the most diverse marine systems on earth—­and still find no support for saturation. Specifically, if local competition, predation, and other interactions limit diversity to some ceiling within a community, this ceiling should be most easily detected where the diversity of potential colonists and the rates of competition, predation, and other biological interactions are highest, as on coral reefs. The absence of saturation in t­ hese studies strongly suggests that local interactions are insufficient on their own to explain community diversity (Ricklefs and Schluter 1993).

Metabolic theory and species diversity On large spatial scales that aggregate many local communities, temperature is the single most consistent predictor of species richness in marine animals (Tittensor et al. 2010, Edgar et al. 2017). The metabolic theory of ecol­ogy is thus a potentially power­ful foundation for explaining diversity in that it predicts broad relationships among body size, abundance, life history, and trophic interactions (chapter 5). Can it also help resolve the central question in ecology—­what determines the diversity and distribution of species? Like most ecological phenomena, species richness is a product of intrinsic organismal traits interacting with the environment, and metabolic theory speaks to both of ­these. Among environmental f­ actors, temperature drives metabolic pro­cesses, which in turn shape population demography, species interactions, and speciation. All of ­these are fundamental influences on species richness. Among organismal traits, body size is similarly and consistently related to metabolism, demography, and species interactions (Andersen et al. 2016). Metabolic theory has addressed the challenge of predicting species richness based on temperature and body size (Allen et al. 2002). The argument begins by accepting as an assumption the energetic equivalence rule that total energy flux (BT) through populations is similar across body size classes (chapter 6). It then incorporates the well-­documented scaling of individual metabolic rate with temperature and body size to proceed through a rather convoluted chain of assumptions to arrive at the conclusion that species richness should increase with temperature. Alas, a general metabolic explanation for patterns in global species richness remains elusive, at least for well-­studied terrestrial organisms. Among 46 data sets spanning a wide range of plant and animal taxa, the predicted linear relationship between temperature and log richness (Allen et  al. 2002) was found in only 9, whereas richness was unrelated to temperature in about a third of the studies (Hawkins et  al. 2007). The relationship between temperature and richness varied widely both among taxa and geo­graph­i­cally, confirming that the predicted universal metabolic influence of temperature on diversity was swamped by the wide range of other environmental and historical influences that shape the distribution of species (Hawkins et al. 2007). This conclusion is further supported by global analy­sis of terrestrial birds and mammals that compared the many influences on diversity and found that temperature emerged as only a minor predictor of diversity relative to regional evolutionary history (Belmaker and Jetz 2015). Clearly related to the metabolic hypothesis for species diversity is the f­amily of species-­energy hypotheses. In many kinds of communities, species richness correlates positively with proxies for availability of energy, including temperature, standing biomass, or (on land) w ­ ater (Currie 1991). Generally, species-­energy hypotheses purport to explain patterns in species richness as a function of energy input, typically solar radiation. But species-­energy hypotheses face several challenges (Clarke

Chapter 8 Ecological Communities

and Gaston 2006). First, energy is a complex concept: it exists in fundamentally dif­fer­ent forms that might affect diversity in dif­fer­ent ways. The most basic form, photosynthetically active radiation (PAR), is used only by plants. Chemical energy, in the form of organic compounds in organismal tissues, can be utilized by all heterotrophs but is very dif­fer­ent in form and dynamics than PAR, so it is difficult to compare them. Second, the link from energy to diversity is indirect: PAR and chemical energy clearly influence biomass and abundance, but it is less obvious how t­ hese translate to diversity. Fi­nally, a central challenge is that PAR is also tightly related to environmental temperature, making it difficult to distinguish the two. In sum, energy affects abundance via dif­fer­ent mechanisms in plants and animals, and ­these affect diversity only indirectly, so the role of energy in species diversity remains largely unresolved (Clarke and Gaston 2006).

Space and species diversity We saw in chapter 3 that species richness scales with habitat area. The often strong relationship between regional and local diversity (see box 8.3) similarly hints at the importance of space in shaping local species diversity. The major global gradient in richness, decreasing with latitude away from the tropics, parallels not only declining temperature but also a decline in available habitat area. This is ­because the earth’s s­ pherical shape geometrically constrains the total area at dif­fer­ent latitudes, resulting in roughly fivefold more tropical habitat than polar habitat. Species richness increases with habitat area at multiple spatial scales in empirical data (chapter 3), and this pattern emerges from neutral models as well (Hubbell 2001). Given the robust species-­area relationship, the geometric constraint of declining habitat with latitude might alone be expected to produce the observed latitudinal decline in species richness. Building on the neutral theory, a spatially explicit metacommunity model that accounts for both area and temperature found that the relationship between speciation rate and area was multiplicative, producing an order of magnitude gradient in species richness from tropics to poles across the world ocean (Tittensor and Worm 2016). Th ­ ese results illustrate that temperature and geographic area could interact to produce latitudinal diversity gradients, such as ­those common around the world, but leaves open the specific mechanisms responsible.

Linking Communities to Ecosystems The activities of diverse organisms interacting in communities drive the biogeochemical and energy fluxes that are the heart of functioning ecosystems, including primary production, nutrient and carbon cycling, and trophic flows (Chapin et al. 1997, Stachowicz et al. 2007, Estes et al. 2011). ­These pro­cesses are sensitive to the kinds of species pre­sent and depend on their functional traits—­body size, growth rate, photosynthetic capacity, feeding specificity—­which mediate their population dynamics, along with interactions among species, probability of arriving at a site from the regional pool, and even probability of speciating.

Functional structure of communities A list of species tells us who lives in a community but l­ittle about what they are d­ oing. Describing a community in terms of the traits of its member species makes community ecol­ogy more mechanistic and links it more naturally to ecosystem ecol­ogy. In princi­ple, functional traits can link environmental change through shifts in community structure to effects on ecosystem pro­cesses, such as production, consumption, and biogeochemical cycling (Lavorel and Garnier 2002, Balvanera et al. 2005). Initial approaches to defining community functional structure divided species informally into functional groups of similar morphology and physiology that are assumed to use resources or interact with the environment in a par­tic­u­lar way. The most obvious functional grouping, long central to ecol­ogy,

211

212

Ocean Ecology

Box 8.3. ​Regional influences on local community diversity Is community diversity set by local competition and predation, or by dispersal from the surrounding region? Two large-­scale studies of marine systems provide evidence to answer this question. First, coral reefs are the

most diverse communities in the ocean and therefore provide an especially power­ful test of local versus regional control of diversity. Ron Karlson and colleagues (2004) sampled corals along a 10,000 km transect that spanned a 180° 1,000 km 0°

(B)

(A)

80

ds Isl ty

So

cie

n ica er

m

an

oa So

m

an Am

on

Isl

Gu Ne ua

Pa p

(C)

ds

ea

0

in

250

w

100 150 200 Regional number of species

ne

50

do

0

20

In

0

40

lo

30

60

So

60

Flat Crest Slope

sia

Flat Crest Slope

Number of species

Local number of species

90

Local species richness

250 200 150 100 50 0

0

200

400 600 800 1,000 Number of species in region

1,200

Figure B8.3.1. Regional control of local diversity. (Top) Reef corals: Map of sampling locations across the tropical Pacific Ocean. (A) Coral species richness within a site increases strongly with the size of the regional species pool (i.e., the number of coral species recorded from the larger region). (B) Nevertheless, local environment also plays an impor­tant role: within regions, reef slopes consistently support more species than reef flats (­after Karlson et al. 2004). (C) Rock walls: Regional control of local diversity in rock wall invertebrate communities from around the globe. Across regions, local richness (mean within quadrats) r­ ose linearly with regional richness, implying that local interactions—­competition, predation, and so forth—­generally have l­ittle influence on benthic species richness (­after Witman et al. 2004).

Chapter 8 Ecological Communities

gradient eastward from the diversity hotspot of the West Pacific. The team surveyed three islands in each of five regions, mea­sur­ing corals along four transects each on the reef slope, reef flat, and reef crest on each island (figure B8.3.1). They estimated local richness as the mean number of coral species on a transect, and regional richness as the total number of coral species recorded across all transects from the three islands in that group. Regional coral richness declined eastward along the gradient into the central Pacific, and ­there was a corresponding decrease in local coral richness within each reef habitat (slopes, crests, and flats) (figure B8.3.1A, B). This pattern implies that the number of coral species that coexist on a given reef flat is ­limited not by the interactions among competitors and consumers that occur ­there but by the availability of recruits from the larger region. This conclusion does not imply, however, that local interactions or niche requirements are not impor­tant. The same study illustrates this: corals ­were consistently more diverse on the slope than on the reef flat across all regions despite the fact that ­these two habitats occur in close proximity to one another within each region. In other words, coral species are adapted to par­tic­u­lar conditions and zones of the reef, and community composition in each zone reflects the match between a species’ habitat requirements and local environmental conditions (habitat

213

filtering). The number of species in each zone (local richness), however, is evidently not affected much by local biological interactions, such as competitive exclusion, but instead reflects the number of species in the region that are adapted to living in that zone and capable of dispersing into it. In ­these coral communities, then, pro­cesses at both the regional scale (dispersal from the larger species pool) and local scale (environmental filtering) interact to mold local species richness and composition. A second illustration of such regional enrichment comes from the sessile communities of invertebrates and seaweeds that grow on submerged rock walls throughout the world (figure B8.3.1C). Jon Witman and colleagues (2004) surveyed ­these rock wall communities at 12 sites spanning the world ocean from the equator to polar seas. The pattern that emerged was very similar to that seen on Pacific coral reefs: the average number of species on a local rock wall was linearly related to the richness of the surrounding region. ­These results are especially robust due to the identical sampling methodology used at all sites, the high-­resolution sampling of richness in the plots, and the coverage of epifaunal communities worldwide. Similar relationships between regional and local richness are now known from a range of organisms (Harrison and Cornell 2008).

organizes species by trophic level. Within trophic levels, functional groups have been described based on growth form, physiology, and feeding mode. As we have seen, functional classifications have been suggested for seaweeds (Littler and Littler 1980, Steneck and Dethier 1994), herbivorous mollusks (Steneck and Watling 1982), corals (Darling et al. 2012, Madin, Anderson, et al. 2016), plankton (Le Quéré et al. 2005, Kruk and Segura 2012), and fishes (Villéger et al. 2017), among ­others (see figures 2.11, 5.11). Historically, functional groups have been defined informally based on intuition or expert knowledge. More recently, functional ecol­ogy has become more rigorous, producing metrics and quantitative analyses aimed at characterizing the functional structure of communities and how that structure is influenced by environmental change and h­ uman impacts (Cadotte et al. 2008, Flynn et al. 2011, Pavoine and Bonsall 2011). For example, the trophic structure of a community influences the strength of top-­down control, transfer of energy and materials through the food web, and biogeochemical cycling. As might be expected, functional structure of communities often appears more stable in the face of environmental change than does taxonomic structure. Sébastien Villéger and colleagues (Villéger, Ramos Miranda, et al. 2008) characterized variation in trophic diversity of coastal fishes and large decapods (nekton), mea­sured as the richness and evenness of nekton biomass across trophic levels at 37 stations in the southern Gulf of Mexico. They found substantial variation (turnover) of species among sites and seasons, whereas trophic diversity varied much less: 50%–60% of nekton biomass was made up of diet generalists within a narrow range of trophic levels. Variation in trophic structure across sites and seasons was small, unrelated to taxonomic structure, and only weakly related to environmental par­ameters. Thus, from a functional perspective, change in the structure of this community was substantially deterministic and stable, with community membership strongly influenced by feeding ecol­ogy.

214

Ocean Ecology

Phyloge­ne­tic structure of communities Two related challenges to functional trait-­based approaches in community ecol­ogy are the difficulty of identifying the traits most impor­tant to community interactions and finding sufficient data on ­those traits, which often requires detailed knowledge of the species’ natu­ral history. One potential way around t­ hese challenges involves using phyloge­ne­tic relatedness as a proxy for functional similarity. The premises are that phenotypic traits determine habitat use and ecological interactions and are generally similar among related species. If ­these two conditions are met, phyloge­ne­tic relatedness can provide a proxy for functional similarity (Cavender-­Bares et  al. 2009, Srivastava et  al. 2012, Madin, Hoogenboom, et al. 2016). One way this phyloge­ne­tic approach has been applied is to infer the pro­cesses of community assembly from the evolutionary relatedness among species in a community relative to that of a random draw of species from the regional pool. The appeal of the phyloge­ne­ tic approach is that it takes advantage of the rapidly growing stream of molecular data on phyloge­ne­ tic relationships, which in most taxa contrasts with the dearth of data on the traits involved in ecological interactions. Exploiting phyloge­ne­tic data for understanding community assembly and dynamics requires probing the assumptions: Is it true that functional traits are similar in related species? Functional traits are indeed often conserved among related species, although traits can also evolve rapidly and obscure such relationships (Wiens et al. 2010). A striking example of such phyloge­ne­tic conservatism comes from a global analy­sis of the effects of herbivory on marine benthic communities. Synthesis of more than 600 herbivore exclusion experiments found that marine herbivore impacts w ­ ere strong, reducing producer abundance by 68% on average, but ­there was ­little or no significant influence of latitude, temperature, or other environmental gradients on herbivore impact (Poore et al. 2012). Instead the most consistent predictors of herbivore impact w ­ ere producer taxonomic and morphological groups (see figure 2.12), which are closely related ­because plant morphology is strongly conserved among related taxa. This means that major trends in marine plant-­herbivore interactions are governed by traits that are many millions or even billions of years old. For example, cyanobacteria are the oldest photosynthesizing organisms on earth, having arisen more than three billion years ago. They are consistently avoided by herbivores and ­were the only group of marine primary producers that ­were actually more abundant on average in the presence of grazers (Poore et al. 2012). Within such broad groups, however, closely related species can differ substantially in ecol­ogy (Losos 2011).

Communities in the Anthropocene Communities consist of many species interacting with one another in many ways, and their membership and structure are sensitive to a range of environmental and biological influences. Thus, it’s easy to understand that the multiple interacting impacts of h­ uman activities are profoundly changing communities (chapters 4, 7). We discuss how communities are changing in Earth’s major marine systems in ­later chapters. ­Here we highlight several general trends.

Climate change and communities Temperature strongly affects metabolic rates (chapter 5), but ­those effects often differ among species such that rising temperatures disrupt or intensify interactions (chapter 7). We have seen that diversity is consistently higher in warmer places and times in the ocean. As anthropogenic climate warming proceeds, most areas of the ocean are likely to become more diverse as species from low latitudes move poleward. But warming also affects trait distributions and community interactions in some consistent ways (chapter 7), and ­these suggest consequences for humanity’s dependence on marine ecosystems. First, respiration rates rise more rapidly with temperature than do biomass production

Chapter 8 Ecological Communities

rates, which can result in both individual organisms and populations becoming smaller at high temperatures (Cheung et al. 2012). At the community level, t­ hese lower average population sizes ­under warming may create opportunities for additional species to establish and persist in communities, facilitated by the changing bound­aries of species ranges u­ nder way. Thus, theory predicts more diverse communities with smaller average body sizes as the ocean warms. ­These effects are already evident among marine zooplankton (Beaugrand et al. 2010) and bacteria (Morán et al. 2015), in which long-­ term warming trends have been accompanied by shifts to more diverse communities of smaller-­ bodied taxa. Warmer w ­ ater also holds less dissolved oxygen, which is expected to strengthen the decline in average body sizes of marine fishes, with consequences for fisheries (Cheung et al. 2012). Combined ocean warming and declining oxygen are expected to broadly alter the distribution and abundance of marine organisms (Deutsch et al. 2015, 2020). Warming of marine ­waters is also broadening the geographic ranges of warm-­water species, and impacts of ­those changes are beginning to manifest at community and ecosystem levels. A striking example involves expansion of warm-­water herbivorous fishes into areas that formerly lacked them in Japan. This phenomenon appears already to be denuding some former macroalgal forests, creating barren areas and shifting some former macroalgal forests to coral cover (Vergés et al. 2014) (see figure 3.19). Similarly, the king crab Paralomis birsteini (see figure 4.13D) was observed on the Antarctic slope in 2003 and had established a large population by 2010 (Aronson et al. 2015). This and other shell-­crushing predators have been absent from Antarctica for tens of millions of years, during which a rich and diverse benthic community has evolved with few defenses against such predators. The rich and relatively undefended communities of the Antarctic shelf are now vulnerable to predation and likely to change substantially with the expansion of the king crab.

Marine defaunation and trophic skew Large species at high trophic levels generally have low abundances and metabolic rates, hence low population growth rates. This slow demography makes ­these predators especially vulnerable to overexploitation and extinction in impacted marine systems, often leading to trophic skew, that is, a change in shape of the trophic pyramid resulting from more extreme decline of top predators relative to species lower in the food web (Pauly et al. 1998, Duffy 2002, 2003). Trophic skew in marine fisheries was formalized in the concept of “fishing down the food web,” whereby virgin fisheries are first depleted of large predators, a­ fter which fishing effort shifts lower in the food web (Pauly et al. 1998). Subsequent global analyses have shown that fishing down the food web is less common than previously believed due to the dynamic and multifarious activities of marine fisheries, which target a wide variety of organisms for dif­fer­ent purposes (Branch et al. 2010). But in marine ecosystems generally, large animals declined sharply over the latter twentieth ­century (Myers and Worm 2003, McCauley et al. 2015), and surveys show that trophic structure has indeed skewed away from top predators as a result of both losses of carnivores and invasions of species at lower trophic levels in several regions of the world (Friedlander and de Martini 2002, Byrnes et al. 2007, figure 4.14). Response traits govern which species are most affected by a stressor, and effect traits govern their impact on ecosystem properties. Large body size, and potentially the correlated trait of high trophic level, are both response and effect traits. The large predators and herbivores most vulnerable to overexploitation also tend to have strong per capita impacts on communities and ecosystem pro­cesses, and disruption of t­ hose top-­down interactions often has far-­reaching consequences for communities and ecosystems (Estes et al. 2011). This trophic skew is a clear example of how ­humans have especially strong impacts on ecosystems when response traits and effect traits of species coincide. ­These impacts include the consistently strong influence of fishing and other ­human activities on marine ecosystems, such as increases in herbivores like sea urchins and cascading declines in macroalgae where ­humans have overexploited top predators.

215

216

Ocean Ecology

­Future Directions Developing a general understanding of how species assem­ble into communities, and the controls on their composition and diversity, remains among the most difficult prob­lems in ecol­ogy. Better quantification and mechanistic understanding of effect and response traits is a key to advancing this frontier. Using traits to quantify the functional structure of communities should open the door to many exciting questions: Is functional structure of communities generally more conserved and stable across space and time than taxonomic structure? Can variation in community functional structure reveal general mechanisms or provide predictive power in understanding community assembly and disassembly, including trophic skew and exotic invasions? ­There is exciting opportunity for using metabolic ecol­ogy as part of a unifying foundation to integrate ecol­ogy from traits through to organisms and ecosystems b­ ecause the theory is well developed, data on environmental temperature are rich and widely available, temperature can be manipulated in the lab and to some degree in the field, and temperature-­mediated biological changes have impor­tant implications for pressing practical issues related to global change. The key research need ­here is at the community level: developing a general understanding of how temperature changes interactions among species and thus community structure, functional trait distributions, and ecosystem pro­ cesses (Bruno et al. 2015, Deutsch et al. 2020). Environmental temperature is a major environmental driver, but of course ­there are many ­others. Promising directions include linking variation in functional traits to environmental gradients, which set the fundamental niches of organisms, and to the web of species interactions within a community; and using mea­sur­able per­for­mance currencies, such as energy intake and reproductive output, to quantify t­ hese links (McGill et al. 2006).

Summary A community is a group of species that co-­occur within a defined area. What determines the numbers, identities, and relative abundances of its species is the central and enduring theme of community ecol­ogy, and indeed of ecol­ogy generally. Resolving t­ hese questions begins with the set of species inhabiting the larger region—­the species pool available to colonize a site. That regional species pool is a product of earth and evolutionary history, specifically the arrangements of continents that determined habitat area and climate history and therefore the composition of the regional biota. The local community at a site is then sculpted from this raw material by environmental forcing and interactions among its species. In short, the assembly and dynamics of local communities are influenced by four classes of pro­cesses: (1) speciation, the creation of new species via evolutionary change; (2) dispersal of organisms through space; (3) ecological drift, resulting from stochastic changes in species abundances; and (4) ecological se­lection among species based on deterministic, niche-­based differences in fitness. Species enter communities via speciation and dispersal, and their abundances and per­sis­tence in the local community are then s­ haped by ecological drift, se­lection, and continued dispersal. With the pos­si­ble exception of ecological drift, ­these pro­cesses are mediated by functional traits interacting with the abiotic environment and the traits of other species. The plethora of concepts and theories in community ecol­ogy differ mainly in their relative emphasis on t­ hese four classes of pro­cesses and the interactions among them. Historically, community ecol­ogy has focused on ecological se­lection pro­cesses at the local scale of interacting organisms. A long history of experimental research shows that competition, herbivory, predation, and facilitation pervasively influence relative abundances and dynamics of species. ­These effects are often mediated by complex webs of indirect interactions, and in some cases, such as keystone predation, they can alter the entire character of an ecosystem. Positive interactions, such as facilitation by foundation species, often have equally pervasive system-­wide effects. The other, nonselective pro­cesses shaping communities—­speciation, drift,

Chapter 8 Ecological Communities

and, to some extent, dispersal—­have been the province of largely separate traditions in evolutionary biology and biogeography. These nonselective processes appear to be the main determinants of local species richness. The development of neutral theory has shown that regional speciation rates and dispersal can strongly affect local species richness and that competitive exclusion is often a very slow pro­cess in such metacommunities, particularly among demographically similar species. However, the composition and relative abundances of species within local communities rarely conform to predictions of neutral theory, confirming that ecological se­lection pro­cesses usually override ecological drift in shaping local community structure. Integrating regional, neutral, and local deterministic pro­ cesses is an exciting frontier in building a mechanistic understanding of community structure and dynamics and has potentially valuable applications in global change research. Understanding the assembly and dynamics of communities is in turn necessary to understand emergent pro­cesses of energy and materials flow through ecosystems.

217

9 H

218

Ecosystems

ow much fish can be harvested from the ocean sustainably? How does that limit vary among regions and why? How much of the fossil carbon we are injecting into the atmosphere can the ocean absorb and where does it go? And how can such pro­cesses be expected to change as the world warms and species shift their distributions and interactions? ­These questions concern ecosystem pro­cesses, that is, aggregate fluxes of energy and materials mediated by organisms interacting across communities and seascapes. Whereas ecological communities are defined primarily by their membership—­the numbers and kinds of species—­the ecosystem is defined by the work t­ hose communities do: the kinds and amounts of materials used, the rates at which they are pro­cessed, and their fates. This work is what we now refer to as ecosystem functioning, and it is the source of the ser­vices humanity depends on from nature, including crop and fishery production, climate regulation, shoreline protection, and waste pro­cessing. It is such complex adaptive systems of diverse organisms interacting with one another and with the abiotic environment over long stretches of time that created Earth’s oxygenated atmosphere (table 1.1), fixed and buried the colossal quantities of organic carbon that fuel our global industrial metabolism ­today (figures 4.4, 4.5), allow coral reefs to bloom in the oligotrophic open ocean (chapter 12), and support global fisheries that feed billions of ­people each year. Mechanistically, the functioning of an ecosystem amounts to the metabolism of its biological communities—­their combined respiration, consumption, growth, materials cycling, and, perhaps most importantly, biomass production. All of t­ hese pro­cesses are fluxes, changes in some state variable per unit time. Thus, ecosystem ecol­ogy is fundamentally a science of accounting for the stocks and fluxes of energy and materials among organisms and the environment, and what influences them. ­These properties—­the shape of the biomass pyramid, total primary production, nutrient fluxes, and so on—­are sometimes referred to as emergent properties of the ecosystem b­ ecause they result (emerge) from the combined actions and interactions of the numerous kinds of organisms pre­sent, ­shaped by the environment, and cannot usually be assigned to individual species. The distinction between community and ecosystem ecol­ogy is fuzzy in practice b­ ecause structure and function are intimately related. Productivity, trophic transfer, and cycling of materials—­the central pro­cesses in all ecosystems—­cannot be fully understood without knowing the types of species involved, ­whether they have specialized or generalized diet and habitat requirements, their demographics and rates of dispersal, and other aspects of community structure. Conversely, community structure is often strongly affected by how foundation species and ­others change the flow of energy and materials locally. Communities and ecosystems are separated h­ ere mainly to simplify pre­ sen­ta­tion. Despite the connections between community ecol­ogy and ecosystem ecol­ogy, for most of their history the two fields developed along separate tracks with l­ ittle interaction. This changed in the 1990s when growing concern about threats to biodiversity generated intense interest in the consequences of changing species composition and diversity for how ecosystems function (Schultze and Mooney 1994, Loreau et al. 2001).

Chapter 9 Ecosystems

In this chapter we trace the evolution of the ecosystem concept as an integration of biological and abiotic pro­cesses, from its origins and maturation in a systems engineering framework to the view of the ecosystem as a complex adaptive system. We sketch the fundamental pro­cesses of biological production, what controls it, and what happens to it as it is consumed by heterotrophic organisms. We concentrate on photosynthetic primary production, herbivory, and decomposition—­the interacting pro­cesses that together form the most basic module in the complex adaptive ecosystem. We then explore patterns in the structure and functioning of marine ecosystems, emphasizing general princi­ples and commonalities. We consider the workings of par­tic­ul­ ar ecosystems in subsequent chapters.

History of the Ecosystem Concept The En­glish botanist Arthur Tansley (1935) introduced the concept of the ecosystem: The more fundamental conception is, as it seems to me, the w ­ hole system (in the sense of physics), including not only the organism-­complex, but also the w ­ hole complex of physical f­actors forming what we call the environment of the biome—­the habitat ­factors in the widest sense [emphasis in original]. Though the organisms may claim our primary interest, when we are trying to think fundamentally we cannot separate them from their special environment, with which they form one physical system. It is the systems so formed which, from the point of view of the ecologist, are the basic units of nature on the face of the earth.

The essence of an ecosystem, therefore, is the interaction among all the organisms living in some area with the abiotic environment and with one another, considered together as a ­whole, and the pro­ cesses that emerge from their interactions. Th ­ ese interactions include stocks and flows of energy and materials and give rise to emergent features, such as trophic structure, distributions of biomass, and ecological efficiency. The concept of an ecosystem has evolved considerably over time, but two themes have remained central. First is the integration of living and nonliving components identified by Tansley. Second is the recognition that the interacting organisms in an area do work, that is, they use materials and energy, transform them into other materials, and in the pro­cess transform their environment. The integrated concept of the ecosystem was operationalized by Raymond Lindeman (1942) in his classic “trophic-­ dynamic” study of Cedar Bog Lake, Minnesota, USA (figure 9.1). Lindeman recognized that plants and animals, typically studied by dif­fer­ent ecologists with dif­fer­ent worldviews, interact intimately. He studied them together, organ­izing producers (“plants”) and consumers (“animals”) into trophic levels reflecting passage of energy through the ecosystem. He also argued, echoing Tansley, that “the discrimination between living organisms as part of the ‘biotic community’ and dead organisms and organic nutrients as part of the ‘environment’ seems arbitrary and unnatural” (Lindeman 1942). Lindeman’s work was pioneering in seeking to quantify the integration and interaction of the biological and abiotic components of nature. For example, he demonstrated that loss of energy to respiration increases up the food chain, and introduced a quantitative rigor to the changes in productivity and efficiency during succession that w ­ ere of ­great interest in plant ecol­ogy at the time. This study was also a key stimulus to the influential work of E. P. and H. T. Odum and their students, who essentially in­ven­ted modern ecosystem ecol­ogy during the 1950s and 1960s. Development of the ecosystem concept was closely tied to the phenomenon of ecological succession, the characteristic sequence of changing species composition at a disturbed site through time. Studies of succession w ­ ere central to the origin of ecological science generally, providing a natu­ral laboratory for observing how organisms associate in their environment. Especially before the widespread use of experimentation beginning in the 1960s, ecol­ogy’s primary modus operandi was the comparative approach. The succession of ecosystem states through time was generally deduced from

219

220

Ocean Ecology

Solar radiation

Solar radiation External Dissolved nutrients

Internal Phytoplankters

Pondweeds

Bacteria Zooplankters

Browsers

OOZE Bacteria

Plankton predators

Benthic predators

Swimming predators

Figure 9.1. ​An early graphical summary of an ecosystem from a trophodynamic perspective. This study of a Minnesota bog launched the modern science of the ecosystem as an integration of abiotic and biotic components, united by flows (represented ­here by arrows) of energy and materials (­after Lindeman 1942).

comparisons in space ­under a space-­for-­time assumption, and this development of the community or ecosystem was often considered analogous to the ontogeny of an individual organism. ­These successional changes ­were ­adopted as a central theme of Eugene P. Odum’s (1969) influential approach to ecosystems and formalized in his proposed “strategy of ecosystem development” (­table 9.1). The argument for strategy is reflected in the language applied to the succession of ecosystem states t­ oward “maturity” and the characteristics of ecosystems associated with pioneer and climax states (see also Margalef 1963). This concept of a “strategy” of ecosystem development was and remains controversial ­because it seems to imply a purpose and direction above the level of natu­ral se­lection operating on individuals within populations, at the ecosystem level. In Odum’s words, “It is an orderly pro­cess of community development that is reasonably directional and, therefore, predictable. . . . ​It culminates in a stabilized ecosystem in which maximum biomass (high information content) and symbiotic function between organisms are maintained per unit of available energy flow. In a word the ‘strategy’ of succession is . . . ​increased control of, or homeostasis with, the physical environment in the sense of achieving maximum protection from its perturbations.” The marine ecologist Ramon Margalef (1963) similarly argued that in the absence of disturbance, ecosystems change in a progressive and orderly way ­toward what he also referred to as “maturity,” recognized as an increase in complexity and a decrease in production per unit biomass. Since that time, it has become clear that autonomous pro­cesses acting at the level of individuals, principally natu­ral se­lection but also species sorting (chapter 10), can produce some of the same structure and pro­cesses highlighted in Odum’s scheme, as we address l­ater in this chapter. Nevertheless, laboratory experiments have indeed produced intriguing evidence that

Chapter 9 Ecosystems

­TABLE 9.1 ​Odum’s summary of the “strategy of ecosystem development” Ecosystem attributes

Developmental stages

Mature stages

Community energetics Gross production/standing crop biomass (P/B ratio)

High

Low

Net community production (yield)

High

Low

Food chains

Linear

Weblike

Community structure Total organic m ­ atter

Small

Large

Species diversity

Low

High

Life history Niche specialization

Broad

Narrow

Size of organism

Small

Large

Life cycles

Short, ­simple

Long, complex

Nutrient cycling Mineral cycles

Open

Closed

Nutrient exchange rate

Rapid

Slow

Se­lection pressure Growth form

For rapid growth (“r-­selection”)

For feedback control (“K-­selection”)

Production

Quantity

Quality

Overall homeostasis Internal symbiosis

Undeveloped

Developed

Nutrient conservation

Poor

Good

Stability (re­sis­tance to external perturbations)

Poor

Good

Entropy

High

Low

Information

Low

High

Source: Odum (1969).

entire biological communities can coevolve in a way that responds to selective pressure on ecosystem-­ level pro­cesses. For example, laboratory soil and aquatic ecosystems selected to maximize plant production and acidity, respectively, exhibited heritable variation at the community level that changed through time in the direction selected, despite being composed of complex microbial associations (Swenson et al. 2000). The superorganismal view of ecosystems reached its most grandiose development in the concept of Gaia, which views the planetary biosphere in its entirety as an adapted self-­ regulating system that, in at least some version of the hypothesis, actively seeks a homeostatic equilibrium (Lovelock 1972).

221

222

Ocean Ecology

The degree to which ecosystems conform to the analogy with individual organisms is debatable, but ­there is no doubt that the abiotic environment strongly influences the character and acOcean a b c d e tivity of the biological community in an area, and that organisms can in turn transform the environment through their collective Crabs Annelids activities. The first formal application of Lindeman’s trophic-­ Crustacea Coral polyps Urchins dynamic approach in the ocean was the landmark study of EnFleshy, calcareous ewetak Atoll in the mid-­Pacific by the b­ rothers H. T. and E. P. Algae in corals encrusting algae Odum (1955). This work grew out of the US atomic bomb test0 200 400 600 800 ing program at nearby Bikini Atoll. Together with the analy­sis of Dry biomass in grams per sq. meter the freshwater ecosystem of Silver Springs, Florida, USA (Odum 1957), it marked the birth of the systems approach in ecol­ogy (B) and ushered in the new field of community metabolism, in which aggregate mea­sure­ments of O2, CO2, and nutrient exchanges arisLoss on ignition 50 ing from the w ­ hole community of species are used to estimate the production and respiration of entire ecosystems (figure  9.2). The study at Enewetak cleverly exploited the fact that the trade winds drive ­water across the coral atoll in a unidirectional flow 40 % Plant for much of the year, such that ­there is an input and an output side to the physical system as a ­whole and, in princi­ple, the net system-­level fluxes and transformations of materials can be esti80 Sand mated by subtracting the values mea­sured at the downstream end from the upstream end. To achieve this, the Odum team mea­sured biomass of vari­ous primary producers and consumers 4 Radioactivity along a transect across the reef and mea­sured chemical changes in the ­water before and ­after it flowed across the reef. The Odums’ coral reef study produced several noteworthy 0 results. First, it showed that biomass on the reef was in fact dominated by plants—­including the symbiotic zooxanthellae inhabFigure 9.2. ​The first quantitative study of a marine ecosystem, iting the coral animals, as well as filamentous algae inhabiting at Enewetak Atoll in the central Pacific. The study exploited the unidirectional flows of ­water over the reef from open ocean (a) coral skele­tons, and vari­ous benthic algae—­just as in terrestrial to lagoon (e). (A) The composition of the biomass pyramid on systems (figure 9.2A). This was a major finding ­because algae are the reef crest (point a). (B) Concentrations of organic ­matter often inconspicuous on pristine coral reefs, and it had been a (loss on ignition) and other properties across the reef transect mystery since Darwin’s time how such a vibrant community could (­after Odum and Odum 1955). thrive in the nutrient desert of the tropical ocean. The major functional groups of benthic primary producers at Enewetak, zooxanthellae and microalgal turfs, turned out to be characteristic of reefs generally. Macroalgae (seaweeds) ­were far less abundant. We see in chapter 12 that many reefs worldwide have transitioned from coral to macroalgal dominance over the late twentieth c­ entury. Mea­sure­ments of oxygen and organic content of the w ­ ater flowing over the reef (figure 9.2B) showed that primary production on the reef was consumed almost entirely by the benthic community, the first evidence that coral reefs are highly efficient systems that maintain high production in the oligotrophic ocean as a result of tight recycling. As more such studies of community metabolism w ­ ere done, energy flow came to dominate the entire science of ecosystem ecol­ogy, largely through the influential work of Howard T. Odum, who built a formal thermodynamic basis and engineering meta­phor for ecosystems, along with a graphical approach for characterizing system flows based on electrical circuitry (figure 9.3). John Teal’s (1962) study of a Georgia salt marsh illustrates the general approach of building an ecosystem energy bud­get by compiling a wealth of data on plant production and decomposition and on animal abundance, diet, assimilation, and respiration to estimate paths and magnitudes of energy flux. Both the ×105 counts min/m3

mg/m3

%

mg/m3

(A)

Chapter 9 Ecosystems

522650

Phytoplankton 1 139940 86250

225637

288273

290802

Watercolumn detritus 2, 34, 35

186594

15024

Benthic diatoms 4

750

775 17969 1818

209064

Benthic suspension feeders 11–13

7132

868059

… 260720

Combined zooplankton 5–10

247727

35807

Sediment POC 3, 36

77803

68864

10889

Suspension feeding fish 20–24

432

3501 6320

960.1

9 296655

753

8283 6150

332526

18086

Benthic deposit feeders 14–19

31885

27832

93171

921.4

1555 219847

105.4 1308.2

174.1 14.5

Carnivorous fish 25–33

158.5 585.2

360.8 Exports Returns to POC

Figure 9.3. ​The ecosystem as an energy cir­cuit, a format borrowed from engineering and pop­u­lar­ized by H. T. Odum. This example uses the graphical format of an energy cir­cuit for the Chesapeake Bay mesohaline ecosystem. Standing stocks (within symbols) are in mg m−2, fluxes (arrows) are in mg m−2 yr−1 (­after Baird and Ulanowicz 1989).

Enewetak reef and salt marsh studies w ­ ere intensive efforts, assembling the information necessary to capture ­these pro­cesses, but ­these early studies ­were essentially accounting ledgers, tallying inputs and outputs in an effort to understand ­whether entire ecological systems are net producers or consumers of energy, how efficient they are at converting sunlight into plant biomass compared with, say, a crop field, and so on. That is, they ­were descriptive. ­These pioneering ecosystem studies firmly established the quantitative approach envisioned by Lindeman for integrating biotic and abiotic components of an integrated ecosystem, and set the stage for the more sophisticated simulation modeling common in applied ecol­ogy ­today.

Evolution of the ecosystem concept The long debate about the nature of an ecological community—­whether it is an integrated entity or merely a transient assemblage of autonomous actors (chapter 8)—­has parallels at the level of the ecosystem. The concept of the ecosystem as a homeostatic machine flowered during the mid-­to late twentieth ­century with the rise of systems-­engineering thinking in society generally, and became increasingly formalized (and esoteric) as the electrical cir­cuit approach developed by H. T. Odum was elaborated by o­ thers (Ulanowicz 1986, Baird and Ulanowicz 1989) (see figure 9.3). Even the word system illustrates the subliminal power of terminology, implying an integrated, self-­regulating entity that tends ­toward equilibrium. The concept of the ecosystem directed, and sometimes ­limited, thinking in ecol­ogy. Several suggestions for a more realistic approach to ecosystems have been offered (O’Neill 2001). The first involves recognizing that ecosystems are generally out of equilibrium, contrary to a central assumption that was initially a mathematical con­ve­nience but grew into a basic

223

224

Ocean Ecology

philosophical theme of ecosystem ecol­ogy. A hallmark of real ecosystems is path dependency—­the sensitivity of ecosystem development to initial conditions and often a tendency ­toward rapid shifts among alternative states (Holling 1973, Levin 1998). Second, early ecosystem work glossed over spatial heterogeneity u­ nder the implicit assumption that the ecosystem is self-­contained. But all ecosystems are open to varying degrees, exchanging organisms and materials with surrounding systems, and ecosystem pro­cesses operate at a hierarchy of spatial scales (O’Neill et al. 1986, Levin 1992). Understanding an ecosystem requires incorporating this range of pro­cess scales, from regional climate forcing, to individual plant growth and demography, to the large home ranges and active movement of top predators. Fi­nally, most traditional ecosystem ecol­ogy ignored evolution, the central pro­ cess in biology. Strangely, although ecosystems have generally been considered homeostatic and self-­regulating, the mechanism was rarely specified. Natu­ral se­lection is the obvious candidate. On a practical level, the historical reliance on ­simple, linear equations, born of the computational limitations and ­limited knowledge of a prior age, has progressed to richer models of ecosystems. Th ­ ese include linked interactions among primary production, herbivory, and decomposition, and parallel “classical” and microbial food webs as networks with feedbacks among components.

Ecosystems as complex adaptive systems Ecosystems are complex adaptive systems. They are complex in that the dynamics and fate of primary production involve intimate interactions, often with feedback loops, among many species of producers, herbivores, and decomposers—­the microbes that metabolize dead organic ­matter and release its components back into inorganic forms available to primary producers. Ecosystems are adaptive in that the magnitude and fates of production can adjust rapidly to changes in environmental forcing as a result of shifts in species composition of organisms as well as evolutionary ge­ne­tic change. Links among ­these components are bidirectional: the herbivores and microbes that consume primary producers not only channel primary production up the food chain, they control the biomass, species composition, evolution, and production rates of autotrophs (see figure 5.1), and therefore the structure and functioning of the ecosystem and rates of biogeochemical pro­cesses. Especially in the ocean, where the dominant producers are nutrient-­rich algae, primary producer distribution and activity are affected so profoundly by consumers that it is difficult to understand production without them (Geider et al. 2001, Poore et al. 2012, Laws 2013). A growing appreciation that ecosystems are dynamic, nonequilibrium, multiscale entities promoted a new view of them as complex adaptive systems (Holling 1973, Levin 1998). This view began to bridge the gulf between ecosystems as engineering diagrams on one hand and as interacting populations of living organisms on the other, by recognizing that ecosystem-­level pro­cesses depend strongly on species composition and diversity, that mechanistic understanding of ecosystems depends on how the pro­cesses are mediated by organismal traits, and that traits are molded by continuous natu­ral se­ lection within populations. The essential components of a complex adaptive system are “sustained diversity and individuality of components; localized interactions among ­those components; and an autonomous pro­cess that selects from among ­those components, based on the results of local interactions, a subset for replication or enhancement” (Levin 1998). The value of this perspective is that it moves beyond a description of system components to focus on mechanisms by which organisms interact to produce the emergent structure and flows in ecosystems. Ecosystems are constantly changing as a result of interactions between the environment and the diverse organisms inhabiting them. ­Because they are made up of living organisms, ecosystems adapt, that is, they respond to environmental change actively via natu­ral se­lection, which results in a degree of optimization. The functioning of the ecosystem—­its gross productivity, efficiency of fluxes among components, and tendency ­toward stability—is determined by interactions between environmental ­drivers and the traits of the biological community. As we saw in chapter 8, the distribution of traits within a community changes as a result of

Chapter 9 Ecosystems

species sorting, se­lection, and immigration. The capacity of the community to adapt to environmental change is proportional to its trait diversity and thus to the diversity of species and genotypes in the community (Norberg et al. 2001), just as the capacity of a population to evolve is proportional to its ge­ne­tic variance. Thus, the environment affects ecosystem functioning both directly, for example, as temperature influences pro­cess rates, and indirectly, as environment changes the conditions for the species sorting and thus the trait distributions in the community (Norberg 2004). Recognizing ecosystems as complex adaptive systems has real-­world implications (Gaichas 2008): outcomes of ecosystem pro­cesses are much less deterministic than formerly assumed, requiring more precaution in management.

Primary Production Production and consumption are the yin and yang of all ecological and biogeochemical pro­cesses. The autotrophic (“self-­feeding”) organisms that produce organic m ­ atter support the heterotrophs that consume them, so autotrophic production is the foundation of all ecosystem pro­cesses. Herbivores consume plant production, and they are the producers for the next level that consumes them (predators), and so on (see figure 5.1). Decomposers, both microbes and larger organisms, consume what’s left and recycle it back into the inorganic forms that plants can use. ­These chains of linked production and consumption form ecological networks through which energy and influence flow, and the resultant biogeochemical transformations form parallel biogeochemical networks. Primary production is the conversion of inorganic materials—­CO2, w ­ ater, and mineral nutrients— into the organic compounds of living biomass, powered by a source of energy. Primary production is so called b­ ecause it creates living from nonliving m ­ atter and is thus the first step in ecological pro­ cesses. The energy that powers the earth’s biosphere is sunlight, and the engines that convert it into forms usable by the myriad components of the global ecosystem are photosynthetic organisms, a highly diverse collection of lineages from all three domains of life—­Archaea, Bacteria, and Eukarya (chapter 2). For con­ve­nience, we often use the word plants to refer to all primary producers (both vascular plants and algae) that make organic m ­ atter in sunlight via oxygenic photosynthesis. But most of the primary production in the sea is conducted by unicellular organisms (grouped informally ­under the name algae), primarily prokaryotes. About half of Earth’s annual net primary production is accomplished by marine phytoplankton despite making up only about 1% of Earth’s photosynthetic biomass (Field et al. 1998). The reason for this outsized role of phytoplankton is the highly efficient resource use and growth pos­si­ble in single-­celled organisms. Although the g­ reat majority of Earth’s primary production involves photosynthesis, chemoautotrophy is impor­tant in certain circumstances (chapter 2). We focus first on the foundational pro­cess of photosynthetic production and its fate. We then consider the other major classes of producer-­consumer interactions, building on general princi­ples of their bidirectional interactions that underpin pro­cesses at the scales of populations, communities, and ecosystems considered in ­later chapters.

Light and photosynthesis In the sea, as on land, the limiting ­factors for organic production are primarily light from above and inorganic nutrients from below (chapter 5). Light is absorbed by seawater and thus attenuates through the ocean’s ­water column, with low-­energy (long-­wavelength) light at the red end of the spectrum absorbed first, leaving deep ­water with a blue tint. The rate of attenuation varies greatly as a function of the amounts and kinds of dissolved and particulate material in the w ­ ater. Photosynthetically active radiation may penetrate to nearly 200 m in the clearest oceanic ­waters, and a few algae and phototrophic corals reach ­these depths in such ­waters. In coastal and especially turbid estuarine ­waters, light attenuates much faster and often limits system productivity. In the stratified upper layer of the open ocean, where the w ­ ater is generally transparent, light is in abundant supply and phytoplankton tend to grow

225

226

Ocean Ecology

u­ ntil they draw down nutrient concentrations to the point that biomass production is ­limited by the inorganic nutrient(s) in shortest supply, a princi­ple known as Leibig’s law of the minimum.

Nutrient uptake and use The organic compounds fixed by photosynthesis serve two purposes: they store energy for ­doing the cell’s work, and they provide the structural backbones of biomolecules. But a host of other ele­ments are also required to build and sustain functional cells (see chapter 5). The principal of t­ hese are nitrogen, a major component of proteins and nucleic acids that constitutes around 7% of cellular biomass, and phosphorus. Large-­scale patterns of primary producer biomass and production closely reflect gradients in nutrient availability, as is evident in maps of surface ocean productivity (see figure 1.4). In ­these maps, built from remotely sensed data on ocean surface color, areas of high phytoplankton biomass are vis­i­ble around the continental margins and equatorial upwelling zones where nutrients are supplied from the continents and the deep ocean, respectively, whereas the central ocean gyres far from sources of nutrients have very low phytoplankton biomass. Over much of the ocean, nitrogen availability limits primary production and the distribution of primary producer biomass. Th ­ ere are two complementary sources of evidence for this. First, nitrogen is highly depleted in surface ­waters over much of the low-­latitude open ocean, as is phosphorus (figure 9.4). Second, phytoplankton biomass and productivity in t­ hese ­waters generally increase when nitrogen alone is added experimentally, whereas other nutrients do not stimulate production as long as nitrogen concentrations remain low (Moore et al. 2013). In seawater, nitrogen is found primarily in four forms: nitrate (NO3), nitrite (NO2), ammonium (NH4), and urea. The first two oxidized forms must be reduced to be assimilated into the cell, and this requires energy. Hence NH4 and urea are preferentially taken up from the environment by phytoplankton. In the open ocean, NO3 is supplied primarily by upwelling and mixing from deep ­water (in coastal regions, much NO3 also enters from terrestrial sources), whereas NH4 and urea are produced by excretion of heterotrophs, including animals, protistan micrograzers, and bacteria. ­Because most NH4 and urea originate in plant tissue that has been grazed and excreted by herbivores, production fueled by ­these reduced forms is often called regenerated production, whereas production fueled by ­ ater or land sources is termed new inputs of NO3 from deep w production (Dugdale and Goering 1967). ­Under steady-­state conditions in the upper ocean ­water column, new production from NO3 inputs is balanced by losses of production (and associated nitrogen) through sinking into the deep ocean and harvest of biomass. Thus, if a steady state can be assumed, uptake of NO3 by the phytoplankton assemblage provides an estimate of the export of production to the deep ocean and, potentially, of the production that can be harvested sustainably.

80°

40°

0°

–40°

Herbivory

–80° 50°E

150°E

110°W

10°W

Figure 9.4. ​Phytoplankton nutrient limitation in the world ocean. Background colors indicate average surface concentrations of nitrate, from low (light blue) to high (green). Center circles indicate the primary, and outer circles the secondary, limiting nutrients as inferred from the artificial addition of N (green), P (black), Fe (red), Si (orange), Co (yellow), Zn (cyan), and vitamin B12 (purple) (­after Moore et al. 2013).

Herbivory, or grazing, is the consumption of plant (or algal) tissue by other organisms. Herbivory plays two central roles in ecosystems (see figure 5.1). First, from a bottom-up perspective, herbivores transfer energy and materials captured by primary producers to the rest of the food web. Second, from the top-­down perspective, herbivore grazing not only can limit plant biomass accumulation but also shapes the species diversity and functional composition of primary producers that provide habitat

Chapter 9 Ecosystems

and structure for the community. The functional diversity of marine herbivores is arguably even greater than that of producers, ranging from parasitoid protists that can be even smaller than their phytoplankton prey, through a wide range of grazing and suspension-­feeding invertebrates, to highly active schooling herbivorous fishes and megafaunal vertebrates (figure 9.5). A

B

D

C

E

F

Figure 9.5. ​A sampling of the diversity of marine herbivores. (A) A heterotrophic dinoflagellate pallium-­feeding on a diatom chain. (B) A tintinnid ciliate. (C) Copepods, the dominant mesozooplankters of the ocean. (D) Parrotfish on a coral reef. (E) Amphipod crustaceans on a bryozoan-­encrusted seagrass blade. (F) Sea urchins on the remnants of overgrazed kelp stalks.

227

228

Ocean Ecology

Our understanding of marine primary production and its fate has evolved rapidly in recent de­cades. Perhaps most fundaCoastal Indian mentally, the history of strong emphasis on abiotic control of biological pro­cesses in oceanography is shifting to incorporate Coastal Pacific top-­down control as a central component. The importance of Polar Arctic, Polar Atlantic, and Polar Pacific grazing in controlling primary producers has assimilated slowly into biological oceanography, which is puzzling given that Polar Southern classic empirical and modeling studies from as early as 1935 Trades Atlantic concluded that most planktonic production was rapidly Trades Indian grazed (A.  R. Longhurst 2007). Current reviews reinforce ­these early findings, concluding that the median fraction of Trades Pacific water-­column primary production grazed is similar among Westerlies Atlantic regions of the world ocean at 49%–77% (Schmoker et  al. 2013) (figure 9.6). In benthic systems the figure is similar—­ Westerlies Pacific experimental exclusion of herbivores reduces biomass of benWesterlies Southern thic marine producers by an average of 68% (Poore et  al. 2012). A second advance in ecosystem modeling has been the 0 50 100 150 200 250 300 350 % Primary productivity grazed evolution of the initial focus on bulk or aggregate properties of life—­constructs such as phytoplankton and zooplankton—to the recognition that functional diversity within such Figure 9.6. ​Most of the world ocean’s primary production is broad categories is comparable to, and sometimes even rapidly grazed in a day. Bars show the percentage of primary production grazed daily across biogeographic realms of the greater in importance than, abiotic ­drivers in influencing bioopen ocean: black squares, boxes, and whis­kers represent the logical pro­cesses (Hood et al. 2006, Ptacnik et al. 2008, Momedian, 25th, and 75th quartiles, respectively, and open symbols kany et al. 2015, Duffy et al. 2017). are outliers (­after Schmoker et al. 2013). A key component of the feedback between herbivores and plants is nutrient regeneration. When an herbivore consumes plant tissue, some portion of the energy (carbon) and nitrogen fuel metabolism and growth, and the remainder is excreted in inorganic form and thus returned to the environment where it is again available for plant uptake and growth. As we have seen, the fraction of total production supported by this regenerated nitrogen defines the potential for export or harvest. In general, oligotrophic systems, such as the open-­ocean gyres and coral reefs, are supported in large part by regenerated nitrogen. In such systems, productivity can be very high even though plant biomass is low b­ ecause primary production is rapidly grazed and regenerated, that is, ­there is rapid turnover. By the same token, such oligotrophic systems cannot sustain high rates of biomass harvest ­because nitrogen to support new production is very ­limited. At the other extreme, upwelling systems characterized by abundant inputs of new NO3 from below the pycnocline support high levels of new production and consequently can sustain high levels of fishing or export, ­whether into the deep ocean, to fisheries, or onto adjacent land by seabirds depositing ­great quantities of guano. Coastal Atlantic

Control of Biomass Distribution and Productivity in Marine Ecosystems The green world hypothesis What controls the biomass and productivity of an ecosystem is among the most fundamental questions in ecol­ogy. An impor­tant landmark was the “green world hypothesis” of Hairston, Smith, and Slobodkin (1960), often abbreviated HSS, which stimulated de­cades of debate about the role of consumers in ecosystem structure and dynamics. The issues addressed by HSS ­were t­ hese: Given some level of abiotic resource input, what pro­cesses control the distribution of biomass at dif­fer­ent levels

Chapter 9 Ecosystems

229

of the food web? And how far do bottom-up and top-­down forces penetrate through the stack of trophic levels? Starting from some s­ imple observations of terrestrial systems, HSS proposed a very general theory of what they termed “population regulation,” which was implicitly extended to the aggregate abundances or biomass of entire trophic levels. The green world argument is as follows. First, the global biomass of life as a ­whole is ­limited by resources (solar energy). Any population within that system that is not resource-­limited must by definition be ­limited to a level below that set by resources. Second, as a rule, terrestrial plants are abundant and not appreciably grazed, that is, “the world is green.” This suggests that plant populations are generally l­imited by the availability of resources, not by grazing. Third, since herbivores can clearly destroy vegetation when they are protected by h­ umans or introduced without enemies, it’s reasonable to conclude that most herbivores are normally ­limited by predation. Fi­nally, by controlling herbivore populations, predators (and parasites) are also drawing down their own resources, and thus are l­imited by ­those resources. In summary, ­whether a population is ­limited by resources or predators depends on its trophic level (figure 9.7): plants are l­imited by resources (from the bottom up), herbivores by their predators (from the top down), and predators by resources (bottom up). Note that this argument implicitly assumes that an ecosystem has three relatively discrete trophic levels and that the influence of predators extends through the herbivore trophic level to affect plants, an indirect interaction now known as a trophic cascade (Paine 1980) (chapter 7). The green world hypothesis is strikingly s­ imple, which may be considered both its genius and its Achilles’ heel. One of its virtues is that the theory makes two testable predictions (figure 9.7). The first and more general prediction is that alternate trophic levels should respond differently to resource enrichment and predator removal: increasing resources to stimulate productivity, for example, should increase biomass of populations at the top trophic level and at alternate levels below it, whereas intervening trophic levels should show no change in biomass but faster turnover rates. Second, standing biomass of plants should be high in food chains with an odd number of links (plants, grazers, and primary predators) and low in chains with an even number of links (an additional level of top predators). We explore empirical support for Carnivores ­these predictions below and in ­later chapters.

Bottom-up control of biomass and productivity by resources What pro­cesses control the quantity of biomass in a system and its distribution among trophic levels? The endpoints are commonly characterized as bottom-up control, in which biomass is set by resource availability (the green world), and top-­down control, in which it is ­limited by consumer pressure. The question has fundamental practical importance as the ocean increasingly ­faces major changes in both bottom-up forcing via massive nutrient subsidies from industrial fertilizer and top-­down forcing via widespread depletion of large consumers by fishing. At the most fundamental level, all ecosystems are controlled from the bottom up, insofar as ­there would be no ecosystem without an external energy source, the sun, and resources usable by primary producers (Hairston et al. 1960, Power 1992). Primary production ultimately determines every­thing ­else that happens in an ecosystem: the accumulation of plant and animal biomass, their relative proportions, the fractions thereof that are sustainably harvestable, the rates of carbon uptake and storage, and so on. Identifying what controls biomass in ecosystems begins with the basic physical ­drivers of all ecological pro­cesses: physics, temperature, and resource availability. The w ­ hole character of marine life and ecosystem dynamics is set by the greater density of w ­ ater than air, a fundamental difference between terrestrial and pelagic systems (chapter  2). The

Herbivores

Autotrophs

Figure 9.7. ​The green world hypothesis (Hairston, Smith, and Slobodkin 1960) proposes that biomass of each trophic level, represented by the size of the circles, is controlled by resources available to the highest level of the food chain. Thus, biomass of successive trophic levels down the food chain is controlled by top-­down predation and bottom-up resource availability (see text). Specifically, plant biomass is controlled by resources in the absence of animals (column 1) and where top predators depress herbivore abundance (column 3), but is controlled by herbivory in the absence of predators (column 2).

Ocean Ecology

density of ­water allows primary producers to float and capture light without the need for the large, metabolically inert structural tissues that land plants need to compete for light (Steele 1991). Thus, the dominant autotrophs of the ocean are microscopic algae. They are fast-­growing, and highly nutritious relative to land plants, but by the same token they are highly vulnerable to grazing, leading to stronger top-­down control generally and, more particularly, stronger trophic cascades in marine systems (Steele 1991, Cebrián 1999, Shurin et al. 2002). Global-­scale patterns of primary production vis­i­ble from satellites (see figure 1.4) flow up the food web to produce similar global distributions of animal biomass and production. Compilations of published data show that consumption by herbivores generally keeps pace with primary production across dif­fer­ent ecosystems (figure 9.8). This is b­ ecause the variability in primary production across systems exceeds the variation in percentage consumed or decomposed (Cebrián and Lartigue 2004). Thus, on global or regional scales, the signal of bottom-up control is clear. The rate of primary production is the fundamental control on flux of carbon up the food web and through the ecosystem. Net primary production is also closely correlated with herbivore biomass in global comparisons across ecosystem types (figure 9.8B) and with fish production on regional scales. For example, along the margin of western North Amer­i­ca, annual fish catch rises with the fraction of primary production retained in the coastal region, as estimated from satellite-­derived chlorophyll (Ware and Thomson 2005) (figure 9.9). Correlations among chlorophyll, zooplankton, and fish yield along coastal British Columbia confirmed strong bottom-up trophic linkages across three levels between phytoplankton, zooplankton, and resident fish. Similar trends are evident in the covarying abundances of dif­fer­ent trophic levels through time in the North Sea (Aebischer et al. 1990). But while the global and regional-­scale distribution of primary productivity is closely related to ocean physics, the fates of that primary production differ greatly among marine ecosystems dominated by dif­fer­ent functional groups of producer taxa (Duarte and Cebrián 1996, Cebrián 1999). That is, physics and biodiversity jointly mediate the basic structure and functioning of marine ecosystems, with the influence of abiotic ­factors declining and that of biodiversity increasing as our spatial scale narrows. Compilations of data show that direct grazing by herbivores is highest on microalgae,

(A)

(B)

Herbivore biomass (g C m–2)

100 Primary production consumed (%)

230

10

1

0.1

0.01 0.01

1 100 10,000 Net primary production (g C m–2 yr –1)

10

0.1

0.01

1

100 10,000 Net primary production (g C m–2 yr –1)

Figure 9.8. ​On a global scale, plant consumption and herbivore production are controlled from the bottom up. Data are compiled from a wide range of terrestrial and aquatic ecosystems. Open symbols represent aquatic systems and filled symbols represent terrestrial communities (­after [A] Cebrián and Lartigue 2004; [B] Cebrián 2004).

Chapter 9 Ecosystems

(B) 120

80 60 40 20 0 2 4 6 8 Mean chl-a concentration (mg • m–3)

Mean resident fish yield (metric tons km–2)

100

0

(C) 4

Mean resident fish yield (metric tons km–2)

Mean zooplankton biomass (mg dry wt. m–3)

(A)

231

2

0

0

20 40 60 80 100 120 Mean zooplankton biomass (mg dry wt. m–3)

2.0

1.0

0.0

0 2 6 4 Mean chl-a concentration (mg • m–3)

Figure 9.9. ​Regional fishery yield is controlled by bottom-up resource flux in the Northeast Pacific. (A) Phytoplankton biomass (chlorophyll a) predicts zooplankton biomass, which (B) predicts fish yield, such that (C) fish yield is strongly related to chlorophyll (­after Ware and Thomson 2005).

both pelagic phytoplankton and benthic microalgae, at > 40% of net primary productivity (NPP). As a result, microalgal-­based marine ecosystems pro­cess nearly all production within the system and average very low rates of export and storage of carbon. Conversely, grazing is lowest on higher plants, such as seagrasses, marsh grasses, and particularly mangroves (see figure 5.3). ­Because of their low vulnerability to herbivory, coastal systems dominated by higher plants tend to export a substantial fraction of their NPP as detritus and store 10%–17% of NPP in sediments. For this reason, conservation and restoration of coastal mangroves and seagrasses have attracted intense interest in the context of carbon and climate policy. ­These patterns of variation in the fate of carbon among producer functional groups are mirrored in an in­de­pen­dent meta-­analysis of experimental data on herbivore impacts on benthic plants (Poore et al. 2012) (see figure 2.12). ­These similarly show that herbivore impacts on benthic plant biomass are best predicted by plant functional groups, with ­little relation to large-­scale gradients in temperature or nutrient supply. The key to this variation among producer groups appears to be their nutrient content, which emerges as the single most impor­tant organismal characteristic controlling the fate and pathways of production in ecosystems. Producers with high nutritional quality (high ratios of N:C and P:C) are more heavi­ly grazed and more rapidly decomposed, channel less production to detritus, and store proportionally less in sediments than do less nutritious plants, particularly the higher plants that dominate coastal marshes, seagrass beds, mangrove forests, and most terrestrial ecosystems (Cebrián 1999, Cebrián et  al. 2009) (see figure 5.4). To summarize, we may distinguish ecosystem controls at dif­fer­ent scales. At the largest scale of the globe or comparisons among regions, the major patterns of production and flux are determined by environmental gradients—­ultimately, by physics of the ocean and atmosphere; and absolute magnitudes of consumption by herbivores and detritivores follow ­these patterns. Ecosystem state is fundamentally set from the bottom up.

Top-­down control of biomass and productivity by consumers At large scales, then, ecosystem biomass and productivity are set by temperature, resource availability, and ocean physics. But within t­ hese abiotic limits, the relative biomass of producers and consumers often differs strongly as a function of the activities of consumers. More generally, as we move from regional to local spatial scales, the structure and functioning of ecosystems are more strongly influenced by biodiversity, broadly including the number and discreteness of trophic levels, and the number and identity of

232

Ocean Ecology

3

C1

”Vertical” biodiversity

Top carnivore

O1

O2

Omnivores H1

2

H2

Herbivores

1

P1

P2

P3

Edible plants

P4

P5

P6

Inedible plants

”Horizontal” biodiversity

Figure 9.10. ​Vertical and horizontal components of biodiversity in a schematic food web. Vertical diversity is mea­sured by a mean food chain length of 2.58 (averaged across all eight food chains in the web), and illustrated by the presence of species with smaller (herbivores) and larger (omnivores) vertical niche breadth. Components of horizontal diversity include, at the basal level, two functional groups; and consumer species with narrow (eating one prey type) and broader (eating three prey types) horizontal niche breadths. For clarity, competitive interactions are not shown (­after Duffy et al. 2007).

functional types of organisms within t­hose levels (Duffy et  al. 2007) (figure 9.10). We focus first on the vertical component, trophic structure, and l­ater on functional diversity within levels. The first rigorous test of the green world hypothesis came from the pelagic communities of lakes. In the 1970s new phosphate detergents began flooding into freshwater ecosystems, causing thick blooms of nuisance algae. Yet comparisons among a large number of lakes showed a curious pattern: supply rates of phosphorus explained less than half the variance in primary production, and lakes with similar P supply differed up to a thousandfold in productivity (Schindler 1978). Attention began to focus on the pos­si­ble role of variance in the food web. Experimental removal of predatory fishes from ­whole lakes ultimately produced strong evidence for trophic cascades, sometimes penetrating through four trophic levels. In the Upper Midwest of North Amer­i­ca, stocking lakes with top predators, such as largemouth bass, reduced biomass of the small planktivorous fishes they ate, increased biomass and body sizes of grazing zooplankton, and reduced phytoplankton biomass relative to lakes without t­ hese top predators (Carpenter et al. 1987). In essence, top predators buffered the lakes against the effects of nutrient pollution.

Trophic cascades in the ocean

In the ocean, the most celebrated and best-­documented example of a system-­wide trophic cascade comes from the patchwork of lush kelp beds and urchin barrens across the remote Aleutian Islands of the Northeast Pacific (box 7.2). In the 1960s James Estes realized that the urchin barrens ­were found mainly on islands without sea otters, which had been hunted to extinction on ­those islands accessible to Rus­sian fur traders in the eigh­teenth c­ entury. ­These local extinctions had far-­ reaching consequences for ecosystem structure and functioning. In the absence of otters, populations of their sea urchin prey exploded, and urchins in turn grazed down the dense kelp beds to bare rock surfaces. The urchin barrens ­were largely devoid of fishes, and the flow of kelp detritus dis­ appeared, removing support for the other­wise rich assemblage of benthic suspension-­feeding invertebrates (Duggins et al. 1989) as well as the subsidy of marine production to nearby terrestrial habitats (Anthony et al. 2008). All of this happened with no obvious change in ocean physics or external supply of nutrients. The changes ­were clearly wrought by the loss of a top predator—­they ­were controlled from the top down. This system also provides among the most spectacular confirmations of the green world prediction that plant biomass should differ in food chains with odd versus even numbers of links (Hairston et al. 1960). During the 1990s killer ­whales began feeding on sea otters in the Northeast Pacific (Estes et  al. 1998), and the shift in top predator feeding habits appears to have strongly depressed the sea otter population, again initiating a trophic cascade that released grazing urchins from predator control and decimated kelp beds that had flourished for years. Biomass declined at odd levels below the top predator and increased at even levels. Hundreds of experiments have now manipulated consumers and resources, in nearly ­every type of ecosystem, in attempts to assess the generality of top-­down control. Marine herbivores often strongly reduce primary producer biomass by 60%–70% on average in all major types of marine pelagic ecosystems (Calbet and Landry 2004) and in benthic systems (Poore et al. 2012; see figure 2.12). ­There is also abundant evidence for the next link in the chain, predator control of herbi-

Chapter 9 Ecosystems

vores, and for penetration of this control through the food chain to influence plant biomass (Borer et al. 2006). Cascading effects of experimental predator removal are strongest in lakes, ponds, and benthic marine communities but relatively weak in the marine pelagic (Shurin et al. 2002). Cascading top-­down control of primary producers is supported by time series data from a range of marine systems, which suggest that bottom-up and top-­down control alternate among adjacent trophic levels (figure 9.11). In the pelagic ocean, gelatinous zooplankton—­including jellyfish, ctenophores, and salps—­appear to be especially impor­tant predators in such cascades (Verity and Smetacek 1996). ­These animals have voracious appetites and grow rapidly, which allows them to respond quickly to favorable environments. In the northeastern USA, ctenophores (comb jellies) are voracious predators of other zooplankton and often reach very high densities in summer. A six-­ year field study showed generally synchronous but opposite fluctuations among trophic levels consistent with top-­down control of zooplankton by ctenophores and of diatoms by crustacean zooplankton, creating a cascade whereby ctenophores increased the abundance of phytoplankton (Deason and Smayda 1982). Similar time series evidence for trophic cascades comes from kelp beds (Estes et al. 1998, Davenport and Anderson 2007), coral reefs (Dulvy et al. 2004), seagrass beds (Baden et  al. 2012, B.  B. Hughes et  al. 2013), estuaries (Eriksson et  al. 2011), open-­ocean plankton (Shiomoto et al. 1997), and the demersal (Worm and Myers 2003, Myers et al. 2007) and pelagic communities (Frank et  al. 2005) of continental shelves. Th ­ ese examples involve a wide range of organisms and environments, and suggest that systematic depletion of predators from the oceans is producing far-­reaching top-­down impacts on the structure and functioning of many marine ecosystems (Baum and Worm 2009). For addressing the relative importance of bottom-up and top-­down forcing, the most valuable studies manipulate both resources and consumers factorially. Elizabeth Borer and colleagues (2006) synthesized data from 173 such factorial experiments, along with patterns of covariation among trophic levels along productivity gradients in nature. Perhaps surprisingly, fertilization increased plant biomass but had no effect on herbivores, on average, whereas predator removal consistently changed biomass of both herbivores (directly) and plants (indirectly). In other words, bottom-up influence declined more quickly through the food chain than did top-­down influence, and this result did not vary appreciably among terrestrial, freshwater, and marine systems. The abundant evidence for widespread trophic cascades in marine ecosystems has impor­tant implications in a human-­dominated ocean ( Jackson et al. 2001). The small populations and slow growth rates of top predators raise their risk of extinction relative to species at lower levels and may lower their resilience to demographic perturbation. This demographic vulnerability is compounded by ­human harvesting, which disproportionately targets large animals near the top of marine food chains (Botsford 1997, Pauly et al. 1998, Jackson et al. 2001). B ­ ecause most ecosystems support only a few species of apex predators, this combination of demographic vulnerability and targeted ­human pressure means that marine (and other) ecosystems ­under h­ uman influence are most vulnerable to loss of an entire functional group or trophic level at the top of the food web (Pauly et al. 1998, Jackson et al. 2001, Duffy 2002). The result is trophic skew, a vertical compaction of the trophic pyramid due to losses of higher-­level species (Duffy 2003, Byrnes et al. 2007) with consequent weakening of top-­down control.

Detritus-­consumer interactions Primary producer biomass that is not grazed while the plant is alive enters the food web as detritus, that is, nonliving organic ­matter (see figure 5.1). A large proportion of primary production in systems dominated by vascular plants meets this fate. Interactions between detritus and its consumers differ from ­those between plants and herbivores in the impor­tant distinction that detritus is dead and therefore its supply rate and quality are not influenced by the consumer. Such systems are

233

(A)

(B)

10,000 8,000 6,000

50 25 0

2,000

800

0

20 40 60 80 100 120 140 160 180 Zooplankton (mg m–3 mo–1)

200

–

600

–

400 200

180 160

0

140

4,000

120 Spartina biomass (g m–2)

Zooplantkton (mg m–3 mo–1)

75

4,000

Snail density (adults m–2)

Skeletonema (cells ml–1 mo–1)

1972 1973 1974 1975 1976 1977

Snail mortality (% loss day–1)

100

12,000

100 80 60 40 20 0

0

2

4 6 8 10 Ctenophores (m–3 mo–1)

12

2 3 4 Ctenophores (m–3 mo–1)

5

3,000 2,000 1,000 0

Tall zone

Short zone

Skeletonema (cells ml–1 mo–1)

3,500 3,000 2,500 2,000 1,500 1,000 500 0

0

1

Figure 9.11. ​Trophic cascades from predators to primary producers in dif­fer­ent coastal ecosystems. (A) In Narragansett Bay estuary, USA, zooplankton control the phytoplankter Skeletonema (top), and predatory ctenophores control zooplankton (­middle), such that ctenophore blooms are associated with higher phytoplankton abundance (bottom) (­after Deason and Smayda 1982). (B) In salt marshes of Georgia, USA, grazing snails in the cordgrass tall zone nearest open w ­ ater suffer strong predation by shell-­crushing crabs and terrapins, reducing their density and allowing much greater accumulation of cordgrass (Spartina) biomass (Silliman and Bertness 2002).

Chapter 9 Ecosystems

Particulate detritus (morphous)

Animals digest morphous particles Colonization by microbes

Animals digest microbes

Dead plant tissue Uptake by microbes Formation of amorphous particles

Dissolved organic matter

Animals digest amorphous particles

Physicochemical precipitation

Figure 9.12. ​Pathways of detritus formation from decaying coastal vegetation and subsequent consumption (­after Mann 1988).

sometimes called donor-­controlled. Detritus can be classified into two forms: morphous detritus, mainly fragments of dead plant or animal tissue, and amorphous detritus, formed by the precipitation of dissolved organic ­matter onto surfaces or aggregates (Mann 1988) (figure 9.12). Detritus is more than simply a low-­quality food that accumulates where plants are not grazed. Detritus of vari­ous sorts plays important and varied roles in energy flow, community structure, and dynamics in many systems as a result of both its food value and its provision of habitat. On a global scale, a far larger share of energy and carbon move through the detritus food web than through the green food web, with the most obvious exception being pelagic systems. The detritus-­based food web—­often referred to as the brown web—­dominates the green web (based on grazing of living plants) as the primary conduit of production to higher trophic levels in habitats dominated by vascular plants, including most terrestrial systems, mangrove forests, salt marshes, and many seagrass meadows. But detritus also dominates flows of production in some macroalgal systems, such as kelp beds, where the dominant plants are ­little grazed. Detritus is the source of food in both the ­water column and benthos of the deep ocean below the photic zone. Indeed, the largest habitat on planet Earth is the lightless deep ocean where t­ here is no photosynthesis and the g­ reat majority of organisms feed on detritus, often highly degraded, that has drifted down from surface w ­ aters. Even in well-­lit surface w ­ aters dominated by phytoplankton, amorphous detritus can be a significant source of food for heterotrophs b­ ecause vigorously growing phytoplankton leak substantial quantities of dissolved organic ­matter (DOM) and zooplankton release phytoplankton cell contents while feeding. Much of this DOM is rapidly taken up by bacteria, but a significant part of it precipitates into suspended aggregates (“marine snow”) or ­settles out on particle surfaces in the w ­ ater column, where it becomes available as food for other heterotrophs. Aggregation of vari­ous forms of detritus—­amorphous material, dead phytoplankton cells, zooplankton fecal pellets—­can also facilitate sinking and export of fixed carbon to the deep ocean. The source of plant detritus strongly affects its availability to consumers and its ecosystem fate, and is most strongly related to nitrogen content. Vascular plant detritus entering the food web tends to be low in nutritional quality owing to its low N content and high composition of structural carbohydrates that are poorly digested by most animal consumers, whereas detritus from macroalgae and

235

236

Ocean Ecology

especially from microalgae is much more readily assimilated by detritivores. Plant detritus changes with time in the environment: particle size is gradually reduced and chemical composition alters as more labile components are initially consumed, leaving refractory (indigestible) components of aged detritus, which serve primarily as habitat for small organisms (Mann 1988). Assimilable material is extracted from detritus mainly by bacteria and fungi that colonize and digest structural lignin and cellulose that are unavailable to most metazoan consumers. Th ­ ese microbes often also enrich the nitrogen content b­ ecause many bacteria are able to take up inorganic nitrogen from the environment. The passage of detrital carbon through food webs is facilitated by this pro­cess of detrital conditioning, and the microbes themselves often provide a significant part, but by no means all, of the nutrition to detritivores. Most marine detritivores are deposit-­feeders, so named ­because they consume organic ­matter that has deposited on the seafloor from the overlying ­water column. Deposit-­feeders feed ­either from surface sediments, ingesting recently deposited and therefore relatively fresh organic m ­ atter, or from below the sediment surface. In e­ ither case, they may select certain types of particles passively (e.g., ­those that adhere to mucus or resist flow) or actively (by sorting with their mouthparts before ingestion). Passive se­lection is prob­ably most impor­tant in most deposit-­feeding invertebrates. Deposit-­ feeding detritivores, mainly worms, face somewhat dif­fer­ent prob­lems than herbivores that consume live phytoplankton or macroalgae. For deposit-­feeders, food is generally plentiful but low in quality, specifically in nitrogen, so they are often even more N-­limited than herbivores. The rate-­limiting step in deposit-­feeder energy gain is digestive assimilation, often of protein nitrogen (Tenore 1983). A common adaptation to this situation includes a voluminous gut—up to 80% of body volume—­that allows extraction of nearly all usable nutrition from the food before egestion. Food se­lection by deposit-­feeders often ­favors small particle size ­because smaller particles have more surface for microbial and organic coating, and low specific gravity, which indicates higher organic relative to mineral content. While research on detritus has focused mostly on the brown food web resulting directly from detritus consumption, detritus has a broader role in ecosystems, affecting the green food web as well, with effects spread across all trophic levels. Both experimental additions of detritus and field surveys show that increasing detritus generally stimulates primary production, likely through release of nutrients, and increases biomass not only of animals that feed on detritus but of living plants, herbivores, and predators by providing both food and habitat (Hagen et al. 2012). One key and relatively unique role of detritus in communities and ecosystems is that it does not consume energy and thus can serve as a reservoir of energy and nutrients available to supplement the needs of consumers. Depending on source and composition, components of this reservoir may have residence times of many years (Moore et al. 2004).

Functional Structure of Marine Ecosystems Organismal traits in ecosystems To understand production, consumption, nutrient use, and decomposition, we need to know what kinds of organisms populate the ecosystem, specifically their functional traits (see figure 5.9). As is also true of community assembly, a frontier in ecosystem ecol­ogy is developing a predictive framework for translating functional variation among types of organisms into system-­level productivity, efficiency, and stability (Norberg et al. 2001, Loreau 2010). The distribution of such functional traits in a community results from the interplay of vari­ous sorting pro­cesses forced by environment and biogeography and mediated by physiological tolerances, competition, and dispersal (chapter  8). Thus, environment and biodiversity are tightly linked to one another and to ecosystem pro­cesses as a complex adaptive system (Levin 1998, Norberg 2004).

Chapter 9 Ecosystems

Two of the traits that most consistently influence ecological pro­cesses are body (or cell) size and stoichiometry. Smaller-­bodied organisms generally have higher metabolic rates, shorter generation times, eat smaller prey, are more vulnerable to consumers, are more abundant, and produce more biomass (chapter 5). Tissue nutrient content, particularly the relative requirements for N and P, are consistently related to competitive ability ­under nutrient limitation, vulnerability to herbivory, and decomposition rate. Plant functional groups differ in size, nutrient stoichiometry, and investment in support structures, profoundly affecting trophic structure and pathways and magnitudes of ecosystem carbon flow (see figures 2.12, 5.3, 5.4). Within functional groups, variation in traits can also have far-­reaching consequences for ecosystem pro­cesses. For example, the nutrient requirements of phytoplankton differ among higher taxa of algae, reflecting anciently evolved differences in stoichiometry (Quigg et al. 2003). Although algal species can adjust nutrient requirements to some degree as environmental conditions change, much of the variance in phytoplankton nutrient needs is genet­ ically determined. As a result, dif­fer­ent algal taxa are favored ­under dif­fer­ent nutrient conditions (Sommer 1994) (see box 5.3), and phytoplankton species composition strongly influences CO2 drawdown from the atmosphere (Arrigo et al. 1999) and carbon sedimentation into deep ­water ­(Sieracki et al. 1993). Thus, variation in the magnitude and fate of marine primary production is affected strongly by the kinds of phytoplankton that dominate. Moving up the food chain, primary consumers similarly include a range of functionally diverse taxa. The functional traits of grazers influence trophic transfer and the fate of production. In the open pelagic ocean, for example, zooplankton graze algae and aggregate their remains into fecal pellets that sink much faster than individual algal cells do, accelerating export of production into the deep ocean. ­ hether crustaceans or gelatinous grazers The efficiency of that biological pump depends strongly on w (salps) dominate. The biological pump is much stronger when salps dominate b­ ecause of their ability to bloom rapidly, feed on the very small picoplankton that dominate oligotrophic w ­ aters, and produce large, rapidly sinking fecal pellets. A model of the oligotrophic Sargasso Sea, where production is dominated by picoplankton such as Prochlorococcus, found that flux of organic ­matter to the deep sea roughly tripled during salp blooms, relative to the normal circumstance where grazers are primarily protists and crustaceans (Michaels and Silver 1988). At the highest levels of the food web, the cascading influence of predators on ecosystem structure and plant biomass depends strongly on feeding selectivity and the degree of omnivory (Duffy et al. 2007, Bruno and Cardinale 2008).

The size spectrum The general scaling of metabolism and interactions with body size implies that the distribution of body sizes within a community influences ecosystem pro­cesses. Understanding this scaling begins with an empirical pattern that is consistent across many types of organisms and systems: if we mea­ sure and plot the sizes of all the individuals of all species in a community, their log abundances generally decline linearly with log body mass according to roughly the −¾ power (figure 9.13). This reflects the greater energy requirements of larger individual organisms and the consequently low densities of large species that a given amount of resources can support (chapter 5). This relationship between abundance and body size within a community is known as the size spectrum, a pattern that takes a ­limited range of forms across communities and systems, suggesting that it is underlain by fundamental biological rules. Explaining size spectra involves three princi­ples. First, as already noted, a given amount of resource can support more small organisms than large ones. Second, larger organisms generally eat smaller ones, so larger organisms tend to occupy a higher level in the food chain than smaller ones do ( Jennings, Warr, et al. 2002, Cohen et al. 2003) (see figure 6.5B). Third is the princi­ple of ecological or trophic efficiency: most of the energy an animal ingests is lost to respiration and excretion, whereas only a fraction—­typically considered 10% as a rule of thumb—is converted into its

237

Ocean Ecology

(A) 1.E+12

Marine abundance, all 175 samples Maximum terrestrial abundance Maximum marine abundance, each of 11 years

1.E+10 Abundance

1.E+08

β = – 0.76

1.E+06 1.E+04 1.E+02 1.E+00

1.E–02 1.E–10 1.E–08 1.E–06 1.E–04 1.E–02 1.E+00 1.E+02 1.E+04 1.E+06 1.E+08 Body mass (B) β = – 0.77 Log abundance

4 2 0 –2 –3

–2

–1 0 Log body mass

1

2

(C) β = – 1.13 Log abundance

238

10

5

0

–5

0 Log body mass

5

Figure 9.13. ​(A) Population abundance scales negatively with body mass across a range of organisms, often but not always to the −¾ power (log-­log). ­These include land plants and marine phytoplankton (­after Belgrano et al. 2002), (B) Chilean rocky shore invertebrates (­after Marquet et al. 1990), and (C) animals of the Ythan estuary, UK (­after Reuman et al. 2009).

own biomass (chapter 5). In most cases (but not all, as we ­will see), ­these three princi­ples operate together such that the base of the pyramid consists of a large biomass of small organisms low in the food chain, and the apex of the pyramid is large-­bodied, but rare, top predators. The pioneering ecologist Charles Elton (1927) first drew attention to ­these regularities as the “pyramid of numbers” (figure 9.14). The general relationship between an organism’s body size and its ecological function motivated the long tradition of characterizing communities by body size in both marine plankton (Sheldon

Chapter 9 Ecosystems

Log (M)

M or TL Biomass (B)

Log (B)

Abundance (N)

N ∝ M˜–1.2

Inverted biomass pyramid

Biomass M or TL

Body mass (M) or trophic level (TL)

Numbers

Log (N)

(C)

(B)

B ∝ M0

Log (M)

Figure 9.14. ​Pyramids of numbers and biomass in marine communities. Schematic illustration of (A) typical pyramid of numbers and (B) typical and (C) inverted pyramids of biomass resulting from dif­fer­ent combinations of trophic transfer efficiency and ratio of consumer to prey body mass (­after Trebilco et al. 2013).

et al. 1972) and benthos (Schwinghamer 1981, Gerlach et al. 1985) (see figures 2.8, 2.10). Classical size spectra are clearest in aquatic systems for two reasons. First, compared with terrestrial systems, consumer size and prey size are more tightly correlated in aquatic systems, in which most consumers are gape-­limited (­limited by mouth dimensions), ingesting their prey w ­ hole. In aquatic systems, small organisms are generally eaten by larger organisms and the ratio of (log) predator to (log) prey size shows strong regularities, although it varies among systems and functional groups (Brose et al. 2006) (see figure 7.7). Size spectra capture ­these individual size distributions and, ­because trophic ecol­ogy is correlated with body size, they tend to capture trophic structure as well (Cohen et  al. 2003). The second reason that size spectra are clearest in aquatic systems is that upper trophic levels are dominated by fishes, which generally have indeterminate growth, meaning that they grow through a wide range in body mass during their lifetimes. The ecol­ogy of marine fishes changes accordingly, from eating microscopic plankton as newly hatched larvae to eating other larger animals as adults. As a consequence, ­there is generally a strong relationship between body size, trophic level, and abundance in fishes ( Jennings et al. 2001) (see figure 6.5). Size spectra are more consistently pyramid-­shaped for metabolic pro­cesses, such as production, than for state variables, such as biomass. ­Because the first and second laws of thermodynamics (conservation of energy and increasing entropy, respectively) dictate inefficient transfer of energy from prey to consumers, pyramids of production are always bottom heavy, with greater primary production by plants than secondary production by herbivores, which is greater than tertiary production by predators, and so on. For this reason, size spectra can be used to describe the flux of energy from primary producers through food webs ( Jennings et al. 2008). The shape of the biomass spectrum can be strongly influenced by the distribution of individual body sizes within the community. In extreme cases, the higher specific (i.e., per unit biomass) productivity of small organisms can result in the biomass of upper trophic levels exceeding that of plants (Trebilco et al. 2013). This is ­because metabolism, which includes rate of biomass production, scales as a negative power function of body size (chapter 5). In other words, although a small organism is less productive and requires less energy per individual than a larger one, small organisms are more productive per unit biomass. Thus, a small

239

240

Ocean Ecology

standing biomass of short filamentous algae, turning over rapidly ­under heavy grazing, can in theory sustain a large biomass of consumers, resulting in an “inverted” biomass pyramid (figure 9.14C). This is the explanation often advanced for the abundance of large sharks (with low metabolic rates) on remote, unfished coral reefs, where algae are microscopic (hence with rapid metabolism) and appear scarce (Stevenson et al. 2007). However, energetic considerations suggest that such inverted biomass pyramids occur only ­under extreme combinations of predator-­prey size ratios and trophic transfer efficiency values, which are unlikely among larger animals (Trebilco et al. 2013). The anomalously high predator abundance on oceanic reefs prob­ably often results from aggregation of predators supported by pelagic production in the surrounding ocean.

The macroecol­ogy of trophic interactions The energetic equivalence hypothesis (chapter 6) proposes that species sharing the same resource generally use a similar fraction of that resource regardless of body size (i.e., the populations are energetically equivalent). But applying this princi­ple more broadly to food webs requires at least two modifications. First, where larger organisms occupy higher trophic levels than smaller ones, as is generally true in marine systems, they do not share a common resource base, and the community-­wide scaling of energy flux with body mass is expected to diverge from energetic equivalence. In this case, species at higher trophic levels have smaller population abundances than predicted ­because the energy available to them is constrained by inefficient energy transfer through the food chain. Systems with multiple trophic levels, in which larger animals occupy higher trophic levels, thus show steeper community-­wide scaling of abundance with body mass than predicted in an assemblage sharing a common resource pool (a single trophic level). Large marine fishes usually occupy higher trophic levels than small fishes, and trophic level is more accurately predicted by individual size than by species ( Jennings et al. 2001). Thus, N scales with M−1.2 in marine fishes, rather than with M−0.75 as predicted by basic metabolic scaling for a guild of species using the same resources ( Jennings and Mackinson 2003), and departures from this relationship can signal impacts of fishing, which typically targets larger size classes (see figure 5.12). A second modification is required to accommodate parasites, which are inconspicuous but impor­tant players in most ecosystems. Parasites are unusual in food webs in being typically small-­ bodied yet occupying high trophic levels. ­Because of their anomalously high trophic level for their size, parasites diverge strongly from the ¾ scaling of abundance with body mass. But when the standard allometric equations are amended to include trophic level in addition to body mass and temperature as predictors, parasites fall in line with free-­living organisms. This was documented in a study of three southern California estuaries, where ecosystem-­wide sampling of all organisms showed that abundance scaled uniformly with body mass, with an exponent close to the predicted ¾, but only ­after accounting for trophic level (Hechinger et al. 2011) (figure 9.15). Expanding the metabolic theory of ecol­ogy to incorporate trophic level and life history, as well as body size and temperature, offers promise for bridging levels of ecological organ­ization, from individual to ecosystem. It may ultimately provide a key element in a general framework for modeling ecosystems mechanistically from first princi­ples. The covarying relationships are well illustrated by the intensively studied pelagic community of Tuesday Lake in Wisconsin, USA, which shows strong and consistent relationships among body mass, abundance, and trophic level: small-­ bodied species are low in the food web and numerically abundant, whereas larger-­bodied species are less abundant and occupy higher trophic positions (Cohen et al. 2003). Body mass remains the best predictor of abundance in this community, but trophic level improves the relationship and, as seen above for marine fishes and parasites, helps reconcile deviations from the allometric relationship between body size and abundance.

Chapter 9 Ecosystems

(B) 8

Log abundance (Ntemp hectare–1)

Log abundance (N hectare–1)

(A)

6 4 2 0 –2 –8

–6

–2 0 –4 Log body size (g)

2

4

8 6 4 2 0 –2

β = –0.74

–8

–6

–2 0 –4 Log body size (g)

2

4

Figure 9.15. ​Relationships between abundance and body size differ for parasitic and free-­living species in a southern California estuary. (A) A single regression line does not fit the data for abundance versus body size; parasitic (circles) and free-­living species (other symbols) diverge. (B) When abundance is statistically controlled for temperature and trophic level, parasites fall in line with free-­living species. The slope of abundance versus body mass is consistent with the −¾ power predicted by metabolic scaling (circles = parasites; crosses = free-­living invertebrates; squares = fishes; diamonds = birds) (­after Hechinger et al. 2011).

Biodiversity and the Functioning of Ecosystems Historically, community ecol­ogy has focused on the number and composition of species, and interacted l­ittle with studies of the fluxes of energy and materials through ecosystems. But in the 1980s growing awareness of threats to biodiversity stimulated a surge of research aimed at understanding how the changing numbers and composition of species influence the flows of energy and materials through ecosystems (Schultze and Mooney 1994, Loreau et al. 2001). Losses or gains of par­tic­ul­ar species can often change production, decomposition, trophic transfer, and nutrient cycling of entire communities (Chapin et al. 1997), as w ­ e’ve seen for the wide-­ranging trophic cascades triggered by predator loss (Estes et al. 2011) and for invasions by nonnative plants and animals (chapter 11). This recognition led to a vigorous research effort to find the common patterns and mechanisms linking ecological structure and functioning. Can we draw generalizations about how assemblages of interacting species influence emergent fluxes at the scale of ecosystems? In other words, how does biodiversity influence ecological pro­cesses? This is a macroecological prob­lem in the sense that it asks how emergent patterns arise from interactions among species that follow fundamental laws of biology. Like metabolic ecol­ogy (chapter 5), characterizing links between biodiversity and ecosystem pro­ cesses is based on mechanistic theory but focuses at a level above the interacting agents (individuals or species in this case) to explain properties of ecosystems as a function of the number of actors (species).

Biodiversity and ecosystem functioning: Theory Species that coexist stably within a community must by definition differ functionally. Therefore, as the number of species in a community increases, so should the diversity of their resource use traits (Tilman 1999, Loreau 2000). If this is true, then diversity of species can be used as a proxy for diversity of ecological functions. Monitoring data from kelp beds on 16 rocky reefs in California, USA, showed that species richness did indeed correlate with functional diversity (Micheli and Halpern

241

242

Ocean Ecology

2005). So species diversity in this case appears to be a reasonable proxy for the range of functions pre­sent in an ecosystem. But the strength of the correlation depends strongly on how narrowly functional groups are defined, that is, on how many traits form the basis of the classification (Petchey and Gaston 2002). Given t­hese functional differences between species, theory predicts that, in a heterogeneous landscape, a more diverse community w ­ ill be more efficient in exploiting resources, and therefore more productive, through one or both of two mechanisms (Tilman et al. 1997). First is a statistical sampling phenomenon: If species are distributed randomly among plots, the plots with more species have a higher chance of including the one species that performs best u­ nder the conditions of that plot. Over time this well-­adapted species ­will grow to dominate the plot, yielding more biomass. In contrast, plots of low diversity are less likely to include the species that does best ­there. ­Under ­these conditions, the high biomass yield of (initially) diverse plots results from the traits of the single species that takes over, not from interactions among the species. In contrast, the second mechanism involves complementarity among species, traditionally known as niche partitioning. If phytoplankton species, for example, grow best at dif­fer­ent light or nutrient levels, then the community of species w ­ ill often use resources more efficiently and produce more biomass than any species does alone (see box 5.1). In essence, more effective niche partitioning leads to more efficient pro­cessing of resources at the ecosystem level. W ­ hether the mechanism is sampling, complementarity, or some combination thereof, the diverse community uses a greater fraction of resources and is more productive, on average, than a community with fewer species (Tilman et al. 1997) (figure 9.16).

Biodiversity and ecosystem production: Empirical evidence Beginning in the early 1990s a large number of experiments sought to test the proposed effects of biodiversity on ecosystem functioning by manipulating species richness and examining how it affected ecosystem pro­cesses, primarily biomass production. Experiments in terrestrial grasslands confirmed that plots with higher plant diversity indeed produced more biomass (reviewed in Tilman 1999). Similarly, in the ocean, diverse assemblages of herbivores (Duffy et al. 2003, 2005) and predators (Byrnes et al. 2005, Douglass et al. 2008) also achieved greater productivity and resource use, and diverse benthic communities better resisted invasion by nonnative species (Stachowicz et al. 1999). ­These effects of diversity can ­ripple through food webs. Studies in a kelp forest in California, USA, explored how predator diversity affected the trophic cascade through sea urchins to kelp abundance (Byrnes et al. 2005). In this system, the major grazers w ­ ere kelp crabs, turban snails, and sea urchins, and the predators ­were two crabs and a sea star. Field surveys in the Channel Islands showed that, as predicted, sites with more predator species had lower herbivore abundance and higher kelp abundance (figure 9.17A). Conversely, predator abundance did not explain kelp or herbivore abundance. To explore the pos­si­ble role of predator diversity in producing t­ hese patterns, the researchers manipulated the number of predator species in mesocosms, stocking some tanks with a single type of predator and ­others with all of them, and taking care that all contained a similar total biomass of predators. They found that, as in the field, kelp indeed performed better where multiple predator species w ­ ere pre­sent (figure 9.17B). The reason appears to be that dif­fer­ent herbivores ­were vulnerable to dif­fer­ent predators, so that grazing on kelp was most strongly suppressed when multiple predator species ­were pre­sent. Many such experiments have since been conducted in a wide range of systems. Meta-­analysis of over a hundred experimental manipulations of diversity supports the conclusion that, on average, loss of species from a system generally reduces production and resource use in a surprisingly consistent way across a range of trophic groups and habitats (Balvanera et al. 2006; Cardinale et al. 2006, 2011; Griffin et al. 2013), including in marine systems (Stachowicz et al. 2007, Gamfeldt et al. 2014). Thus, we can conclude that diversity within trophic levels—­horizontal diversity—­generally increases

Chapter 9 Ecosystems

(A)

Complementarity

(C)

Sampling (selection)

Niche axis 2 (NO3 concentration)

10

Plant biomass

8

r

6 4 2

0

0

ar

0

1

2

3

Niche axis 1 (temperature)

4

5

6

7

8

9

10

11

9

10

11

Original species richness

(B)

(D) 10

Resource

Total plant biomass

8 6 4 2

0

5

10

15

20

Species richness

25

30

0

0

1

2

3

4 5 6 7 8 Original species richness

Figure 9.16. ​Diversity can increase community biomass and productivity in two ways. (A) Individual plant species (circles) grow best at unique combinations of environmental conditions (e.g., temperature and NO3). (B) In a heterogeneous environment, a community of multiple species (dots) occupies more of this two-­dimensional niche space and therefore produces more biomass than any one species does. That is, ­there is complementarity among species. Species combinations are more productive than even the best single species (rising red line). (C) When one species (red dot) is more productive than any combination, more diverse communities are more productive on average ­because they are more likely to contain (“sample”) that most productive species; in this case, the most productive community at each level of richness contains the dominant species and achieves similar productivity (flat red line). (D) Communities that produce the most biomass also tend to be most efficient, reducing resource levels more (­after Tilman et al. 1997).

average resource use and biomass production in many kinds of ecosystems. The mechanisms ­behind ­these positive effects of diversity involve both sampling and complementarity. Quantitative models built from such experiments found that biomass increases with species richness as a power function with mean β = 0.26, and that this relationship was robust to experimental design across trophic levels and kinds of ecosystems (O’Connor et al. 2017). Perhaps surprisingly, aquatic and terrestrial primary producers followed similar relationships, despite their very dif­fer­ent structure and life habits. The strength of the diversity-­biomass relationship did differ, however, among trophic levels. In aquatic systems, the slope of biomass on richness was nearly twice as high for herbivores and detritivores as for primary producers (O’Connor et  al. 2017), consistent with expectations (Duffy 2002). The steeper diversity effect among herbivores also holds for diversity effects on multifunctionality, that is, effects on multiple ecosystem pro­cesses considered together (Lefcheck et al. 2015).

243

Ocean Ecology

(A)

(B) 8 7 6 5 4 3 2 1 0 0.5

10 Change in kelp mass (g)

Kelp abundance (No m–2)

244

0.7 0.9 1.1 1.3 Predator diversity (Shannon-Weiner)

0

0

1

2

3

–10 –20 –30 –40

m p r

1.5 Number of predator species

Figure 9.17. ​The strength of the trophic cascade increases with predator species richness in a kelp forest community. (A) Field observations in California, USA, show that sites with high predator diversity have more kelp. (B) A mesocosm experiment showed a similar pattern: treatments with higher predator diversity resulted in better kelp survival b ­ ecause more diverse predator treatments ate more herbivores. Filled circles are mean values for each level of predator diversity, open circles are single predator species treatments, and open square had no herbivores (­after Byrnes et al. 2005).

As experiments confirmed a consistent positive influence of biodiversity on biomass production, attention turned to w ­ hether the same situation holds in the more complex situation of wild nature. Field observations have now confirmed a quite similar pattern in nature (figure 9.18). Among primary producers, data from freshwater and brackish phytoplankton show that resource use efficiency (estimated as total algal biomass per unit phosphorus) is positively related to phytoplankton diversity (Ptacnik et  al. 2008) (figure  9.18A, B), consistent with experimental results from Baltic plankton (Gamfeldt et al. 2005). Presumably, this is b­ ecause dif­fer­ent phytoplankton species perform best ­under dif­fer­ent combinations of light, nutrients, and other environmental f­actors (Sommer 1994; see box 5.3) such that diverse communities perform well across the range of ­these conditions. Most interestingly, phytoplankton diversity was a better predictor of resource use efficiency in ­these field studies than ­were environmental f­actors, including pH, temperature, salinity, and lake morphometry (Ptacnik et al. 2008). Positive effects of biodiversity on production have also been documented in other marine systems. Deep-­sea sediments support highly diverse food webs of microbes, invertebrates, and fishes that are key players in ocean biogeochemical cycles. Analy­sis of 116 deep-­sea sites around the globe found that production by benthic macrofauna and prokaryotes, and resource use efficiency, increased exponentially with faunal diversity (Danovaro et al. 2008). Along the world’s rocky and coral coasts, surveys of more than 4500 fish communities found that fish species richness and functional diversity ­were strong predictors of fish biomass (figure 9.18C, D), especially for the large-­bodied species preferred by fishers (Mora, Aburto-­Oropeza, et al. 2011, Duffy et al. 2016). Together sea surface temperature, biodiversity, and ­human influence explained almost half the global variation in fish biomass. ­These results have practical importance ­because fishes play a variety of impor­tant functional roles in marine ecosystems and provide protein for billions of p­ eople, especially in the developing world. The general message from both experiments and observational studies across a range of marine, freshwater, and terrestrial systems is that biodiversity—­variation in the kinds and numbers of species—­has just as strong an effect on ecosystem structure and pro­cesses as do climate and nutrient supply (Hooper et al. 2012, Duffy et al. 2017). Variation in ecosystem pro­cesses at local and even regional scales was often better explained by the kinds of organisms in the system than by variance in such abiotic ­factors. Of course, ­these ele­ments are not independent—­biodiversity has been molded over evolutionary and ecological time by ­those same abiotic ­drivers, resulting in well-­documented

Chapter 9 Ecosystems

(A)

(C)

Log fish biomass (g 500 m–2)

Algal biomass (C) P–1

100.0 20.0 5.0 2.0

FI SE NO BS

0.5 10

6

4

2 Tropical Temperate

0

20 50 Diversity (number of genera)

0

1 Log estimated richness

(D)

(B)

an m

Hu

ra T( SS

m T( SS

s

–0.25

rsi ty

MD

ive

Cont SA

ts

T

en

pH

od

NP

tri

TP

Bi

BD

ea

0

0.00

e)

10

0.25

Nu

20

0.50

ng

30

Direct Indirect

n)

Standardized effect size (β)

0.75 40 % of R2

2

Figure 9.18. ​Biomass and production increase with diversity in dif­fer­ent marine ecosystems. (A) Production efficiency of aquatic and estuarine phytoplankton in Scandinavia rises with diversity, and (B) biodiversity (BD) contributes more strongly to production than several environmental variables (% of R2 = percentage of variation explained by each variable; cont = continentality or the degree of inland isolation; MD = ln(mean depth); NP = ln(N/P); T = temperature; TP = total phosphorus; SA = lake/estuary surface area) (­after Ptacnik et al. 2008). (C) Global reef fish biomass increases with species diversity; and (D) diversity effects are comparably impor­tant to temperature (SST) and h­ uman impacts, effect sizes are sums of both direct and indirect paths of causality (­after Duffy et al. 2016).

correlations between latitude, temperature, biological diversity, and other ­factors (Tittensor et  al. 2010). But abiotic forcing is not enough to understand the structure and major flows of materials through ecosystems.

Biodiversity and ecosystem stability The stability of ecosystems—­their ability to maintain and restore structure and function in the face of disturbance—is of long-­standing interest in both basic and applied ecol­ogy. Naturalists from Darwin to MacArthur (1955) have concluded that diverse communities are more stable and better able to resist disturbances than species-­poor communities. The general reasoning was that, as in h­ uman economies, a community with varied ways of making a living is better able to weather change ­because its members respond differently to change, such that some of them thrive ­under any par­tic­u­lar set of conditions. Building on ­earlier qualitative arguments, theory has identified several mechanisms by

245

246

Ocean Ecology

which biodiversity can increase stability of communities and ecosystems (McCann 2000). The most general is s­ imple statistical averaging, also called the “portfolio effect,” drawing an analogy between species in a community and financial stocks: as long as the fluctuations of co-­occurring species through time are partially in­de­pen­dent of one another (i.e., not perfectly in sync), the variation in their combined abundance w ­ ill be lower in the face of environmental change compared with the average variability of the individual species (Tilman et al. 1997, Doak et al. 1998). Higher diversity can also stabilize ecosystems against perturbations by increasing the system’s ability to absorb a stress without changing (re­sis­tance) or by increasing the rate at which it returns to its original state a­ fter perturbation (resilience). This may happen in at least two ways. First, the broader range of species functional traits in diverse communities makes it more probable that at least some of ­these species ­will thrive as environmental conditions change. Second, the greater number of species in diverse communities means that ­there ­will generally be more species with any given functional trait. This phenomenon is known as functional redundancy and can enhance stability against species extinctions; ­because if a species is lost from a diverse community, t­ here are likely to be other (redundant) species that can do a similar job as the one that was lost. The latter two phenomena are sometimes combined as the insurance hypothesis (Yachi and Loreau 1999). Meta-­analysis of experiments spanning a range of systems shows that, on average, increasing diversity indeed enhances the stability of community properties through time ( Jiang and Pu 2009, Gross et al. 2014). Consistent with the theory, observational data also show that diverse phytoplankton communities have more stable average resource use efficiency (Ptacnik et  al. 2008), as mentioned in the previous section. Observational evidence from fishery science also supports the prediction that diversity can enhance stability of fishery yield in the face of fishing and environmental variation (box 9.1). In nature, species interact not only via competition with other species but also with predators at higher levels and with species below them in the food chain. Theory predicts that ­these horizontal and vertical components of diversity interact, often indirectly, to affect stability. At the most basic level, high prey richness should buffer the prey assemblage from predator impacts. Theory suggests that a prey assemblage (­whether plants or animals) with more species is more likely to contain at least one species resistant to consumption that can come to dominate in the presence of a consumer, such that a more diverse prey community w ­ ill maintain higher aggregate biomass u­ nder predation (Leibold 1989, 1996). This is a variant of the sampling mechanism described above. Prey richness can also buffer against predator impact via complementarity: if dif­fer­ent prey species are resistant to dif­fer­ent predators, then a diverse prey community can maintain higher biomass. Th ­ ese general predictions are consistent with meta-­analyses of experiments examining the effects of consumers on algae (Hillebrand and Cardinale 2004, Edwards et al. 2010) and on benthos of marine hard substrata (Edwards et al. 2010) (figure 9.19). ­These meta-­analyses synthesized data from experiments across natu­ral gradients in diversity, but w ­ ere consistent with experiments that manipulated species richness directly. The importance of diversity to stability differed strongly between experiments with single versus multiple trophic levels: systems with more species w ­ ere more stable at both population and community levels in multitrophic systems, whereas t­ here was no consistent effect of diversity in experiments using a single trophic level ( Jiang and Pu 2009). Th ­ ese patterns w ­ ere broadly equivalent across experimental and observational studies and across terrestrial and aquatic studies. They provide an intriguing hint that the vertical component of diversity—­the presence of multiple trophic levels—­increases ecosystem stability. The role of diversity in stabilizing multitrophic marine ecosystems is well illustrated by the case of coral reefs. The dominance of corals on tropical reefs often depends on strong grazing pressure by herbivores, which maintain low abundance of algae that might other­wise outcompete the corals for living space. A role for herbivore functional diversity in maintaining coral dominance was first

Chapter 9 Ecosystems

247

Box 9.1. ​The portfolio effect in the Alaskan salmon fishery

(A)

Streams

Window of all stocks combined

Rivers

Narrowest window of individual stock

Lakes 13 Jul

27 Jul 10 Aug 24 Aug

7 Sep

21 Sep

5 Oct

Date (B)

0.7 0.6

Low diversity

0.5 Probability

Many experiments suggest that more diverse communities are more productive and use resources more efficiently, but most have been conducted in simplified systems. Do ­these patterns hold up in the more complex world of wild nature? Several lines of evidence support theoretical predictions (Tilman et al. 1997) that diversity can enhance stability of production (or catch) in a multispecies fishery. For example, in North Sea fishes, trends in community biomass through time are less variable than t­ hose of individual fish species, on average (Jennings and Kaiser 1998). The best evidence for such portfolio or insurance effects of diversity comes from responses of Alaskan salmon to decadal-­scale climate variation. Sockeye salmon (Oncorhynchus nerka) in Bristol Bay, Alaska, have supported the most valuable fishery in the US for de­cades, with a landed value of nearly $8 billion between 1950 and 2008 (Schindler et al. 2010). B ­ ecause sockeye salmon return to the streams or lakes of their birth to spawn, populations show strong ge­ne­tic differentiation among areas. Several hundred discrete ge­ne­tic populations or salmon “runs” differ substantially in life history and ecol­ogy, and behave like distinct species, breeding in dif­fer­ent rivers and habitats at dif­fer­ent times and responding differently to climate forcing. The history of ­these runs provides evidence that their ecological and life history diversity produced a “portfolio effect,” that is, diversity reduced the variance in combined salmon catch over the long term. Variability in annual returns of Bristol Bay salmon w ­ ere lower by half, and fisheries closures ­were lower by 90%, over the last 50 years compared with expectations if the species consisted of a single homogeneous population (figure B9.1.1). The food and economic stability provided by this diversity is crucial to a range of stakeholders, including both commercial and subsistence fishers and the communities that depend on them. A similar pattern emerges at the global scale, where variability in total fishery catch declined with species richness across the world’s 64 large marine ecosystems (Worm et al. 2006). In par­tic­u­lar, diversity appears to stabilize fishery yield against climate variability. Across the globe, diverse fish communities sustained higher biomass ­under variable temperatures than did species-­ poor communities (Duffy et al. 2016). And this effect translated to fishery yield: in multispecies fisheries, diversity in thermal biology among the harvested species raised yield in the face of varying temperatures, compared with less diverse fisheries (Dee et al. 2016). ­These links between stock diversity and stability in a range of systems are consistent with theory showing that niche differentiation can stabilize aggregate biomass and production of communities (Loreau et al. 2001).

0.4 0.3 0.2

High diversity

Fishery closure Swamped capacity

0.1 0

1.5 0 0.5 1.0 Variability in catch through time (coefficient of variation)

Figure B9.1.1. Diversity increases stability and economic returns in the Alaskan salmon fishery. (A) Spawning win­dows for individual stocks. The white column shows narrowest win­dow of an individual stock; blue shading shows win­dow for all stocks combined. (B) Results of a model showing probability of fishery closure due to too few spawners (blue line) or swamped capacity due to overload of fleets (green line) ­under scenarios of low and high stock diversity (­after Schindler et al. 2010).

Ocean Ecology

(A)

(B) –6

2.0

–5 Consumer effect size (log response ratio)

248

–4

1.5

–3

1.0

–2 –1

0.5

0 0.0

1 2

0

10

20 30 40 50 Algal species richness

60

70

2

5 10 Prey species richness

20

40

Figure 9.19. ​Consumers have weaker effects on more diverse prey assemblages. Consumer effect strength is plotted as log(prey biomass exposed to consumers/prey biomass protected from consumers) for (A) herbivores feeding on algae (­after Hillebrand and Cardinale 2004) and (B) consumers feeding on sessile prey on marine hard bottoms (­after Edwards et al. 2010). Note that the consumer effect is shown as negative in (A) and positive in (B).

suggested by the observation that overharvesting of herbivorous fishes during the twentieth ­century had ­little effect on algal biomass u­ ntil sea urchins, the other major herbivore group, suffered mass mortality from disease in the 1980s. Deron Burkepile and Mark Hay (2008) directly tested the hypothesis that herbivore diversity buffers coral reefs against algal overgrowth using field experiments that allowed access by ­either single herbivore species, mixed species of herbivores, or no herbivores. Compared to treatments with single herbivore species, mixed herbivore assemblages reduced macroalgal abundance by > 50%, enhanced cover of the crustose coralline algae that are preferred recruitment sites for corals, increased coral cover by > 20%, and prevented coral mortality. Experiments demonstrated that ­these diversity effects arose from complementary feeding by herbivorous fishes. Specifically, the two major groups of herbivores, parrotfishes and surgeonfishes, fed on dif­fer­ent types of algae and both w ­ ere necessary to keep the substratum clear, allowing corals to ­settle and thrive. Stability is related not only to the range of functional traits in a community but also to the diversity of interactions among species. In the Bahamas, experiments showed that damselfish (Stegastes partitus) recruiting to coral patch reefs established much more stable populations when both competitors and predators ­were pre­sent than with ­either competitors or predators alone, or in their absence. The mechanism appeared to be that competitors and predators together induced density-­ dependent mortality among the damselfish recruits, reducing variation in their abundance by an order of magnitude (Carr et  al. 2002). Similarly, field experiments found that populations of the planktivorous damselfish Chromis cyaenea ­were also stabilized by complex interactions: juvenile damsels experienced density-­dependent (stabilizing) mortality only when exposed to both reef-­ resident predators and transient predators attacking from the ­water column (Hixon and Carr 1997). Fi­nally, experiments on an Irish rocky reef found that removal of ­either strong or weak interactors reduced temporal and spatial stability of the community (O’Gorman and Emmerson 2009). Together ­these experiments imply that the diversity of species interactions characteristic of intact marine communities regulates the sizes of fish and invertebrate populations, stabilizing the structure and functioning of the system. An impor­tant implication is that overfishing may destabilize t­hese systems by reducing predator abundance and diversity, with potentially cascading impacts (Hsieh et al. 2006).

Chapter 9 Ecosystems

Alternative Stable States and Regime Shifts in Complex Adaptive Ecosystems We have seen that the presence or absence of a predator can shift an ecosystem between two quite dif­ fer­ent states, as when the hunting of sea otters flipped northeastern Pacific rocky reefs from kelp forests to urchin barrens dominated by coralline algal crusts (see box 7.2). How such shifts between alternative states are triggered, and w ­ hether they are stable or resistant to change, have been topics of g­ reat interest for many years. ­There are limits to stability in all systems, and understanding them has taken on new urgency as the accelerating pace of global change provides abundant evidence of instability in ecosystems (T. P. Hughes et al. 2013). The perceived danger of “tipping points,” or sudden shifts among alternative community states, or regimes, is frequently highlighted in manifestos on conservation and management, for example, in the context of proposed planetary bound­aries (Rockström et al. 2009). How might such tipping points work? One of the characteristics of complex adaptive systems, including ecosystems and the biosphere as a ­whole, is path dependency—­the dependence of a system’s development on its starting state (Levin 1998), which can result in nonlinear changes, including rapid phase shifts among alternative states. A key prob­lem is to understand the conditions that foster such alternative states and their dynamics (Sutherland 1974, Scheffer et al. 2001). Regime shifts can be defined as relatively rapid transitions between distinct and relatively long-­lasting, semistable states (regimes), or alternative attractors, within a system (Steele 2004, Knowlton 2004). Regime shifts have been described in a number of marine ecosystems, and in pelagic systems often appear to track decadal-­scale climate variation (Overland et al. 2006). The key question from both a basic science and a practical management perspective is the extent to which t­hese regime shifts are simply responses to long-­period environmental forcing versus alternative states that are stabilized against change by internal biological pro­cesses. That is, are abrupt shifts between states forced by gradual changes in conditions? And are ­these shifts reversible? Answering ­these questions requires recognizing that regime shifts vary along a continuum; three general types have been described (Scheffer et al. 2001, Collie 2004) (figure B9.2.1A): (1) a smooth regime shift, defined by a roughly linear relationship between a forcing variable and a response variable; (2) an abrupt regime shift, in which the relationship is clearly nonlinear; and (3) a discontinuous regime shift, in which not only is the relationship nonlinear but the trajectory of the response variable differs when the forcing variable is declining compared with when it is increasing. The latter situation results in two pos­si­ble states of the response variable at a given value of the forcing variable, referred to as hysteresis, and is equivalent to what ecologists call alternative stable states. Regime shifts are inherently difficult to diagnose ­because of the complex dynamics thought to underlie them and the long time series required to recognize them against the background variability of the ecosystem. Nevertheless, where strong evidence is available, a regime shift can be identified by applying a series of criteria together (Scheffer and Carpenter 2003). In field data, ­these include abrupt shifts in time series, a bi-­or multimodal frequency distribution of states in a time series, and dual (or multiple) relationships between ecosystem state and a forcing variable. Experimental evidence includes dependence of final state on initial state (e.g., order of colonization during succession); shift ­toward a distinctly dif­fer­ent stable state ­after a pulse perturbation; and hysteresis, that is, change of the ecosystem along dif­fer­ent pathways when the forcing variable is increased versus decreased. Several convincing examples of alternative stable states have been identified in marine ecosystems (box 9.2). Yet well-­documented cases of alternative stable states are few, and systematic reviews of the lit­er­a­ture have found ­little empirical evidence that they are common ( Jones and Schmitz 2009), including in estuarine or freshwater ecosystems (Mac Nally et al. 2014). For example, the transition between kelp forest and urchin barrens driven by sea otters is not a transition between alternative stable states since the states are not stable—­recovery of sea otters allows recovery of kelp.

249

250

Ocean Ecology

Box 9.2. ​Alternative stable states in marine fisheries? and Carpenter (2003) for a regime shift between alternative stable states, possibly forced by overfishing (Collie 2004) (figure B9.2.1B–­E). First, the time series shows a discrete step in average stock biomass, which dropped abruptly ­after a spike in catch due to an influx of foreign fishing fleets in the early 1960s, ­after which biomass remained low

Proving that an ecosystem exhibits alternative stable states is difficult and demands intensive data. The best evidence comes from commercially exploited species that are well monitored. A pos­si­ble candidate is Georges Bank haddock. Data on age-­structured abundance of haddock for the period 1931–2000 appear to meet the criteria of Scheffer

(C)

(B) Biomass Catch

Exte

al ern Int cture stru

Smooth

rnal fo Abrupt rcing

Thousand tonnes

Abundance

Discontinuous

200 100 1940

1960 1980 Year

1931–1965 1966–2000

60 40

20

20

0

0

0.0

0.2 0.4 0.6 Fishing mortality

Planktivorous fish

Catch (kt)

40

10 5

0

100

200 300 BIomass

400

(F) 80

60

15

0

2000

(E) 80

Catch (kt)

300

0

(D)

0.0

2.0 1.5 1.0 0.5 0.0 –0.5 –1.0 –1

0.2 0.4 0.6 Fishing mortality

85

77 2000

82 97

88

94 90 91 0 1 2 Fishing mortality

(H)

(G) 75

–0.5

85

80 78 89

–1.0 –1.5 –1

83 2000 93

98 0

2.0 Phytoplankton

0.0 Zooplankton

400

Thousand tonnes

(A)

1

Gelatinous plankton

2

1.5

97 99

1.0 0.5 0.0 –0.5 –1.0 –1.5

94

91

2000 79

89

83

87 –1 –0.5 0 Zooplankton

81

0.5

Figure B9.2.1. Empirical evidence consistent with discontinuous regime shifts in marine ecosystems. (A) A conceptual model suggests that change in organismal abundance mediated by environmental change or disturbance varies from smooth in s­ imple systems (­little “internal structure”), to abrupt, to discontinuous in complex systems (high internal structure). Data are from (B–­E) Georges Bank haddock and (F–­H) the Black Sea pelagic food web. In haddock, discontinuous regime shifts are recognized by (B) a discrete shift from higher to lower abundance ­after ~1965, (C) a bimodal distribution of biomass across the time series, (D) dif­fer­ent functional relationships between fishing mortality and catch in the dif­fer­ent periods, and (E) hysteresis in this functional relationship in simulations of a model fit to empirical data. Hysteresis is also evident in (F–­H) changing relationships among trophic levels over time (numbers indicate years) in the Black Sea ([A–­E] ­after Collie et al. 2004; [F–­H] a­ fter Daskalov et al. 2007).

Chapter 9 Ecosystems

for most of the late twentieth ­century. Second, the distribution of biomass values was bimodal, indicating a high-­biomass state and low-­biomass state. Third, catch showed a dif­fer­ent statistical relationship to fishing mortality before and ­after the shift in the 1960s, suggesting some change in the ­factors controlling population size. ­These three observations support the existence of two distinct regimes during the time series. Simulations of a population model fit to empirical data also support the hypothesis, showing that catch followed dif­fer­ent trajectories when fishing mortality was increased versus decreased; that is, the system exhibited hysteresis, a key piece of evidence for a discontinuous regime shift between alternative semistable states (Collie 2004). Another example of a regime shift, again evidently forced in part by fishing, was documented at the ecosystem scale in the Black Sea (Daskalov et al. 2007).

251

­ ere strong fishing pressure on top predators drove H their decline and eventual collapse in the 1970s, which was accompanied by cascading changes in lower trophic levels, leading to phytoplankton blooms and nutrient depletion. A subsequent change in the focus of fisheries to smaller planktivores, like sprat and anchovy, then led to a subsequent collapse of the planktivorous fishes and corresponding increases in the jellyfish that compete with them (Daskalov 2002). Plotting the time trajectories of the vari­ous trophic groups shows that, for several interactions, the relationships between consumer and prey abundances differed in early versus ­later years, suggesting the hysteresis characteristic of a discontinuous regime shift (figure B9.2.1F–­H). Models confirm that overexploitation can trigger such shifts between alternative states (Daskalov 2002, Collie 2004).

Empirical evidence for regime shifts in marine ecosystems Regime shifts have been reported from both pelagic (see box 9.2) and benthic systems and may be an impor­tant pro­cess in the widespread degradation of submersed vegetation in estuarine and coastal systems. Studies in shallow lakes provide evidence for two alternative states resistant to transition (Scheffer 2004): a clear-­water state dominated by rooted macrophytes and a turbid-­water state lacking macrophytes. Macrophytes bind and stabilize sediments and reduce ­water movement that resuspends sediments. When t­hese plants are lost, sediments become more mobile, stay in suspension, and interfere with the establishment of macrophytes, creating positive feedback that resists transitions between the alternative states. Similar pro­cesses may also explain variation in eelgrass occurrence across coastal seascapes. Analy­sis of data from 83 sites across western Eu­rope identified a positive feedback between sediment conditions, light conditions, and seagrass density (van der Heide et al. 2011). As in lakes, seagrasses trap and stabilize suspended sediments, improving ­water clarity and seagrass growth conditions. Eutrophication reduces light conditions, tipping the system into the turbid-­water phase, thus degrading seagrass dominance. In general, eutrophication appeared to be the dominant control on seagrass distribution in sheltered estuaries, whereas the feedback between seagrass, sediments, and light dominated in exposed areas (van der Heide et al. 2011).

Mechanisms of marine regime shifts Several mechanisms potentially can produce regime shifts between alternate attractors, or semistable states (Collie 2004, Knowlton 2004). At the population level, the most general mechanism involves the Allee effect, that is, positive density dependence at low population sizes; called depensation in the fisheries lit­er­a­ture. For ­simple predator-­prey systems, the range in types of regime shifts can be generated from the same model simply by changing par­ameters, particularly the ratio between prey carry­ing capacity and predator half-­saturation constant (K/D), the relationships between maximal predation rate and prey growth rate, and the minimum time scale for shifts (Collie 2004). Regime shifts in predator-­prey interactions may also be mediated by the ontoge­ne­tic shifts in trophic level common in both benthic (Barkai and McQuaid 1988) and pelagic ecosystems (Bakun 2006). In many such systems, the dominant species in one of the alternative states feeds on early life history stages of species that dominate the

252

Ocean Ecology

alternative state, generating an unstable feedback loop that suppresses the alternative dominant from gaining abundance. A potential example of this phenomenon involves the collapse of cod in the Baltic Sea, which was accompanied by a shift to dominance by the cod’s prey, planktivorous herring and sprat. Cod failed to rebound from their initial collapse throughout much of the North Atlantic, despite severe fishing restrictions, prob­ably in part b­ ecause abundant herring and sprat feed heavi­ly on cod eggs and larvae. Thus, for pelagic marine food webs, the ontoge­ne­tic size structuring of trophic interactions may be a key ­factor in mediating commonly reported regime shifts between alternative stable states (Bakun 2006). As noted above for seagrasses, regime shifts can also result from effects of organisms on the environment, particularly in benthic sedimentary habitats. H ­ ere ecosystem engineers often modify the environment such that it becomes less hospitable to species characteristic of the alternative regime. Certain types of infaunal invertebrates, for example, are both more tolerant of sediment resuspension and more active in resuspending it; ­these animals may interfere with the establishment of species that would other­wise dominate in stable sedimentary environments (Rhoads and Young 1970, Peterson 1984, van Nes et al. 2007). Regime shifts may be promoted by low diversity, as suggested by the observation that rapid transitions in ecosystem state are best documented in relatively low-­diversity systems, including temperate lakes, the Black Sea (Daskalov et  al. 2007), and the North Atlantic (Frank et al. 2007). Such a link is consistent with experiments showing lower stability in systems with low diversity. Invasion of marine communities by exotic species, for example, can trigger irreversible shifts in ecosystem structure and function, and is more frequent in communities of low diversity, apparently ­because such communities frequently leave limiting resources unused, allowing invaders to establish (Stachowicz et al. 1999, Stachowicz, Fried, et al. 2002).

Applications of Marine Ecosystem Modeling in Fisheries and Management A primary motivator for scientific understanding of marine ecosystems is the need for data to inform fishery management. The simplest approach to modeling fish production is entirely empirical, modeling production as a function of environmental and other forcing variables. An innovative application of this approach comes from a forensic analy­sis of global fisheries catch statistics collated by the United Nations’ Food and Agriculture Organ­ization (FAO). Reg Watson and Daniel Pauly (2001) used a general linear model to predict fisheries catch as a function of depth, primary productivity, ice cover, surface temperature, latitude, distance from shore, and upwelling index. Over most of the world ocean, the modeled catch matched well with the catch that countries reported to the FAO. The exception was along the Chinese coast, where reported catches w ­ ere higher than predicted to be pos­si­ble. The authors noted that Chinese fishery officials w ­ ere rewarded for large catches, and concluded that they had been providing inflated catch statistics to the FAO from the mid-1980s ­until 1998 when, ­under domestic and international criticism, the government proclaimed a “zero-­growth policy” explic­itly stating that reported catches would remain frozen at their 1998 value. China is among the major world consumers of fish, and an impor­tant consequence of the misreported catch is that global fishery catches appeared to increase throughout the reporting period when the reanalysis suggested that in fact world fishery catch reached a plateau around 1980. This analy­sis thus revealed that, unbeknownst to the rest of the world, the global supply of wild-­captured fish may have reached its limit during that time. A more mechanistic approach to modeling energy flows through a food web is based on the princi­ple of mass balance, whereby the model aims to account for all inputs (of, say, carbon or nitrogen) to the system and reconcile them with transformations and outputs. Although early models of ecosystems ­were static snapshots of (assumed) equilibrium states, management needs and growing computational power stimulated a vigorous evolution of dynamic and simulation approaches to modeling marine ecosystems. Indeed, much of the history of marine ecosystem modeling has been driven by fishery management needs. The influential Ecopath mass balance model was originally de-

Chapter 9 Ecosystems

veloped to estimate the energy flows within the relatively pristine coral reef ecosystem of French Frigate Shoals in the Northwestern Hawaiian Islands (Polovina 1984). Among other rationales, the study was motivated by the paradox that coral reef ecosystems appear to be highly productive yet many reef fisheries decline ­under relatively light fishing pressure. The Ecopath model helped explain this paradox by showing that only ~20% of the production of reef fishes and benthic animals was consumed by higher predators, the rest being consumed by “internal predation” by other species within the same trophic group. Thus, sustainable yields of fisheries targeting higher predators are considerably lower than might be expected based on gross primary production in the system. A major advance in the application of this approach in marine fisheries science came with the introduction of the Ecopath II program (Christensen and Pauly 1992) and its ­later development, Ecopath with Ecosim, or EwE (Christensen and Walters 2004), based on Polovina’s original Ecopath model. EwE provided a tool for building mass balance models of w ­ hole marine ecosystems based on a core module for estimating biomasses of components within a food web u­ nder the assumption of a steady state. The EwE uses as input data on the dietary links and energetic par­ameters (P/B, assimilation efficiency, ­etc.) among the components, the latter often derived from general empirical functions based on body size. The energy flows among components are estimated using general empirically derived functions of feeding, assimilation, respiration, and growth rates with body mass, and mass balance is achieved by simultaneous solution of a system of linear equations, one for each component (species or functional group) in the system u­ nder the assumption of a steady state (Pauly 1995) (figure  9.20). In the pro­cess of estimating the flows and balance, Ecopath II also aimed to estimate and evaluate several derived metrics meant to capture Odum’s (1969) concept of ecosystem maturity—or proximity to the system’s climax state. ­These include the recycling index and several more esoteric metrics of energy flow, such as ascendency.

4

Tiger fish

Nile perch

0.4 Small pelagics

Trophic level

3

6.3

5.6

Catfish

Zooplankton

0.1

0.1

Catch

Catch

0.1 Benthic fish

20 0.6

0.1

.02

0.03 0.1

0.03

207 2

0.002

Catch

0.002

0.2

0.1 Catch

0.3

0.1

1 R

PP

1,300 1

Phytoplankton

3,900

6

2.8 2.5 Detritus

2

1

+

3

+

10

=1

+

Figure 9.20. ​Estimation of the primary production (right, in gC km−2 yr−1 ) required to support a fishery, using an ecosystem mass balance approach. This example of a s­ imple food web is from Lake Turkana in Africa (­after Pauly 1995).

253

254

Ocean Ecology

In the ensuing de­cades, EwE has grown into the most widely used approach for modeling marine ecosystems. The original mass balance food web module was supplemented with components for dynamic simulations through time (Ecosim) and space (Ecospace) (Christensen and Walters 2004). Ecosystem models, including EwE, are now widely used in a variety of applications for simulating not only changing fishing pressure but impacts of ecological restoration (Frisk et al. 2011) and projected impacts of climate change. In Australia, for example, a study used EwE to link models of projected climate change, through primary production by phytoplankton and benthic plants, to changes in biomass, catch, and economic value of fishes and vertebrates of conservation concern (Brown et  al. 2010). The authors concluded that climate warming is likely to increase primary production around Australia, producing bottom-up increases in fisheries catches and value, on average, as well as enhanced abundance of threatened turtles and sharks. EwE is one example of the f­ amily of “end-­to-­end” models that aim to represent not only the biophysical basis of the ecosystem but the components of ­human society and industry that interact with it, incorporating the dif­fer­ent scales of ­these vari­ous pro­cesses and, most importantly, focusing on dynamic two-­way interactions between ­these ecosystem components (Fulton 2010). End-­to-­ end models are inherently quite complicated, most obviously b­ ecause of the large number of par­ ameters and complex interactions required to capture every­thing from climate forcing to phytoplankton to top predators, along with influences of, for example, fishing fleets and tourism. But they are also complicated b­ ecause ­these dif­fer­ent components of the system often require quite dif­fer­ent modeling approaches, which must be dynamically coupled. So-­called aggregate models, like EwE, consist of trophodynamic or bioenergetics networks among components of the system. “Hybrid” models (figure 9.21) might, for example, link a biogeochemical model of environmental forcing; Department of Transport

Connections

Environmental Protection Agency

Shipping

Oil and gas production

Fisheries management authority

Salt production

Influence Flow Conservation authority

Ports Logger Dredging

Spotter planes

Contaminants

Recfishing

Prawn trawl

Trawl fishing

Oyster lease Coastal development

Acid suplhate soil

Seagrass

Rainfall

Components Governance Monitoring Industries Fisheries Animals Habitat Physical Natural disasters Sediments Bathymetry Model types

Nemipterus Mangroves

Trap fishing

Sharks

Turtles

Prawns

Survey vessel

Affect all

Saurida

Lutjanus

Lethrinus

Macroalgae

Light

Wind

Currents

Cyclones Bathymetry

Reef habitat Sediments

Governance Contaminants Logger IBM Vessel Industry Expression Footprint Metapopulation Gridded data Population

Figure 9.21. ​A hybrid, end-­to-­end ecosystem model of Ningaloo Reef, Australia. The flowchart shows the structure and model types used in the InVitro model, including both biophysical (natu­ral ecosystem) components and socioeconomic components. Recfishing = recreational fishing. Only major flows or connections are shown (Fulton 2010).

Chapter 9 Ecosystems

box or size-­spectra models of lower trophic levels coupled to individual-­based, age-­structured population models of large, commercially harvested fishes or protected vertebrates; and socioeconomic models of fishing fleets, each with a characteristic spatial and temporal scale. In general, end-­to-­end models are intended not to find optimal solutions for specific cases but to evaluate alternative hypotheses, typically proposed management strategies. The toughest challenge is incorporating h­ uman components, in part b­ ecause biophysical and social scientists use very dif­fer­ent approaches and come from dif­fer­ent worldviews, which are only beginning to bridge (Fulton 2010, Salomon et al. 2011). Nevertheless, it is now clear that incorporating both biological and social interactions in an ecosystem approach to fisheries is pos­si­ble and in fact critical to effective management (Gaichas 2008).

Models of the Global Ecosystem The ultimate success of ecosystem ecol­ogy, indeed of ecol­ogy generally, would be production of a realistic model of the global biosphere from first princi­ples, that is, by applying mechanistic laws of biology to patterns of environmental forcing across the earth. Do we know enough to produce such a model? Only within the last de­cade have understanding of functional biology and the massive computational power required to tackle the prob­lem converged to bring this goal within reach. Several efforts have made exciting pro­gress in this area, initially with specific taxa and more recently for multitrophic ecosystems as a w ­ hole. An early effort to build a virtual ecosystem from scratch focused on oceanic phytoplankton, the base of the global ocean food web (Follows et al. 2007) (figure 9.22). This began with a set of empirically derived equations of basic physiology—­observed relationships between cell size, growth rate, and so on as a function of light, temperature, and nutrients—­and applied them to the abiotic conditions of the real (mea­sured) ocean in a simulation model. The model communities consisted of populations of cells, drawn from a realistic range of physiologically pos­si­ble traits. The populations ­were then allowed to grow, virtually speaking, according to the empirically derived equations relating cell size, light and nutrient physiology, and growth rate, and to drift out through the virtual ocean and compete with one another for resources. As they responded to this competition and environmental forcing, the populations (species) best suited to local conditions thrived, o­ thers withered, and ultimately a community structure and biogeography emerged. The set of functional types that emerged in a region defined its community, and the variation of ­these communities across space constituted biogeography. ­These properties ­were emergent in the sense that the community structure and biogeography ­were not programmed or constrained in any way, but instead arose from growth and interactions of the individual populations. The turning of this virtual world produced striking results (see figure 9.22): the phytoplankton communities that emerged consisted of a range of cell types with strong physiological similarity to real species of the cyanobacteria Prochlorococcus, which dominate the world’s upper ocean. Moreover, their distributions across space and depth w ­ ere similar to the observed distributions of Prochlorococcus strains or species (Follows et al. 2007). This study illustrated that basic physiological and metabolic relationships captured in relatively ­simple equations can produce communities with traits (in this case light harvesting, cell size, and temperature tolerances) quite similar to ­those observed in nature. In short, the work reproduced a complex adaptive system from ­simple rules applied to individual organisms interacting with one another and the environment. The first and most ambitious global mechanistic simulation of the earth ecosystem was the General Ecosystem Model, applied across the full range of marine and terrestrial environments (Harfoot et al. 2014). As in Follows’s phytoplankton model, basic organismal pro­cesses w ­ ere based on functional forms and pa­ram­e­ter values derived from the lit­er­a­ture. The model then took ­these equations and values and applied a vast computational power to simulate the fate of all organisms with body masses between 10 mg and 150,000 kg across the globe, over land and sea. As the model churned, properties emerged at the level of the individual organism (e.g., growth rate), community (e.g., functional

255

Ocean Ecology

Observations NO3

(A)

Model NO3

(B)

Depth (m)

30

100

20

100

10

200

200 Total Prochlorococcus

Depth (m)

Total Prochlorococcus

100

100

200

200

5 4 3 2 1 Model ecotype m-e1

Depth (m)

Prochlorococcus ecotype eMIT9312

100

100

200

200

5 4 3 2 1 Model ecotype m-e2

Depth (m)

Prochlorococcus ecotype eMED4

100

100

200

200

5 4 3 2 1 Model ecotype m-e3

Depth (m)

Prochlorococcus ecotype eNATL2A

100

100

200

200

5 4 3 2 1 40°S

20°S

0

20°N

40°N

Prochlorococcus ecotype eMIT9313 Depth (m)

256

100

200

40°S

20°S

0

20°N

40°N

Figure 9.22. ​A model based on basic physiological equations reproduces distributions of pelagic primary producers. (Left) Observed and (right) modeled distributions of Prochlorococcus ecotypes along the AMT13 cruise track encompassing the Atlantic Ocean (­after Follows et al. 2007).

type richness), ecosystem (e.g., trophic pyramids, biomass turnover rates), and macroecological scales (e.g., global patterns of trophic structure) that ­were in general agreement with observed patterns. As in the phytoplankton example, t­ hese properties emerged from a relatively small number of equations specifying the biology of, and interactions among, individual organisms with no direct constraints on the higher-­order organ­ization of communities and ecosystems. The authors’ arresting

Chapter 9 Ecosystems

conclusion was that “ecologists have gathered sufficient information to begin to build realistic, global, and mechanistic models of ecosystems, capable of predicting a diverse range of ecosystem properties and their response to ­human pressures” (Harfoot et al. 2014). ­These types of studies have reproduced reasonably well the global patterns of biomass and distribution among trophic levels, producer functional groups, and geographic regions. The challenge now is refining them to the point where resolution is sufficient to help address specific conservation and management issues.

Marine Ecosystems in the Anthropocene Applied ecol­ogy and management have increasingly shifted focus from individual threatened species to the broader ecosystem. In fishery management in the USA, this shift was mandated by law in the form of the ecosystem approach to fisheries required by the Magnuson-­Stevens Fishery Conservation and Management Act. This approach is also growing in the conservation community with recognition that entire ecosystems are threatened and that understanding how ­those systems work as a ­whole is impor­tant to effectively managing and protecting the species that live within them. Climate change is shuffling species into new combinations—­novel (or emergent) ecosystems (Williams and Jackson 2007, Hobbs et al. 2009). Efforts are ­under way to develop standard global assessments of risks at the ecosystem level modeled on the IUCN’s influential Red List of Threatened Species. A conceptual model for ecosystem risk assessment based on ecological theory (D. A. Keith et al. 2013) identifies four symptoms of ecosystem risk as a basis for assessment: (1) rates of decline in ecosystem distribution; (2) restricted distribution of an ecosystem with continuing decline or threats; (3) rates of environmental degradation; and (4) rates of disruption to biotic pro­cesses. ­These separate components are tied together in an overall criterion, a quantitative estimate of the risk of ecosystem collapse, intended to provide the basis for ecosystem red-­listing. Application of the protocol to a range of ecosystem types has shown that the approach appears workable based on available data, and the results are consistent with other assessments by local experts.

Eutrophication Marine ecosystems face many challenges, including both bottom-up and top-­down pressures. Productivity throughout much of the ocean is l­imited by usable nitrogen, and in t­ hese systems nitrogen influx increases ecosystem productivity and biomass accumulation. Since anthropogenic nitrogen is delivered mainly by runoff from the land, N-­stimulated increases in production are most intense in estuaries and the coastal ocean where rivers drain continental watersheds. The resulting blooms of algae, both phytoplankton and benthic macroalgae, have increased dramatically over recent de­cades throughout the world (Korpinen et al. 2007, Anderson et al. 2012). This so-­called cultural eutrophication has transformed the character of estuaries, in par­tic­u­lar by fueling large blooms of algae that escape grazer control, accumulate, and decompose (chapter 11). ­These blooms typically arise in summer, when the ­water column is stratified and the oxygen consumed by decomposition is not replenished from above, generating hypoxic “dead zones” that reduce w ­ ater quality and are unsuitable for the living communities that support fisheries (Diaz and Rosenberg 2008). A second effect of nitrogen eutrophication is a shift in the dominant primary producer types. In pelagic systems, eutrophication has produced large increases in the distribution and frequency of harmful algal blooms (HABs) dominated by phytoplankton species resistant to grazers and often protected by noxious or toxic chemical components dangerous to h­ umans and wildlife (Heisler et al. 2008). Our understanding of the complexity of harmful algal bloom dynamics and management has changed considerably in the relatively short time since the 2008 consensus statement on eutrophication and HABs, which does not even mention grazing (Heisler et al. 2008). More recent reviews recognize that grazer interactions with the algal community are a potentially central part

257

258

Ocean Ecology

of the multifaceted controls on HAB dynamics (Anderson et al. 2012), and that not only nutrient loading but cascading effects of overfishing can generate algal blooms in both pelagic and benthic communities (Casini et al. 2008, Eriksson et al. 2009). Eutrophication similarly has changed the character of benthic systems, leading to the replacement of macrophytes by ephemeral algae. In eutrophied systems, large perennial seaweeds, such as fucoid rockweeds that have historically dominated intertidal shores, give way to fast-­growing algae, such as Enteromorpha and Ulva (Worm and Lotze 2006), in part ­because ephemeral algae absorb nutrients and grow more efficiently than perennial macroalgae. Field observations of this transition are supported by experiments, which show that larger macroalgae in temperate systems generally do not respond to nutrient enrichment, whereas filamentous and turf algae increase with nutrient enrichment. The fast-­growing ephemeral algae are suppressed by herbivores (Burkepile and Hay 2006), but the ability of herbivores to control them declines as nutrient enrichment increases.

Defaunation and trophic skew The importance of consumers in maintaining ecosystem structure and functioning also extends to higher levels in food webs, and t­ hese large predators are u­ nder unpre­ce­dented pressure in the Anthropocene ocean. It is now widely recognized that marine food webs have been dramatically altered from their pristine states nearly everywhere in the ocean, largely as a result of overfishing. A major stimulus for this change in worldview was the assembly of a diverse range of historical information—­from the fossil rec­ord, archaeology, sixteenth-­century nautical charts, and the like. Together with integrated global fishery data, ­these sources documented steep and sustained declines in large animals throughout the ocean that have followed ­humans wherever we have gone, even in small populations with primitive technology ( Jackson et  al. 2001). The picture provided by this historical-­ecological approach helped begin to reverse the shifting baseline phenomenon (Pauly 1995), originally invoked in fisheries science but broadly applicable to environmental science (chapter 4). In addition to the substantial economic impacts of such changes in fisheries, the ecological extinction of large species, many of them top predators or megaherbivores, also has pervasive cascading effects on the ecosystems they ­were part of. This disproportionate impact of h­ uman activities on large animals drives systematic changes in trophic structure in human-­impacted systems, that is, trophic skew (Duffy 2003), that have cascading impacts on marine communities and ecosystems, as we learn in subsequent chapters.

­Future Directions A major theme in ecological research for de­cades has been the relative importance of bottom-up and top-­down control in ecosystems. Hundreds of experiments exploring this issue have been conducted, dissected, and meta-­analyzed. What are their practical implications? The ocean is now in the midst of the largest such experiment in earth history, happening on a global scale and over far longer time spans than any previous experiment. Global variation in fishing pressure and nutrient loading provides opportunities for rigorous comparative studies to understand how t­ hese pro­cesses play out in nature, over large scales and involving the time lags and indirect interactions difficult to parse out in short controlled experiments. Taking advantage of ­those opportunities ­will require integrating coordinated observations, targeted experiments, and modeling over substantial time and space scales. The results ­will be critical to informed and effective management of the new ocean. ­There is a growing imperative to understand the role of biodiversity in ecosystem pro­cesses ­because ­human influence is disrupting the natu­ral links between environmental ­drivers, the structure of communities, and their mediation of energy and materials fluxes. ­These historical relationships have evolved and adjusted over millions of years but are beginning to unravel as a result of h­ uman impacts, emphasizing that understanding changes in ecosystem pro­cesses requires understanding how both the

Chapter 9 Ecosystems

environment and the functional composition of biological communities are changing, how they respond to stressors, and how ­those responses translate to altered ecosystem pro­cesses and ser­vices. An exciting development that can catalyze and or­ga­nize advances in t­hese areas is the rapid growth of computational capacity to support power­ful models and simulations of complex ecosystems up to the global level. Th ­ ere is a potentially symbiotic, two-­way relationship between the evolution of such models, which depend on empirically mea­sured par­ameters as well as field studies that can provide ­these data and that can be guided by model predictions.

Summary An ecosystem consists of all living organisms interacting with one another and their abiotic environment in some defined area. The structure of an ecosystem is defined by the quantities and distributions of biomass among trophic levels and functional groups. Its functioning is defined by the flows of energy and materials, notably the biologically active ele­ments carbon, nitrogen, and phosphorus, among organisms and the nonliving environment. Energy and materials enter the living part of the ecosystem through primary production, move through it via herbivory, decomposition, predation, and export (advection), and are ultimately respired back to inorganic form or stored in sediments and deep ­water. The broad outlines of ecosystem structure and function are determined by global gradients in abiotic ­drivers, notably temperature and nutrient availability, and by biological diversity, which in turn is an evolutionary product of ­those ­drivers. Marine primary production is highest along continental margins and in polar seas where inorganic nutrients are supplied from land or mixing from the deep ocean. This production flows up the food web and is mirrored in geographic patterns of fish production, yielding a strong signature of bottom-up control of ecosystem structure at the global scale. Flows of energy and materials are also strongly determined by biodiversity, including its vertical aspect defined by the number and discreteness of trophic levels, which mediate top-­down control, and its horizontal component comprising the variety of functional groups within trophic levels. At regional and local scales, biological diversity often explains as much or more variation in ecosystem properties than does abiotic forcing. Historically, ecosystems ­were conceptualized as production machines regulated around equilibrium. This view is giving way to understanding ecosystems as complex adaptive systems, characterized by diverse, locally interacting components, subject to se­lection pro­cesses that result in an assemblage of species well suited to local conditions. Complex adaptive ecosystems self-­ organize into emergent structures but show l­ ittle evidence of equilibrium and often respond in a nonlinear way to perturbations, sometimes resulting in shifts among alternative states. Within ecosystems, productivity and stability often increase with biological diversity due to complementary resource use among species. This diversity is based on functional traits, including organism size, stoichiometry, and mobility. Primary producer nutrient content is an evolutionarily conserved trait that varies systematically among higher taxa and functional groups and strongly determines ecosystem structure and pro­ cesses. Nutrient content is highest in planktonic microalgae, in which most production is grazed, and lowest in higher plants, such as seagrasses and mangroves, in which most carbon flows to detrital food webs, with significant quantities stored in sediments. Models of ecosystems based on the princi­ple of mass balance and general metabolic equations are evolving in sophistication t­ oward end-­to-­end models that can link disparate submodels of physical forcing, successive trophic levels, and h­ uman industry, which require dif­fer­ent model formulations, in a spatially explicit context. Th ­ ese models are increasingly useful in simulating ecosystem dynamics for forecasting and scenario analy­sis in fishery management and global change impacts. Marine ecosystems are being transformed rapidly and fundamentally from both the bottom up, primarily as a result of massive new nitrogen and carbon inputs, and the top down, as a result of fishing that depletes large animals, particularly predators, worldwide. Managing marine ecosystems in the Anthropocene requires attention to both bottom-up and top-­ down forcing and their still poorly understood interactions.

259

10 F

260

The Open Ocean

or most of ­human history, the open ocean has been a forbidding wilderness—­vast, unexplored, and fraught with danger. But to ­those who mastered it, the open horizon also offered ­great riches in harvestable fish, routes to new territories and markets, and strategic advantages over adversaries. Early in the second millennium BCE, Polynesian voyagers mastered open-­ocean navigation, a remarkable feat for their ­simple technology, and spread out in small outrigger canoes from Southeast Asia across the expanse of the tropical Pacific Ocean—an area comparable to the continent of Asia—­colonizing Polynesia, New Zealand, the Hawaiian Islands, eventually even reaching Rapa Nui (Easter Island) off the coast of South Amer­i­ca by 700–1200 CE. It is easy to understand how some, notably the adventurer Thor Heyerdahl, believed that Polynesians and South Americans ­were in contact prior to Eu­ro­pean colonization of the New World, and some population ge­ne­tic evidence appears consistent with this idea. Eu­ro­pe­ans came ­later to open-­ocean exploration but left a bigger mark. Crews on the early Eu­ro­ pean voyages famously feared that they might sail off the edge of the world, and even up to the seventeenth ­century world maps w ­ ere embellished with mythical sea monsters, highlighting the prevailing ignorance of the open ocean. To ply their trade, the developing empires and nation-­states of Eu­rope needed reliable information on currents, distances, coastal geography, and populations and resources of foreign lands, so they began including naturalists aboard their expeditions. The systematic study of the open ocean—­oceanography, as it came to be known—­grew out of ­these symbiotic military, commercial, and dawning scientific ambitions as the Eu­ro­pean powers vied for world domination from the seventeenth through the nineteenth centuries. A notable milestone was the voyage of Captain James Cook, age 39 at the outset, in command of HMS Endeavour, which departed ­England in 1766 on the first long-­ range voyage officially commissioned in the ser­vice of science. The goal was to rec­ord the transit of Venus across the sun, which the Royal Society desired for the purpose of devising a means, still lacking at that time, to estimate longitude and therefore distances across the ocean. Cook sailed west across the Atlantic, rounded the tumultuous Cape Horn into the Pacific, successfully reached Tahiti, and made the mea­sure­ments, whereupon he unsealed additional ­orders to search for the still mythical ­great southern continent. He succeeded in this task too, mapping much of the east coast of Australia and New Zealand before proceeding westward to complete the circumnavigation by returning to E ­ ngland. Accompanying Cook as naturalist aboard Endeavour was the botanist Joseph Banks, who made g­ reat contributions to natu­ral history as a result of his collections and observations during the three-­year expedition. Cook’s voyage was the beginning of the golden age of natu­ral history as the world’s navies and merchant marines, particularly the British, carried scientists around the globe. We owe much of our modern knowledge of the order of nature to the decision in 1831 of Captain Robert FitzRoy (himself a meteorologist and originator of the term forecast) to invite a 22-­year old youth named Charles Darwin as ship’s naturalist aboard HMS Bea­gle’s round-­the-­world voyage of discovery. This saved the lad from his trajectory ­toward an uneventful life as an En­glish country vicar, and the rest is history.

Chapter 10 The Open Ocean

The first major voyage explic­itly dedicated to the study of the ocean was that of HMS Challenger, described in chapter 3. But modern oceanography, founded on a detailed knowledge of ocean physics, was born during the Second World War, which pitted nations against one another across all the world’s oceans, demanding mastery of tides and navigation and, in the new world of submarine warfare, knowledge of the open ocean beneath the surface for the first time in history. ­These demands and the technology spawned in the fever of the war not only revolutionized ocean science but forever changed fishing and other maritime industries (chapter 4). Other discoveries followed in increasingly rapid succession, notably the use of detailed satellite mapping of the ocean’s plant biomass, and power­ful molecular ge­ne­tic approaches that revealed a vast, hidden world of microscopic ocean life, much of it ancient. We consider ­these and other advances in this chapter.

Physical Forcing of Pelagic Ecosystems All ecosystems are structured fundamentally by the physical forces acting on them, but that influence is clearest in the open ocean, where solar heating, wind mixing, and geostrophic currents act on a mass of ­water that is other­wise homogeneous. This action forms the world ocean into a collection of ­water masses that are chemically and biologically distinct, and separated from one another by surprisingly sharp bound­aries. Thus, the major features of oceanic ecosystems are traceable to physics and its effects on biological activity and the chemistry of seawater (chapter 3). The range of pro­cesses structuring pelagic ecosystems was captured by the physical oceanographer Henry Stommel (1963) in a graphic summary that became famous as the Stommel diagram (figure 10.1), showing the relative importance of ocean pro­cesses as a function of the temporal and spatial scales of observation. The diagram illustrates that pro­cesses acting over large spatial scales generally also develop over long time periods; for example, as ice ages affect the entire world ocean. But ­there are con­spic­u­ous departures from this rule that illustrate one of our central themes: biology can drive as well as respond to the abiotic environment in shaping ecosystems. The most striking example in the open ocean is the phenomenon of diurnal vertical migration, in which a large proportion of the pelagic fauna rises ­toward the ocean surface to feed each night and descends again into the depths by day in a massive, basin-­wide movement of biomass between the epipelagic and mesopelagic zones. This phenomenon is vis­i­ble as a sharp ridge in the Stommel diagram.

The global distribution of ocean productivity The first detailed picture of the world ocean’s biological productivity emerged in 1979 when the Coastal Zone Color Scanner began beaming back data on the color of the surface ocean from its perch aboard the earth-­orbiting Nimbus 7 satellite. Given some assumptions about w ­ ater transparency, depth distribution, and optical properties, t­ hese data on fluorescence from algal pigments in surface ­waters could be translated into the geography of phytoplankton biomass (Longhurst et al. 1995). Technology advanced rapidly a­ fter this debut, and by 2000 the SeaWIFS satellite was able to produce a composite view of global plant biomass on both land and sea, sketched in the vivid (synthetic) colors of photosynthetic pigments (see figure 1.4). The striking image of global plant biomass captures the major features of the world ocean’s biology and serves as a logical starting point for understanding the distribution and functioning of Earth’s pelagic ecosystems. ­These features include (1) the expansive blue deserts of the central gyres, permanently stratified and chronically depleted of the nutrients needed for plant growth; (2) much higher productivity in the polar regions of the open ocean; (3) narrow zones of high productivity along the continental coasts, particularly in the upwelling zones along the eastern margins of ocean basins and around the mouths of major rivers like the Amazon; and (4) a narrow equatorial band of enhanced productivity, reflecting the upwelling of

261

262

Ocean Ecology

A B C D E F G H I J K

“Micro” patches Swarms Upwelling Eddies and rings Island effects “El Niño” type events Small ocean basins Biogeographic provinces Currents and oceanic fronts—length Currents—width Oceanic fronts—width

10,000 km

Ice age variations

Diel vertical migration

1,000 km

I

Annual cycles H

C

F

D

J

100 km

G 10 km

9

K

E

100 m

B 6

1m

g Lo gth

len )

(cm

Biomass variability

A

12

3

Y

0Y

00

,00

1,0

r Yea

te nu Mi

0

y Da

2

Time

nth Mo k e We

4

ry ntu Ce

6

Space

10

10 8

1 cm

sec)

riod (

e Log p

Figure 10.1. ​Temporal and spatial scales of major ocean pro­cesses (the Stommel diagram), developed from Henry Stommel’s (1963) original graphical depiction of relevant scales of spatial and temporal variation of physical pro­cesses in the ocean, and adapted to biological pro­cesses.

nutrients from deep ­water along the equator, where the Coriolis effect switches sign and generates diverging surface currents. As we saw in chapter 3, t­ hese patterns of ocean plant biomass are ultimately generated by global-­ scale physics—­the circulation of the ocean set in motion by atmospheric heating and the latitudinal pattern of wind speed and direction, interacting with the continental margins. Th ­ ese physical pro­ cesses in turn drive the distribution and flux rates of inorganic nutrients needed for plant growth. Large-­scale patterns of primary producer biomass and production closely reflect gradients in nutrient availability, with nitrogen limiting marine primary production over much of the ocean, and iron impor­tant in certain oceanic surface ­waters far from the continents.

Vertical structure of the pelagic ­water column As spectacular as the satellite images of ocean productivity are, they literally only scratch the surface. What is not vis­i­ble from satellites is the g­ reat volume of ocean ­water below the top meter. The patterns of surface pigments are, however, clearly related to the vertical structure of the ­water column. As discussed in chapter 3, that structure is forced primarily by solar heating of the upper ocean, which generates a vertical density gradient that strongly influences functioning of the pelagic ecosystem. The relatively sharp density discontinuity that develops between warm sunlit surface ­water and the cold, dense ­water below—­the pycnocline—is the most impor­tant ecological boundary in the ocean, dividing Earth’s aquatic realm into two layers with profoundly dif­fer­ent ecologies (figure 10.2). The

Chapter 10 The Open Ocean

(A)

Continental shelf

Epipelagic Zone (the Sunlight Zone) Mesopelagic Zone (the Twilight Zone)

200 m 1,000 m 2,000 m

Bathypelagic Zone (the Midnight Zone)

Continental slope

3,000 m 4,000 m

Abyssopelagic Zone (the Abyss) Continental rise

7,000 m

26,300 ft

10,000 m

29,600 ft 32,800 ft 36,100 ft

11,000 m (B) Light 0%

200 m

16,300 ft 23,000 ft

Hadal Zone (the Trenches)

9,000 m

100 m

9,900 ft

19,700 ft

Ocean basin

8,000 m

0m

6,600 ft 13,100 ft

5,000 m 6,000 m

3,300 ft

Nutrient limitation 1%

Euphotic zone 1%

Particulate organic carbon (μmol/kg) 100% 0 10 20 30 40

DCM

Mixed layer Pycnocline Thermocline (slow mixing)

Light limitation

300 m 0 2 4 6 0 100 200 300 16°C 26°C N concentration Chlorophyll a Temperature (μmol/kg) (ng/kg)

Figure 10.2. ​Vertical structure of the open ocean. (A) Depth zones of the open ocean (https://­ manoa​.­hawaii​.­edu​/­exploringourfluidearth​/­physical​/­ocean​-­depths​/­depth​-­zones). (B) Concordant vertical gradients in light, temperature, nutrients, and physical mixing in the subtropical open ocean (https://­www​.­nature​.­com​/­scitable​/­knowledge​/­library​/­the​-­biological​-­productivity​-­of​-­the​ -­ocean​-s­ ection​-­70631104).

overlying epipelagic zone is illuminated by the sun and extends to roughly 200 m depth in clear open-­ ocean ­water. It is roughly coincident with the mixed layer, maintained by wind-­induced stirring of the surface ­water. The re­sis­tance of the surface layer to mixing with the colder, denser w ­ ater below retains algal cells in the sunlit epipelagic zone where they are able to photosynthesize and grow. Phytoplankton growth in turn depletes nutrients in the epipelagic, which are incorporated into phytoplankton cells, detritus, and the feces of zooplankton that eat them, and which gradually sink into the deep ocean. In summary, solar forcing at the surface creates a characteristic one-­dimensional structure of the ocean ­water column that can be described as a set of interrelated depth profiles (figure 10.2B): a vertical section through the ­water column of the open ocean generally shows strong, roughly coincident gradients in temperature, density, and nutrient concentrations at the base of the pycnocline. Phytoplankton cells, which require light from above and nutrients mixed up from deep ­water below, tend to aggregate near the interface between t­ hese two fluxes, forming a characteristic deep chlorophyll maximum in the lower euphotic zone of many regions. Biologically, the epipelagic zone is the engine of the ocean.

263

264

Ocean Ecology

Depth (m)

Annual primary production sedimented (%)

Below the pycnocline lies the other 98% of the ocean’s volume. The organisms of the deep ocean depend entirely on subsidy of material that is produced in the epipelagic zone above and rains down, with the impor­tant exception of chemosynthetic vent communities, discussed ­later in the chapter. The deep ocean is fur10 ther divided into layers, although they are not as clearly defined as that separating the epipelagic from the rest. In the open ocean, some light penetrates dimly below the permanent thermocline. Although it is not sufficient for appreciable photosyn1 thesis, ­there is enough light that organisms adapted to this realm can see. This twilight zone, which extends ~200–1000 m, is the mesopelagic zone. As detritus from the epipelagic ecosystem falls to the deep ocean, mesopelagic consumers get the first 0.1 crack at this food supply, and the mesopelagic zone is therefore the most biologically 0 200 1000 10.000 active layer of the deep sea. As the organic ­matter slowly sinks, it is consumed along Depth (m) with oxygen by bacteria that respire the particulate carbon and draw down dissolved oxygen in the upper mesopelagic zone. This consumption results in an exponential Figure 10.3. ​The decline and fall of marine decline in flux of organic m ­ atter and respiration with depth (figure 10.3). One organic ­matter. The proportion of the surface result is a characteristic oxygen minimum zone in the upper mesopelagic, extendocean’s primary production that reaches the ing roughly from the base of the photic zone at ~200 m down to ~1000 m. Oxygen benthos declines exponentially with depth is higher above this zone due to its production by photosynthesis and higher below (­after Parsons et al. 1984). ­because organic ­matter is too scarce to draw down oxygen via respiration. The mesopelagic zone is the primary feeding area for some of the ocean’s largest animals, including the sperm ­whale (Physeter macrocephalus), which feeds on squids that congregate ­there. The basin-­wide phenomenon of vertical migration mentioned above appears to be driven primarily by the mi­grants’ opposing needs for food, which is abundant in surface w ­ aters, and escape from predators, which are less abundant and effective at hunting in the dark, cold depths. Many pelagic animals have resolved this trade-­off by rising into epipelagic ­waters to feed at night and descending again to the relative safety of the mesopelagic zone by 0 day. B ­ ecause many such animals migrate together, vertical migration results in aggregation into dense shoals of small, midwater 40 fishes, squids, and crustaceans. ­These shoals ­were first detected in 80 the mid-­twentieth ­century by acoustic signals (sonar) deployed by warships to reveal ­enemy submarines. The so-­called deep scat120 tering layer was defined by reflection of the sonar signal back 160 from the gas-­filled swim bladders of aggregated fishes. The large concentration of animal biomass in the deep scattering layer rep200 resents a rich food source for predators able to exploit it, as the 240 ­great w ­ hales do (figure 10.4). Indeed, exploitation of deep-­water prey aggregations appears to have been a major driver in the evo280 lution of w ­ hale body size, morphology, and functional ecol­ogy 320 0 2 4 6 8 10 12 14 16 (Goldbogen et al. 2017, Slater et al. 2017). Time (min) Fi­nally, below the reach of the last light, starting at ~1000 m in the clearest ocean ­water, lies the bathypelagic zone, constituting the g­ reat bulk of the ocean’s volume. Th ­ ere is no sunlight at Figure 10.4. ​Blue ­whale feeding on deep prey aggregations. Yellow line shows the trajectory of a blue ­whale foraging dive all ­here and very ­little food, most of which is consumed on the through a deep aggregation of krill. Red, blue, and white way down. Most of the ­water in this deep reservoir has been out indicate highest, lower, and lowest krill densities, respectively. of contact with the surface for hundreds or thousands of years and The green circles indicate feeding lunges by the w ­ hale at the is therefore very cold, generally a few degrees above freezing. bottom of its foraging dive (­after Goldbogen et al. 2013). 100

Chapter 10 The Open Ocean

The bathypelagic zone is relatively rich in inorganic nutrients accumulated over time as organic ­matter sinking from the surface is consumed and remineralized. Despite this forbidding environment, the deep ocean is not devoid of life, as once thought. Indeed, although sparsely populated, the deep-­sea benthos is paradoxically more diverse than that of many shallow environments, as we see ­later in the chapter.

The spring bloom Among the most con­spic­u­ous large-­scale biological rhythms of the ocean is the development and dissipation of the spring phytoplankton bloom in north temperate ­waters. Phytoplankton are said to bloom when they grow rapidly to achieve a high biomass that visibly colors the w ­ ater. Blooms are fundamentally impor­tant in pelagic ecosystems ­because the concentration of algal cells is critical to efficient feeding by grazers, which channels production up the food chain. In temperate systems, and particularly the North Atlantic, the spring bloom initiates a relatively predictable sequence of events that ­ripple through the ecosystem and provide an impor­tant win­dow into the biological working of marine ecosystems. The canonical spring bloom, roughly the “nutrient-­limited spring production peak” in Longhurst’s (2007) terminology, is a predictable greening of the surface ocean that occurs quite rapidly across much of the North Atlantic, driven ultimately by the seasonal increase in solar irradiance. In the North Atlantic, the spring bloom lasts about a month and can achieve 30%–40% of annual primary production during that time. The pioneering efforts to understand this ocean-­scale biological phenomenon ­were based on fieldwork by Gordon Riley (1946) on Georges Bank in the Northwest Atlantic and conceptual synthesis by Harald Sverdrup (1953). At this latitude, deep mixing by violent winter winds replenishes the surface ocean with nutrients from below the pycnocline but, together with the low light, prevents significant phytoplankton growth. T ­ oward the end of winter the lengthening days initiate the bloom sequence in two ways. First, increasing day length crosses a threshold where photosynthesis by the phytoplankton community exceeds its respiration, allowing net growth of phytoplankton biomass. Second, warming air temperatures create a layer of buoyant ­water that resists mixing with the denser, colder w ­ ater below, increasing stratification and maintaining phytoplankton cells in the photic zone. A month or two ­after the winter solstice ­these two ­factors interact to substantially raise the daily light dose experienced by a phytoplankton cell, and phytoplankton growth increases, often rapidly, ­after crossing this threshold. Primary productivity peaks in mid-­spring. If not disrupted by wind, the warm surface layer stabilizes, algal cells remain in light throughout the day, and phytoplankton biomass accumulates. A key ­factor controlling bloom dynamics is decoupling between phytoplankton growth and grazing. The reason that blooms are characteristic of higher latitudes is that cold winter temperatures result in low abundances and metabolic activity of metazoan zooplankton grazers and slower population growth rates relative to their algal prey. Typically, the spring bloom declines in early summer as grazer populations catch up with the rich food abundance and as limiting nutrients are depleted in the stable, stratified ­water column. Sverdrup (1953) synthesized this general model of the spring bloom in a form that became central to biological oceanography. He defined the key concept of critical depth as the depth of mixing at which net growth of the phytoplankton community (or “population” in his words), integrated through the mixed layer, exactly matches net losses. He argued that the seasonal bloom begins when light and temperatures rise to the point that the ­water column stratifies, ­after which phytoplankton begin to grow and accumulate. This critical depth paradigm of pelagic ecosystem seasonality reigned for the latter half of the twentieth ­century. But the situation is more complex than envisioned by Sverdrup, so much so that some have argued his classic model needs to be abandoned entirely. ­There are two main challenges. First, Sverdrup implicitly lumped phytoplankton respiration, zooplankton grazing, and zooplankton respiration into the single term “population respiration,” confounding the quite distinct

265

266

Ocean Ecology

physiological pro­cess of phytoplankton respiration with the community-­level pro­cess of consumption by animals (Smetacek and Passow 1990). Second, the emphasis on the onset of stratification as initiating the bloom has not borne up as the highly complex and patchy physics of upper ocean mixing have become apparent (Behrenfeld 2010). Specifically, phytoplankton blooms often begin before clear stratification develops, apparently as a result of uneven mixing and inherent buoyancy of bloom-­ forming algal cells that counteracts sinking. The nature and timing of blooms depend on both physical forcing and the functional traits of algae and grazers, including their buoyancy but also relative growth rates, vulnerability to grazing, and nutrient requirements (Marañón 2015). The algae that bloom most densely in the ocean, and that drive the classic spring bloom, are typically diatoms of intermediate cell size. A key trait facilitating the dominance of diatoms is the vacuole, which increases the effective ratio of cell surface to metabolizing cytoplasm, allowing high rates of nutrient uptake and exploitation of nutrient pulses via so-­called luxury uptake—­uptake beyond the cell’s immediate needs—­and storage. Nutrient storage allows diatoms to continue to grow ­after nutrients have been depleted from the ­water column and to outcompete other phytoplankton u­ nder the intermittent resource supply of mixing surface w ­ aters. Diatoms also are less vulnerable to grazing by the microzooplankton as a result of their siliceous frustule (shell) and spines. Indeed, bloom-­forming phytoplankton taxa generally share a low vulnerability to grazing by microzooplankton as a result of large size, coloniality, spines, and/or toxic compounds (Irigoien 2005). Examples include not only siliceous diatoms but also the colony-­forming haptophyte Phaeocystis of high-­latitude w ­ aters and the toxic dinoflagellates that form red tides. In general, blooms appear to depend as much on disruption or avoidance of grazer control as on physically forced thresholds in algal physiology.

High-­nitrogen low-­chlorophyll (HNLC) regions Algal cells require iron in several biochemical pathways, but iron has very low solubility in seawater and is especially scarce in high-­nitrogen low-­chlorophyll (HNLC) regions (chapter 5). ­These observations motivated the argument that phytoplankton production in such regions might be l­ imited by iron availability (Martin and Gordon 1988). The suggestion proved controversial, but resolving it had to await the invention of scrupulously clean culture techniques (imagine trying to avoid contamination with minute amounts of iron on a 100-­ton ship—­made of iron—in the corrosive environment of the ocean). Such experiments ­were done initially in culture ­bottles and confirmed that iron indeed strongly stimulated primary production in oligotrophic ocean ­water (Martin and Fitzwater 1988). But the case was sealed by a series of iron enrichment field experiments in the open ocean. The first of t­ hese, IronEx I, added iron to a large swath (64 km2) of the equatorial Pacific upwelling zone and confirmed spectacularly that iron addition stimulated a bloom, doubling phytoplankton biomass and quadrupling primary production (Martin et al. 1994). Since then, several subsequent iron addition experiments have been conducted in the Southern Ocean, the North Pacific, and the Northeast Atlantic, all fertilizing blooms that increased chlorophyll by roughly an order of magnitude (Boyd et al. 2007). Synthesis of t­hese experimental data now confirms that iron-­limited HNLC regions cover about 30% of the ocean surface, and thus that low-­latitude oligotrophic systems and HNLC regions form two broad regimes in the open ocean, primarily ­limited by nitrogen and iron, respectively (Moore et al. 2013) (see figure 9.4).

Organisms and Traits The phytoplankton: Major functional types All ecosystem pro­cesses begin with primary production. In the open ocean, the primary producers are phytoplankton. Although marine phytoplankton make up less than 1% of Earth’s photosynthetic biomass, they contribute almost half of its primary production, making the ocean phytoplankton as

Chapter 10 The Open Ocean

267

Nutrients level

impor­tant in the global carbon cycle as all land plants combined (Field et al. 1998). It’s often said that the oxygen produced by ocean phytoplankton provides e­ very other breath we take. The phytoplankton is an ecological rather than a phyloge­ne­tic group, meaning that it includes a collection of diverse organisms grouped together mainly by their small size and ability to photosynthesize. They range from the minute cyanobacterium Prochlorococcus that dominates the world’s open-­ocean ­waters, through diverse nanoflagellates, calcified coccolithophores, a wide range of silica-­encased diatoms, and nitrogen-­fixing cyanobacteria such as Trichodesmium, which forms characteristic golden clumps resembling sawdust and vis­i­ble to the naked eye in the blue ocean gyres. Tiny picoplanktonic cyanobacteria are by far the most abundant phytoplankton, responsible for up to half of total phytoplankton biomass and production in the oligotrophic open ocean. Compared with terrestrial and benthic plants, phytoplankton are very small, with rapid population growth rates, and their extreme diversity means that, collectively, phytoplankton are adapted to a wide range of conditions. The oceanographer Ramon Margalef was a pioneer in developing a conceptual foundation for the functional ecol­ogy of marine phytoplankton and the ecosystems they inhabit. Margalef (1978) argued that the most impor­tant determinant of phytoplankton species and functional composition is energy input to the w ­ ater column, specifically turbulent mixing, which ultimately controls availability of both nutrients and light to individual algal cells. He or­ga­nized phytoplankton community composition along an axis in the two dimensions of turbulence and nutrient supply, from well-­mixed, nutrient-­rich ­waters dominated by diatoms to stratified, nutrient-­depleted environments dominated by dinoflagellates (figure 10.5). Subsequent research has expanded the major axes of phytoplankton ecol­ogy to include traits involved in light and nutrient acquisition and use, temperature tolerance, interactions with enemies, cell morphology, temperature sensitivity, and reproductive mode (figure  10.6) (Litchman and Klausmeier 2008, Litchman et al. 2013). No organism can optimize all of t­ hese traits si­mul­ta­neously, so the functional structure of plankton communities emerges from balancing trade-­offs among them along environmental gradients. The trade-­offs arise from fundamental biophysical constraints, including cellular scaling relationships and enzyme kinetics. Suites of functional traits differ among major taxa of marine phytoplankton and contribute to their differential importance along environmental gradients, as we ­will see. A primary trait mediating phytoplankton responses to physical forcing is cell size (Andersen et al. 2016). Cell size influences sinking rate, and thus the ability of cells to remain in sunlit surface ­waters, as well as nutrient uptake b­ ecause smaller cells have higher High Diatoms surface per unit volume. Cell size also influences susceptibility to grazRed tide ing since metazoan zooplankton generally can graze only large cells in dinoflagellates Thalassiosira the nano-­to microplankton range, while smaller picoplankton are inGonyaulax Chaetoceros stead grazed by protists and remain in the microbial loop. Thus, phyChaetoceros toplankton size structure fundamentally determines the proportions Coccolithophorids gy r of energy flowing through the classical food chain—­net plankton to ate r t C. fusus S copepods to fishes—­versus the microbial pathway, and how much C. furca K Ceratium genus carbon is recycled within surface ­waters versus exported to the deep ocean. Subsequent studies have generally corroborated Margalef ’s Ornithocercus scheme: small phytoplankton cells typically dominate in oligotrophic, Dinoflagellates Low stratified ­waters, where nitrogen is available primarily as ammonium Low High regenerated via recycling, whereas large cells dominate better-­mixed Turbulence level ­waters with a supply of new nitrate, resulting in greater export to the deep ocean. A formal empirical model provided quantitative support Figure 10.5. ​A classic summary of how dif­fer­ent for Margalef ’s conceptual model and confirmed that large cells indeed phytoplankton taxa dominate u­ nder dif­fer­ent increasingly dominate as mesoscale vertical motion (upward velocity) combinations of turbulence and nutrient availability increases (Rodriguez et al. 2001). (­after Margalef 1978).

Ocean Ecology

(A)

Ecological function Reproduction

(B) Feeding

Cell size

Morphological

Ecological function

Resource acquisition Predator avoidance Morphological

268

Cell shape Coloniality

Behavioral Sexual/asexual reproduction Resting stages

Survival

Body size Volume to biomass ratio Food particle size range Mixotrophy

Offspring size and number

Maximum growth rate

Stoichiometric requirements/content Feeding efficiency/ clearance rate

Defenses Transparency Bioluminescence Starvation tolerance Basal metabolic rate Longevity

Fecal pellet production Escape responses

Feeding mode

Life history

Motility

Physiological

Environment

Behavioral

Toxin production

Trait type

Physiological

Protein synthesis

Nutrient uptake, requirements

Mixotrophy

Life history

Trait type

Photosynthesis

Growth and reproduction

Motility (pattern, speed) Vertical migration Sexual/asexual reproduction Reproduction frequency Size at maturation

Resting stages/ dormancy

Figure 10.6. ​Trait-­based schemes for classification of plankton. Mea­sure­ments of dif­fer­ent traits (green boxes) allow classification of planktonic organisms into functional types. (A) Phytoplankton (­after Litchman et al. 2008) and (B) zooplankton (­after Litchman et al. 2013).

Most plants and algae interact with grazers, with impor­tant consequences at the ecosystem level. ­Until recently, the potential importance of defenses against herbivory had been largely overlooked in pelagic systems, but this is changing. Unicellular algae are especially vulnerable to grazers as they are small and tend to be highly nutritious. Thus, strong se­lection for defense might be expected. The major enemies of phytoplankton are micrograzers, specifically flagellates and other protozoa, as well as viruses. One reason for the dominance of diatoms in the phytoplankton of many regions may be that the silica cell wall (frustule) provides a relatively inexpensive and effective defense that protects them against such micrograzers (Smetacek 1999). The frustule may also serve as a defense against viruses, which are barred from the living surface of the cell membrane. The outer “skin” surrounding colonies of the planktonic haptophyte Phaeocystis may serve a similar purpose as it appears impervious to viral infection. The plasticity of Phaeocystis in response to enemies (chapter 7) further supports the hypothesis that colony formation is an adaptation to thwart enemies.

Grazers: Major functional types The herbivores of pelagic systems include a wide range of protists and small metazoans, as well as a few larger animals (chapter 5). The most diverse, impor­tant pelagic grazers globally are the microzooplankton, mostly comprising vari­ous taxa of flagellates and ciliates. Calanoid copepods are the dominant mesozooplankton through most of the world ocean and prob­ably the most abundant animals on earth. Among the few larger herbivores of the plankton, the most impor­tant are euphausiid crustaceans (krill)—­which feed on large algal cells, particularly ­those associated with polar sea ice—­ and salps, delicate and transparent relatives of benthic sea squirts that filter bacteria and other picoplankton, primarily in oligotrophic seas (see figure 2.7). ­Because the ocean’s dominant primary producers

Chapter 10 The Open Ocean

are picoplanktonic cyanobacteria, the quantitatively most impor­tant grazers are the heterotrophic protists capable of feeding on ­these tiny organisms. A large part of primary production in the open ocean flows through this microbial loop.

Grazing So far we have focused on the ­factors required for growth of primary producers. But production of biomass is a product of growth and reproduction of many individuals and is therefore a population pro­cess. Growth of any population represents a balance between births and deaths (chapter 6). Thus, controls on biomass production must account for both the requirements for growth and the ­factors imposing mortality. Historically, the science of primary production has focused largely on the first of ­these, the limitation of phytoplankton growth by light and nutrients. Surprisingly, recent research has found that many phytoplankton grow vigorously even when nutrient concentrations in the w ­ ater are below detection limits (Laws 2013). Adoption of the clean sampling and incubation methods that allowed rigorous tests of iron limitation also resulted in removal of vari­ous trace contaminants in lab apparatus that depressed phytoplankton growth. As a consequence, since around 1980 estimates of phytoplankton growth have averaged roughly twice ­those reported ­earlier. The high efficiency of phytoplankton growth in oligotrophic conditions has directed attention to the second term in the s­ imple population equation: mortality. Grazing rates on phytoplankton have often been mea­sured in b­ ottle incubation assays using the dilution method. In ­these dilution assays, phytoplankton biomass growth is mea­sured in a series of aliquots of raw seawater containing plankton that are diluted with increasing fractions of filtered seawater. The assumption is that the lower densities of plankton in the diluted aliquots reduce grazer encounter with phytoplankton cells but do not other­wise change phytoplankton growth rates (Landry and Hassett 1982). In other words, dilution reduces phytoplankton losses to grazing but not gains through phytoplankton growth, and so allows calculation of grazing impacts as the difference between them. Synthesis of a large number of such dilution experiments concluded that microzooplankton graze 49%–77% of pelagic primary production and that this fraction is relatively constant among regions of the ocean. Summed globally, microzooplankton graze 62% of total primary production, five times higher than that of mesozooplankton (Schmoker et al. 2013, figure 9.6). The emerging picture is that phytoplankton accumulation in the open ocean is more generally l­imited by zooplankton grazing than by nutrient limitation (Laws 2013). Thus, understanding the magnitude, distribution, and fate of marine primary production requires bringing herbivores into the equation.

Structure and Organ­ization of Pelagic Communities The tremendous diversity of the plankton contrasts sharply with the seeming monotony of the open ocean’s abiotic environment. This disconnect inspired Hutchinson’s (1961) influential essay “The Paradox of the Plankton.” As we saw in chapter 8, the paradox is how so many species that use similar resources can coexist. Three general solutions, not mutually exclusive, have been offered to resolve it. The first possibility is that plankton species are more specialized than they appear, meaning that they do not in fact use the same resources and instead specialize on par­tic­u­lar combinations of resources and environmental conditions, thereby reducing competition. The second possibility is that competitive exclusion is continually interrupted by disturbances that allow competitively inferior species to persist. A third pos­si­ble explanation, not considered by Hutchinson, is that complex webs of interactions generate inherently chaotic dynamics that prevent competitive exclusion. ­There is evidence that each of ­these mechanisms contributes to pelagic marine diversity patterns. We explore each of them in the subsections that follow.

269

270

Ocean Ecology

Specialization and resource partitioning The competitive exclusion princi­ple states that species with identical resource use and responses to the environment (i.e., identical niches) cannot coexist at equilibrium (chapter 7). Conversely, species can reduce competition and coexist if they use dif­fer­ent types or combinations of resources. Such specialization appears more common among pelagic marine organisms than envisioned when Hutchinson described the paradox of the plankton. A major component of the niche involves requirements for and means of obtaining resources. In phytoplankton, the principal resources are light and nutrients, so resource partitioning involves specialization on par­tic­u­lar segments of the light spectrum or in obtaining nutrients (see boxes 5.1, 5.2). The globally abundant cyanobacterium Prochlorococcus comprises hundreds of genet­ically distinct populations that appear to have diverged millions of years ago, suggesting ancient and stable divergence in their niches (Kashtan et al. 2014). Th ­ ese Prochlorococcus populations differ in pigment content, growth rates, and light and nutrient use patterns, and are therefore referred to as ecotypes (­whether they are distinct species is a difficult question for such asexual organisms). Along a north-­south transect covering most of the Atlantic Ocean from the surface to 200 m, six major ecotypes differ substantially in distribution by temperature (latitude) and depth ( Johnson et al. 2006). Th ­ ese complementary distributions match well with their dif­fer­ent temperature optima and tolerance ranges confirmed by lab experiments, and suggest mechanisms for avoiding competitive exclusion. For example, one ecotype grew better than the other at their optimum temperature, but the slower-­growing ecotype performed better at low temperatures. Thus, the ecotypes partition the pelagic environment, allowing the community of ecotypes to grow u­ nder a greater range of conditions than any single ecotype could. Th ­ ese empirical patterns of resource partitioning w ­ ere also reproduced with a mechanistic model based on functional traits of Prochlorococcus ecotypes (Follows et al. 2007), as discussed in the last chapter (see figure 9.22). Such functional differentiation can operate not just among species but at the level of higher taxa as well. For example, diatoms and dinoflagellates often respond differently to environmental forcing, as Margalef argued. In the En­glish Channel, diatoms grow better at lower light levels, cooler temperatures, and higher nutrient concentrations compared with dinoflagellates. In this case, t­hese higher taxa serve as reliable functional groups, and a common suite of traits can be used to characterize species within each group (Mutshinda et al. 2016). A second axis of niche differentiation involves vulnerability to consumers. Many zooplankton are highly selective feeders, and phytoplankton parasites are even more specialized. Models in which zooplankton graze preferentially on par­tic­u­lar types of phytoplankton predicted three times greater phytoplankton diversity than where grazing was nonselective, and the model predictions agreed well with observations. The mechanism appears to be that variable grazing creates refuges for less competitive phytoplankton types, foiling competitive exclusion (Prowe et al. 2012).

Nonequilibrium dynamics The second suggested solution to the paradox of plankton diversity is the possibility that equilibrium is never actually realized in the constantly changing world, and therefore competitive exclusion is continuously interrupted (Hutchinson 1961). The best-­known expression of this idea is the intermediate disturbance hypothesis (chapter  8), which suggests that diversity is highest at some intermediate point between no disturbance, which eventually allows a dominant competitor to exclude ­others, and a disturbance rate so high that most species are unable to grow fast enough to maintain v­ iable populations (Connell 1978, Huston 1979) (see figure 8.4). In the plankton, Hutchinson argued, weather-­driven fluctuations in the environment could continually disrupt competitive exclusion.

Chapter 10 The Open Ocean

Physical oceanography has corroborated Hutchinson’s hunch that the homogeneous conditions assumed by the competitive exclusion princi­ple rarely if ever occur in the ocean, which is instead characterized by spatial and temporal complexity resulting from mesoscale vortices and fronts. We might then expect constant change in plankton communities. Surprisingly, one of the most comprehensive data sets on marine plankton diversity, from the North Pacific central gyre, found l­ ittle evidence that nonequilibrium conditions maintain plankton diversity (McGowan and Walker 1985). Theories that invoke disequilibrium in maintaining diversity predict shifts in species composition of communities among patches or through time, reflecting the dif­fer­ent states in succession initiated by disturbance or changing conditions. Contrary to such predictions, the order of dominance of North Pacific copepod species was nearly constant across samples collected over large time and space scales, despite episodic changes in the physical regime. ­These patterns suggest that this community, at least, is resilient and that its structure is likely regulated by species interactions, albeit still poorly understood, rather than by physical ­factors.

Chaos Fi­nally, a distinct mechanism not anticipated by Hutchinson is chaos. In contrast to the colloquial sense of that word, which connotes a complete absence of order, chaos in ecol­ogy is deterministic, meaning that it arises from completely specified pro­cesses, which can be described by equations for single or interacting species, in a homogeneous, undisturbed environment. Chaos has been defined verbally as “a sensitive dependence on initial conditions” (Hastings et al. 1993) and can entail large and unpredictable fluctuations in species abundances. This “disturbing” aspect of ecological dynamics (May  1974) was first widely recognized when Robert May reported that even the simplest nonlinear difference equations of single-­species population dynamics could produce stable equilibria, stable cycles, or chaotic fluctuations, depending on the intrinsic rate of population growth. Chaos in ecol­ogy may help resolve a long-­standing puzzle posed by competition theory, which predicts that the number of species coexisting at equilibrium cannot exceed the number of limiting resources. For phytoplankton, as Hutchinson argued, t­ here are only a few such resources: light, nitrogen, phosphorus, silicon, iron, inorganic carbon, and perhaps a few trace metals  (Scheffer et al. 2003). Yet tens to hundreds of species commonly coexist in the plankton. Chaos theory potentially resolves this paradox by showing that, even in a homogeneous and constant environment, species composition and abundances never s­ettle to equilibrium; rather, their interactions can produce irregular oscillations. Specifically, species competing for three or more resources can produce oscillations that allow many species to coexist on only a few resources (Huisman and Weissing 1999). Experiments support the theory. The clearest example comes from a long-­term laboratory experiment with a plankton food web from the Baltic Sea, consisting of bacteria, phytoplankton, zooplankton, and detritivores (Benincà et  al. 2008). The experiment was maintained for more than six years—­hundreds of generations of plankton organisms—­and sampled twice a week. Despite constant external conditions, the species continued to fluctuate over several o­ rders of magnitude as a result of food web interactions. Th ­ ese results demonstrate that interactions among even a small number of species can generate chaos. This raises the sobering prospect that, at the species level, the dynamics of plankton (or any community) might be inherently unpredictable, rendering long-­term predictions impossible. But this outcome seems inconsistent with, for example, the highly stable community structure of plankton in the North Pacific central gyre (McGowan and Walker 1985) described in the last subsection. A frontier for ecol­ogy involves understanding what environmental and biological characteristics determine which of ­these scenarios a community follows.

271

272

Ocean Ecology

Functioning of Pelagic Ecosystems Pelagic ecosystem dynamics are driven ultimately by physical forcing in the form of solar heating and wind, as discussed in chapter 3. Physical forcing selects for par­tic­ul­ ar functional types of organisms, and ­these physical and biological components together mediate ecosystem pro­cesses. The most fundamental ecological pro­cess is primary production. An influential early effort to capture t­ hese links from physics through biodiversity to pelagic ecosystem functioning was Margalef’s (1978) classification of phytoplankton types along the gradient of nutrient concentrations and turbulence, discussed above (see figure 10.5). In this scheme, physical energy governs both the replenishment of nutrients from below the pycnocline and, by controlling stratification, the average light dose experienced by phytoplankton cells. With subsequent advances in physical and biological oceanography, A. R. Longhurst (2007) considered Margalef’s classification too simplistic and argued for a larger number of such “cases,” forced as in Margalef’s model by physical mixing and nutrient concentrations, but with the addition of seasonal changes in light. The variation in t­ hese d­ rivers with latitude (as well as salinity-­mediated stratification driven by ice melt at high latitudes) defines the cases. We considered one of the most impor­tant of ­these cases, the iconic spring bloom, above.

Pelagic food webs: The microbial loop Historically, the ocean’s pelagic food web was thought of as a relatively ­simple food chain beginning with (net) phytoplankton, particularly diatoms and dinoflagellates. Th ­ ese formed the meadows of the open sea and w ­ ere grazed by metazoan zooplankton, dominated in most regions by calanoid copepods and at higher latitudes by euphausiids (krill) (figure 10.7). ­These crustacean zooplankton in turn fed fishes, and marine mammals and seabirds fed on both krill and fish. In the late 1970s this picture began to change. We now know that this classical food chain is only a part, and in some places a minor one, of the ocean’s pelagic food web, the other part being the microbial loop. New technologies revealed that primary production by net phytoplankton like diatoms is commonly only a small fraction of total primary production, the bulk of which is a product of the tiny organisms that pass through fine mesh nets ( 125 µm casses also support communities of specialized invertebrates Organic C that may persist for de­cades (Smith 1992) (see figure B2.2.1B). 1 Benthic animals respond to food falls nearly as quickly as microbes do, behaviorally, metabolically, and reproductively. Camera deployments show that deposit-­feeding sea urchins and holothurians aggregate quickly at patches of sedimented phytodetritus, and meat baits attract dense swarms of fish and 0.1 scavenging amphipods that can strip the bait within hours. The AM J J A S ON D J F M AM J J A S ON D J FM AM J J A windfalls from above are sufficiently predictable that many 1978 1979 1980 deep-­sea animals synchronize their reproductive periods with seasons when high-­quality food deposits on the bottom. Such Figure 10.11. ​Vertical flux to the deep ocean tracks seasonality patterns are now known in a wide variety of deep-­sea animals, in surface production. Seasonal changes in flux of total particulate ­matter and of its components to a sediment trap at 3200 m off including crustaceans, echinoderms, and sponges, although the Bermuda (­after Deuser et al. 1981). trend is far from universal (Gooday 2002). In such cases, ­either yolk deposition in eggs or release of juveniles or both are synchronized with periods of maximum food availability on the bottom, generally the sedimentation of blooms during late spring and summer.

Deep-­sea biodiversity In the early 1800s the abyssal ocean was understandably assumed to be devoid of life. The Challenger expedition dramatically shattered that notion, and exploration with new technologies in the mid-­ twentieth ­century proved that, although sparsely populated, the deep sea is among the most diverse environments on earth. Sampling along a transect through the Northwest Atlantic with a newly devised benthic sled, Robert Hessler and Howard Sanders (1967) found to their surprise that deep-­sea diversity was higher than in shallow ­water and second only to that of coral reefs (figure 10.12). ­Later intensive sampling on the continental slope and rise off New Jersey, USA, pushed estimates of deep-­sea diversity through the roof, turning up 898 species from more than 100 families and 12 phyla, with the most abundant species seldom comprising more than 8% of individuals (Grassle and Maciolek 1992). The emerging picture presented an analogue to the famous paradox of the plankton: How can so many species coexist in such a seemingly monotonous environment? To resolve it, several hypotheses w ­ ere advanced that correspond roughly with t­ hose suggested for plankton (Hutchinson 1961). Hessler and Sanders ­were the first to wrestle with the paradox. They proposed that environmental stability in the deep sea, and a long history without disturbance, had fostered the evolution of specialized

280

Ocean Ecology

Number of polychaete and bivalve species

resource use in the face of competition. They called this scenario “biological accommodation,” essentially equivalent to resource partitioning, and contrasted it with the suggested dominance of 75 adaptation to physically dynamic habitats in shallow w ­ aters. This hypothesis reflected the strong focus on local competition and 50 coevolution within communities that prevailed throughout the 1960s and 1970s, flowering from the seeds planted by Robert MacArthur (chapter 8), but the emphasis on evolutionary di25 versification over large, stable areas aligns well with the modern understanding of long-­term drivers of diversity. The biological 0 accommodation hypothesis is difficult to test, but it is not en0 500 1,000 1,500 2,000 2,500 3,000 Number of individuals sampled tirely consistent with the finding (Rex 1981) that diversity of several benthic taxa peaks on the continental slope and rise rather Tropical shallow water Outer continental shelf Tropical estuary Boreal shallow water than in the presumably more stable abyss. Biological accommoSlope (deep sea) Boreal estuary dation must have a limit—­likely set by the low food availability on the abyssal plain, where few macrofaunal species are able to Figure 10.12. ​The deep-­sea biodiversity surprise. Species maintain ­viable population sizes. accumulation curves for polychaetes and bivalves on a A related hypothesis emerged as further exploration deep-­sea transect across the Northwest Atlantic continental showed that the deep seabed is much more heterogeneous slope into the abyss, compared with other marine benthic than previously supposed, with wide variation in sediment habitats (­after Sanders and Hessler 1969). type, physical disturbance, and quantity and nature of food fall from above. Some of this variation stems from physical processes—­scouring currents, flows of sediment and organic m ­ atter down undersea canyons to abyssal deltas, formation over eons of manganese nodules, clusters of rock outcrops that create complex landscapes, and of course hydrothermal vents. Another part of the variation is biological: populations of xenophyophores—­giant, bush-­like protists—­and other organisms create structure on other­w ise featureless sediments (see figure  2.4A), bonanzas of phytodetritus fall from productive surface ­waters, and oases of specialized life colonize and persist for de­cades on the carcasses of dead ­whales (see figures B2.2.1B, 10.10E). Such habitat heterogeneity can enhance diversity if species use dif­fer­ent microhabitats preferentially or are superior competitors in dif­fer­ ent microhabitats. As suggested in the discussion of the paradox of the plankton, the maintenance of high diversity by spatial heterogeneity depends on species specializing on dif­fer­ent microhabitats (i.e., resource partitioning) or responding differently to disturbance. ­There is evidence of both pro­cesses in the deep sea. Microhabitat specialization is illustrated by animal communities around deep reefs and other features that differ from t­ hose in ambient sediments (Levin and Sibuet 2012). Infaunal invertebrates also tend to specialize on sediments of dif­fer­ent grain size, which influences both habitat quality and food availability. For example, diversity of invertebrate species increases with sediment grain size diversity in benthic communities of the West Atlantic continental slope, and this effect is more impor­tant than that of depth (Etter and Grassle 1992). Evidence that deep-­sea species respond differently to disturbance is seen in benthic communities of the abyssal Northeast Atlantic, where freshly deposited phytodetritus patches supported abundant, low-­diversity assemblages of foraminifera dominated by species other­w ise rare in surrounding sediments, suggesting that ­these species are opportunists that specialize on patchy accumulations of rich food (Gooday 1988). Collectively, ­these cases show that deep-­sea species span a range in microhabitat use and response to disturbance, and support the idea that a mosaic of habitat and disturbance patches contributes to the maintenance of high diversity in the deep sea. Thus, the situation is consistent with Hutchinson’s (1961) suggested solutions to the paradox of the plankton, involving specialized niches and perhaps also disruption of competitive exclusion by disequilibrium. 100

Chapter 10 The Open Ocean

Chemosynthetic Ecosystems: Vents and Seeps The most spectacular moment in twentieth-­century oceanography was the discovery in 1976 of lush oases of life around deep hydrothermal vents amid the desert of the deep-­sea floor (figure 10.13). Geologists had long suspected that mid-­ocean spreading ridges produced hot hydrothermal discharges, but they had never been observed. On an expedition to the Galápagos rift zone with a deep-­ towed vehicle, they confirmed t­ hese plumes of warm w ­ ater, but also captured images of an unpre­ce­ dented landscape of yellow and white chemical precipitates on the rocks, with populations of clams and anemones at densities far higher than typical of the deep sea (Lonsdale 1977). No one had seen anything like it before. A second expedition was soon mounted to the mid-­ocean ridge of the East Pacific Rise off the coast of Mexico using the manned submersible Alvin, this time including biologists. The results ­were even more astounding. The submariners came upon clusters of big clams, dense groves of the largest tube worms ever seen, and scuttling crustaceans amid geysers shooting out black clouds of mineral-­rich ­water. Many of the dominant animals ­were from genera, families, and even o­ rders new to science. It was obvious that something very strange was g­ oing on: animal biomass was ­orders of magnitude higher than in the surrounding desert of the deep sea. Microbial biomass was dense and vis­i­ble to the naked eye as white mats resembling the sulfide-­oxidizing bacterium Beggiatoa coating rocks and sediments around the vents. What ­were they ­doing ­here? ­These early observers speculated that the density of apparent suspension-­feeders might reflect entrainment of bottom currents bearing food t­ oward the ridge by the rising plume of warm vent w ­ ater, but they also raised the possibility that chemosynthetic bacteria might contribute.

A

B

C

D

Figure 10.13. ​Life at deep-­sea chemosynthetic vents and seeps. (A) Aggregations of the shrimp Rimicaris hybisae at deep hydrothermal vents on the mid-­Cayman rise. (B) Crabs aggregate on a mussel bed at a deep methane seep in the northern Indian Ocean. (C) The ­giant tube worm Riftia pachyptila, powered by symbiotic sulfur-­reducing bacteria, clustered around a plume of hot hydrothermal w ­ ater. (D) An aggregation of methane ice worms burrowed into methane hydrate “ice” at a cold seep, deep in the Gulf of Mexico.

281

282

Ocean Ecology

Hydrothermal vents Solving the mystery of what supports ­these anomalous oases of high productivity begins with the unique physical setting of vents. The earth’s crust is divided into a number of tectonic plates that move slowly, evolve in shape, and jostle against one another over geologic time driven by convection cells in the earth’s hot mantle (chapter 3). That convection brings hot molten rock to the surface along seams between the plates, especially where the upward pressure drives them apart, accreting new rock at the trailing edges of the plates as they cool. The hot mantle material below them buoys ­these areas up above the surrounding ocean floor, where they are detectable as mid-­ocean ridges, linked in a network of diverging crustal plate bound­aries around the globe (figure 10.14). At t­ hese spreading centers, where the first hydrothermal vents w ­ ere discovered, w ­ ater heated by molten rock close to the surface interacts with the basaltic rock to produce hot (> 300°C) fluids, rich in H2S and other reduced compounds, that rise and escape through vents in the rock. Bacteria are very dense in the hot ­water, most suspended in clumps or associated with mineral particles. The bacteria teeming in the mineral-­rich vent ­waters form the base of the unique food web of hydrothermal vents. ­These bacteria are chemoautotrophic, and the dominant types obtain energy by oxidizing the hydrogen sulfide spewing from the vents, using dissolved oxygen in the ­water as an electron acceptor and using the energy produced to fix carbon (table 2.3). The vis­i­ble microbial mats growing in the reduced, mineral-­rich ­waters provided the first clues that vent communities might be supported by such chemosynthetic bacteria. Experiments confirmed that bacteria w ­ ere able to grow and produce in the almost inconceivably hot ­waters shooting out of the “black smoker” chimneys. At the intense pressure of the deep sea, ­water remains liquid to at least 460°C. Bacteria isolated from ­these hot vent ­waters and cultured in pressurized titanium vessels grew chemoautotrophically at in situ conditions of 265 atm and at least 250°C, in fact growing faster at 250°C than at surface temperatures (Baross and Deming 1983). Chemoautotrophic bacteria also sustain the profusion of macroscopic life through symbiosis. The most con­spic­u­ous and abundant macrofauna at the East Pacific vents are tube worms and clams that resemble shallow-­ water suspension-­ feeders (see figure  10.13C). But the resemblance is

Figure 10.14. ​Global distribution of hydrothermal vent fields.

Chapter 10 The Open Ocean

superficial—­most are instead sustained by sulfide-­oxidizing bacteria within their tissues. The iconic tube worm Riftia pachyptila that dominates East Pacific vents is a ­giant—up to 3 m long and 5 cm in dia­meter—­but belongs to a group of other­wise small, aberrant polychaetes characteristic of sulfide-­ rich coastal sediments. ­These bizarre animals have neither mouth, gut, nor anus. Instead, most of the Riftia body consists of a mass of tissue called the trophosome filled with bacteria as well as crystals of elemental sulfur—­a telltale sign of sulfide oxidation (Cavanaugh et al. 1981). The symbiotic bacterial community can fix carbon and reduce inorganic nitrogen (Felbeck 1981). Similar symbioses with chemosynthetic bacteria are found in several other vent invertebrates, including bivalves, gastropods, and apparently also shrimp. Thus, hydrothermal vent ecosystems are essentially self-­sufficient, requiring no food from the surface. They are founded not only on chemosynthesis but, like coral reefs, also on an intimate symbiosis between habitat-­forming invertebrates and microbes. Most of the dominant vent animals are symbiotic with bacteria that use hydrogen sulfide, methane, or even elemental hydrogen to power chemosynthesis (Petersen et al. 2011). The clouds of suspended bacteria and dense populations of symbiotic invertebrates around ­hydrothermal vents strikingly illustrate the importance of bottom-up control by resources in the deep ocean. Although vent communities typically have ­orders of magnitude greater biomass than the surrounding deep-­sea floor, diversity at vents is much lower, as is typical of habitats with very high resource availability or extreme environmental conditions. As in other marine systems, predation also is impor­tant in shaping the biomass and species composition of the vent communities. The organisms of vents are unique, but the functional groups are familiar from shallow-­water communities: sessile invertebrates competing for space, limpets grazing the rock surfaces (albeit feeding on bacteria), and abundant predatory fishes and crabs. Despite the formidable logistical challenges, experimental manipulations at 2500 m on the East Pacific Rise showed that predators structure the vent community both directly by eating gastropods and indirectly by reducing gastropod grazing and disturbance of recruiting sessile invertebrates (Micheli et al. 2002). The top-­down effects of predation are strongest near the vents, where productivity and abundances of organisms are highest, even though environmental conditions are most extreme near the vents. ­These experiments confirmed the generality of several community organ­izing pro­cesses familiar from shallow habitats: predators strongly affect community composition through both direct and indirect interactions, and ­those interactions are often stronger in more productive habitats. Since the first discovery of hydrothermal vents in the 1970s, lush communities have been identified around hydrothermal vents throughout the world ocean, but their character differs greatly among regions (Van Dover et al. 2002, Ramirez-­Llodra et al. 2007) (see figure 10.13). In contrast to the tube worm groves of the East Pacific, hydrothermal vents on the mid-­Atlantic ridge are dominated by dense swarms of the unique shrimp Rimicaris exoculata, which appears to be nourished by absorption of soluble organic products from symbiotic, chemoautotrophic bacteria living in its gill chambers (Ponsard et al. 2012). In the West Pacific, the vents are populated primarily by limpets, barnacles, and mussels. Indian Ocean ridge communities have a mix of ­these taxa.

Cold seeps Cold seeps refer to areas of the seafloor where sediment pore ­water, rich in methane and sometimes other hydrocarbons, leaks out of the sediment as a result of pressure forcing ­these substances upward from deep subsurface reservoirs. Seeps occur scattered along the world’s continental margins and are especially abundant in the Gulf of Mexico. The methane-­rich sediments are typically anoxic and highly reduced, and support dense microbial communities dominated by anaerobic Archaea, which oxidize the methane in conjunction with sulfate-­reducing bacteria (Ruff et  al. 2015). Like their counter­parts at hydrothermal vents, t­hese microbes include both free-­living cells and species involved in intimate symbioses with the dominant invertebrates, often of the same taxa found at hot

283

284

Ocean Ecology

vents, including siboglinid tube worms and large bivalves. The vigorous production of t­ hese chemosynthetic microbes forms the base of the food web (MacAvoy et al. 2005). The microbes also indirectly provide habitat for sessile organisms by precipitating carbonate as a result of coupled methane oxidation and sulfate reduction, and in some cases by altering seeping hydrocarbons. The seep carbonates form a hard substratum that is colonized by bivalves and tube worms, creating three-­ dimensional structure. Like hydrothermal vents, cold seeps support dense accumulations of invertebrate megafauna that greatly exceed that of the surrounding deep-­sea floor away from the seeps (Levin 2005). Cold seeps feature some of the more bizarre adaptations in the animal kingdom. The g­ iant tube worm Lamellibrachia luymesi grows in large clumps in hydrocarbon-­rich areas along the slope of the Gulf of Mexico, where it can live at least 170 years—­and may be the longest-­lived noncolonial invertebrate known (Bergquist et al. 2000). Uniquely among its kin, this species produces long extensions of its tube and tissues, termed roots, that grow into the under­lying sulfidic sediments and can be longer than the above­ground part of the animal. Experiments showed that the term roots is apt—­the worm acquires sulfide through its roots from the surrounding highly reduced sediments, and its symbiotic bacteria oxidize the sulfide to generate the energy required to fix inorganic carbon, which is absorbed from the overlying, oxygenated ­water (Freytag et al. 2001). Arguably even stranger is the methane ice worm (Hesiocaeca methanicola; see figure 10.13D) that inhabits gas hydrate outcrops in the Gulf of Mexico. Gas hydrates, also known as methane clathrates or, more colorfully, fire ice, are ice-­like crystalline solids composed of methane and other gases entrapped within a lattice of w ­ ater molecules and formed u­ nder the high pressures and low temperatures of deep continental margins. Methane hydrates occur as deposits that can exceed 2 m in thickness in ­these deep ­waters. They are estimated to contain large quantities of methane, which are both attractive to modern society as a potential fuel source and a worrisome climate h­ azard due to the instability of ­these hydrates in warming conditions and methane’s high climate warming potential as a green­house gas. Surfaces of methane hydrate outcrops in the Gulf of Mexico are inhabited by dense populations of the methane ice worm, which feeds on the methane-­oxidizing bacteria that coat the surface of the ice, as confirmed by stable isotope values (Fisher et al. 2000).

Macroecol­ogy of the Open Ocean The open ocean’s pelagic and benthic habitats are the largest ecosystems on earth and, b­ ecause of challenging logistics, among the least well sampled. For t­ hese reasons, application of macroecological princi­ples offers promise for understanding the structure and functioning of the open-­ocean ecosystem. We have already mentioned applications of macroecol­ogy in modeling global fish production and developing null distributions for mea­sur­ing fishing impacts (chapter 9; see figure 5.12). ­Here we explore macroecological patterns of diversity, energetics, and ecosystem pro­cesses in open-­ocean communities.

Controls on biodiversity in the open ocean On land, plant species diversity at the local scale is controlled primarily by resource availability (often confusingly referred to as productivity) in many systems (Waide et al. 1999). Plant diversity commonly shows a hump-­shaped relationship, with maximal diversity at intermediate levels of resource availability. Similar patterns are also common among animals (Rosenzweig and Abramsky 1993). Explaining the mechanisms for and variation in such productivity-­diversity relationships is among the oldest challenges in ecol­ogy. In the ocean, evidence similarly supports a hump-­shaped relationship between biodiversity and resource availability in both oceanic plankton and deep-­sea benthos. Phytoplankton of the Northwest Atlantic have been extensively sampled using flow cytometry,

Chapter 10 The Open Ocean

which sorts cells by size and the optical properties of their pigments. Cytometric diversity calculated from such samples is a mea­sure of functional diversity and is maximal at intermediate biomass levels (figure 10.15) (Li 2002). Making the common assumption that standing biomass is a proxy for resource availability, this equates to a hump-­shaped relationship of diversity with resource availability. Both diversity and composition of phytoplankton are also related to water-­column stratification, providing large-­scale support for Margalef ’s (1978) view that physical energy input to the ­water column is a major control on plankton community composition and function. Picoplankton dominated ­under the nutrient-­depleted conditions of warm, highly stratified w ­ aters, with larger cells increasing as mixing broke down stratification. At intermediate levels of mixing, all size classes ­were similarly abundant, yielding maximal diversity, consistent with the intermediate disturbance hypothesis, with mixing acting as a disturbance that prevents competitive exclusion (Li 2002). ­These patterns are corroborated by a compilation of data from marine pelagic systems around the world, which confirmed that species richness of phytoplankton and microzooplankton ­were each unimodally related to resource availability, proxied as total biomass (Irigoien et al. 2004), paralleling the common pattern in terrestrial vegetation. The functional composition of phytoplankton also varied predictably along the gradient: as in the Northwest Atlantic example, unproductive areas with low phytoplankton biomass w ­ ere typically dominated by a few species of small pico-­or nanoplankton, whereas large cells dominated at higher biomass, and assemblages with very high biomass ­were nearly always dominated by a single large phytoplankton species (see figure  10.15). Surprisingly, ­these unimodal patterns linking diversity and biomass w ­ ere consistent across multiple regions with quite dif­fer­ent environmental conditions, suggesting a very general macroecological relationship. Among animals, population growth is generally l­ imited by chemical energy in the form of organic ­matter (food). Species-­energy hypotheses (chapter 8) posit that species richness rises with the availability of such food energy (Evans et al. 2005). A challenge to testing the species-­energy hypothesis is that temperature and chemical (food) energy availability are often closely correlated at large geographic scales and therefore difficult to separate (chapter 3). The deep sea provides an opportunity to disentangle them. H ­ ere temperature is low and varies relatively ­little, whereas availability of food energy varies widely—­food is severely ­limited far from sources of surface production, such as ­under central gyres, but is more plentiful u­ nder productive surface w ­ aters. Several syntheses (Rex 1981, Levin

(A)

(B) 100 Cytometric diversity (exp H’)

Phytoplankton (cells m–2)

10

14

1013 1012 10

11

1010 109 10

8

107 0 10

19 17

80

15 13

60

11 9

40

7 5

20

3 1

10

1

10 Chlorophyll (mg m–2) 2

10

3

10

4

0

1

2

3 4 Log (phytoplankton ml–1)

Figure 10.15. ​Macroecological patterns of plankton diversity. (A) Abundance versus total chlorophyll concentration of pico-­(blue), small nano-­(yellow), and large nanophytoplankton (green). (B) Cytometric diversity (i.e., based on cell size and chlorophyll content, rather than taxonomy) versus phytoplankton abundance; color scale denotes density of observations (­after Li 2002).

5

6

285

286

Ocean Ecology

et al. 2001) have documented hump-­shaped relationships of deep-­sea species richness to depth (chapter 3) and have suggested the decline in richness in the deepest w ­ aters of the abyss is explained by low food supply t­ here. Potential mechanisms could include the difficulty of adapting to chronically severe food shortage and/or chronic inability to find mates (Allee effect) at the very low population abundances of the abyss. The species-­energy hypothesis is supported for several taxa of deep-­sea animals. Species richness of brittle stars (Ophiuroidea), which are common deposit-­feeders in the deep sea, is greatest at high latitudes and near continental margins in areas of high organic flux from the surface, supporting the species-­energy hypothesis (Woolley et al. 2016). Similarly, deep-­sea mollusk diversity is unrelated to temperature and instead peaks at intermediate levels of chemical energy (particulate organic carbon) flux (McClain et al. 2012). Deep-­sea benthic foraminifera show a contrasting pattern: variation through cores over the past 130,000 years showed that richness increased with bottom-­water temperature but was unrelated to productivity (Hunt et al. 2005). One complication with interpreting such patterns is that if the under­lying relation of diversity with productivity is hump-­shaped, as much evidence suggests, the correlation within any single study might be negative, neutral, or positive depending on the range of productivity studied. On balance, the evidence suggests that food energy is an impor­tant driver of deep-­sea diversity, though the mechanisms responsible remain uncertain.

Global controls on microbial diversity Most of the pelagic ocean’s organisms are microbes. Explaining large-­scale distribution of microbial diversity represents a special challenge, which has seen a surge of new interest as a result of the genomic revolution. Two schools of thought prevail in efforts to understand biogeographic distributions (chapter 3)—­ one is ecological, seeking explanations in environmental ­drivers, and the other is historical, emphasizing chance-­based dispersal. Microbiologists have long emphasized ecological control, based on the huge populations and ease of transport of microbes, and assumed that variation in microbial communities must be dictated by responses of an essentially global species pool to local conditions. This view was famously described as “every­thing is everywhere, but the environment selects,” attributed to the Dutch microbiologist Lourens Gerhard Marinus Baas Becking in the 1930s (de Wit and Bouvier 2006). ­There are two central questions. The first concerns pattern: How do microbial diversity and composition vary across the globe? The second is about under­lying pro­cess: Is spatial variation caused by current environmental conditions, by historical contingencies such as dispersal, or both? Addressing the first question, many microbes are indeed widely distributed, but microbial communities clearly differ strongly through space. ­There are many examples of microbial communities molded by environment, as we have seen (e.g., figure 10.5). For example, the species composition of planktonic diatoms in the Atlantic Ocean is strongly associated with nutrient availability, and this relationship has been stable over the past 250,000 years through major climate-­related biogeographic changes (Cermeño et al. 2010). Regarding the second question, t­here is growing evidence of provincialism among microbes, contrary to the “every­thing is everywhere” hypothesis. This is vis­i­ble in the clustering of microbial species by geographic distance and is generally taken to indicate historical dispersal limitation (Martiny et al. 2006, Sul et al. 2013). Among marine planktonic eukaryotes, abundant species tend to be widely distributed, whereas the numerous rare species show much stronger geographic clustering (Logares et al. 2014). Microbial dispersal may be less universal than historically believed for several reasons. Microbes may be unable to tolerate the stress accompanying dispersal, or they may be poor competitors against natives in the new site. ­There is growing evidence that marine bacteria follow several of the macroecological patterns familiar from larger organisms, notably latitudinal gradients in diversity. Bacterial richness declines with latitude and increases with temperature, supporting the pattern and hypothesis familiar from larger organisms (chapters 5, 8) that metabolic activity is a fundamental f­ actor in driving patterns of

Chapter 10 The Open Ocean

diversity (Fuhrman et al. 2008, Sul et al. 2013). Bacteria also have narrower latitudinal ranges in the tropics than at higher latitudes (Sul et al. 2013), conforming to the pattern known as Rapoport’s rule (Stevens 1989) among macroscopic organisms. The emerging picture suggests that major global patterns of diversity are indeed similar among microbes and macroscopic organisms.

Macroecol­ogy of open-­ocean ecosystem pro­cesses The metabolic balance of the open ocean, ­whether net autotrophic or heterotrophic on large scales, remains a subject of debate. Metabolic theory suggests that the balance between photosynthetic production and respiration in a system depends strongly on environmental temperature, as respiration increases more steeply than photosynthesis with temperature (chapter 5). Metabolic theory also defines allometric scaling relationships that connect body size, population density, and temperature within ecosystems. Compilation of a large data set on marine plankton confirmed the prediction from metabolic theory that respiration increases more steeply with temperature than does photosynthesis. Application of ­these relationships to an in­de­pen­dent, global data set of more than 1000 mea­ sure­ments from marine systems confirmed a close match between empirical data and metabolic theory: the mea­sured ratio of plankton community respiration to production increased with temperature at close to the predicted rate, and the threshold primary production required for metabolic balance also increased with temperature (López-­Urrutia et al. 2006) (figure 10.16). Although this approach cannot resolve the question of ­whether the ocean is currently net autotrophic or heterotrophic, it yields two general conclusions. First, the balance is highly sensitive to environmental temperature, shifting ­toward heterotrophy with warming. Second, it suggests that ocean communities mediate a positive feedback on climate warming: as the ocean warms, the lower net production at high temperatures results in less CO2 flux into the ocean, which is instead retained in the atmospheric green­ house. First-­order calculations based on this analy­sis suggest that the epipelagic ocean ecosystem ­will capture 21% less CO2 by the end of the current ­century (López-­Urrutia et al. 2006). ­These sobering conclusions emphasize the critical importance of further research to understand the controls on the net carbon balance of the ocean (Ducklow and Doney 2013). The metabolic theory of ecol­ogy has also produced general inferences about energetics in deep-­ sea ecosystems, where energy fluxes are difficult to mea­sure. Synthesis of extensive data on marine vertebrates and invertebrates showed that the metabolic rates of deep-­sea animals scaled consistently with temperature and body mass and in fact followed an allometric relationship indistinguishable from that of shallow-­water species, with a mass exponent (−0.20) only slightly lower than the −0.25 predicted by metabolic theory (McClain et  al. 2012). Metabolic rate was unaffected by depth—­a proxy for the availability of food energy from surface ­waters—­but instead varied among taxonomic groups. Carbon flux did, however, predict body mass, with areas of higher flux supporting larger organisms as well as higher abundance and larger population biomass. The surprising conclusion appears to be that, in the deep sea, individual animals are metabolizing and growing as fast as temperature allows, given their body size, and that energy (food) availability is impor­tant mainly in controlling total abundance and biomass (McClain et al. 2012). The consistency of ­these allometric relationships suggests that they can be usefully employed in models of deep-­sea response to warming and altered carbon flux. We have seen that carbon flux influences diversity in the deep sea. What about the opposite effect, that of biodiversity on ecosystem pro­cesses? Evidence from deep-­sea benthos appears to support the general rule that biodiversity often increases ecosystem productivity and efficiency of resource use. Across more than 100 deep-­sea sites around the globe, several mea­sures of ecosystem functioning ­rose with the diversity of nematodes, which dominate benthic invertebrate communities, but w ­ ere unrelated to ambient temperature, depth, or carbon flux (Danovaro et al. 2008). ­These ecosystem pro­cesses included benthic bacterial production, organic ­matter recycling rate, and meiofaunal biomass. The proposed mechanism involves enhanced bioturbation by benthic invertebrates. The sur-

287

Ocean Ecology

(A)

(B) 6 Predicted gross primary production

Predicted community respiration

3 2 1 0 –1

4 2 0 –2 –4 –6 –8

–1

0

1

2

3

–8 –6 –4 –2

In situ O2 consumption In (mmol O2m–3d–1)

0

2

4

6

In situ O2 production In (mmol O2m–3d–1)

(C)

(D ) 2

4 Threshold GPP In (mmol O2m–3d–1)

In (production/respiration)

288

2 0 –2 –4 –6

1 0 –1 –2 –3

0

10

20

Temperature (°C)

30

0

10

20

30

Temperature (°C)

Figure 10.16. ​Temperature affects the metabolic balance of the ocean. Mea­sured estimates of (log) plankton (A) community respiration and (B) gross primary production (GPP) versus predictions of metabolic theory. Dashed lines represent the 1:1 relationship. (C) Temperature dependence of the ocean’s metabolic balance (GPP/community respiration) based on application of metabolic theory to empirical data. (D) Temperature effects on change in threshold GPP that separates heterotrophic (negative values) from autotrophic (positive values) communities (­after López-­Urrutia et al. 2006).

prising exponential relationship between richness and functioning in this ecosystem has also been hypothesized to reflect facilitation of positive interactions among species in more diverse communities (Mora et al. 2014).

Deep-­Sea Fisheries As the industrial boom of the late twentieth ­century’s ­Great Acceleration (chapter 4) depleted fish stocks throughout the world’s continental shelves, the hunt turned to the last frontier. The distant and forbidding deep sea had long been protected from fishing, but once the eyes of technology penetrated the depths, they quickly located large aggregations of deep-­water fish congregating around seamounts, where habitat and food are concentrated. Th ­ ese aggregations proved an irresistible target and set off a frenzy of exploitation. Global fishery landings of bottom-­associated marine fishes have rapidly shifted to progressively deeper species over the last 50 years (figure 4.18C), and ­these fisheries are currently exploiting the last refuges on earth for commercially harvestable fish species (Morato et al. 2006). The trend was exacerbated by the location of many seamount populations in high seas beyond national jurisdictions, where a lawless atmosphere has long prevailed (Merrie et al. 2014). B ­ ecause

Chapter 10 The Open Ocean

the cold, food-­poor deep sea selects for late maturation, slow growth, and low fecundity, deep-­sea fishes are highly vulnerable to exploitation, and such fisheries are often depleted within 5–10 years (Koslow 2000). The impacts are not just economic: surveys of five deep-­water fish species in Canadian ­waters showed population declines of 87%–98% over a 17-­year period—­less than a single generation—­qualifying ­these species as critically endangered in the Northwest Atlantic by IUCN criteria, which have not historically been applied to deep-­sea fishes, largely ­because of po­liti­cal sensitivity (Devine et al. 2006). With such slow population growth and recovery from disturbance, deep-­sea species rarely if ever support fisheries that are both sustainable and eco­nom­ically ­viable. It has frequently been noted that harvesting such species is more akin to mining than to the usual case of harvesting a renewable resource. Perversely, “like old-­growth trees and ­great ­whales, their biomass makes them tempting targets while their low productivity creates strong economic incentive to liquidate their populations rather than exploiting them sustainably” (Norse et al. 2012). The poster child for the dilemma of deep-­sea fisheries is the orange roughy (Hoplostethus atlanticus) (box 10.2). But other deep-­sea fisheries have seen similar rapid boom-­and-­bust histories based on exploiting spawning aggregations (Roberts 2002). ­These include the pelagic armourhead (Pseudopentaceros wheeleri) fished in international ­waters northwest of Hawaii beginning in the late 1960s, which peaked in 1976, collapsed the following year, and never recovered. In the North Atlantic, fisheries for blue ling (Molva dypterygia) followed a similar trajectory, with survival of the fishery dependent on continually locating new aggregations ­after rapid depletion, reminiscent of slash-­and-­burn agriculture. On top of the unsustainable harvest of the target stocks themselves, deep-­sea fishing imposes tremendous “collateral damage” (Roberts 2002). This includes bycatch of nontarget fishes, which often constitute a large fraction of the catch and are invariably killed when hauled up from the deep, as well as destruction of sensitive biogenic habitats by the heavy trawls used at ­these depths. The latter include unique reefs of cold-­water corals (see figure 10.10C) that are centuries or even millennia old and often rich in endemic species. Th ­ ese coral habitats are used as nurseries by deep-­sea fishes, and b­ ecause of their extremely slow growth rates, are likely to require centuries to recover from trawling damage. Destruction of such corals represents an “externality” cost that is generally ignored in the narrowly focused analyses of economic sustainability typical of fisheries.

The Open Ocean in the Anthropocene The open ocean is Earth’s last g­ reat wilderness (Ramirez-­Llodra et al. 2011). The forbidding high seas that challenged mari­ners for millennia have been transformed within two generations by massively expanded industrial fishing, pollution (especially by per­sis­tent plastics), a steadily warming climate, and climbing concentrations of dissolved CO2. B ­ ecause of the world ocean’s ­great volume, ­these developments have far-­reaching implications for the dynamics of the earth system as a ­whole and for ­human society. In recent de­cades, humanity’s strongest impacts on the deep sea have shifted from pollution by waste disposal in the past, to exploitation of resources (including fisheries and mineral extraction) at pre­sent, to projected warming and acidification in the ­future.

Climate and the Anthropocene ocean The most far-­reaching aspect of global change is climate warming, since the pervasive role of fossil fuel in the global economy and consequently rising anthropogenic green­house gas concentrations have set us on a course that is unlikely to be reversed for de­cades. Roughly a quarter of the CO2 emitted by ­human activities each year diffuses into the surface ocean, where it is already altering ocean temperature and acidity, trends that are reasonably well documented and understood (Doney et al. 2011). Models predict that by 2100 the surface ocean may be, on average, 2°C–3°C warmer, 0.2 pH

289

290

Ocean Ecology

Box 10.2 ​The tumultuous ­career of the orange roughy The deep-­sea fish Hoplostethus atlanticus lives in large aggregations around the deep slopes of seamounts throughout the world ocean. Originally known by the colorful name “slimehead,” H. atlanticus was rebranded as “orange roughy” by the US National Marine Fisheries Ser­vice in the 1970s to make it more marketable. Orange roughy was first fished intensively in that de­cade, and the concentrated aggregations and relatively large size of the fish (up to 75 cm) rapidly attracted intense fishing pressure through the 1980s, particularly around southeastern Australia and New Zealand, where early years of the fishery produced catches of spectacular abundance (figure B10.2.1), and it quickly became a highly valuable fishery. The life history of orange roughy is typical of deep-­sea fishes: they can live to at least 125 years and possibly to nearly 150, and individuals generally do not reach sexual maturity u ­ ntil the tender age of 25–30. ­These extremely slow demographic rates make orange roughy (and other deep-­sea fishes) highly vulnerable to industrialized fishing, as does their habit of aggregating around seamounts,

A

where they are efficiently harvested in large numbers. Estimates suggest that the maximum sustainable yield of orange roughy is about 2% of its virgin biomass (Clark 2001). The fragility of deep-­sea fish populations is illustrated by the Irish commercial fishery for orange roughy, which opened in 2001, peaked the following year, declined rapidly, and was closed in 2005. A bioeconomic analy­sis concluded that, perversely, the fishery would not have been ­viable even for this brief period without government subsidies (Foley et al. 2011)—­a well-­meaning but counterproductive effort, common to many countries, meant to help fishers left dry by the collapse of overfished shallow-­water species. Orange roughy is still fished in New Zealand, and it appears that some stocks have remained stable u ­ nder a regime of s­ imple harvest control rules (Doonan et al. 2015). To what extent the continued production of t­ hese stocks is evidence of good governance versus replenishment of fished stocks from elsewhere is unclear due to poor understanding of recruitment.

B

Figure B10.2.1. The deep-­sea fishery for orange roughy. (A) Hoplostethus atlanticus, the species formerly known as slimehead. (B) Hauling a catch of orange roughy aboard a fishing vessel off Tasmania.

units more acidified, 2%–4% lower in oxygen concentration, and 2%–20% less productive, a magnitude of change that has not been seen for 20 million years (Mora et al. 2013). The ocean’s biological pump is a key control on Earth’s climate, ­because changes in both primary production in surface ­waters and export flux to the deep ocean strongly influence atmospheric CO2 concentrations over geologic time. Primary production and export fluxes depend in turn on physical forcing that determines mixed-­layer depth, nutrient fluxes to and within the ocean, and food web structure (Falkowski et al. 1998). Accurate predictions of ­future climate ­will depend on better understanding of ­these pelagic ecosystem interactions. What difference does it make if the world warms up a few degrees? Warming directly influences ocean circulation and stratification, which in turn affect biological productivity and oxygen concen-

Chapter 10 The Open Ocean

trations via the magnitude of exchange across the pycnocline (Behrenfeld et al. 2006). In the open ocean, a principal effect is increasing water-­column stratification due to warmer average surface ­waters as well as freshening, particularly at high latitudes due to melting of ice caps (Doney et al. 2011). Stratification of the w ­ ater column reduces upwelling and mixing of w ­ ater from the deep ocean, but t­ hese conditions are likely to have dif­fer­ent effects on production in dif­fer­ent regions. In the permanently stratified w ­ aters of low-­latitude central ocean gyres, stronger stratification is expected to intensify existing nutrient scarcity, favoring smaller over larger phytoplankton cells. The shift to smaller cells would likely reduce the flow of production to higher trophic levels through the classical food web, and reduce the strength of the biological pump’s sequestration of fixed carbon to the deep ocean. At high latitudes, in contrast, the stronger stratification resulting from warming may enhance primary production by reducing mixing of phytoplankton below the euphotic zone (Marañón 2015). A more indirect physical effect of warmer climate on the open ocean stems from the possibility of increased desertification on land and consequent changes in the delivery of aeolian dust, which alters inputs of iron to the open ocean. Satellite data show that altered phytoplankton biomass is already detectable over large swaths of the open ocean, declining especially in the central ocean gyres (Antoine 2005), with similar trends apparent over the last c­ entury based on ocean transparency mea­sure­ments (Boyce et al. 2010). This lower ocean productivity, together with a trend ­toward smaller phytoplankton cells (Morán et  al. 2010), is expected to flow upward to reduce pelagic fish production. Warming effects are also penetrating below the epipelagic ocean. The most pervasive impact is on subsurface oxygen concentration. The oxygen minimum zone that extends throughout much of the world ocean at upper mesopelagic depths is expanding, especially in the tropics and adjacent to continental shelves, and dissolved oxygen in this layer has declined over wide areas. Th ­ ese changes reduce the volume of the deep ocean where large animals can live (Stramma et al. 2010). Earth system models forced by projected green­ house gas emission scenarios predict the largest changes along continental margins, with oxygen decreasing over large regions of the seafloor, particularly near the poles, highlighting the probability of synergistic ecosystem responses and reor­ga­ni­za­tion of global biodiversity patterns (Mora et al. 2013). Climate change impacts are extending into the deep sea as well. Abyssal ocean temperatures could increase by 1°C by the end of the twenty-­first ­century, accompanied by widespread reduction in dissolved oxygen and declining pH. The flux of particulate organic ­matter to the deep-­sea floor is also expected to decline (Sweetman et al. 2017). More challenging is predicting the indirect effects of warming as changes in abiotic forcing and circulation r­ ipple through the ocean’s food webs to influence community composition, productivity, and biogeochemical pro­cesses, including carbon export to the deep ocean. Th ­ ese biologically mediated effects depend on how functional composition of the biological community responds to climate. Is such prediction pos­si­ble? Surprisingly, ­simple models based on responses of species to temperature— an essentially one-­dimensional thermal niche—­have considerable predictive power, at least for oceanic zooplankton. Empirical data support predictions from theory that co-­occurring species tend to shift their biogeo­graph­i­cal ranges roughly in synchrony ­because they have similar niche requirements and therefore community reconfiguration resulting from climate change is ­limited (Beaugrand and Kirby 2018). Th ­ ese authors suggest that major changes in community structure are expected only where species vary strongly in the degree of eurythermy (temperature tolerance), where species are connected by specialized interactions, and/or where the community depends on dominant foundation species, none of which is common in pelagic communities—­although they are clearly impor­tant in coastal benthic systems, such as coral reefs and seagrass beds (chapters 11, 12). For zooplankton, models of community diversity and geographic range shifts forced solely by observed trends in temperature tolerances of individual species reproduce well the observed shifts in zooplankton diversity in the North Atlantic over the latter twentieth ­century and between the Last Glacial Maximum and the pre­sent (Beaugrand et al. 2015). The predictions for zooplankton are similar to ­those made for

291

292

Ocean Ecology

fishes ( Jones and Cheung 2015), suggesting that diversity w ­ ill decline in warmer w ­ aters while increasing in temperate and polar regions. ­These thermal niche models can be extended based on relatively ­simple assumptions to predict changes in geographic and seasonal distributions with climate change. Assuming a Gaussian thermal niche (i.e., a normal distribution of per­for­mance around some optimum temperature), the predicted distribution and abundance of a species can be mapped as a function of observed sea surface temperature into phenological zones in which phenology and distribution respond differently to warming depending on the match between the organism’s physiological tolerances and the prevailing temperature regime (Beaugrand and Kirby 2018). More specific predictions of global change depend on how it influences the functional composition of pelagic communities. The major traits of interest are phytoplankton cell size and elemental stoichiometry, which often respond predictably to environmental conditions and follow rules linking environment, growth rates, and food web interactions to biogeochemical cycling (Finkel et al. 2010). As w ­ e’ve seen, changes in turbulence and nutrient ratios can influence the relative per­for­mance of primary producer functional groups, shifting dominance among diatoms, dinoflagellates, coccolithophorids, and nitrogen-­fixing cyanobacteria, with large consequences for the magnitude of carbon flux to higher trophic levels and through the biological pump to the deep ocean. Inferences from theory and the paleo-­oceanographic rec­ord suggest that warming w ­ ill tend to disfavor diatoms b­ ecause of stronger stratification (Hood et al. 2006).

Ocean acidification The injection of anthropogenic carbon into the atmosphere also reduces pH as it dissolves into the surface ocean. The first ocean acidification alarms ­were raised by laboratory experiments showing that seawater with reduced pH e­ tched the shells of pteropods, planktonic mollusks impor­tant in pelagic food webs, particularly in the subarctic North Pacific (Orr et al. 2005). The risk of dissolution of calcareous plankton, corals, and shellfish (especially larvae) in a more acidified ocean is now widely appreciated (chapters  4, 12). But in open-­ocean ecosystems acidification may also have impor­tant indirect impacts by affecting iron availability. We have seen that in large areas of the open ocean away from continents—­the high-­nitrogen low-­chlorophyll regions—­primary production is l­ imited by iron availability (chapter 5). Lab experiments show that acidification changes the iron speciation and reduces the ability of diatoms and coccolithophores to use iron, such that ongoing acidification w ­ ill likely increase iron stress of phytoplankton populations in HNLC areas (Shi et al. 2010).

High-­seas fisheries The Anthropocene has witnessed the first significant impacts of open-­ocean fisheries exploitation in ­human history, both in the pelagic and the deep sea. It is within the last half ­century that Homo sapiens has become the top predator throughout the ocean’s global volume. The strongest current impacts of h­ umans in the ocean are a result of deep-­water trawling felt on the upper continental slopes, seamounts, and through reefs of deep-­water coral (Ramirez-­Llodra et al. 2011). Fishing impacts extend to the ocean’s largest animals: since whaling began, the g­ reat ­whales have declined by an estimated 66%–90%, likely with pervasive impacts on the structure and functioning of their ecosystems (Roman et al. 2014) (see box 2.2), potentially including trophic cascades that have led to population collapses of smaller marine mammals (Springer et al. 2003). Fortunately, some w ­ hale populations are already recovering, and this may be pos­si­ble in ­others as well. The ongoing and expected evolution of the Anthropocene ocean is of far more than academic interest. The poorest ­people in the world, numbering 470–870 million, depend intimately on marine

Chapter 10 The Open Ocean

ecosystems for food, livelihoods, and revenues. The countries in which they live are also ­those with the most fragile governance and least capacity to cope with pervasive change in ocean environments and ecosystems (Mora et al. 2013). Recognizing ­these social-­ecological dependencies highlights the importance of identifying the bright spots—­situations in which h­ uman and environmental health are better than expected (Cinner et al. 2016)—­and analyzing them for general insights into more effective management and policy.

­Future Directions The ­future of Earth’s climate and its expected interaction with altered physical forcing in the ocean are becoming ever clearer. A central challenge now is honing models of how t­ hese changes affect the functional composition and dynamics of biological communities, how ­these in turn translate to ocean productivity, vertical flux, trophic transfer, and elemental cycling, and how t­ hese pro­cesses are projected to vary through space and time. A key question is, To what extent are major properties of pelagic (and other) ecosystems predictable based on extrapolating s­imple species niche models ( Jones and Cheung 2015, Beaugrand and Kirby 2018) versus showing emergent properties or chaotic dynamics that are not predictable from the linear interactions of their component species? Answering this question w ­ ill require basic autecological studies of key taxa, realistic experimental manipulations of stressor impacts, detailed field time series, and modeling that integrates all of ­these components. This is a tall order but the stakes are high. Making the connections that explain and help predict marine ecosystem dynamics w ­ ill also require integration with terrestrial ecol­ogy, not only ­because the land influences estuarine and coastal ­waters but to understand how t­ hese linkages influence the dynamics of the earth system as a ­whole. We understand reasonably well how warming strengthens stratification of the w ­ ater column, with concomitant reductions in mixing of nutrients into the surface ocean from below. But warming of terrestrial landscapes also has implications for ocean productivity by influencing the hydrological cycle and thus the degree and distribution of aridity on the landscape. Since wind-­borne iron is a major control of productivity over large expanses of the open ocean, the moisture dynamics of terrestrial landscapes is likely to have strong impacts on marine primary production (Falkowski et al. 1998). Biogeochemical pro­cesses are mediated largely by the diverse and poorly understood microbial biosphere—­the invisible ocean (see box 2.1). A frontier for biological oceanography is building on the major advances in description of ­these diverse microbes to understand what they are ­doing, particularly how ­these organisms and their interactions respond to altered temperature, CO2, dissolved oxygen, and nutrient supply in the changing ocean (Arrigo 2005). Central to this challenge are the organisms of the microzooplankton, which exert strong controls on primary producer biomass ­because of their ability to graze the tiny picoplankton that dominate the oligotrophic ocean as well as their rapid growth and population dynamics. Yet microzooplankton remain a black box in many ways, as we know ­little about basic physiology for more than a few species and about the top-­ down influence of predation on their populations (Caron and Hutchins 2013). This is a specific instance of the larger, frequently noted challenge that oceanographers and climate change researchers have focused largely on primary producers despite clear evidence that grazing is a major control on producer biomass and productivity in many ocean regions. Climbing higher in the food chain, pelagic trophic cascades have been demonstrated in experiments and are consistent with some observational data, but depend on length of the food chain in ways that may obscure their importance in nature (Stibor et al. 2004). What is the role of top-­down forcing in marine regime shifts, and how does it interact with simultaneous changes in physical forcing and other stressors (Baum and Worm 2009, Pershing et al. 2014)? Again, answering this question w ­ ill require integration of natu­ral history information, experiments, time series, and modeling.

293

294

Ocean Ecology

Summary More than in any other habitat, the structure and dynamics of open-­ocean pelagic ecosystems are forced by ocean and atmospheric physics and downstream effects on seawater chemistry and biological activity. The distribution of marine plant biomass, revealed by satellites, includes expansive oligotrophic areas in the central gyres, permanently stratified and chronically depleted of nutrients; seasonally high productivity in polar regions; narrow bands of high productivity along the continental coasts; and productive upwelling zones along the eastern margins of ocean basins and along the equator. Biomass distribution of deep-­sea benthos largely mirrors t­ hese patterns, reflecting bottom-up control by organic ­matter flux from surface w ­ aters. The global distribution of phytoplankton biomass can be explained in large part by the influence of solar heating and atmospheric circulation on the distribution and stability of ­water masses, and their implications for biological depletion of nutrients within them. Phytoplankton biomass distributions are also driven by the ocean’s vertical stratification: the most impor­tant ecological boundary in the ocean is the pycnocline, the sharp density discontinuity between the warm, sunlit epipelagic skin and the cold, dense w ­ ater of the deep sea’s interior that constitutes 98% of the ocean’s volume. The ocean ­water column generally shows roughly coincident gradients in temperature, density, nutrient concentrations, and phytoplankton biomass at the base of the pycnocline. The re­sis­tance of the surface layer to mixing retains algal cells in the epipelagic zone, where phytoplankton growth depletes nutrients and incorporates them into biomass, which then passes to microbial and metazoan food webs, a fraction of which is exported to the deep sea. Thus, the epipelagic zone is the engine of the ocean and the bathypelagic zone below is its storage reservoir. The oceanic plankton are classically divided into groups by log body size, which differ in nutrient kinetics, growth rate, vulnerability to grazing, and sinking rate. The planktonic primary producers of the open ocean are mostly minute, and, although they make up less than 1% of Earth’s plant biomass, marine phytoplankton contribute almost half of its primary production. A large proportion of the ocean’s primary production is attributable to tiny picoplanktonic cyanobacteria (Prochlorococcus and Synechococcus spp.). The major phyloge­ne­tic groups of phytoplankton tend also to differ functionally in terms of light and nutrient acquisition, interactions with enemies, and responses to environmental forcing, notably turbulence. Physiological trade-­offs among t­ hese capacities define contrasting ecological strategies, which often can explain the distributions of major groups across environments. The ­great majority of ocean life is invisible, comprising an extraordinary diversity of Bacteria, Archaea, tiny eukaryotes, and viruses. Much of the ocean’s productivity cycles through this “microbial loop,” grazed by protistan microzooplankton, without ever making its way to metazoan animals. The ecol­ogy of microzooplankton is a frontier of biological oceanography. The familiar animals (macro-­ and megafauna) of the ocean are a minute fraction of its total biomass but of outsize importance in ­human economies and, in some areas, as agents of top-­down control on ecosystem biomass and community composition. In temperate regions, particularly the North Atlantic, physical forcing of ecosystem dynamics is illustrated by the spring bloom, a predictable, often basin-­wide increase in phytoplankton biomass initiated by increasing light dose, more rapid growth of phytoplankton than of grazer populations, and (sometimes) warming-­induced stratification. The concentrated algae fuel zooplankton growth and in turn fish recruitment, and the bloom declines as grazing and nutrient depletion in the stable ­water column emerge during summer. In tropical oceans, where all ­these ­factors are less seasonally variable, plankton biomass and productivity are more uniform through the year. The rich diversity of ocean plankton, contrasting with the seeming monotony of their open-­ water habitat, appears to be maintained by a combination of specialized resource and environmental requirements and interactions, differential responses to disturbance, and complex interactions that can generate strong dynamics in the absence of environmental variation. The deep-­sea benthos is also paradoxically diverse, prob­ably maintained by similar mechanisms and a relatively stable envi-

Chapter 10 The Open Ocean

ronment. But biomass is very sparse in the deep sea ­because ­these communities depend on food from the surface, which declines exponentially with depth. The exceptions are the dense communities of hydrothermal vents and cold seeps, which are nourished by bacterial chemosynthesis, both among free-­living cells and symbionts of the dominant invertebrates. The most far-­reaching aspects of global change in the open ocean are climate warming and fishing. Since the mid-­twentieth c­ entury, technological advances and depletion of coastal and shelf fisheries have led to exploitation of deep-­sea fishes that aggregate around seamounts. The ensuing fisheries typically followed a rapid boom-­bust cycle due to the very slow growth and long generation times of species in ­these cold, food-­poor environments. Deep-­sea fishing is more akin to mining than managing a renewable resource and has been disastrous for deep-­sea ecosystems ­because of the demographic vulnerability of slow-­growing fishes and the destruction by trawling of slow-­growing biogenic habitat, including cold-­water corals on which the fish populations and many other deep-­sea species depend. Robust models predict that the ocean in 2100 w ­ ill be, on average, warmer, more acidified, and less productive than at any time in the last 20 million years and that impacts ­will fall most heavi­ly on poor nations with ­little capacity to address them. Managing the consequences of ­these changes is a key scientific and societal challenge, the thorniest part of which is understanding and predicting how the altered abiotic forcing w ­ ill ­ripple through the complex interactions of the ocean’s food webs to influence biological productivity and biogeochemical pro­cesses.

295

11

Estuaries and Coastal Seas

The Edge of the Sea For most of us, as terrestrial beings, the ocean is inextricably associated with the shore—­the vast expanse of open w ­ ater stretching away from our vantage at its edge. From a global perspective, this coastal zone is a small and quite anomalous part of the ocean. Yet it is of outsized importance to humanity b­ ecause of this proximity. Where the ocean meets the land, their mutual influence on one another creates a unique region, the coast, distinct from both parents. The coastal ocean extends from the waterline to the edge of the continental shelf at a depth of roughly 200 m. This narrow band around the world’s continents and islands is home to a diverse range of ecosystem types, from the turbid organic-­rich sediment bottoms of estuaries to the massive reefs built by corals and their symbiotic algae in the transparent, nutrient-­poor ­waters of the tropics. What sets t­ hese coastal ecosystems apart from the open ocean are the strong links between the seabed, the ­water column, and the adjacent land. The primary producers that dominate h­ ere are large plants rooted to the bottom, including marsh grasses, mangrove trees, submerged seagrasses, seaweeds—­and corals, which are animals but whose symbiotic algae render them net producers and therefore honorary plants. Th ­ ese large plants—­macrophytes—­produce both food and habitat and thus serve as foundations for entire ecosystems. Along the coast, the land delivers sediments, freshwater, nutrients, and other substances to the ocean via rivers, runoff, and even wind, supporting vigorous biological production. As a result, coastal regions are hotspots of both biodiversity and biogeochemistry. Despite making up  90% of the global marine fishery catch, 80% of global carbonate production, 50% of global denitrification (Millennium Ecosystem Assessment 2006a), half the carbon buried in marine sediments (Duarte et al. 2013), and a disproportionate fraction of the ocean’s biodiversity. For ­these reasons, and b­ ecause roughly 40% of the world’s population lives within 100 km of the ocean, coastal ecosystems are central to h­ uman well-­being. For most of humanity, coastal w ­ aters are the ocean. In this chapter we begin with the general features of coastal geomorphology and environment; trace how they influence the characteristic life forms and biogeochemical pro­cesses of major coastal habitats; review how coastal systems are changing in the Anthropocene epoch; and, fi­nally, explore promising directions for ­future research.

Interacting ocean and continents

296

The coast is a boundary. From a global perspective, its most impor­tant feature is that emergent land alters the ocean’s large-­scale circulation. Currents are deflected as they approach a coast, contributing to the geostrophically driven circulation gyres characteristic of the modern ocean. Together with the shape and arrangement of the continents, ­those circulation patterns determine the general biogeography

Chapter 11 Estuaries and Coastal Seas

and character of coastal ecosystems (chapter  3). The earth’s continents are arranged in a general north-­south orientation. As a result, the Atlantic and Pacific Oceans extend nearly from pole to pole and are bounded on the east and west by land. In the Northern Hemi­sphere in both oceans, a classic clockwise gyre circulation prevails, resulting in moderate climate on the eastern sides of the Atlantic and Pacific basins and a more variable climate on the western sides. Th ­ ese circulation features have been relatively stable for about three million years, when the Isthmus of Panama ­rose above w ­ ater and blocked the equatorial current that formerly flowed uninterrupted around the world tropics, and they have strongly affected the evolution and biological composition of marine communities. The earth’s rotation displaces the current gyres t­ oward the west, producing intense boundary currents on the western sides of oceans, such as the Gulf Stream in the Northwest Atlantic and the Kuroshio Current of the Northwest Pacific. Along the eastern sides of ocean basins, currents are weaker and more variable. As a result, wind-­driven currents play a stronger role in the coastal zone on the eastern sides of the basins and generate intermittent upwelling conditions along the west coasts of North Amer­ic­ a, South Amer­i­ca, and southern Africa. On regional scales, the oceanography of the coasts is more complex than that of the open ocean, ­shaped by both the basin-­scale circulation just mentioned and the complex topography of the coastline, which differs substantially between passive and active margins (chapter 3). Active margins, such as ­those around most of the Pacific Ocean, feature narrow, steep shelves cut by submarine canyons. ­Here rocky shores are common and the open ocean’s influence penetrates close to shore. In contrast, passive margins, such as ­those of the Atlantic, generally have wide, shallow shelves buried in thick accumulations of sediment. Estuaries are common and often extensive on the shallow shelves of passive margins, and the coastal ocean is more distinct from the offshore ocean than it is along the narrower active continental margins. In many parts of the world, a shelf-­break front separates open-­ ocean ­water from the richer, more dynamic ­waters overlying the continental shelf. The ecol­ogy of nearshore ­waters is strongly s­ haped by the presence and variability of coastal upwelling and the dynamics of tides and internal waves that can move w ­ ater, with associated nutrients and larvae, on-­or offshore. Prominent headlands and other coastal features interact with coastal currents, creating offshore jets or circulation cells that can isolate stretches of coastline from one another and produce quite dif­fer­ent oceanographic conditions over short distances. The climate of the adjacent continent also shapes the character of its coastal ecosystems. ­Water has a high heat capacity relative to air, meaning that the ocean absorbs heat from the atmosphere in summer and releases it in winter, moderating temperature variation of both ocean and overlying atmosphere. The land has much lower heat capacity, and therefore temperatures generally fluctuate more over continents than over the ocean. Winters are colder and summers hotter in the continental interiors, and daily temperature variation is also greater on land. Shallow coastal w ­ aters feel this influence and are more variable than ­those of the open ocean. Continental topography also interacts with atmospheric circulation to drive the magnitude and seasonal pattern of rainfall. Fi­nally, at the local scale, the characteristic structure of a community on the shore is strongly governed by the rhythm of the tides, the alternating rise and fall of the ocean’s surface caused by the changing gravitational pull of the moon (primarily) and sun as they move relative to one another. The seashore, intermittently submerged and exposed by changing tides, is a unique environment. ­Because it is the most accessible part of the ocean, ­people have foraged in the intertidal zone throughout ­human history. As we have already seen, the intertidal zone has also been a productive laboratory for learning general rules about how communities work. At the interface between land and sea, the intertidal is a zone of gradients in multiple environmental ­factors: ­water depth, wave exposure, air exposure, light, temperature. On g­ ently sloping sediment bottoms, t­ hese gradients occur over long horizontal distances and produce parallel bands of vegetation or habitat between the ocean and upland, whereas on steeper rocky shores they occur over a narrower vertical range. Th ­ ese environmental gradients ultimately drive the characteristic zonal organ­ization of nearshore communities, which

297

298

Ocean Ecology

is modified and shifted in one direction or another by interactions among species. B ­ ecause of the alternating exposure and submersion, the intertidal zone is a challenging environment for organisms of both terrestrial and marine ancestry. But t­ hose organisms that have adapted to it often flourish and grow into dense populations ­because waves and currents bring a rich flux of nutrients for primary production and allochthonous organic ­matter from both terrestrial and oceanic sources.

Estuaries Where rivers draining a continent meet the ocean, a unique type of ecosystem forms, blending characteristics of the ocean and freshwater (figure 11.1). An estuary is a semienclosed coastal body of w ­ ater with a connection to the sea, where seawater is diluted with freshwater. Estuaries may contain any of the habitat types discussed in this chapter and range from modest areas at the mouths of creeks, to large-­scale coastal features, such as the river deltas of the Mississippi, Amazon, and Nile, to enclosed ­water bodies, such as the Baltic Sea and Chesapeake Bay. Estuaries are supplied by the rivers that drain surrounding lands, carry­ing freshwater, sediments, nutrients, and organic ­matter, and thus they tend to be turbid and nutrient-­rich. The turbidity often restricts primary productivity to the top few meters of the ­water column, or even less. ­Because of their shallow depths, unconsolidated sediments, and close (A)

(B) 1 Offshore wind

Land

Upwelling

Warm, mixed layer Cool, nutrient-rich layer Nutrients

2 Entrainment

Freshwater Saline water

Nutrients 3 Low tide

Nutrients High tide

Nutrients

Figure 11.1. ​Estuaries. (A) The Murray River estuary, Queensland, Australia (freeaussiestock​.­com). (B) Three modes of estuarine circulation (see text) (­after Mann 1982).

Chapter 11 Estuaries and Coastal Seas

299

Number of species

connections to the land via watercourses, most estuaries are dynamic environments, with highly variable environmental conditions on all time scales. Diurnal variation in temperature, pH, and dissolved oxygen in temperate estuaries often exceeds annual variation in the open ocean. Within the estuary, the primary driver of biological community structure is salinity, which is identical to that of the ocean at the mouth Freshwater of the estuary, grading to freshwater at its upstream head. The biological animals communities of estuaries accordingly vary, from ­those dominated by marine species at the mouth to freshwater species in the tidal freshwater upper reaches. The low and fluctuating salinity of estuaries is stressful to both marine and freshwater organisms, reducing the numbers of species that survive ­there compared with the surrounding coastal ocean or upMarine animals Brackish land freshwater. This concept is recognized in the classic paradigm in water estuarine ecol­ogy of the diversity minimum in brackish w ­ ater of 5–7 animals PSU, introduced by Adolf Remane (1934) based on distributions of benthic invertebrates in the Baltic Sea (figure 11.2). The diversity mini0 5 10 15 20 25 30 35 mum is thought to reflect a meeting zone between organisms physiologSalinity (PSU) ically adapted to osmoregulation in marine versus freshwaters, a line difficult to cross (Telesh and Khlebovich 2010). Subsequent work has Figure 11.2. ​Schematic trends in species richness and found, however, that estuarine diversity gradients vary widely among composition along the estuarine salinity gradient. The taxa and geographic regions and that in many cases ­there is no recognizestuarine diversity minimum is illustrated as the dip in able diversity minimum in brackish regions (Whitfield et al. 2012). In proportion of species derived from fresh, brackish, any case, for ­those species that can tolerate estuarine conditions, the and marine change along the salinity continuum. Minimal diversity is assumed to occur between 5 and 7 rich supply of nutrients and organic m ­ atter fosters vigorous growth, PSU (­after Whitfield et al. 2012). high productivity, and biomass, and estuarine habitats are therefore generally dominated by extensive near monocultures of marsh grasses, mangroves, seagrasses, and/or oyster beds (figure 11.3). Much of the character of estuaries is defined by their unique circulation patterns and associated nutrient delivery that emerge from the interaction of freshwater and seawater and drive their vigorous productivity. Three general pro­cesses at the interface between land and sea can pump nutrients from deeper sources into well-­lit surface ­waters, increasing primary productivity (see figure 11.1). First, in shallow coastal ­waters, wind-­induced mixing stirs nutrients up from sediments or deeper oceanic ­water to the surface where plants can use them to photosynthesize. A special case of this wind mixing is coastal upwelling, in which per­sis­tent winds blow surface w ­ ater away from shore and deeper nutrient-­rich ­water rises to the surface near the coast to replace it. The second form of nutrient pumping is the characteristic estuarine circulation, driven by river runoff. Freshwater flowing out to sea is less dense than ocean w ­ ater and thus rides over it, resulting in a w ­ ater column stratified by salinity. But the fresh and salty ­waters mix across the pycnocline, and the fresh surface ­water outflow entrains large quantities of under­lying salty w ­ ater. To make up for the w ­ ater pulled up and out by the surface flow, nutrient-­rich seawater flows in along the bottom from offshore. The third and final form of nutrient pumping occurs in areas with a large tidal range, where the daily rise and fall of the tides mixes the ­waters, bringing nutrient-­rich ­water from the sediments up to the surface where it can be used in photosynthesis. Estuaries support high animal (secondary) productivity fueled by both high local production by plants and algae and large inputs of allochthonous organic m ­ atter drained from the watershed. As a result of both the protected physical environment and high fishery productivity, estuaries have attracted dense h­ uman settlement for millennia, and ~70% of the global coastal population lives within 50 km of an estuary (Millennium Ecosystem Assessment 2006a). In summary, geomorphology and climate interact to determine the type of ecosystem that develops along a coast. The key features are the nature of the continental margin, ­whether active or

300

Ocean Ecology

A

C

B

D

Figure 11.3. ​Some types of coastal ecosystems: (A) salt marsh, Georgia, USA; (B) mangrove; (C) seagrass meadow, British Columbia, Canada; (D) oyster reef, Mary­land, USA.

passive; under­lying geology, particularly ­whether the coast is bedrock or accumulated sediment; precipitation regime and supply of freshwater, nutrients, and sediments from the watershed; and temperature regime. Generally speaking, similar abiotic conditions produce similar types of coastal ecosystems in dif­fer­ent biogeographic regions. Examples include forests of large macroalgae, such as kelps, at high latitudes in both Northern and Southern Hemi­spheres and seagrass meadows and mangrove forests in shallow, soft sediment environments throughout the world.

Coastal life and communities Coastal ecosystems differ biologically from t­ hose of the open ocean primarily in the dominance of benthos over pelagic organisms and in the central role of benthic foundation species, primarily macrophytes—­ large, multicellular plants and macroalgae attached to the substratum (see figure 11.3)—in providing the physical habitat. Thus, the dominant primary producers of estuaries are rooted flowering plants along the margins of the shore, including marsh grasses in temperate regions, mangrove trees in the tropics, and seagrasses below the tide line. Primary production in coastal regions is accomplished by both macrophytes and phytoplankton, their relative importance varying with depth, w ­ ater clarity, and other conditions. Generally, macrophytes dominate in intertidal areas and clear shallow w ­ aters—up to perhaps 20 m depth depending on the area—­where sufficient light reaches the bottom for photosynthesis. Phytoplankton dominates in systems where turbidity limits light penetration to the bottom.

Chapter 11 Estuaries and Coastal Seas

In turbid estuaries with abundant sediment input or resuspension, primary production is often ­limited by light availability. On rocky shores, dominant macrophytes include kelps and other macroalgae (see figure 2.3), whereas sediment shores support seagrasses and emergent marsh grasses and mangrove trees that ­ aters root in marine sediments but have emergent canopies in the air (see figure 11.3). In tropical w with ­little nutrient or sediment input, the dominant coastal primary producers are corals with their symbiotic zooxanthellae (chapter 12). Both macrophytes and corals typically achieve ­orders of magnitude higher biomass than the phytoplankton of the open ocean (Duarte and Chiscano 1999) (­table  11.1), contributing to the far higher average productivity of coastal versus oceanic ­waters. Both the size and attached habitat of macrophytes make ­these ecosystems more similar to terrestrial than to pelagic marine systems, being built around the physical structure of sessile foundation species anchored to a two-­dimensional surface and competing for light. Marine macrophytes, as well as larger sessile invertebrates, such as oysters and corals, act as foundation species, providing physical structure and habitat heterogeneity that shelter a wide range of associated organisms. Flowering plants, such as seagrasses, marsh grasses, and mangroves, invest a lot of energy and biomass in nonphotosynthesizing support tissues made of cellulose and lignin, which have l­ittle nitrogen and are physically tough, thus making them difficult for herbivores to digest. As a result, coastal flowering

­TABLE 11.1 ​Average biomass and net primary production of dif­fer­ent plant communities Community

Biomass

Production

Forests Tropical

45,000

5.2

Temperate

35,000

3.4

Boreal

20,000

2.2

Savanna

4,000

2.4

Temperate

1,600

1.6

Tundra and alpine

600

0.4

Swamps and marshes

15,000

5.5

1,000

1.8

Grasslands

Cultivated land Phytoplankton

9.2

Microphytobenthos Coral reefs Macroalgae Marsh plants

0.13 2,000 40.7 767

Mangroves Seagrasses

0.35

0.8 1.0 3.0 2.7

461

2.7

Source: Duarte and Chiscano (1999). Note: Biomass mea­sure = g dry mass m−2; production mea­sure = g dry mass m−2 day−1.

301

302

Ocean Ecology

plants are much less grazed than phytoplankton, and much of their production enters the food web as detritus or is buried (Cebrián et al. 1998) (figures 5.3, 5.4). Thus, key traits of coastal macrophytes affecting their ecological roles are attachment to the substratum, large size, structural complexity, and low nutritional quality. Whereas the herbivores of the open ocean are primarily microbes and tiny metazoans such as copepods, herbivores of coastal benthic systems span a wide range of sizes and animal taxa. Most are benthic, associating with the attached primary producers on which they feed. Major coastal herbivores include sea urchins, gastropod mollusks, vari­ous crustacean groups, and several families of fishes in tropical and warm-­temperate w ­ aters (see figure 9.5). Many coastal regions, particularly in the tropics, also support populations of large herbivorous vertebrates, including sea turtles and sirenians—­manatees, dugongs, and the now extinct Steller’s sea cow. Most of t­ hese megaherbivores have been greatly reduced from their primeval abundances by hunting and habitat change. In coastal communities dominated by macrophytes, positive interactions in the form of facilitation and ecological engineering play impor­tant roles in community organ­ization and maintenance of diversity. In the intertidal zone, for example, the canopy of plants and macroalgae reduces heat and desiccation stress, allowing species that shelter u­ nder them to persist and thrive. Such communities support the hypothesis that positive interactions often contribute to community diversity and ecosystem structure in stressful environments (Bertness and Callaway 1994). The dominance of perennial foundation species is often in turn facilitated by selective feeding of herbivores on faster-­growing and more palatable “ephemeral” algae. ­These algae often overgrow the macrophytes when unchecked by grazing, but they provide poor habitat for other organisms. Dominance of macrophytes, including corals, often depends on a delicate balance of top-­down and bottom-up control. Herbivores may become enemies, especially when foundation plants are stressed. In salt marshes, for example, drought stress can trigger outbreaks of snail grazing that damage the plants and kill them (Silliman 2005). In California, kelps stressed by warm temperatures have been overgrazed by amphipod crustaceans (Tegner and Dayton 1987). In both of ­these cases, ­there are suggestions that the herbivore outbreaks result from relaxed predation pressure that would normally control them. That is, the dominance of foundation species may depend on trophic cascades—­top-­down control emanating from predators. In such cases, overfishing or other perturbations to the top of the food web can ramify through the food chain to fundamentally change the system. Syntheses of data from vegetated coastal ecosystems show that grazers are capable of reducing plant biomass, but that cascading effects of predators can keep grazers in check and thereby maintain high plant growth and biomass. Grazing impacts are often amplified by nutrient inputs and disturbance, and grazing impacts of ectothermic herbivores are greater u­ nder the warmer temperatures at low latitudes (He and Silliman 2016). On North Atlantic shores dominated by eelgrass (Zostera marina) and rockweed (Fucus spp.), trophic cascades are similarly impor­tant: on average, small fishes that feed on invertebrates (i.e., mesopredators) double the biomass of ephemeral algae by suppressing feeding by herbivorous crustaceans. Moreover, ­these top-­down effects are generally of similar magnitude to the effects of nutrient loading and are stronger ­under eutrophication (Östman et al. 2016). In some cases, predator effects cascade through four levels, from piscivorous fishes through mesopredators, to herbivorous invertebrates, to algae, although few such studies have been conducted. In summary, cascading trophic control is common in coastal systems and suggests that managing fish stocks can be impor­tant to conserving coastal vegetation.

Coastal ecosystem pro­cesses Coastal wetlands, and estuarine ecosystems generally, have long been considered classic examples of bottom-up control. At large scales, this is supported by the strong positive relationship between nitrogen input, or primary production, and fish production in estuaries and coastal ­waters (Nixon 1982, Herman et al. 1999) (figure 11.4). Dense benthic communities in coastal w ­ aters are favored by

Chapter 11 Estuaries and Coastal Seas

(B) 70

160

60

140 Fish landings (kg ha-1 year –1)

Macrobenthic biomass (g AFDM m–2)

(A)

50 40 30 20 10 0

0

100 200 300 400 500 600 700 Primary production (g C m–2 year –1)

120 100 80 60 40 20 0

0

100 200 300 400 500 Primary production (g C m –2 year –1)

Figure 11.4. ​Animal (secondary) production increases with plant (primary) production in estuarine and coastal ecosystems. Primary productivity versus (A) system-­averaged macrobenthic biomass (Herman et al. 1999) and (B) fishery production (Nixon and Buckley 2002), across shallow well-­mixed estuarine systems.

the high inputs of nutrients and organic ­matter from rivers and by the shallow ­water column. Organic ­matter delivery to the seabed drops off exponentially with ­water depth and, since continental shelf ­waters are one to two o­ rders of magnitude shallower than the deep-­sea floor, shelf sediments receive comparably greater organic ­matter from surface ­waters (chapter 9). Accordingly, the shallow shelf supports much higher biomass and activity of benthic organisms nourished by that food supply than does the floor of the open ocean. The pelagic and benthic communities of coastal systems are also strongly coupled, both by energy and materials flows and by linked life history stages of their organisms. Animals generally are far more abundant and achieve greater biomass in coastal regions than in the open ocean due to both higher productivity and the physical habitat structure provided by foundation species, accounting for the much higher productivity of coastal than oceanic fisheries (Pauly and Christensen 1995). As a result, coastal ecosystems are much more impor­tant on a global scale than would be expected from the small area they occupy along the fringes of continents. Coastal zones are also impor­tant in global carbon and elemental bud­gets as conduits and modifiers of land-­ derived production and biogeochemical fluxes from the land to the ocean. The dominant primary producers of pelagic and coastal benthic systems are phytoplankton and benthic macrophytes, respectively. Many of the differences between ­these systems are explained by differences in edibility of t­hose dominant autotrophs, which influence the rates and fates of plant production, including losses to grazing, flux through the detritus pathway, rates of decomposition, and accumulation in sediments. The flowering plants that dominate coastal ecosystems have tougher tissues and lower nitrogen content than algae, and are therefore less grazed (Cebrián 1999), so that much of the production of seagrasses, marshes, and mangroves flows through the detritus pathway, and a substantial fraction of the ungrazed carbon is stored in sediments and plant biomass (see figure 5.4). Coastal vegetation provides several valuable ecosystem ser­v ices, including fishery production, shoreline protection from storms, and carbon sequestration in the organic-­rich sediments. Mangroves and salt marshes form a protective barrier that shelters coastal land from erosion and storm surges by attenuating waves and reducing property damage and ­human deaths. Not only the living plants but also the accretion of their dead biomass as peat strengthen shorelines and provide a robust barrier that protects coastal land against sea level rise (Gedan et  al. 2011). And

303

304

Ocean Ecology

coastal wetlands are major seafood factories b­ ecause of both their high productivity and the habitat complexity that creates substratum and predation refuge for fishes and the organisms on which they forage.

Rocky Shores The ecol­ogy of rocky intertidal shores is arguably better understood than that of any other marine habitat as a result of their accessibility, sharp gradients in environmental conditions, and the relative ease of experimentally manipulating the sessile and slow-­moving organisms that live ­there. We focus ­here on intertidal habitats b­ ecause they are best studied, but similar interactions and outcomes are common on rocky subtidal reefs (Witman and Dayton 2001). The most striking feature of rocky shores throughout the world is zonation, the sequence of more or less clearly defined bands dominated by dif­fer­ent organisms that replace one another from high to low on the shore (see figure 8.1). In temperate regions, t­ hese zones are generally predictable, with a dark band of lichen and cyanobacteria at the highest limit of wave splash, a band of grazing littorine snails below, grading into a white band of barnacles in the upper intertidal zone. Below this, in the mid-­intertidal, is a mixed zone of sessile invertebrates and seaweeds, dominated by mussels in many areas. At the lowest tide line and in the oscillating ­water beyond can be glimpsed kelps and other large brown seaweeds that flourish on the subtidal rocks. On hot tropical shores, intertidal life is generally much sparser. On the Pacific coast of Panama, for example, crustose coralline algae dominate intertidal rocks, apparently ­because desiccation, heat stress, and intense grazing and predation exclude most other algae and sessile invertebrates (Lubchenco et al. 1984). The sharp vertical gradient in marine influence across the shore suggests an explanation for the banded zones of intertidal life. By the mid-­twentieth ­century several syntheses had cata­loged the characteristic distributions of seaweeds and invertebrates on rocky shores and established a conceptual model explaining their distribution according to their physiological tolerances for the increasing heat and desiccation stress from low to high intertidal (Lewis 1964). Rocky intertidal communities seem the archetype of a landscape structured by the abiotic environment, with the highest levels ­limited by exposure to harsh terrestrial conditions and successively lower bands dominated by physiologically less hardy organisms. In the early 1960s a series of experiments began to change this view of s­ imple environmental control. The first was Joseph Connell’s (1961) demonstration that the band of Chthamalus stellatus barnacles in Scotland was restricted to the highest intertidal not by the barnacle’s preference for physical conditions but ­because another barnacle, Balanus balanoides, competitively excluded it from the more favorable conditions lower down (see box 7.1). Shortly thereafter, Robert Paine became curious about what the big predatory sea stars (Pisaster ochraceous) ­were ­doing on the wave-­beaten rocks of Washington State, USA, so he removed them from a swath of shore. He then watched the shore transform over a few years from a diverse community of seaweeds, barnacles, and limpets to a near monoculture of mussels. This experiment convinced him that the predatory sea star’s preference for the competitively dominant mussel prevented it from displacing other species and thus that the sea star played a keystone role in maintaining diversity of the intertidal community (Paine 1966, 1974). A wave of such experimental studies followed, demonstrating that the zonation of rocky shores arose from a complex web of interactions among competitors, herbivores, and predators—as well as their dif­fer­ent physiological tolerances (Dayton 1971, Menge 1976, Lubchenco 1978, Lubchenco and Menge 1978). ­Later studies documented the equally critical importance of positive interactions between species, especially the role of foundation species in facilitating associated diversity (Suchanek 1979, Bertness et al. 1999, Lafferty and Suchanek 2016) (chapters 7, 8).

Chapter 11 Estuaries and Coastal Seas

Geomorphology and environment A key feature of rocky shores is the two-­dimensional rock surface to which seaweeds and sessile invertebrates must attach to survive. Competition for this living space is a central pro­cess in organ­ ization of ­these communities, especially where larval supply is strong and promotes high population densities. On rocky shores, the primary environmental ­drivers—in addition to temperature—­are light, desiccation, and wave stress, all of which take low values deep in the subtidal and increase up through the intertidal zone. Rocky shore communities also are strongly s­ haped by the productivity of the adjacent coastal ocean in two major ways (Connolly and Roughgarden 1999, Menge et al. 2003). First, the strength of community interactions depends on the supply of larvae, which sets recruitment intensity and thus density of competing individuals. Second, productivity of the adjacent ocean sets limits on the growth and demographic rates of intertidal suspension-­feeding invertebrates, whch feed primarily on phytoplankton. Much of the experimental research that established our understanding of community interactions on rocky shores (chapters 7, 8) was conducted in productive ­waters, notably the cool-­temperate Pacific and Atlantic coasts of North Amer­i­ca, where larval supply is abundant and sets the stage for intense competition among recruits. In areas where larval supply is lower, competition is accordingly weaker (Underwood and Fairweather 1989). Thus, coastal oceanography and larval supply modulate the strength of biological interactions among species. The transport and retention of pelagic larvae in benthic habitats is also influenced by coastal currents and upwelling patterns (Connolly and Roughgarden 1999). Many coastal species have evolved life histories in which larvae exploit predictable features of flow to deliver them to appropriate habitats. Many intertidal crabs, for example, release larvae at night, during tides of maximal amplitude, which moves them quickly through the gauntlet of nearshore predators and out to sea (Morgan and Christy 1995). Delivery of pelagic larvae back to their adult habitats on the shore is enhanced in some regions during periods of relaxed coastal upwelling (Wing et al. 1995) or onshore transport by internal waves, which buoyant larvae exploit by rising to the surface and traveling shoreward in the associated surface slicks that propagate ­toward shore (Shanks 1983, Pineda 1991). Wave energy, dictated in part by coastal orientation and fetch, strongly affects rocky shore community and ecosystem structure, both by moving ­water and delivering nutrients and food to sessile organisms, and by mediating accessibility to predators. Wave energy increases productivity of rocky shore seaweeds, presumably by breaking down diffusive bound­aries and increasing the efficiency of nutrient and light capture (Leigh et al. 1987), and delivers food to suspension-­feeders. Mobile predators, such as crabs, are often more active and effective in quiet ­waters, whereas waves suppress their activity (Peterson 1979). Among rocky shores around the world, total invertebrate biomass increases strongly with wave exposure, prob­ably reflecting both greater production and lower predation in wave-­swept areas (Ricciardi and Bourget 1999) (figure 11.5).

Organisms and traits Rocky intertidal communities are dominated by seaweeds (macroalgae), sessile suspension-­feeding invertebrates nourished by plankton, and sluggish consumers. Since t­hese communities have been studied mostly when the tide is out and the rocks are exposed, ­there has been ­little attention to the role in community organ­ization of fishes and other mobile animals that invade at high tide. But the few studies that have focused on t­ hese consumers show they are impor­tant, especially in warmer regions where herbivorous and predatory fishes are abundant and active (Bertness et al. 1981, Menge and Lubchenco 1981, Menge et  al. 1986). A key biological trait molding rocky shore community structure and function is the two-­phase life history of most of its inhabitants, consisting of a pelagic larval phase and a benthic adult (see figure 6.1A). The pelagic larval phase has two related conse-

305

Ocean Ecology

(A) Rocky shores

(B) Sediment shores 3

1000

Biomass (g m–2)

2 Biomass (g m–2)

306

100

1 0 –1 –2

0

Sheltered

Semi-exposed Exposed Exposure category

–3

Sheltered

Semi-exposed Exposed Exposure category

Figure 11.5. ​Physical energy regime influences macroinvertebrate biomass differently on (A) rocky versus (B) sedimentary shores. Exposure categories increase in wave energy from left to right (­after Ricciardi and Bourget 1999).

quences for community organ­ization. First, as noted above, larvae depend on food and transport in the ­water column, meaning that coastal oceanography strongly influences the recruitment and density of organisms on the shore (Roughgarden et al. 1988, Menge et al. 2003). Second, local population dynamics are often decoupled from local conditions ­because larval supply tends to come from the broader region’s contribution to the pelagic larval pool rather than from adults on that stretch of shore (Cowen and Sponaugle 2009). Thus, population dynamics are strongly influenced by conditions in the coastal ocean where their young stages develop.

Community organ­ization and key interactions As in all habitats, the organ­ization of rocky intertidal communities is set by several interacting pro­ cesses. First, the species pre­sent are drawn from a regional biota, or species pool, s­ haped by historical pro­cesses over evolutionary time (chapter 3), which sets upper limits on the number and functional types of species that interact in the community (chapters 7, 8). Second, the intensity of larval recruitment, forced by productivity and physics of the adjacent coastal ocean, sets the density of interacting organisms and the pace and intensity of their interactions. Third, the physiological tolerances of ­those species and biological interactions often change in strength and sign along the predictable environmental gradients from low to high intertidal. The main interactions among organisms ­after they ­settle on the shore include competition for two-­dimensional living space on the rock surface, herbivory and predation that can e­ ither relax or intensify competition by removing individuals, and facilitation by large seaweeds and other foundation species that create favorable conditions for ­others by reducing abiotic stress or hiding prey from their predators. Competitive interactions tend to reduce diversity by favoring one species over another, as shown by Connell’s experiments with barnacles mentioned above. But competitive exclusion is slowed by the regular influx of larvae from the surrounding metacommunity and can be delayed or forestalled by events that kill dominant competitors, ­whether nonselective disturbance or selective predation. Keystone predation is a special case of the pro­cesses that interrupt competition: predators are called keystones when they target competitive dominants and thereby increase prey diversity (see box 7.1), as illustrated by the original keystone predator, the sea star Pisaster ochraceous, which feeds preferentially on the competitively

Chapter 11 Estuaries and Coastal Seas

dominant mussel on rocky shores of the Northeast Pacific (Paine 1966, 1974). In contrast, where consumers feed preferentially on competitively inferior prey, they have the opposite effect, accelerating competitive exclusion. On rocky shores of New ­England, USA, for example, littorine snails graze ephemeral algae, which are competitively inferior, and thereby accelerate the rise of perennial red and brown macroalgae to dominance (Lubchenco 1978). Pathogens may be the ultimate keystones in that they often strongly reduce par­tic­ul­ar host species, with far-­reaching impacts on communities and ecosystems, despite having negligible biomass themselves. A striking example involves sea star wasting syndrome, first reported in the 1970s and now known to be caused by a densovirus (Hewson et al. 2014). The disease advances rapidly and has infected at least 20 species of sea stars and several other echinoderms in the Northeast Pacific, including the iconic keystone predator Pisaster ochraceous, causing mass mortality. In 1978 wasting syndrome appeared in the sun star Heliaster kubiniji in the Gulf of California, then the commonest intertidal sea star ­there, and within two weeks nearly all ­were dead throughout the region; at some sites, its preferred barnacle prey subsequently increased and diversity of other sessile species declined (Dungan et al. 1982). Unusually high ­water temperatures ­were implicated in the sun star epidemic, and evidence suggests that many other disease outbreaks are also exacerbated by climate change (Harvell et al. 2002, Altizer et al. 2013). Time w ­ ill tell how such extinctions r­ ipple through their respective communities. Appreciation for positive interactions among species historically lagged well b­ ehind that of the antagonistic interactions of competition and predation (Bruno et al. 2003). Yet it is clear from natu­ ral history that many species depend on the shelter and protection of ­others to thrive. The band of blue mussels on mid-­intertidal shores of the Northeast Pacific was seen in early studies of community organ­ization as a monoculture (Paine 1966), and this is approximately true if one considers only sessile macro-­organisms occupying the rocks. But the mussels also harbor a rich variety of small invertebrates in the spaces between them and the mat of byssal threads below, such that total diversity in the intertidal is substantially higher where blue mussels dominate (Suchanek 1979). Similarly, seaweeds often facilitate other algae and invertebrates beneath their canopy by protecting them from desiccation, overheating, wave stress, and consumption (Hay 1986, Bertness et al. 1999). Experiments in rocky intertidal communities launched a revolution of sorts in ecol­ogy, revealing that the distributions and abundances of organisms are not a ­simple consequence of environmental control but are also strongly influenced by interactions among the species. But it is worth emphasizing that the vertical zonation of rocky shore communities—as well as their variation across latitude—is still ultimately forced by gradients in abiotic conditions. What the experiments changed is the recognition that ­those distributions are mediated not only by physiological responses of individual organisms to the environment but also by interactions among them. Integration of environmental and biological interactions into a theory of community organ­ization began with the models of Bruce Menge and John Sutherland (1976, 1987). They argued that physiological stress and larval supply set the under­lying template for (post-­recruitment) interactions among organisms on the shore. Competition and predation ­were hypothesized to more strongly shape community structure where high larval supply increases density of recruits and favorable conditions foster survival and rapid growth. Benign conditions also f­ avor intense predation. In contrast, as stress increases, mobile consumers are expected to decline first as a function of their (inferred) higher sensitivity to stress, which releases sessile species to compete more strongly. In the most stressful environments, diversity is expected to be low due to direct effects of stress. What t­ hese models w ­ ere largely missing was positive interactions, which can change model predictions both quantitatively and qualitatively, depending on the strength of facilitation and which species are facilitated (Bertness and Callaway 1994, Bruno et al. 2003) (figure 11.6). Throughout most of the long and influential history of rocky shore ecol­ogy, studies of community organ­ization have generally taken the pool of local species as given. As ­we’ve seen, that pool is

307

Ocean Ecology

(B)

Predation Competition Abiotic stress

Relative importance

(A)

Relative importance

308

Stress Competition amelioration Predation

Abiotic stress

Associational defenses Low

Medium Environmental stress

High

Low

Medium

High

Environmental stress

Figure 11.6. ​General models of community organ­ization as a function of environmental stress, without and with positive interactions. Models predict the relative importance of negative interactions (predation, competition, abiotic stress) and positive interactions (amelioration of abiotic stress and associational defenses) as a function of environmental stress. Models assume high recruitment of the basal (prey) taxa. (A) The original Menge-­Sutherland model (1987); (B) inclusion of interspecific facilitation (­after Bruno et al. 2003).

strongly ­shaped by historical biogeography (chapters 3, 8). Both composition and diversity of species within the community can have impor­tant consequences for interactions, diversity, and ecosystem pro­cesses. For example, on exposed rocky shores of the Northeast Atlantic, large patellid limpets are influential herbivores, strongly reducing biomass of intertidal macroalgae, particularly the dominant perennial fucoids (Coleman et al. 2006). Th ­ ese limpets are absent from the Northwest Atlantic ( Jenkins et al. 2008), where periwinkles (Littorina) are the major grazers. But the smaller periwinkles are not abundant on exposed shores, and they appear unable to control larger perennial algae (Lubchenco and Menge 1978). Thus, the diversity of t­ hese local communities is ­shaped significantly by the apparent historical absence of large limpets. Community diversity in turn predictably shapes the stability and strength of trophic interactions: among 57 experiments that removed consumers from marine hard substrata, the strongest predictor of the intensity of consumer effects was species richness of the prey community ­after controlling for habitat type and taxonomic composition. Exclusion of consumers increased prey abundance on average by 1200% in communities with only 2 species but by only 200% in communities with 37 species, even though predation intensity appeared strong across this diversity gradient (Edwards et al. 2010) (see figure 9.19B). More diverse prey communities appear to resist predation ­because they are more likely to contain resistant prey species. Outcomes of community interactions are often mediated by the physical environment. Rocky intertidal communities have produced some of the best examples of how environmental forcing influences the strength and outcomes of biological interactions, including recruitment, competition, facilitation, and predation. Community interactions commonly change across gradients in physical energy or ­water flow. In Maine, USA, pelagic larvae recruited strongly to rocky shore communities at high flow sites, developing dense barnacle and mussel cover and more abundant grazers and predators. Sites with low flow, in contrast, had much more bare space. Higher flow increased growth of some species and reduced predation intensity, modulating the strength of both top-­down and bottom-up forces. In sum, environmental stress models w ­ ere most successful in explaining patterns at low flow sites. Thus, the top-­down effects of consumers ­were the dominant force structuring communities in low flow, whereas bottom-up controls of recruitment and growth dominated ­under high flow (Leonard et al. 1998) (figure 11.7). Fi­nally, the environment changes interactions that influence community organ­ization on a geographic scale as well as along gradients across the intertidal zone (Sanford 2014).

Chapter 11 Estuaries and Coastal Seas

Low flow sites

High flow sites

Crabs

?

?

Crabs ?

Grazers

Diatoms

Nutrients

Grazers

?

Mussels

Larvae

Barnacles

Plankton

Diatoms

Nutrients

Whelks

?

Mussels

Larvae

Barnacles

Plankton

Figure 11.7. ​­Water flow changes community interactions and structure on a rocky shore. Upward arrows indicate energy transfer; downward arrows indicate interaction effects. Arrow width is proportional to strength of the effect; font size is proportional to abundance; dashed lines represent larval input. Flux of nutrients, larvae, and plankton increases in regions of higher flow; whelks are absent from low flow sites. In general, top-­down forces dominated low flow sites, while bottom-up forces dominated high flow sites (­after Leonard et al. 1998).

Ecosystem pro­cesses and ser­vices Rocky shores are the dominant habitat of the coastline in many parts of the world, particularly along tectonically active or recently glaciated continental margins. The high productivity of seaweeds and sessile invertebrates fueled by the confluence of terrestrial and oceanic nutrients and wave energy makes them hotspots for feeding and spawning of many marine animals, and resting sites for seals and other pinnipeds (Thompson and Crowe 2002). B ­ ecause of their exposure to the coastal ocean and the pelagic dispersal of most species living ­there, rocky shore communities are linked to pelagic and other coastal ecosystems demographically and energetically (Connolly and Roughgarden 1999, Menge et al. 2003). The forests of kelps and other macroalgae that dominate many subtidal rocky habitats are highly productive, supporting abundant populations of both herbivores and detritivores as well as impor­tant fisheries (Tegner and Dayton 2000). Indeed, kelp forests of the northeastern Pacific formerly supported the largest-­known herbivorous marine mammal, Steller’s sea cow (Hydrodamalis gigas). This ­giant 8–9 m relative of the dugong was hunted to extinction within a few de­cades of its first encounter with Eu­ro­pe­ans in 1741, although this was likely only the final death blow in a decline initiated by aboriginal American hunters who had already extirpated it from all but a few remote islands in the North Pacific (Estes et al. 1989).

Rocky shores in the Anthropocene ­ ecause of their accessibility, rocky intertidal shores have provided food for gleaning h­ umans since B prehistoric times (Erlandson and Rick 2010). Mussels, oysters, abalones, snails, seaweeds, and even barnacles and tunicates are eaten by ­people in vari­ous regions. Such sites are often strongly depleted of harvestable species compared with less accessible shores, and rocky shores are subject to many of the same stressors impinging on other coastal ecosystems, including pollution, w ­ ater quality deterioration, and habitat modification (Thompson and Crowe 2002). Productive coastal w ­ aters supply larvae and food to rocky shores and thus tend to boost density, growth rate, and interaction intensity in shore communities. But eutrophication—­unusually high input of nutrients from ­human sources—­ not only results in greater plant biomass but often shifts in community composition. Specifically, the growing flux of nitrogen and phosphorus into coastal ­waters over the last half ­century, along with other stressors, has increasingly shifted rocky shore communities from the historical dominance by

309

310

Ocean Ecology

perennial, canopy-­forming seaweeds to ephemeral algae (Moy and Christie 2012, Strain et al. 2014). Grazing can counteract effects of eutrophication to some degree b­ ecause grazers generally feed preferentially on ephemeral algae. But the capacity of grazers to control ­these blooms is l­imited. Field experiments on rocky shores of the Northwest Atlantic and Baltic Sea showed that the ephemeral green alga Enteromorpha was ten times more productive at eutrophied sites. Grazing reduced Enteromorpha by 80%, on average, but could not keep pace with the higher algal productivity u­ nder eutrophic conditions. As a result, perennial algae declined by 30%–60% across a gradient of eutrophication, giving way to dominance by ephemeral algae and suspension-­feeding invertebrates. Eutrophic sites also w ­ ere less diverse, often dominated by single species of algae or suspension-­feeding invertebrates (Worm and Lotze 2006). Th ­ ese experiments show that high nitrogen inputs from h­ uman activity overcome the capacity of grazers to buffer marine vegetation from natu­ral variation in nutrient supply. Invasions by nonindigenous species are also widespread on rocky shores, particularly around harbors and urbanized estuaries, and have transformed rocky shore communities in many areas (Grosholz 2002, Thompson and Crowe 2002). Littorina littorea is the most abundant intertidal snail in northeastern North Amer­i­ca, and its grazing on algae strongly affects community organ­ization (Lubchenco 1978, Bertness 1984). Rockweed (Fucus serratus) is a dominant seaweed on the same shores. Both species are iconic members of the rocky intertidal throughout the region, but molecular data and shipping rec­ords indicate that both species ­were introduced to the region within the last few centuries, apparently from Britain and Ireland (Brawley et al. 2009). The Mediterranean mussel Mytilus galloprovincialis was introduced accidentally to South Africa in the 1970s and has come to dominate rocky shores along much of the southwestern part of the African continent, evidently as a result of both vigorous growth and its ability to escape from native predators and parasites (Griffiths et al. 1992, Branch and Steffani 2004). Among the strongest interactors globally is the Eu­ro­pean green crab, Carcinus maenas, which has now established populations in Australia, South Africa, Japan, and both coasts of North Amer­ic­ a (Carlton and Cohen 2003) and often has strong impacts on local prey populations, albeit mostly in soft sediments (Grosholz and Ruiz 1996). As a result of both invasions and ­human harvesting and disturbance, rocky shore communities close to ­human populations tend to have very dif­fer­ent structure than more isolated sites, often with lower biomass, fewer large individuals, and communities shifted ­toward fewer slow-­growing species (Keough et al. 1993).

Sediment Bottoms Soft sediment bottoms are the most widespread marine habitat, not only in coastal w ­ aters but covering most of the deep-­sea floor. They range from high-­energy ocean beaches of well-­sorted clean sand to quiet intertidal mudflats and organic-­rich anoxic depositional basins. Passive continental margins, such as t­ hose of midlatitudes on both sides of the Atlantic Ocean, have accumulated extensive sediment plains over millions of years of continental erosion that deposited the eroded sediments along the coastal plains and continental shelves. Sediment bottoms are fundamentally dif­fer­ent from rocky shores in being fluid, shifting substrata that provide a three-­dimensional habitat, allowing organisms to live both in and on the substratum and to conduct diverse biogeochemical pro­cesses in the complex chemical environment of the pore ­waters between sediment grains. Marine sediments are complex mixtures of dif­fer­ent types and sizes of mineral grains, with vari­ous quantities of organic ­matter from a range of sources, often bound by microbial mats on the surface.

Geomorphology and environment The primary environmental ­drivers shaping marine sediment habitats are physical energy regime and organic ­matter input. Th ­ ese are partly correlated. Sediments are most commonly classified by grain size, ranging from coarse gravel to increasingly finer sand, silt, mud, and clays. Both grain size

Chapter 11 Estuaries and Coastal Seas

Sediment surface Eh

pH

O2 5

10

CO2

Fe*** 15

NO3– Yellow layer NO

– 2

Gray layer

Fe** –200

0

200 mV

400

H2S 6

7 mV

0

8 0

100

200 mg

300

400

2 %

4

NH3 CH4

Figure 11.8. ​Vertical geochemical zonation within marine sediments. ­These steep gradients may span only a centimeter or so in organic-­rich fine sediments. Eh is redox potential, the system’s relative capacity for oxidation (positive values) versus reduction reactions (negative values) (­after Lenihan and Micheli 2001).

distribution and organic content are strongly influenced by physical energy regime. The turbulence of wave-­swept coasts allows only larger, heavier mineral grains to s­ ettle out, forming clean, coarse sandy beaches with l­ ittle organic m ­ atter. Sheltered shores, in contrast, accumulate the fine silts, clays, and low-­density organic particles that s­ ettle out of suspension only in quiet w ­ ater. The sediment column is characterized by sharp vertical gradients in geochemistry (figure 11.8; see also figure 2.9), often recognizable as layers of dif­fer­ent colors, as a result of l­imited diffusion of oxygen from the overlying ­water column, interacting with gradients in chemical and microbial composition through the upper few centimeters of the sediment column. W ­ ater depth influences both physical energy regime and organic fluxes: compared with deep sites, shallow bottoms are physically more dynamic, on average, and receive a higher percentage of the overlying water-­column productivity and advection of organic ­matter from land. Shallow sediment bottoms also can support benthic microalgae, seagrasses, and mangroves, discussed below.

Organisms and traits As on rocky shores, soft sediment habitats support entire food webs of algae (and other plants), grazers, detritivores, and predators. Unvegetated sediment bottoms generally are net heterotrophic systems. It is difficult for algae to establish on mobile sediments and therefore, with the exception of rooted macrophytes, the food webs of sediment habitats are supported mainly by phytoplankton production in the w ­ ater column and organic ­matter advected from the nearby land. In quiet ­water, however, shallow sediments often develop substantial mats of benthic diatoms, cyanobacteria, and other photosynthetic microbes whose production can exceed that of phytoplankton in the overlying ­water column (MacIntyre et al. 1996). Benthic microalgal production in turn can support abundant grazing and deposit-­feeding invertebrates. Mat-­forming microbes on marine sediments created some of the first recognizable ecosystems on the early earth and left fossils remarkably similar to modern cyanobacteria, as well as stromatolites—­layered accretions of mineral and microbial material—in sedimentary rocks at least 3.4 billion years old (Knoll 2003). Microbial mats are formed by cyanobacteria and some eukaryotic algae as they grow and proliferate, secreting extracellular polysaccharides (Decho 1990) that bind and stabilize the sediment surface, presumably allowing the cells to maintain a position at the surface and reduce turbidity that would interfere with photosynthesis.

Black layer

311

312

Ocean Ecology

Figure 11.9. ​Fauna of an estuarine sediment bottom in the Chesapeake Bay, USA. Suspension-­feeders include several clams and the parchment worm. Deposit-­feeders include vari­ous polychaete worms. Fishes and wading birds feed on the fauna.

The invertebrates that inhabit sediments tend to sort into discrete size categories, with biomass peaks among meiofauna (body mass of 0.05—0.5  μg organic carbon) and macrofauna defining a characteristic size spectrum on a log scale (Schwinghamer 1981) (see figures 2.10, 2.13). The macroscopic infauna of sediments comprise mostly deposit-­feeding and suspension-­feeding (i.e., filter-­ feeding) invertebrates. Th ­ ese are dominated by bivalve mollusks and a diverse array of worms, mostly polychaetes, although in some areas crustaceans, sea cucumbers, sand dollars, phoronids, or o­ thers can be impor­tant. Suspension-­feeders consume phytoplankton and other organic material in the overlying w ­ ater column, which they obtain e­ ither by extending feeding structures up into the w ­ ater or by drawing a current of ­water through their buried tubes (figure 11.9). In both cases, the animals capture suspended particulate material on filter-­like appendages or mucus traps, which they then transfer to their mouths. Deposit-­feeders, in contrast, consume organic m ­ atter that has already fallen to the sediment surface. They feed ­either by sorting organic-­rich particles from the sediment surface before ingesting them or by consuming the sediment ­whole, digesting the organic ­matter, and voiding the remaining indigestible mineral material. Some animals, such as certain spionid polychaetes, switch between suspension-­and deposit-­feeding depending on conditions. In addition to the vari­ous modes of suspension- and deposit-­feeding, the polychaete worms that dominate infauna include parasites, carnivores, and even osmotrophs (Fauchald and Jumars 1979, Jumars et al. 2015). Within the sediments, infaunal invertebrates are preyed on by other invertebrates (Commito and Ambrose 1985) and also by fishes and decapod crustaceans. Quiet w ­ aters support greater biomass of inverte-

Chapter 11 Estuaries and Coastal Seas

brates than physically more dynamic sediments—­a pattern opposite that of rocky shores—­prob­ably ­because sediments in quiet w ­ aters are relatively stable and accumulate organic m ­ atter depositing from the ­water column (see figure 11.5). The meiofauna that inhabit interstitial spaces between sand grains are dominated by nematode worms and harpacticoid copepods, along with a diverse array of minor invertebrate taxa (see figure 2.13). Most of ­these animals make their living by grazing bacteria and algae from the sediment grains, or feeding on detritus, but the meiofauna also have their own, tiny top predators (Montagna 1995). Major traits of infaunal organisms impor­tant in ecosystem pro­ cesses include body size, mobile versus sessile habit, feeding mode, and degree of aggregation.

Community organ­ization and key interactions Competition for space can shape the relative abundances of infaunal species in sediments as it does on the two-­dimensional surfaces of rocky shores. In general, competitive exclusion appears less severe in the three-­dimensional volume of the sediments, where organisms have more opportunities to avoid one another than do t­ hose on rock surfaces b­ ecause they can partition the vertical dimension of the sediment column (Peterson 1979). For example, experiments in a California lagoon found that infaunal bivalve species compete only if they live at the same depth in the sediment. When the deep-­ burrowing clam Sanguinolaria nuttalli was transplanted into a muddy sand bottom, its growth was reduced by 80% in the presence of other similarly deep-­dwelling bivalves, and individuals often emigrated from the enclosures. In contrast, Sanguinolaria was unaffected by the presence of a clam that lives shallower in the sediment (Peterson and Andre 1980). ­There are numerous such examples of interference competition among infaunal species with similar feeding modes and positions in the sediment, and stronger evidence of intraspecific competition (Wilson 1990). Segregation among sediment-­dwelling species may result from mechanisms other than competition. In the field, suspension-­feeders and deposit-­feeders often live in dif­fer­ent environments or areas, with l­ ittle overlap. Suspension-­feeders tend to dominate in firm, sandy substrata, while deposit-­ feeders are most abundant in mud. This situation illustrates the idea of trophic amensalism, by which ­these functional groups create conditions unfavorable for one another, increasing their segregation (Rhoads and Young 1970). Deposit-­feeders often strongly disturb the sediment, creating an easily resuspended and unstable surface layer that clogs the feeding structures of suspension-­feeding organisms and may also interfere with settlement of their larvae. ­There is considerable evidence that sediment disturbance can reduce abundances of par­tic­ul­ar species, but a ­simple negative relationship between suspension-­and deposit-­feeders has been elusive, in large part ­because of functional diversity among species within each group (Wilson 1990) and the common switching between t­ hese two feeding modes by vari­ous species (Lopez and Levinton 1987, Jumars et al. 2015). Predation has strongly ­shaped sediment communities in both ecological and evolutionary time. Shallow marine sediments in the Paleozoic era supported flourishing communities of epibenthic crinoids, brachiopods, and other invertebrates, but many of ­these groups declined greatly, dis­appeared, or ­were marginalized to the deep sea during the Cenozoic, coincident with the evolution of power­ful predatory fishes and crustaceans (Vermeij 1978). In contrast, bivalve mollusks radiated in the early Cenozoic, and lineages with siphons and burrowing habits w ­ ere most successful, evidently b­ ecause ­these traits allowed them to escape predators foraging over sediment surfaces (Stanley 1968). In the modern ocean, most life on sediment bottoms is below the surface. Predation is likely a strong influence on larval recruitment, as it is in larger life stages. Many infaunal invertebrates, like their counter­ parts on rocky shores, produce planktonic larvae that develop in the ­water column. Recruitment of larvae back into the benthos can be a dangerous affair as settlers approaching the benthic boundary must avoid being eaten or entrapped by suspension-­feeders. Despite their small size, the invisible animals of the meiofauna can be impor­tant predators on t­ hese recruits. Field experiments in a bay in North Carolina, USA, showed that turbellarian flatworms in par­tic­ul­ ar feed on settling larvae and can

313

314

Ocean Ecology

reduce abundances of macrofaunal polychaetes, bivalves, and amphipods during this critical transition (Watzin 1985, 1986). Fi­nally, facilitation among species is common in sediments and possibly more impor­tant than competition. In the absence of organisms, sediment bottoms have ­little physical structure. Colonization by foundation species, such as large bivalves and tube worms, provides firm substrata, zones of oxygenated sediment, protection from disturbances and predation, and ­water flow that attracts other organisms (Woodin 1976, Reise 2002), often supporting specialized commensal invertebrates. Many sediment-­dwelling invertebrates are apparently attracted to other infaunal species and the structures they build, and field experiments in several parts of the world show that infaunal invertebrates often facilitate the colonization of other species (Gallagher et al. 1983, Thrush et al. 1992). In general, marine sediment communities are structured as much or more by the engineering activities of organisms—­ via bioturbation, sediment stabilization, and creation of biogenic structures—as they are through trophic and competitive interactions (Reise 2002).

Ecosystem pro­cesses and ser­vices The communities of marine sediments illustrate perhaps better than any other habitat the complex bidirectional influences between organisms and the abiotic environment. Marine sediments support a diverse array of microbes that drive intense biogeochemical cycling of ele­ments powered by the flux of organic m ­ atter from the w ­ ater column and, in estuaries, from the adjacent land and watershed. The extensive area of shallow marine sediments, the quantity of organic ­matter that accumulates, and the richness of microbial pro­cesses within the geochemically complex surface layers mean that shallow marine sediments are major sites of carbon and nutrient cycling. Invertebrates, including the tiniest ones, are key players in ­these biogeochemical pro­cesses. Estimated rates of mass-­specific oxygen consumption suggest that metabolism, that is, pro­cessing of organic carbon, by meiofauna and macrofauna is of similar magnitude (Gerlach et al. 1985). Invertebrates can strongly influence energy flux and biogeochemical cycling in sediment habitats through two main classes of mechanisms. The primary influence is through bioturbation, the physical disturbance and mixing of sediments by animals. Deposit-­feeding invertebrates actively burrow through and/or ingest sediment, pro­cess it in the gut, and deposit the spoils usually on the surface (Lopez and Levinton 1987). The burrowing and feeding activities of infauna strongly modify the physical and chemical environment of the sediments (Reise 2002), creating mosaics of microenvironments that catalyze microbial metabolism and biogeochemical cycling. Burrowing organisms move sediment from deep layers to the surface, irrigate other­wise anoxic sediments below the surface with oxygenated ­water, secrete mucus along burrow walls, and sequester organic-­rich particles, all of which contribute to increased activities of both microbes and meiofauna, including enhanced nitrification (conversion of NH4 to NO3) and denitrification (conversion of NO3 to N2) (Aller 1988). The feeding and burrowing activities of infauna thus generate strong feedbacks between the geochemical environment, microbes, and their own populations (Herman et al. 1999). The second way that animals mediate ecosystem pro­cesses in sedimentary environments is by transforming the physical nature of the substratum itself through their own growth and accumulated products (Reise 2002). In subsequent sections we explore the engineering roles of benthic macrophytes, and in chapter 12 of coral reefs. Extreme cases involve suspension-­feeders that can turn soft sediments into hard substrata. Oysters in par­tic­u­lar can form massive aggregations with far-­reaching effects on both under­lying sediments and the overlying w ­ ater column. Prior to intense h­ uman exploitation (figure 11.10), oysters in many coastal areas formed reefs similar in magnitude to ­those of corals and posed similar ­hazards to shipping. In former times, oyster reefs pro­cessed huge volumes of ­water: It’s been estimated that, prior to exploitation, oysters could have cleared the entire volume of the Chesapeake Bay in two to four days (Newell 1988), supporting a now widespread motivation

Chapter 11 Estuaries and Coastal Seas

Figure 11.10. ​The lost kingdom of the oyster. A mountain of discarded shells from the industrial-­scale mining of oysters that once dominated the economy of the Chesapeake Bay region, USA, and many other regions around the world (https://­oysterrecovery​.­org​/­oysters​-­101​/­).

for restoring oysters as a strategy to eliminate nuisance algal blooms and resulting hypoxia. Most such filtration would be restricted to shallow nearshore areas, however (Pomeroy et al. 2006). The strength of bioturbation and biogeochemical pro­cessing depends on the traits of the organisms pre­sent, particularly body size, mobility, and feeding mode (Lopez and Levinton 1987, Wilson 1990). In shallow w ­ aters, bioturbation by animals can mobilize sediment nutrients that increase benthic primary productivity, as shown by experimental manipulations of the large irregular urchin Echinocardium sp. in a New Zealand bay. Plots with more of ­these deposit-­feeders had elevated rates of both ammonium release from the sediment and primary production, despite their grazing on benthic microalgae (Lohrer et al. 2004). Moreover, both field experiments (Thrush et al. 2006) and simulations based on field data (Solan et al. 2004) confirm that large infaunal invertebrates are disproportionately impor­tant in bioturbation and stimulation of biogeochemical fluxes between sediment and ­water column. Since large animals are also generally the most severely affected by h­ uman disturbance, including trawling (Thrush and Dayton 2002) and pollution (Pearson and Rosenberg 1978), ­human impacts on large benthic invertebrates can strongly reduce rates of sediment biogeochemical pro­ cesses. Both diversity (richness) and species composition of infauna also influence intensity of bioturbation (Solan et al. 2004). ­Because marine sediments cover vast expanses of the earth’s surface in both shallow and deep ­water, predator-­prey interactions in t­hese habitats are major routes of energy and materials flux through marine ecosystems. Particularly in shallow w ­ ater, pelagic-­benthic coupling is an impor­tant link to demersal fishes: phytoplankton production is consumed by benthic suspension-­feeders and deposit-­feeders as it ­settles out, and ­these invertebrates are consumed by fishes and crustaceans that support fisheries. On the east coast of the USA, an intensive expert review concluded that unstructured soft sediments are the most impor­tant coastal habitat type supporting fishery production, largely ­because of their extensive area (Kritzer et al. 2016).

Sediment bottoms in the Anthropocene Communities inhabiting marine sediments are changing u­ nder a plethora of h­ uman influences. Infaunal organisms live in intimate contact with the sediments, the composition of which therefore fundamentally influences their biology and community composition. We have seen already that the faunal composition of sediments varies as a function of grain size and organic content. Pollution of sediments by organic materials, metals, and other substances similarly changes community composition.

315

316

Ocean Ecology

During the 1970s growing environmental concern generated numerous field studies of pollution effects on marine benthos. Early syntheses of this work focused on organic enrichment, including sewage, and concluded that benthic community composition varies predictably along gradients in organic enrichment, driven largely by oxygen depletion due to intense decomposition in sediments loaded with organic pollution. Highly polluted sites are anoxic and may have no macrofauna at all, but with declining pollution ­there is often a peak in abundance of small, opportunistic invertebrates nourished by the organic ­matter. Th ­ ese enriched sites are dominated by a few species, whereas more pristine sites support more taxonomically and functionally diverse communities of longer-­lived invertebrates (Pearson and Rosenberg 1978) (figure 11.11). Such marine hypoxia is a growing global prob­ lem (Diaz and Rosenberg 2008, Breitburg et al. 2018); low-­oxygen “dead zones” have spread substantially around the world’s coastal regions as land-­based nutrient runoff stimulates phytoplankton blooms that sink and decompose in bottom ­water, and low-­oxygen zones are also spreading in the deep-­water column of the open ocean, apparently as a result of climate warming, which reduces solubility of oxygen and raises metabolic oxygen demand of organisms. As industrial fishing has expanded in scope and intensity, trawling of the seabed has become the most widespread ­human impact on marine sediment ecosystems. Bottom trawling has been compared with forest clear-­cutting in that it is nonselective and often takes out long-­lived structure-­ forming organisms, which are considered essential fish habitat ­because many other species depend on them. Trawling is estimated to affect an area equivalent to nearly half the global continental shelf each year, more than 100 times the land area clear-­cut annually (Watling and Norse 1998). Trawlers work w ­ aters of nearshore areas, the continental shelf, and even seamounts. Trawls that gouge deeply into the sediment cause more damage, and hydraulic dredges are the most destructive of all, removing 41% of benthic animals on average; recovery from trawling may take more than six years in such

Organic enrichment

Aerobic sediment

2

3

Anaerobic sediment

4 Zone

Normal

Typical Nucula macrofauna Amphiura dominants Terebellides Rhodine Echinocardium Nephrops

Polluted

Transitory Lobidoplax Corbula Goniada Thyasira Pholoe

Chaetozone Anaitides Pectinaria Myriochele Ophiodromus

Capitella Scolelepis

Grossly polluted No macrofauna Surface covered by fiber “blanket”

Figure 11.11. ​Infaunal communities change consistently in functional community structure across a gradient in organic enrichment. Clean, undisturbed sediments develop rich faunas that include large-­bodied invertebrates, whereas polluted sediments support low-­diversity faunas dominated by opportunistic worms (­after Pearson and Rosenberg 1978). This example is from the Baltic Sea.

Depth in sediment (cm)

1

Chapter 11 Estuaries and Coastal Seas

cases (Hiddink et al. 2017). Yet, in heavi­ly fished areas, trawlers may return to the same areas several times each year.

Seagrass Meadows Shallow sediments throughout much of the world’s coastal ocean are, or formerly w ­ ere, carpeted by submerged grasslands—­productive, teeming with animal life, and often overlying organic-­rich soils. The foundations of t­ hese ecosystems are seagrasses, the only group of flowering plants that has evolved the ability to live its full life history submerged in seawater. Seagrasses form dense populations in estuarine and protected coastal ­waters, from the equator to high latitudes on all continents except Antarctica (figure 11.12). The physical structure of the seagrass canopy, and high productivity of associated algae and detritus, supports animal communities that are often considerably more diverse and productive than in surrounding unvegetated sediments (Orth et al. 1984). Seagrass meadows are especially impor­tant as nursery habitat for young stages of fishes and larger invertebrates (Lefcheck et al. 2019). But much seagrass production is ungrazed and flows into detritus food webs, or is buried in sediments, making seagrass meadows impor­tant sites of “blue” carbon burial (Cebrián 1999).

Geomorphology and environment Most seagrass species root in sediment bottoms, usually in protected environments (see figure 11.3), with the exception of the surfgrasses (Phyllospadix spp.) and a few other species that can colonize rocky shores (see figure 8.1). Seagrasses live in shallow w ­ aters where sufficient light penetrates to the bottom to support photosynthesis and net positive growth, generally  5 m year−1 in Halodule ovalis (Duarte 1991, Marba and Duarte 1998), and has consequences for re­sis­tance and resilience to disturbance (O’Brien et al. 2018). Perhaps as a result of the leaf architecture inherited from terrestrial ancestors, seagrasses are more subject to inorganic carbon limitation than are algae (Koch et al. 2012), so seagrasses are likely to benefit from increased carbonization of seawater, for example, ­under simulated ocean acidification (Hall-­Spencer et al. 2008). Seagrasses are by definition the dominant biomass components in seagrass meadows, but algae are often more impor­tant in fueling the food chain. The leaves of seagrasses provide a large surface area that is often colonized by epiphytic (from Greek: “on plant”) algae, as well as sessile invertebrates, such as barnacles, sponges, tunicates, and tube worms. The productivity of epiphytic diatoms and filamentous algae is typically 20%–60% that of the seagrasses they grow on, and can surpass seagrass productivity in some cases (Hemminga and Duarte 2000) despite the usually much smaller biomass of epiphytes. Epiphytic algae are much more intensively grazed than seagrasses. Thus, productivity by the low standing biomass of algae feeds most of the animals within the meadow, as confirmed by stable isotope data (Thayer et al. 1978, Moncreiff and ­Sullivan 2001). Seagrass stands support diverse communities of animals. Waterfowl feed on seagrasses in many temperate areas (Kollars et al. 2017), and in the tropics seagrasses are grazed by sea turtles, sirenians (dugongs and manatees), scarid and sparid fishes, and sea urchins (Thayer, Bjorndal, et al. 1984). In many temperate seagrass communities, the most abundant herbivores are smaller invertebrate mesograzers that feed on algae and detritus attached to seagrass leaves (Valentine and Duffy 2006). Generally, vertebrate herbivores feed primarily on seagrass tissues whereas invertebrates feed on epiphytic algae and detritus, although many fishes graze both (Heck et al. 2000) and sea urchins graze directly on seagrasses (Valentine and Heck 1999).

Community organ­ization and key interactions Seagrass dominance over algae is maintained, or disrupted, by a nexus of interactions among nutrient supply, the seagrass plant, algae, grazers, and sometimes also predators. ­These interactions vary regionally depending on the species pre­sent and their traits (Duffy et  al. 2014). ­Because seagrasses have relatively low requirements for nutrients and are able to obtain them from the sediment, they tend to dominate where ­there is strong light, low nutrients, and healthy populations of grazers that control competing algae. When one or more of ­those conditions is disturbed, the balance can be upset in ­favor of macroalgae or phytoplankton. Herbivory is a key interaction in many seagrass systems, the nature and outcome of which depend on grazer food preferences and vulnerability to predation (Duffy et al. 2014). Top-­down interactions can have ­either positive or negative effects on seagrasses depending on the dominant herbivores and food web structure (figure 11.14). Vertebrate herbivores and sea urchins that feed directly on seagrasses often depress their biomass and fitness, as illustrated by the “halos” of denuded sediment surrounding tropical reefs (see figure 7.5), which reflect grazing by urchins and fishes reluctant to stray farther from the shelter of the reef and risk predation (Ogden et al. 1973, Madin et al. 2011). In contrast, most invertebrates feed preferentially on algae. In the mutualistic mesograzer model (Orth and Van Montfrans 1984), small herbivorous invertebrates feed on epiphytic algae, reducing competition with seagrasses and thus fostering seagrass dominance. Meta-­analysis of grazing experiments in seagrass systems found that seagrass growth is generally stimulated by nutrient addition, especially in the sediments, and that impacts of grazers depend on w ­ hether they feed on epiphytic algae, which generally increases seagrass fitness, or on the seagrasses themselves, reducing seagrass fitness (Hughes et al. 2004). Cascading effects of predators can influence seagrass biomass and production, but the sign of ­those effects depends on which type of herbivore dominates, and on ­whether predators are strict carnivores or omnivores that also graze algae (see figure 11.14). In some systems, small fishes consume

319

320

Ocean Ecology

(A) Mutualistic mesograzers: West Sweden Large predators

Small predators

Invertebrate seagrass grazers

–

Large predators

+

Small Omnivores

+

Invertebrate seagrass grazers

+ –

Fouling animals

Seagrasses

Fouling animals

Small predators

Small Omnivores

Algae

Algae grazers

?

Algae

Large predators

–

–

+

(D) Destructive mesograzers: Southern California

(C) Fouling invertebrates: Southern California

Small predators

Invertebrate seagrass grazers

Seagrasses

Algae

Large predators

Small Omnivores

Small predators

Algae grazers

Fouling animals

Seagrasses

(B) Omnivourous intermediates: Gulf of Mexico

Algae grazers

+

Invertebrate seagrass grazers

Seagrasses

+

Fouling animals

Small Omnivores

–

Algae grazers

Algae

Figure 11.14. ​Interaction pathways in seagrass beds documented by field studies. Plus and minus signs indicate positive and negative effects, respectively, of each functional group on seagrass biomass; question mark indicates that the net effects of a functional group on seagrasses are uncertain; solid and dashed arrows denote trophic and nontrophic paths of influence, respectively. (A) Effects of mutualistic mesograzers, Swedish west coast. (B) Effects of omnivorous intermediate predators, Gulf of Mexico. (C) Effects of fouling organisms, San Diego Bay. (D) Effects of invertebrate grazers on seagrasses, San Diego Bay (­after Duffy et al. 2014).

invertebrate grazers, releasing algae to overgrow seagrasses (Hughes et  al. 2004, Baden et  al. 2010, Reynolds et al. 2014), whereas in other systems omnivorous fishes also consume algae (Heck et al. 2000), thus fostering seagrass growth. Disease organisms can also play key roles in seagrass population and community dynamics, and are a major, often underappreciated, structuring force in many marine ecosystems (Lafferty et al. 2006). The protist Labyrinthula zosterae is common in several seagrass

Chapter 11 Estuaries and Coastal Seas

species and has been implicated in seagrass wasting disease, which caused large-­scale die-­offs of eelgrass throughout the North Atlantic in the 1930s (Muehlstein et al. 1991, ­Sullivan et al. 2013). This “catastrophe of nature” (Addy and Aylward 1944) reportedly wiped out 99% of eelgrass along the east coast of North Amer­ic­ a in 1931–1932, impacting numerous waterfowl species that use eelgrass as habitat and food (Rasmussen 1977) and effectively killing the fishery for bay scallops, juveniles of which are obligate associates of seagrass (Thayer, Kenworthy, et al. 1984). In the 1980s a similar die-­off began in New ­England and was traced to Labyrinthula zosterae (Short et al. 1987, Muehlstein et al. 1991). Labyrinthula is now known from eelgrass populations worldwide (Martin et al. 2016). As in other systems, the outcomes of interactions among plants, grazers, and their enemies depend on environmental forcing. This is illustrated by coordinated experiments crossing fertilization and mesograzer reduction in eelgrass meadows across the Northern Hemi­sphere, which found that grazing impacts on algae tend to be stronger in warmer areas (Duffy et al. 2015). The severity of wasting disease may also be exacerbated by high temperatures. The West Atlantic eelgrass die-­off of the 1930s coincided with unusually warm ocean ­waters throughout the region and the disappearance from the British Isles of several cold-­water plankton taxa (Southward 1960). Interactions between warming and disease have potentially impor­tant implications for the f­ uture of seagrass ecosystems in the current age of ocean warming.

Ecosystem pro­cesses and ser­vices Seagrass habitats support high primary and secondary productivity and create physical structure on other­wise featureless sediment bottoms, and the sediments are often hotspots of biogeochemical pro­cesses. Seagrasses support abundant animal communities through two main mechanisms. First, ­these plants create among the most productive natu­ral habitats on land or sea (Duarte and Chiscano 1999) (see ­table 11.1). In tropical areas, this productivity is, or was historically, grazed directly by green sea turtles and sirenians (manatees and dugongs), which are specialized feeders on seagrasses (Thayer, Bjorndal, et al. 1984), and in some areas also by sea urchins. In temperate regions, migrating waterfowl, such as Canada geese and brant, can feed heavi­ly on intertidal and shallow subtidal seagrass, often reducing biomass and sometimes denuding the substratum (Kollars et al. 2017). But in modern seagrass ecosystems, less than 20% of net seagrass production is grazed, on average, and much of the production goes into the detrital food chain (Cebrián 1999, Valentine and Duffy 2006). The second mechanism by which seagrasses facilitate animal communities is via the dense canopy of shoots and leaves that provide structurally complex habitat for algae, infaunal and epifaunal invertebrates, and fishes, some of them commercially impor­tant. Seagrasses support higher densities and growth of juvenile fishes than other coastal habitat types, especially in temperate and subtropical regions. Much of this benefit appears attributable to provision of food, whereas refuge value is similar to that of other nursery habitats (Heck et al. 2003, McDevitt-­Irwin et al. 2016, Lefcheck et al. 2019). ­These seagrass pro­cesses can support impor­tant ecosystem ser­vices to ­people, particularly in the developing world. For example, the diverse seagrass meadows of Indonesia contribute 54%–99% of daily protein intake for local h­ uman communities, much of this coming from juvenile fishes that use the seagrass beds as nursery (Unsworth et al. 2014). ­There is also l­imited evidence that seagrasses, particularly the larger and more robust taxa, such as Posidonia, Enhalus, and Thalassia, can provide significant coastal protection again storms and erosion (Nordlund et al. 2016). In some regions, seagrass meadows export substantial production as detritus to other habitats, even the deep sea. Seagrass detritus is commonly observed on the deep ocean floor around continents (Wolff 1979), and deep-­sea urchins and sea cucumbers off the Ca­rib­bean island of St. Croix evidently feed largely on the shallow-­water seagrass Syringodium filiforme washed in from nearby coastal habitats, as evidenced by gut contents and stable carbon isotope ratios (Suchanek et al. 1985). But a greater proportion of seagrass production in most systems is buried in situ. Despite making up a

321

322

Ocean Ecology

Ecosytem C storage (Mg Corg ha–1)

1,200 1,000

Living biomass Soil Corg

800 600 400 200 0

Boreal

Temperate

Tropical upland

Oceanic mangrove

Seagrass

Figure 11.15. ​Estimated carbon storage capacity of dif­fer­ent terrestrial and marine vegetated ecosystems. Contributions of plant biomass and soil organic carbon (Corg) to storage in the top meter of soil in seagrass and terrestrial systems (­after Fourqurean et al. 2012).

tiny fraction of the ocean’s area, seagrass meadows are responsible for an estimated 10% of the yearly burial of organic carbon in marine sediments, or roughly twice the average organic carbon storage in terrestrial soils on a per area basis, and an equivalent total storage to that of salt marshes and mangrove forests combined (figure 11.15) (Fourqurean et al. 2012). This high carbon storage capacity appears to result from the combination of high primary production, the tendency of seagrass canopies to filter out suspended particles and deposit them in sediment, and low decomposition rates in oxygen-­poor seagrass sediments. Although the carbon storage capacity of seagrass meadows is high on average, it’s impor­tant to recognize that seagrass species and populations vary widely in this capacity. Even within the single species Zostera marina, organic carbon content of the sediment varies by two ­orders of magnitude (Röhr et al. 2018).

Seagrass meadows in the Anthropocene

Requirements for clear ­water and low nutrient concentrations make seagrasses vulnerable to nutrient pollution and sediment loading that reduce light availability and ­favor faster-­growing algae. The protected embayments in which seagrasses grow best are also prime real estate for coastal and harbor development, exacerbating t­ hese threats. As a result, seagrasses are declining worldwide, and studies suggest that 30% of global seagrass cover has been lost since the first estimates ­were made in the late nineteenth ­century, with loss rates increasing in recent de­cades (Waycott et al. 2009). Of the 72 known seagrass species, 10 are at elevated risk of extinction and 3 are classified as endangered (Short et al. 2011). A principal challenge to conserving seagrass ecosystems is restoring w ­ ater quality to suitable levels. Tampa Bay in Florida, USA, provides one of the success stories ( Johansson and Lewis 1992). Tampa Bay was so degraded with poorly treated wastewater, fertilizer, and industrial effluents during the 1960s that residents complained about the smell. In 1972 Florida implemented new wastewater treatment regulations, which successfully reduced nitrogen loading by an order of magnitude over several years. ­Water clarity improved, harmful algal blooms declined, and seagrass, which had not been seen for de­cades in most parts of the bay, began to come back, expanding rapidly in the late 1980s (figure 11.16). Similarly, long-­term monitoring in Chesapeake Bay, USA, revealed that legislation to reduce nutrient runoff from the watershed has led to gradual recovery of submerged vegetation in this large and heavi­ly populated estuary (Lefcheck et al. 2018). Many temperate seagrasses are sensitive to unusually warm temperatures. Warming-­induced deterioration of seagrass ecosystems has been observed over recent de­cades in the West Atlantic, Mediterranean, and Baltic Sea, with summer temperature spikes often leading to widespread seagrass mortality (Short and Neckles 1999, Reusch et al. 2005, Moore and Jarvis 2008, Jordà et al. 2012). Experiments with the perennial Posidonia oceanica in the Mediterranean Sea found that warming increased seagrass mortality, leaf necrosis, and respiration, and depleted above­ground biomass and the carbohydrate reserves of seeds; seedlings grown ­under warmer conditions ­were more vulnerable to grazers (Tomas et al. 2015). Climate warming is also affecting other components of seagrass ecosystems, notably via tropicalization—­increasing repre­sen­ta­tion of tropical species—­among seagrass-­associated fish communities (Fodrie, Heck, et al. 2009), with the potential to reduce seagrass biomass and habitat complexity as tropical herbivorous fishes increase (Heck et al. 2015). Among the most serious concerns is rising frequency of disease, which is often aggravated by warming (Harvell et al. 2002, Altizer et al. 2013). The widespread die-­offs of eelgrass throughout the North Atlantic in the 1930s are thought to have been related to unusually warm temperatures that triggered wasting disease, and subsequent

Chapter 11 Estuaries and Coastal Seas

323

Halodule coverage (km 2)

Schizothrix filaments (1000 ml–3)

Secchi depth (m)

Chlorophyll a (mg m–3)

Total nitrogen loading (kg N d–1)

(A) experiments confirmed a link between warm ­waters and prevalence of the pathogen Labyrinthula zosterae in eelgrass, Zos5 tera marina (­Sullivan et al. 2013, Kaldy 2014). 4 As global temperatures rise, so does mean sea level. A sea Tampa Bay level model calibrated with empirical data and applied to the 3 Intergovernmental Panel on Climate Change’s (IPCC, 2014) “business-­as-­usual” scenario (RCP 8.5) proj­ects a sea level 2 rise of 52–131 cm by 2100 (Kopp et al. 2016). Small changes in sea level can have large impacts on coastal ecosystems 1 ­because of the flat, gentle slope of much coastal land. By definition, rising sea level increases ­water depth, reducing light 0 1975 1980 1985 1990 penetration, but is also generally accompanied by greater tidal flows and salinity intrusion into upstream estuaries and (B) groundwater (Short and Neckles 1999), which can reor­ga­ 1.2 nize estuarine ecosystems. Although coastal wetlands, in30 Secchi cluding seagrass meadows, are dynamic ecosystems that can counteract sea level rise, their capacity to do so is l­ imited and 1.0 20 affected by many ­human activities (Kirwan and Megonigal 2013). One major challenge is “coastal squeeze,” develop0.8 Chl a ment in the coastal zone that blocks the paths by which wet10 lands could other­wise migrate shoreward as sea level rises. 0.6 Seagrasses are likely to be among the few kinds of marine 0 organisms that perform better in a more acidified ocean. Sea1975 1980 1985 1990 grass growth and productivity can be ­limited by inorganic car(C) bon availability in seawater, suggesting that increasing dissolved carbon dioxide ­will increase seagrass growth and productivity, 50 500 on average (Koch et al. 2012). This prediction is supported by Schizothrix Return of blooms 400 40 Halodule greater growth rates of seagrasses around natu­ral marine CO2 seeps; indeed, ­these seagrass communities sequestered consid300 30 erably more carbon belowground ­under acidified conditions, suggesting a pos­si­ble feedback to reduce the impacts of CO2 200 20 injection into marine ­waters (Russell et al. 2013). The large animals that depend on seagrasses have de100 10 clined substantially. Prior to Eu­ro­pean colonization of the 0 0 Amer­i­cas, large seagrass-­feeding vertebrates w ­ ere extremely 1985 1990 1975 1980 abundant. For example, the Caymans Islands fishery in the Year Ca­rib­bean landed ~13,000 sea turtles each year for de­cades beginning in the late seventeenth ­century ( Jackson 1997), Figure 11.16. ​A success story in coastal restoration. In Tampa Bay, and the number of dugongs along the coast of the G ­ reat Barrier Florida, USA, nitrogen loading from municipal waste was reduced Reef region was much greater than it is ­today (Marsh et al. tenfold in the late 1970s, ­after which phytoplankton biomass (chlorophyll a) declined and ­water transparency (Secchi depth) 2005). Such densities of large herbivores surely had major increased (TON = total organic nitrogen). The system shifted from impacts on seagrasses. Dugongs, which feed on both rhizomes one dominated by blooms of cyanobacteria (Schizothrix calcicola) and shoots, require ~3.5 hectares of seagrass a year per individback to the historical cover of seagrass (Halodule wrightii) (­after Cloern 2001; original data from Johansson and Lewis 1992). ual (Heinsohn et al. 1977). One approach to remedying the decline in seagrass extent is active restoration by planting, which has become a major activity, albeit with decidedly mixed success (van Katwijk et al. 2009). Growing evidence indicates that, in many areas, not only good ­water quality but intact food webs are also impor­tant to healthy seagrass populations (box 11.1). Seagrass restoration has succeeded in several areas with good w ­ ater quality but has failed in many

324

Ocean Ecology

Box 11.1. ​Top-­down control and the restoration of coastal vegetation something unexpected happened. Eelgrass reversed its decline and has increased by 600% since then, despite dissolved nitrate concentrations in the ­water that continue to increase exponentially and are considered highly eutrophic (B. B. Hughes et al. 2013). The paradoxical resurgence of seagrass coincided with sea otters reinvading the estuary ­after over a c­ entury of near absence. As the estuary’s sea otter population grew over the ensuing de­cades, their crab prey declined sharply, grazing invertebrates that are vulnerable to crab predation increased, and epiphytic algae declined, leaving clean

Major expansion of industrial fertilizer use a ­ fter World War II led to a massive global increase in nitrogen availability and widespread eutrophication of the world’s coastal ­waters, accompanied by deteriorating w ­ ater quality and hypoxia (chapter 4). In many regions, dominance by perennial benthic vegetation gave way to dense blooms of phytoplankton and ephemeral algae (Anderson et al. 2002, McGlathery et al. 2007). In Elkhorn Slough, California, USA, eelgrass declined steeply through the 1970s as nitrogen loading to the estuary increased, mirroring patterns seen in many coastal regions worldwide. But in the mid-1980s,

400

40

200

20

0 0 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 Year

Edible crab biomass (g FW CPUE)

Nitrate as N (μM)

60

1.0

Grazer mass Large grazers

0.8 0.6

50

0.2 0

400

150 100

0.4

4

200

–

T

Large mesograzers (>2 cm) m–2

3

600

Sea otter density

80

9 6

100

Grazer biomass (g DW shoot –1 cm–1)

After otters

(C)

Carapace width (mm)

Before otters

12

120

Tomales Bay, no otters/low Elkhorn Slough, nutrients otters/high nutrients 1,800 140 1,500 120 1,200 100 900 80 600 60 300 0 40

Algal epiphytes (mg DW shoot –1 cm–1)

15

Eelgrass cover (Ha)

(B)

Eelgrass Sea otters Nitrate

Aboveground biomass (g DW m –2)

(A)

+

T

–

0

3

–

2

+

1

+

– –

0 100 80 60

200

40

100

20 0

1 2 3 4 Otter density *ha–1)

0

Belowground biomass (g DW m –2)

Aboveground Belowground

+

–

Algal epiphytes C T

300

0

T

+

+/– T Nutrients

Figure B11.1.1. Evidence that sea otters initiated a trophic cascade that facilitated seagrass recovery. (A) Historical trends in annual mean dissolved nitrate, seagrass cover, and sea otter densities in Elkhorn Slough, California, USA. (B) Contrasting trends at alternating trophic levels with increasing sea otter density: crab biomass and size declined, grazer biomass increased, epiphytic algae declined, and eelgrass biomass increased (DW, dry weight; FW, fresh weight; CPUE, catch per unit effort). Elkhorn Slough (green) has sea otters; Tomales Bay (blue) lacks sea otters. (C) Summary of top-­down and bottom-up pro­cesses influencing seagrass cover in Elkhorn Slough. Sea otter predation cascades through mesopredator crabs, to epiphyte mesograzers, to control algal epiphytes that compete with eelgrass. Solid and dashed arrows indicate direct and indirect effects, respectively; plus and minus symbols indicate positive and negative effects (­after Hughes et al. 2013).

Chapter 11 Estuaries and Coastal Seas

eelgrass. ­These trends of change in alternating trophic levels ­were not observed in nearby Tomales Bay, which remained uncolonized by sea otters. Experiments and comparisons between the sites led to the conclusion that a trophic cascade through four trophic levels has rebooted eelgrass dominance in Elkhorn Slough—­despite continuing eutrophic conditions (figure B11.1.1). A similar picture emerges from the seagrass and rockweed (Fucus) habitats that formerly fringed the Baltic Sea. In this region, eelgrass similarly suffers heavy fouling by algae in the absence of grazing by invertebrates, which are in turn controlled by fish predation. Shallow Baltic ­waters have transformed over recent de­cades from dominance by eelgrass and rockweed to accumulations of ephemeral algae (Bonsdorff et al. 1997). Long-­term field

325

monitoring and experiments suggest that exploitation of piscivores such as cod in offshore w ­ aters has released the smaller inshore fishes—­mesopredators—­from top-­down control, and their consumption of grazing invertebrates indirectly leads to algal blooms and decline of perennial seagrasses (Baden et al. 2010, Eriksson et al. 2011). Meta-­analysis of experiments support this explanation, showing that top-­down effects of mesopredators, including sprat, sticklebacks, and gobies, cascaded through reduced mesograzer abundances to increase ephemeral algae, with effects of similar strength to t­ hose of eutrophication (Östman et al. 2016). T ­ hese examples emphasize that restoration of impor­tant coastal vegetation habitats may benefit from management of both w ­ ater quality and fish populations.

o­ thers. Recent research has documented benefits of both ge­ne­tic and species diversity of seagrasses in restoration success. In a field experiment in ­Virginia, USA, eelgrass transplants with higher ge­ne­tic diversity survived longer, increased density faster, and supported roughly twice the areal productivity of eelgrass and invertebrates compared with plots planted at low ge­ne­tic diversity (Reynolds et al. 2012). Review of several such experiments showed that the boost to productivity in more genet­ically diverse seagrass populations was stronger during or ­after periods of stress (Salo and Gustafsson 2016). And in the highly diverse coastal areas of Indonesia, seagrass restoration efforts that planted multiple species resulted in better overall survival (Williams et al. 2017).

Salt Marshes Salt marshes are intertidal prairies, communities of emergent flowering plants with roots in seawater and canopy in the air, generally dominated by a few species of salt-­tolerant (halophytic) grasses and herbs (Bertness and Silliman 2014). Salt marshes form extensive dense stands on protected temperate and boreal shores. ­Toward the tropics, they are replaced by mangrove swamps in such environments. Like coral reefs, mangrove forests, and seagrass beds, salt marshes are habitats that are created by their foundation species. The salt-­tolerant plants that dominate marshes grow in dense stands that trap sediment, stabilize the substratum, attenuate storm waves, oxygenate the sediment by diffusion from their air-­filled root tissues (aerenchyma), and shade and reduce evaporation from the sediment surface. Salt marsh vegetation varies in diversity and composition around the world, but cordgrasses in the genus Spartina are characteristic dominants in many areas, particularly along Atlantic shores. In eastern North Amer­i­ca, where marshes have been most intensively studied, marshes are dominated by salt marsh cordgrass, Spartina alterniflora. Elsewhere in the world salt marshes can be composed of more diverse communities of forbs and grasses, as along the North Sea coast of ­England. Succulent plants are also common—­and sometimes dominant—­components of salt marsh communities, such as Salicornia in southern California, USA (Pennings and Callaway 1992) and Suaeda in some areas of the Mediterranean. Like other coastal communities of flowering plants, including seagrass meadows and mangrove swamps, salt marshes are highly productive, rivaling that of rich farmland. Also like other wetland vegetation, only a small fraction of salt marsh productivity is grazed u­ nder normal circumstances, and most enters the detrital food web ­after senescent plant tissues are colonized by fungi and bacteria and decomposed into smaller particles (Teal 1962). Nevertheless, some herbivores in the system

Ocean Ecology

(A)

(C) 4,000

10

Landings (ton yr –1)

Shrimp yield (kg yr –1)

8

107

106

105 4 10

106 105 Area of vegetated estuary (ha)

3,000

2,000

1,000

0

107

0

30

60 90 120 150 Mangrove fringe (km)

180

0

30

60 90 120 150 Mangrove fringe (km)

180

(D)

(B)

8

–1

Economic benefit ( US$ yr )

1,600

1,200 Annual catch (mt)

326

800

400

0

0

20

40

60

80

Wetland area lost (km2)

100

6

4

2

0

Figure 11.17. ​Wetlands are seafood nurseries. The yield of commercially harvested shrimp along the Louisiana coast of the Gulf of Mexico is (A) strongly related to the area of adjacent coastal wetland, and (B) declined with the area of coastal wetlands lost through time (­after Valiela et al. 2004). In the Gulf of California, (C) landings and (D) economic value of fisheries increased with the area of nearby mangrove forests (­after Aburto-­Oropeza et al. 2008).

feed on the emergent vegetation above ­water, including insects and rasping snails, while some crabs feed on the submerged vegetation. Salt marshes harbor dense animal populations and support high secondary productivity as a result of both their high primary productivity and the shelter from predators provided by vegetation structure. Fishery yields in many coastal areas are clearly related to the area of marsh and other vegetated habitat in the region (figure 11.17) (Valiela et al. 2004). Fi­nally, like other coastal vegetated ecosystems, salt marshes are threatened worldwide by dense and increasing h­ uman populations in the coastal zone and the multiple, interacting stressors that result, including eutrophication, climate warming, food web disruption, and coastal squeeze.

Geomorphology and environment Salt marshes develop in coastal regions where w ­ ater movement slows enough to deposit fine river-­ borne sediments. Major sites of marsh development include estuaries at the mouths of coastal rivers and the protected lagoonal shores backing barrier islands. They are especially extensive on the broad,

Chapter 11 Estuaries and Coastal Seas

Lower limits set by physical stress

Iva frutescens zone

Upper limits set by competition

Juncus gerardii zone

Spartina patens zone

Spartina alterniflora zone

Figure 11.18. ​Zonation of communities and pro­cesses across an intertidal salt marsh in New E­ ngland, USA (­after Bertness and Silliman 2014).

flat shores of sediment-­rich passive continental margins. Salt marsh plants strongly modify their environment and even local geomorphology through a positive feedback pro­cess: marsh plants trap fine sediment, which builds a shallow substratum for expansion of marsh plants, which then further slow ­water movement and capture more sediment (Redfield 1972). Salt marsh vegetation continuously modifies the environment as plant growth interacts with tidal flow to or­ga­nize intertidal sediments into a characteristic landscape featuring a network of tidal creeks in a matrix of marsh vegetation. As in other estuarine habitats, salinity is a major environmental driver of marsh structure and function: vegetation structure varies predictably along the gradient from freshwater at the upper end of tidal influence to characteristic salt marsh vegetation at the seaward end of the gradient (figure 11.18) (Bertness and Silliman 2014). The intertidal zone is a harsh environment for terrestrial and marine organisms alike, and abiotic stresses allow few flowering plants to thrive ­there, accounting for the characteristically low diversity of salt marshes. ­These stressors include variable salinity, which can fluctuate from near fresh u­ nder heavy rains to hypersaline during dry summer conditions, as well as hypoxia and sulfide toxicity in the organic-­rich muddy sediments. Nutrient limitation also appears to contribute to a characteristic feature of the extensive marshes of eastern North Amer­i­ca: bands of tall Spartina alterniflora line the banks of tidal creeks and channels, infilled with ­great expanses of short-­statured S. alterniflora. The greater stature of tall-­form Spartina evidently results from more vigorous w ­ ater flow and nutrient delivery, and reduced salinity stress, as well as better access by marine predators that maintain low densities of snails and crabs that can overgraze the Spartina.

Organisms and traits Salt marshes are defined by their foundation species, halophytic flowering plants adapted to living with roots submerged in saline, often waterlogged soil. Throughout much of the world, salt marsh vegetation is dominated by cordgrasses in the grass (Poaceae) genus Spartina, which contains about 17 species and is most diverse in the West Atlantic. Cordgrasses are clonal, spreading by growth of under­ground rhizomes to form extensive, dense meadows. A key stressor for flowering plants in the intertidal zone,

327

328

Ocean Ecology

both salt marshes and mangrove forests, is the anoxia that develops in the organic-­rich, waterlogged soil. Dominant plants of t­ hese environments, including Spartina alterniflora in eastern American salt marshes, have air-­filled tissues (aerenchyma) that deliver oxygen to their roots (Teal and Kanwisher 1966), allowing them to survive in the anoxic mud where most other plants cannot. ­Because marsh plants are emergent, with leaves above ­water, they support both marine and terrestrial animals. The rich production and decomposition of marsh plant detritus supports dense populations of deposit-­feeders, notably fiddler crabs (Uca) and marsh crabs (Sesarma), as well as suspension-­feeders, including con­spic­uo­ us clumps of mussels (Geukensia). The emergent marsh vegetation supports both herbivorous insects and littorine snails that graze on fungi and senescent leaf tissue (Silliman and Newell 2003). B ­ ecause of their high secondary production, salt marshes and adjacent ­waters often provide critical habitat for shorebirds and ­water birds, especially during seasonal migrations.

Community organ­ization and key interactions Salt marsh vegetation displays a characteristic zonation with tidal elevation just as distinct as that of the rocky intertidal, albeit over a much gentler elevation gradient and therefore a much wider horizontal expanse. This zonation is ultimately forced by the gradient in inundation time, desiccation, and salt stress across the intertidal zone. But it is also structured by strong interactions among species. The field experimental tradition that transformed rocky intertidal ecol­ogy came much l­ater to salt marshes, and has been advanced largely by Mark Bertness and his many students and collaborators in New ­England, USA (Bertness and Silliman 2014). Experimental studies of salt marsh communities have revealed three main themes. First, competition strongly shapes plant distribution across the marsh, generally setting the upper limits of species while abiotic stress sets the lower limit (see figure 11.18) (Vince and Snow 1984, Bertness and Ellison 1987). This pattern is opposite the traditional paradigm of rocky shores, primarily ­because salt marsh plants are derived from terrestrial ancestors, whereas the rocky intertidal is dominated by marine organisms. Zonation of the salt marsh community is a function of competition among plant species with dif­fer­ent physiological tolerances played out along a gradient in salinity and degree of soil moisture. The competition is asymmetric: in general, plant species that dominate the lower marsh, where the soil is saturated, saline, and frequently anoxic, are refugees from better conditions on the upland and fresher marsh areas. In New ­England, USA, for example, salt marsh cordgrass dominates the low marsh ­because it can tolerate the stressful conditions of salty, waterlogged soil. Reciprocal transplant experiments confirmed that Spartina alterniflora actually grows better in less saline soils than in the brackish intertidal w ­ aters where it is normally found, but it is outcompeted by freshwater and upland plants u­ nder the more favorable conditions t­ here and is thus restricted by competition to saline intertidal habitats (Crain et al. 2004). The high-­marsh dominant Spartina patens is in turn outcompeted by marsh elder (Iva frutescens) and needle rush (Juncus gerardii) that grow at higher tidal elevations. Experiments in Alaska (Vince and Snow 1984), California (Pennings and Callaway 1992), and Chile (Farina et  al. 2009) show that competition plays a strong role in salt marsh plant zonation around the world. ­These competitive interactions and the resulting zonation can be further modified by altered environmental conditions, however. The competitive hierarchy can be reversed by nutrient fertilization, which in New E ­ ngland caused both Spartina species to expand landward at the expense of the species that normally outcompete them, prob­ably as a result of relaxed competition for soil nutrients and a switch to competition for light (Levine et al. 1998). The second theme emerging from experiments in salt marshes is that, contrary to the view from ­earlier ecosystem-­focused studies, top-­down control by predation and herbivory can strongly affect

Chapter 11 Estuaries and Coastal Seas

community structure in salt marshes as in rocky intertidal communities. Early studies in salt marshes of Georgia, USA, documented the small proportion of production grazed (Smalley 1960). In the absence of ­human disturbance, this appears to be the norm in many regions, and a relatively low proportion of marsh production is directly grazed (Cebrián and Lartigue 2004). But herbivores are capable of overgrazing salt marsh plants, severely depressing their biomass, and experiments show that ­these herbivores are often controlled by their own predators. When predator abundance is reduced via fishing or other means, grazers can explode to trigger large-­scale salt marsh decline. In eastern North Amer­i­ca, Spartina alterniflora supports dense populations of snails, Littoraria irrorata, that rasp the plant surfaces and graze primarily on fungi growing in and on the leaves. Experimental protection of ­these grazers from their predators, primarily blue crabs, allowed snail populations to increase to the point where their grazing nearly eliminated Spartina (Silliman and Bertness 2002). Another path of trophic cascade has been implicated in salt marsh die-­offs seen throughout New ­England, USA. The banks of marsh creeks in this region are riddled with the burrows of the marsh crab Sesarma reticulata that feeds on plant material and hides in the burrows from its predators. Experimental exclusion of predators from a marsh through the growing season doubled the rate of crab grazing on Spartina and increased unvegetated bare space by > 150% compared with plots accessible to predators (Holdredge et al. 2009, Bertness, Brisson, et al. 2014). The emerging picture is that several groups of salt marsh consumers have the potential to destroy salt marsh vegetation, and the lower level of grazing documented in classical early studies of salt marsh ecol­ogy depends on effective top-­ down control by predators. That ecosystem balance is sensitive to disruption by fishing and other ­human activities. Fi­nally, the third theme highlighted by experiments in salt marshes is the importance of positive interactions in structuring communities (Bertness and Silliman 2014). Although competition among marsh plant species is well documented, facilitation also strongly influences community structure, mainly by reducing environmental stress. Salt marsh vegetation supports predictions of environmental stress models that positive interactions increase in importance in stressful environments (Bruno et al. 2003). Early plant colonists of the low marsh aerate the waterlogged soils, allowing other plants to grow ­there. At the high end of the marsh, established plants shade the soil surface and reduce evaporation, creating a more favorable environment that allows other species to colonize and increases diversity (Bertness and Hacker 1994). The thick canopy of foundation species, such as S. alterniflora, reduces evaporation from the sediment surface, maintaining moisture and low salinity; as a result, plant fitness often increases rather than decreases in the presence of neighbors, ­whether of the same or dif­fer­ent species. Invertebrates can also facilitate marsh plants. In the eastern USA, the h­ orse mussel Geukensia demissa is a common member of the salt marsh community, forming clumps among the Spartina. The mussel clumps retain w ­ ater and also fertilize the soil with the feces and pseudofeces they produce. Experiments by Christine Angelini and colleagues (2016) showed that cordgrass survival during drought was 5–25 times greater when mounds of h­ orse mussels w ­ ere interspersed among the cordgrass ­because the mussels retained ­water and reduced soil salinity at low tide. In both New ­England and Argentinean salt marshes, burrowing crabs increased Spartina production by aerating the soil, which made limiting nitrogen more available in New E ­ ngland (Holdredge et al. 2010) and allowed colonization of plant roots by mycorrhizal fungi in Argentina (Daleo et al. 2007). Thus, as in most ecosystems, salt marsh communities are structured by a web of antagonistic and facilitative interactions. Generally, competitive interactions prevail u­ nder favorable conditions whereas facilitative interactions dominate ­under more stressful conditions, across both time and space (Bertness and Ewanchuk 2002). Abiotic stressors and d­ rivers of community interactions vary predictably with latitude and salinity, and the relative balance of competitive and facilitative interactions varies with environmental temperature and therefore latitude. Intensity of herbivory and plant defense are also higher ­toward the equator (Pennings et al. 2001).

329

330

Ocean Ecology

Ecosystem pro­cesses and ser­vices The science of ecosystem ecol­ogy was strongly ­shaped by research in the salt marshes of Georgia, USA, by the b­ rothers Eugene and Howard Odum and their students, beginning in the 1950s (chapter 9; Bertness and Silliman 2014). Inspired by Lindeman’s (1942) trophic-­dynamic theory of ecosystems, the Odums and colleagues traced links from physical forcing to the vigorous productivity of salt marshes and how that energy flowed up through the food chain. ­These early studies established the influential paradigm of salt marshes as systems dominated by bottom-up control, where biological interactions ­were weak and had ­little impact on community structure (Teal 1962, Odum et al. 1971). As discussed in the previous section, subsequent experimental studies revealed that strong competition and facilitation are impor­tant controls on marsh community structure, and that grazing and predation can radically alter marsh communities and landscapes when triggered by disturbance. ­Under most conditions, however, grazing on marsh plants appears to be maintained at low levels, at least in part by predators. Most salt marsh productivity remains ungrazed and senesces in place, and on a per area basis more production is buried in the sediment of salt marshes than in any other vegetated marine ecosystem (Cebrián 2002). Salt marsh animals are supported largely by detritus and algal production, as evidenced by stable isotope studies (Currin et al. 1995). As a result, salt marshes are considered major blue carbon sinks for atmospheric carbon dioxide. Salt marsh ecosystems provide several benefits to h­ uman society. A major one is fishery productivity: across 27 estuarine regions worldwide, yields of penaeid shrimp w ­ ere closely related to the area of estuarine vegetation but uncorrelated with total area, depth, or habitat volume (Turner 1977). In Louisiana during the twentieth c­ entury, declines in shrimp harvest ­were greatest in areas that had lost the greatest area of adjacent marshland (Valiela et al. 2004) (see figure 11.17). This high secondary production presumably resulted from both high primary productivity of marsh plants and algae, and from the refuge they provide from predators. Marshes and other coastal wetlands also protect coastal ­human communities against erosion and storm damage. Analy­sis of 34 major US hurricanes found that economic damage in the affected areas was reduced by wetland area, and that wind speed and wetland area together explained 60% of the variation in damage; coastal wetlands w ­ ere estimated to provide more than $23 billion per year in storm protection in the USA (Costanza et al. 2008), including $625 million in avoidance of direct flood damage for a single hurricane, Sandy in 2012 (Narayan et al. 2017). Similarly, studies in the UK concluded that maintenance of natu­ral marsh was much less expensive than building and maintaining sea walls (King and Lester 1995).

Salt marshes in the Anthropocene The shallow protected shores where salt marshes form are prime habitat for h­ umans. For this reason and ­because the moist soil and quiet w ­ ater breed biting insects, marshes have been drained, filled, and other­wise engineered aggressively for centuries. Along the east coast of the USA, only 28%–55% of the marshes in the Chesapeake and Delaware Bays remained intact by the early 1990s, and in San Francisco Bay nearly 80% of tidal marsh habitats have been lost to conversion to human-­dominated habitats (Valiela et al. 2004). Rising sea levels also threaten marshes: model simulations based on the IPCC’s projected climate scenarios suggest that salt marshes ­will decline in area by 20%–45% by 2100, largely as a result of coastal squeeze (Craft et al. 2009). But ­human activities also appear to have increased the area of salt marshes since Eu­ro­pean colonization of the Amer­i­cas. Stratigraphic analy­sis and radiocarbon dating show that salt marshes in Mas­sa­chu­setts, USA, expanded rapidly during the eigh­teenth and nineteenth centuries as Eu­ro­pean settlers cleared the region’s forests for farmland, delivering large quantities of sediment to estuaries that ­were then colonized by marsh plants. One interpretation of this evidence is that the expansive marshes of eastern North Amer­i­ca may be his-

Chapter 11 Estuaries and Coastal Seas

torically anomalous, and their regression in modern times is a return to the more typical precolonial state (Kirwan et al. 2011). In addition to direct habitat conversion, salt marsh vegetation is vulnerable to a range of other ­human disturbances, often acting in concert, that can upset the delicate balance of food web interactions and lead to rapid decline. Historical data and experiments in New ­England, USA, show that the major ­factors include eutrophication, drought, climate warming, and trophic cascades initiated by fishing. Along the east coast of the USA, marsh dieback and erosion of creek banks began accelerating during the early 2000s. Surveys, experiments, and historical rec­ords at 14 sites revealed that marsh die-­off in New E ­ ngland resulted from a trophic cascade: grazing marsh crabs (Sesarma reticulatum) that burrow into the sediment along creek banks are normally controlled by fish and crab predation, but areas with intense recreational fishing removed this control, leading to marsh crab outbreaks, overgrazing of cordgrass, and landscape-­scale declines of marshes (Altieri et al. 2012).

Mangrove Forests Mangroves are coastal forests with their roots in the sea, occupying protected sediment shores in the tropics and subtropics (figure 11.19). They are replaced by salt marshes in such habitats at higher latitudes. The plants that dominate t­ hese forests, also called mangroves, are a polyphyletic collection of salt-­tolerant trees and shrubs numbering 57–110 species, from about 21 families, with greatest diversity in the Indo-­Pacific region. Mangroves are adapted to grow in organic-­rich, intertidal marine sediments, with the soil surface intermittently exposed to the air but generally saturated with seawater.

A

C

B

D

Figure 11.19. ​Mangrove forests of the world. (A) Red mangrove (Rhizophora mangle), Ca­rib­bean Sea. (B) Buttress roots of mangrove (Heritiera littoralis), Okinawa, Japan. (C) Gray mangrove (Avicennia marina) encroaching into Sarcicornia marsh, Australia. (D) Pneumatophores of mangroves at Bunaken Island, Indonesia.

331

332

Ocean Ecology

Major environmental controls on mangrove distribution and fitness include temperature, salinity, and sediment nutrient availability. Temperature appears to set the broad geographic distribution of mangroves. The poleward range limits of mangroves, at least in North Amer­i­ca, are determined by low winter temperatures, which periodically kill young recruits and thus limit the poleward distribution of t­ hese primarily tropical plants (Cavanaugh et al. 2013), and mangroves are moving poleward in several areas in response to climate warming. Historically, mangroves have been considered facultative halophytes, restricted to the saltwater fringe primarily by competition from other species better adapted to freshwater and upland soils. Indeed, some mangrove species thrive in freshwater u­ nder laboratory conditions. But a synthesis of the evidence suggests that true mangrove species may be obligate halophytes, since many are unable to grow for extended periods in freshwater, whereas they store Na+ and Cl− ions in their embryos, and have enzymes with higher salt tolerance than do other plants (Wang et al. 2011). Like other intertidal community types, mangrove forests show a characteristic zonation from low to high ­water. In the Amer­i­cas, the dominant species are red mangrove (Rhizophora mangle) and black mangrove (Avicennia germinans), with white mangrove (Laguncularia racemosa) along the upland parts of the swamp.

Geomorphology and environment The environmental ­drivers of mangrove community structure are similar to t­hose acting in salt marshes, the other main community of intertidal flowering plants. Both communities are rooted in intertidal sediments that are saline, usually saturated, rich in organic ­matter, and therefore anoxic. With essentially unlimited sunlight and ­water, mangroves are poised for high productivity, which is therefore usually ­limited by nutrients, particularly in the oligotrophic ­waters and calcareous sediments around coral coasts and islands. Experiments have shown that fertilization with nitrogen and especially phosphorus often substantially increase mangrove growth at sites around the world (Feller 1995, Lovelock et al. 2007, Naidoo 2009).

Organisms and traits The structure and functioning of mangrove ecosystems are strongly ­shaped by the traits and activities of the mangrove trees that dominate them. Mangroves vary from low, sparse bushes in some oligotrophic areas to rainforests of towering trees in Southeast Asia (see figure 11.19). Like other trees, mangroves produce a canopy of fo­liage elevated by large woody trunks, with roots in the sediment. Like salt marsh plants, mangroves face the twin challenges of maintaining osmotic balance and aerating their roots in the salty, hypoxic soil. W ­ ater must be moved against the osmotic gradient from the saline soil, through roots and trunk up to the leaf canopy. This is accomplished via two physiological adaptations. First, leaves are hyperosmotic relative to the vascular system and roots, which draw ­water up from the soil. Second, the roots are able to draw ­water out of the salty pore ­water against a steep osmotic gradient by maintaining a strong negative hydrostatic pressure generated by the upward flow of w ­ ater via transpiration—­essentially sucking w ­ ater out of the soil. Like many salt marsh plants, mangroves meet the challenge of rooting in anoxic sediments by producing specialized structures to deliver oxygen to their roots. The pneumatophores of black mangroves are woody knobs that protrude vertically from roots in the sediment to capture oxygen from the air at low tide. The red mangrove (Rhizophora mangle) of the western Atlantic captures oxygen using prop roots, which are not actually roots at all, but a ring of branches that arch out from the trunk and into the sediment (see figures 11.3, 11.19). Both pneumatophores and prop roots are filled with spongy tissue through which oxygen passes easily, and they bear abundant lenticels that open when exposed at low tide to draw oxygen rapidly into the hypoxic tissue.

Chapter 11 Estuaries and Coastal Seas

­ ittle mangrove production is grazed while alive, averaging less than 5% (Bouillon et al. 2008), L and much of the production enters the detrital food web. Nevertheless, herbivory varies greatly among species and sites, and mass defoliation has been reported from several sites around the world, sometimes lasting for several years (Cannicci et al. 2008). Leaf-­cutting and wood-­boring insects can have strong effects on mangrove fitness disproportionate to the small amount of biomass removed as a result of feeding on young leaves and buds and causing premature leaf abscission. A particularly impor­tant group in many mangrove forests, especially in the Indo-­Pacific, is crabs, mainly in the families Grapsidae, Sesarmidae, and Ocypodidae (Lee 1998). Crabs feed on mangrove leaves and propagules, recycling large quantities of mangrove production. Their copious feces support a food chain of small invertebrates, with a significant fraction exported in the w ­ ater to suspension-­ feeders, and their bioturbation aerates the soil and may alter biogeochemical cycling. The trunks, branches, and fo­liage of mangrove trees and the submerged prop roots create a complex three-­ dimensional habitat that supports diverse and abundant animals both above and below ­water. Together with their high productivity, the rich input of organic m ­ atter, and the warm quiet w ­ ater, mangrove forests create ideal incubators for marine animals, acting as nursery areas for juvenile fishes and shellfish. The thickets of submerged prop roots often swarm with schools of small planktivorous fishes and the juveniles of fish species that occupy nearby reefs as adults, and the prop roots themselves frequently support diverse communities of sessile invertebrates. The emergent trees are home to insects, birds, reptiles, and mammals. Thus, the key traits of mangrove trees are adaptations to salt and anoxia, and perennial woody habit that creates complex physical structure (Kathiresan and Bingham 2001).

Community organ­ization and key interactions Mangrove forests and salt marshes are both intertidal habitats dominated by flowering plants, suggesting that similar pro­cesses may structure the two community types; namely, competition among plant species overlain on a gradient of abiotic conditions across the intertidal, across which relative competitive ability changes predictably. As discussed in the previous section, the interactions among ­these pro­cesses have been documented in salt marshes by an extensive history of experiments. Community interactions in mangrove forests have been less intensively studied due to the much longer life spans of the dominant plants, the logistical challenges of access, and the tropical distribution of mangroves. Nevertheless, a growing body of research suggests strong similarities in the pro­cesses structuring ­these intertidal communities of flowering plants. Experiments show that competition and facilitation among plant species influence mangrove communities (Guo et al. 2013) as they do in salt marshes. In the Gulf of Mexico, the black mangrove (Avicennia germinans) is spreading poleward during milder winters as the climate warms. Experiments that transplanted mangrove seedlings into marsh vegetation and removed marsh plants around mangroves across four sites showed that marsh vegetation facilitated mangrove seedlings at the sites near their poleward distribution limit, but depressed mangrove growth at lower, more favorable latitudes. This result follows a now familiar pattern that facilitation among species is more pronounced in more stressful environments. Thus, the expansion of black mangroves with climate warming is mediated not only by physiology but by interactions among species (Guo et al. 2013). Also as in salt marshes, experiments reveal that the historical focus on bottom-up control by resource availability and physiological stress is incomplete, and that grazing and bioturbation by animals, particularly the typically abundant crabs, can alter both community composition and biogeochemistry of mangrove systems. Some crabs feed preferentially, and heavi­ly in some places, on the propagules of certain species of mangroves, which can influence mangrove community structure (Smith 1987), although the community impacts of crab feeding are difficult to generalize (Cannicci et al. 2008).

333

334

Ocean Ecology

Ecosystem pro­cesses and ser­vices Like other coastal vegetation, the dominant mangrove trees drive pro­cesses within t­ hese ecosystems, including primary production, habitat for fished species and other animals, carbon storage, and biogeochemical cycles. The first quantitative studies of mangrove ecosystems emphasized the vigorous primary production of the trees and the potential export of much production into detrital food webs (Odum and Heald 1975). Subsequent studies generated a long debate about the degree to which mangrove forests are sources of organic carbon, via outwelling of mangrove production to adjacent ­waters, versus carbon sinks, via burial in the thick accumulations of organic soil beneath the trees (Lee 1995). Early compilations concluded that roughly 10% of mangrove production is buried and 40% is exported, with much of the remainder respired within the system. But a synthesis of data available in 2008 concluded that more than half the carbon fixed by mangrove vegetation could not be accounted for in bud­gets, prob­ably owing to large underestimates of in situ mineralization, and suggested that most export from mangroves to adjacent ­waters is in the form of dissolved inorganic carbon (Bouillon et al. 2008). A second key theme in mangrove ecosystem research has been the support of fish production. Several studies have shown that interactions between adjacent mangrove and coral reef habitats are impor­tant for boosting tropical marine fish production, discussed ­later in the chapter. But recent studies at sites around the world suggest that production by mangroves and associated algae provides relatively l­ittle of the nutrition of fishes. Stable carbon and nitrogen isotope data showed that mangrove production provides 12%–72% of fish diet among sites and that about half the studied fish species obtain  1000 Mg carbon per hectare in wood and soil, most of it (49%–98%) in the sediment.

Mangrove forests in the Anthropocene Mangrove ecosystems are threatened. The global area of mangrove forests has declined by 30%–50% since the mid-­twentieth ­century in the face of coastal development, aquaculture expansion, and harvesting (Donato et al. 2011). Much of this deforestation is driven by coastal land clearance for shrimp and other forms of aquaculture (Richards and Friess 2016). But the aquaculture boom has had perverse consequences for local communities. Much aquaculture production is exported to other countries and most of the wealth it generates is privately held, whereas in much of the developing world the mangroves provide local communities with wood and other resources, support of coastal fisheries, and shoreline protection, which are mostly public benefits. When t­ hese benefits and the externalities caused by mangrove destruction and ­water pollution by aquaculture are included in the equation, shrimp farming becomes less attractive than mangrove conservation (Sathirathai and Barbier 2001). Mangrove loss also has global consequences. Many mangrove forests contain large quantities of organic carbon in biomass and soils. The combination of high carbon content of mangrove stands

Chapter 11 Estuaries and Coastal Seas

and the rapid rate of clearance means that mangroves may account for up to 10% of carbon emissions from deforestation globally, despite making up  50% cover) on only 4% of surveyed Ca­rib­bean reefs and only1% of Indo-­ Pacific reefs (figure 12.11). In summary, available data appear to f­ avor the phase shift model rather than alternative stable states on coral reefs. Generally, corals appear to dominate environments where disturbances are primarily natu­ral, such as hurricanes and rare bleaching events. But where ­human impacts prevail, the coral-­dominated community loses resilience and shifts to dominance by

Ocean Ecology

Macroalgal cover (%)

100

75

50

25

0

0

25

50

75

0

50

100

150

100 0 Coral cover (%)

25

50

75

100

100

200

300

400

>5 5 Phase shift index

366

4 3 2 1 0 –1 –2  2 mm, including larger copepods, krill, and some gelatinous animals. Marine spatial planning: A spatially explicit approach to integrated management of the multiple, often conflicting, ­human activities in the sea. Mass balance: A form of quantitative accounting for the fluxes and transformations of ­matter and energy within an individual organism, population, or ecosystem. Maximum sustainable yield (MSY): The maximum quantity of a harvested stock (e.g., fish species) that can be regularly removed from the population while maintaining a stable population; fishery science generally assumes that this occurs at a standing stock biomass roughly half that of the unharvested stock biomass. Meiofauna: The microscopic organisms that live in the water-­filled, interstitial pore spaces among individual sediment grains. Meroplankton: The temporarily planktonic larvae and/or young of larger organisms that live as benthos or nekton as adults. Mesograzers: Herbivorous animals of a size intermediate between microscopic (meiofauana) and typical macrofauna, which use macrophytes as habitat and sometimes also as food. Mesopelagic zone: The subsurface volume of the open-­ocean ­water column in which light is detectable by some animals but insufficient for photosynthesis, between 200 and 1000 m, depending on ­water clarity. Mesoplankton: Planktonic organisms with body size in the range of 0.2–2.0 mm. Metabolic theory of ecol­ogy: The science that unites phenomena across scales of organ­ization ranging from molecules to ecosystems, based on fundamental pro­cesses of metabolism. Metabolism: The biochemical reactions that pro­cess energy and materials by which living organisms sustain themselves. Metacommunity: A network of communities connected via dispersal.

388

Ocean Ecology

Metagenomics: The characterization of ge­ne­tic material from environmental samples rather than from individual organisms. Metapopulation: A group of spatially separated populations connected by dispersal. Microbial loop: The portion of the marine food web composed of microscopic organisms (mostly prokaryotes and protists) that produce, graze, and recycle organic ­matter among themselves with ­little of the energy or materials flowing to macroscopic organisms. Mid-­ocean ridges: The seismically active submarine ridges that mark the junctions of diverging tectonic plates; the mid-­ocean ridges are sites of regular tectonic activity, including seafloor spreading that results in upwelling of magma and outflux of mineral-­rich ­water through hydrothermal vents, which in turn host unique biological communities supported by chemotrophy. Mixotroph: An organism that is capable of both autotrophy and heterotrophy. Modules: Small groups of species whose dynamics are linked closely by direct and indirect interactions. Monophyletic group: The set of all species descended from a common ancestor. Mortality schedule: A population’s age-­specific pattern of survivorship. Multifunctionality: In ecol­ogy, the ability of an ecosystem to si­mul­ta­neously provide multiple functions and ser­vices. N Nanoplankton: Pelagic organisms in the size range of 2–20 μm, including phototrophic and heterotrophic eukaryotes. Natu­ral se­lection: Differential survival and/or reproduction of variants (usually individual organisms) within a population that results in changed gene frequencies within the population. Net plankton: Planktonic organisms within the size range captured by conventional plankton nets, usually with a mesh size of roughly 0.75 mm. Net primary production: The total quantity of chemical energy (biomass) produced from inorganic precursors, ­after the chemical energy used in respiration is accounted for; net primary production is gross primary production minus respiration. Neutral theory: A body of theory in ecol­ogy that aims to explain the species composition and diversity of communities as a result of stochastic, or nondeterministic, pro­cesses, such as random dispersal and ecological drift; neutral theory is often used to provide a null model for evaluating the importance of ecological se­lection pro­cesses (habitat filtering, competition, predation, ­etc.). New production: Primary production fueled by inputs of nitrogen (typically in the form of NO3) into a system, as opposed to regenerated production.

Niche: The set of conditions in which a species can sustain a stable or growing population. Niche construction: A pro­cess in which an organism modifies its conditions of life (i.e., its niche) through its own be­ hav­iors and activities, influencing its own selective regime and thus generating feedback on its own evolution. Niche differentiation: The pro­cess by which competing species use the environment in dif­fer­ent ways, reducing competition and fostering their coexistence. Novel ecosystems: Configurations of species and abiotic conditions that have not historically occurred together and thus have l­ ittle evolutionary history of interaction; usually applied to ecosystems that emerge in areas strongly affected by ­human activities. Null model: In ecol­ogy, a theoretical model that aims to evaluate the importance of a pro­cess (e.g., competition) by omitting it from the model, thus defining the expectation in the absence of the pro­cess, which can provide a baseline against which to compare patterns in empirical data. O Ocean acidification: The decline in seawater pH as a result of diffusion into the ocean of anthropogenic CO2 derived from combustion of fossil fuels. Oligotrophic: Characterized by low concentration or availability of dissolved inorganic nutrients, and supporting low biomass and productivity of autotrophs. In oceanography, usually referring to “blue ­water” regions of the open ocean far from the continents. Operational taxonomic units (OTUs): A generic term to indicate a distinct species-­level taxon, usually detected with molecular ge­ne­tic data, and not yet identified. Overfishing: In fisheries science, overfishing refers to harvesting a fish stock (population) at a rate above that producing maximum sustainable yield. Overfishing debt: The estimated loss of revenue, employment, food security, or other benefit resulting from overfishing. P Parapatric speciation: Origin of a new species by divergence of populations along an ecological transition while the populations remain physically and demographically connected. Paraphyletic group: A set of species descended from a common ancestor, but which does not include all descendants of that ancestor, thus considered an unnatural group in phyloge­ne­tics. Parsimony: The princi­ple that a simpler explanation for an empirical pattern is preferred over a more complex explanation; in phyloge­ne­tics specifically, the criterion that the preferred phylogeny reconstruction among alternative possibilities is the one that requires the fewest evolutionary changes.

Glossary

Pelagic-­benthic coupling: The influence of surface water-­ column pro­cesses on the under­lying benthos, generally as a result of sedimentation of surface production to the bottom. Phages: Viruses that infect bacteria; short for bacteriophage. Phase shift: In ecol­ogy, often used to refer to a transition (also known as regime shift) between alternative stable states of a system, but phase shift is sometimes distinguished from ­these as a change from one condition to another (e.g., from coral to algal dominance) in a system with a single equilibrium state, forced by a per­sis­tent change in environmental conditions. Phenological mismatch: Divergence and reduced overlap in seasonal dynamics of a consumer and its prey as a result of their differential responses to changing conditions, typically environmental temperature. Phenology: The timing of an organism’s life cycle events in relation to seasonal variation in climate and environmental conditions. Phenotype: The totality of an organism’s observable characteristics, including morphology, physiology, and be­hav­ior, resulting from the interaction of its genotype and environment. Phenotypic plasticity: The range in phenotypes that can be realized by a single genotype ­under dif­fer­ent environmental conditions. Photosynthetically active radiation (PAR): The spectrum of vis­i­ble electromagnetic radiation that can be used in photosynthesis, from approximately 300–700 nm. Phyloge­ne­tic systematics: The science of reconstructing the evolutionary history and relationships among organisms based on their shared traits, which may include gene sequences and any phenotypic characteristic. Phylogeny: The evolutionary history of a lineage, species, or par­tic­u­lar feature (trait) of an organism, specifically the lines of descent and relationships among them. Phytoplankton: Microscopic photosynthetic organisms, including a diverse range of cyanobacteria and eukaryotic algae, that live suspended in the ­waters of the photic zone and form the base of the ocean food web. Picoplankton: Planktonic organisms in the size range of 0.2–2.0 μm, mainly prokaryotes. Planktotrophy: Life history in which eggs are small and provisioned with ­little energy (yolk), such that embryos hatch as plankton and must obtain additional energy by feeding in the plankton for some time before metamorphosing into the benthic or nektonic juvenile form. Polygenic: A trait controlled by multiple genes, thus presenting a range of phenotypes that varies more or less continuously rather than discretely as a function of ge­ne­tic variation. Polyphyletic group: A set of species descended from more than one ancestor, and not including all descendants of

389

t­ hose ancestors, thus considered an unnatural group in phyloge­ne­tics. Power function: A mathematical function of the form f(x) = αxp; resulting in a linear relationship on a log-­log scale. Primary production: The synthesis of living organic ­matter from inorganic compounds using a source of energy, such as photosynthesis using light energy; sometimes used to refer to the quantity of biomass produced thereby. Primary productivity: The rate of primary production. Pycnocline: The strong density discontinuity between upper and lower layers of the ocean ­water column, a result of density stratification. R Reaction norm: The distribution of phenotypes produced by a single genotype across a range of environmental conditions. Realized niche: The subset of conditions (i.e., of the fundamental niche) that a species actually occupies in the field as a result of activities of other species that restrict its distribution. Regenerated production: Primary production fueled by NH4+ and urea excreted by herbivores and derived from the plant tissue they consumed. Regime shift: A sudden, nonlinear, and per­sis­tent change in the structure and function of an ecosystem, usually thought to be due to self-­reinforcing feedback; the term overlaps substantially with the concept of alternative stable states, and has also been used in reference to climate, financial systems, and other complex systems. Regional enrichment: The increase in species richness of a local community resulting from influx of species from the surrounding region; vis­i­ble as an approximately linear relationship between regional and local richness across a sample of widely distributed communities ­under similar environmental conditions. Regional richness: The sum of all species recorded from a region. Regional species pool: The collection of species in a biogeographic region that are potentially available to colonize a local community. Reproductive value (vx): An individual’s expected number of offspring, at a given age, from that point to the end of her life. Rescue effect: The reduced risk of extinction of a declining population resulting from immigration of individuals from elsewhere. Resilience: In ecol­ogy, resilience has taken on two meanings. In the narrow sense, from engineering, resilience means the speed with which some mea­sur­able characteristic recovers to its original state ­after a disturbance. A frequently used (and vaguer) broader meaning is the ability of an ecosystem

390

Ocean Ecology

to maintain or return to some state considered normal or desirable in the face of disturbance. Re­sis­tance: The ability of an entity to remain unchanged in the face of a disturbance or challenge. Resource partitioning: The use of dif­fer­ent kinds or combinations of resources by potentially competing species. Respiration: The pro­cess by which energy-­rich compounds are decomposed to provide energy to power the body’s metabolic pro­cesses; in most organisms this requires oxygen, and oxygen consumption is commonly used as a mea­sure of respiration; respiration may be considered the fundamental rate of metabolism. Response traits: Organismal traits that influence how an organism responds to environmental conditions and changes in the ecosystem. S Sampling phenomenon: The situation in which more species-­rich communities are more productive than the average of single-­species communities, not ­because of complementarity among species in resource use but ­because diverse communities are statistically more likely to contain (sample) a particularly productive species. Saturation: In ecol­ogy, an upper limit on local species richness that results from local interactions and is therefore unrelated to the richness of the surrounding region. Secondary metabolites: Chemical compounds with no known function in primary metabolism that are distasteful or noxious to consumers, most prominent in situations where herbivory or predation is intense, and presumed to have evolved as chemical defenses against enemies. Shifting baseline syndrome: The situation in which observers unconsciously use their first memory of a situation as a baseline for ­future comparison, such that long-­term decline (or improvement) is underestimated as each new generation of observers adopts a lower (or higher) baseline for comparison. Size spectrum: The distribution of biomass within a multispecies community divided into logarithmically equal body size intervals. Source-­sink dynamics: The changes in population abundance of a species distributed across two or more patches of habitat, in at least some of which (sinks) populations would decline on their own, but they persist ­because of immigration from populations in more favorable patches (sources) that produce surplus individuals that emigrate. Speciation: The evolutionary origin of new species as a result of divergence among populations of an ancestral species. Species-­area relation (SAR): The increase in the number of species recorded (S) as the cumulative area (A) of sampled habitat increases. Species density: The number of species within a plot of a given size, regardless of total abundance.

Species-­energy hypothesis: The hypothesis that species richness is higher in habitats that receive more of the energy needed to power metabolism. Species pool: The set of species occurring in a certain region that could potentially colonize and inhabit a local community. Species richness: The number of species in a sample. Species turnover: A quantitative estimate of the difference in species composition among samples, based on some formulation of the number of species not found in both samples. Specific production: Ratio of production per unit biomass. Spreading centers: Areas where tectonic plates diverge and new ocean crust is produced, typically associated with mid-­ ocean ridges. Stabilizing mechanisms: In ecol­ogy, traits or pro­cesses that reduce interspecific competition by differentiation in niche dimensions, such as use of resources, which ­favors stable coexistence; contrasted with equalizing mechanisms. Stabilizing se­lection: Natu­ral se­lection that ­favors the mean phenotype in a population, thus decreasing abundance of extreme variants and reinforcing the mean. Stable age distribution: An age distribution of individuals within a population that remains stable across generations. Stenothermal: Having a narrow range of temperature tolerance. Stoichiometry: In ecol­ogy, the quantitative relationships among ele­ments in tissues of dif­fer­ent organisms and the environment, which influence the nature and rates of ecological pro­cesses. Stratification: Separation of layers in a fluid as a result of differences in density; in the ocean, the major stratification is between the sun-­warmed upper (epipelagic) layer and the colder deep ocean, which are separated by a strong density discontinuity (pycnocline). Stromatolites: Dome-­or mushroom-­shaped structures formed in shallow ­water by accretion of layered mineral and microbial materials; dating from Precambrian times, ­these ­were the first biogenic reefs. Succession: The characteristic sequence of changing species presence and dominance that occurs through time on habitat newly created or opened by a disturbance. Suspension-­feeding: Feeding by capturing and ingesting small organisms and particulate ­matter that is suspended in the ­water column, usually involving some sort of filtering structure and/or creation of a current that draws in food-­ laden ­water. Sympatric speciation: Origin of a new species in the absence of physical barriers to interbreeding by divergence of subpopulations that differ in habitat or other resource use, mating be­hav­ior, or some combination thereof. Synapomorphies: In phyloge­ne­tics, a character (trait, gene sequence, or any observable aspect of an organism) that is derived, i.e., evolved within the phyloge­ne­tic tree, and

Glossary

391

shared among a (hypothesized) ancestor and all of its descendant lineages, indicating common ancestry and placing them together in a monophyletic clade. Syngameons: Complexes of partially interfertile species that regularly exchange genes; thought to be common among reef corals.

Trophic skew: Flattening of a community’s typical pyramid of numbers or biomass as a result of ­human pressure on top predators. Tropicalization: Increased repre­sen­ta­tion of tropical species, and their associated traits, in temperate communities as a result of a warming climate.

T Theory of island biogeography: A body of theory, developed by MacArthur and Wilson (1963, 1967) that aims to predict the species richness of island communities as a function of island (i.e., habitat patch) size and distance from the mainland (source community), as they influence immigration and extinction probabilities; an early development of neutral theory in ecol­ogy. Thermohaline circulation: The global movement of ­water through the deep ocean, beginning with sinking of cold saline surface ­water during polar winters, and its slow flow through the deep ocean basins u­ ntil upwelling again thousands of years l­ater; colloquially called the “ocean con­vey­or ­belt.” Trade winds: The per­sis­tent winds that prevail at midlatitudes, blowing ­toward the low-­pressure zones near the equator from the northeast in the Northern Hemi­sphere and the southeast in the Southern Hemi­sphere. Trait: Any phenotypic characteristic of an organism. Trait-­mediated effect: An interaction between species mediated by flexible species trait(s), such as be­hav­ior; often used in ecol­ogy to refer to nonconsumptive effects of predators on their prey, such as when predator presence reduces herbivore activity and feeding, boosting accumulation of plant biomass. Trophic cascade: The indirect influence of a consumer on organisms two or more levels below it in the food web, mediated through an interaction chain, the most common example being increase of plant biomass by a predator that feeds on herbivores, reducing their controlling impact on plants.

U Unified neutral theory of biodiversity and biogeography: A theory predicting the diversity and relative abundances of species in ecological communities (strictly speaking, competitors within a trophic level) as a function of habitat area and rates of speciation, dispersal, and extinction; the theory is neutral in the sense that species are considered equivalent in per capita rates of birth, death, dispersal, and speciation (Hubbell 2001); an example of a null model in ecol­ogy. Upwelling: Upward flux of cold subsurface ­water, usually rich in nutrients, ­toward the surface, typically a result of wind-­induced Ekman transport of surface ­water. V Vertical migration: Active movement of animals between shallow and deeper depths in the ­water column, typically occurring on a diurnal cycle, allowing animals to exploit rich food near the surface at night while evading visual predators by moving to deeper ­water during the day. W Western boundary current: A strong longshore current that develops along the western edge of an ocean basin as geostrophic flow pushes ­water against the continent on the western side of the basin; the ­water then flows poleward in a concentrated current parallel to the coast before turning again to the east at high latitudes; examples include the Gulf Stream, Kuroshio, and Agulhas currents.

Lit­er­a­ture Cited Abram, N. J. 2003. “Coral Reef Death during the 1997 Indian Ocean Dipole Linked to Indonesian Wildfires,” Science 301:952–955. Aburto-­Oropeza, O., E. Ezcurra, G. Danemann, V. Valdez, J. Murray, and E. Sala. 2008. “Mangroves in the Gulf of California Increase Fishery Yields,” Proceedings of the National Acad­emy of Sciences 105:10456–10459. Acosta, C. A. 1999. “Benthic Dispersal of Ca­rib­bean Spiny Lobsters among Insular Habitats: Implications for the Conservation of Exploited Marine Species,” Conservation Biology 13:603–612. Acosta, C. A., and M. J. Butler IV. 1999. “Adaptive Strategies That Reduce Predation on Ca­rib­bean Spiny Lobster Postlarvae during Onshore Transport,” Limnology and Oceanography 44:494–501. Adam, T. C., D. E. Burkepile, and B. I. Ruttenberg. 2015. “Herbivory and the Resilience of Ca­rib­bean Coral Reefs: Knowledge Gaps and Implications for Management,” Marine Ecol­ogy Pro­gress Series 520:1–20. Addy, C. E., and D. A. Aylward. 1944. “Status of Eelgrass in Mas­sa­chu­setts during 1943,” Journal of Wildlife Management 8:269–275. Adey, W. H. 1978. “Coral Reef Morphogenesis: A Multidimensional Model,” Science 202:831–837. Aebischer, N., J. Coulson, and J. Colebrook. 1990. “Parallel Long-­Term Trends across Four Marine Trophic Levels and Weather,” Nature 347:753–755. Agnew, D. J., J. Pearce, G. Pramod, T. Peatman, R. Watson, J. R. Beddington, T. J. Pitcher, and S. A. Sandin. 2009. “Estimating the Worldwide Extent of Illegal Fishing,” PLoS ONE 4:e4570. Airoldi, L., X. Turon, S. Perkol-­Finkel, and M. Rius. 2015. “Corridors for Aliens but Not for Natives: Effects of Marine Urban Sprawl at a Regional Scale,” Diversity and Distributions 21:755–768. Alberti, M., C. Correa, J. M. Marzluff, A. P. Hendry, E. P. Palkovacs, K. M. Gotanda, V. M. Hunt, T. M. Apgar, and Y. Zhou. 2017. “Global Urban Signatures of Phenotypic Change in Animal and Plant Populations,” Proceedings of the National Acad­emy of Sciences 18:201606034. Albouy, C., F. Lasram, L. Velez, and F. Guilhaumon. 2015. “FishMed: Traits, Phylogeny, Current and Projected Species Distribution of Mediterranean Fishes, and Environmental Data: Ecological Archives E096–203,” Ecol­ogy. Albright, R., Y. Takeshita, D. A. Koweek, A. Ninokawa, K. Wolfe, T. Rivlin, Y. Nebuchina, J. Young, and K. Caldeira. 2018. “Carbon Dioxide Addition to Coral Reef ­Waters Suppresses Net Community Calcification,” Nature 555:516–519. Allen, A., J. Brown, and J. Gillooly. 2002. “Global Biodiversity, Biochemical Kinetics, and the Energetic-­Equivalence Rule,” Science 297:1545–1548. Allen, A. P., J. F. Gillooly, V. M. Savage, and J. H. Brown. 2006. “Kinetic Effects of Temperature on Rates of Ge­ne­tic Divergence and Speciation,” Proceedings of the National Acad­emy of Sciences 103:9130–9135. Aller, R. C. 1988. “Benthic Fauna and Biogeochemical Pro­cesses in Marine Sediments: The Role of Burrow Structures.” Pages 301–338 in T. H. Blackburn and J. Sorensen, editors, Nitrogen Cycling in Coastal Marine Environments. Chichester, UK: John Wiley & Sons. Alonso Aller, E., M. Gullström, F. K. J. Eveleens Maarse, M. Gren, L. Nordlund, N. Jiddawi, and J. S. Eklöf. 2014. “Single and Joint Ef-

fects of Regional-­and Local-­Scale Variables on Tropical Seagrass Fish Assemblages,” Marine Biology 161:2395–2405. Alpine, A. E., and J. E. Cloern. 1992. “Trophic Interactions and Direct Physical Effects Control Phytoplankton Biomass and Production in an Estuary,” Limnology and Oceanography 37:946–955. Alter, S., E. Rynes, and S. Palumbi. 2007. “DNA Evidence for Historic Population Size and Past Ecosystem Impacts of Gray Whales,” Proceedings of the National Acad­emy of Sciences 104:15162–15167. Altieri, A. H., M. D. Bertness, T. C. Coverdale, N. C. Herrmann, and C. Angelini. 2012. “A Trophic Cascade Triggers Collapse of a Salt-­ Marsh Ecosystem with Intensive Recreational Fishing,” Ecol­ ogy 93:1402–1410. Altieri, A. H., and K. B. Gedan. 2014. “Climate Change and Dead Zones,” Global Change Biology 21:1395–1406. Altieri, A. H., S. B. Harrison, J. Seemann, R. Collin, R. J. Diaz, and N. Knowlton. 2017. “Tropical Dead Zones and Mass Mortalities on Coral Reefs,” Proceedings of the National Acad­emy of Sciences 114:3660–3665. Altizer, S., R. S. Ostfeld, P. T. J. Johnson, S. Kutz, and C. D. Harvell. 2013. “Climate Change and Infectious Diseases: From Evidence to a Predictive Framework,” Science 341:514–519. Alvarez-­Filip, L., N. K. Dulvy, J. A. Gill, I. M. Cote, and A. R. Watkinson. 2009. “Flattening of Ca­rib­bean Coral Reefs: Region-­Wide Declines in Architectural Complexity,” Proceedings of the Royal Society B-­Biological Sciences 276:3019–3025. Álvarez-­Noriega, M., A. H. Baird, M. Dornelas, J. S. Madin, V. R. Cumbo, and S. R. Connolly. 2016. “Fecundity and the Demographic Strategies of Coral Morphologies,” Ecol­ogy 97:3485–3493. Amin, S. A., M. S. Parker, and E. V. Armbrust. 2012. “Interactions between Diatoms and Bacteria,” Microbiol. Mol. Biol. Rev. 76:667–684. Andersen, K. H., T. Berge, R. J. Gonçalves, M. Hartvig, J. Heuschele, S. Hylander, N. S. Jacobsen, et al. 2016. “Characteristic Sizes of Life in the Oceans, from Bacteria to Whales,” Annual Review of Marine Science 8:217–241. Anderson, D. M., A. D. Cembella, and G. M. Hallegraeff. 2012. “Pro­gress in Understanding Harmful Algal Blooms: Paradigm Shifts and New Technologies for Research, Monitoring, and Management,” Annual Review of Marine Science 4:143–176. Anderson, D. M., P. M. Glibert, and J. M. Burkholder. 2002. “Harmful Algal Blooms and Eutrophication: Nutrient Sources, Composition, and Consequences,” Estuaries 25:704–726. Andersson, J. O. 2005. “Lateral Gene Transfer in Eukaryotes,” Cellular and Molecular Life Sciences 62:1182–1197. Andrews, K. R., W. F. Perrin, M. Oremus, L. Karczmarski, B. W. Bowen, J. B. Puritz, and R. J. Toonen. 2013. “The Evolving Male: Spinner Dolphin (Stenella longirostris) Ecotypes Are Divergent at Y Chromosome but Not mtDNA or Autosomal Markers,” Molecular Ecol­ogy 22:2408–2423. Angelini, C., J. N. Griffin, J. Van De Koppel, L. P. M. Lamers, A. J. P. Smolders, M. Derksen-­Hooijberg, T. van der Heide, and B. R. Silliman. 2016. “A Keystone Mutualism Underpins Resilience of a Coastal Ecosystem to Drought,” Nature Communications 7:12473. Angilletta, M. J., T. D. Steury, and M. W. Sears. 2004. “Temperature, Growth Rate, and Body Size in Ectotherms: Fitting Pieces of a

394

Ocean Ecology

Life-­History Puzzle,” Integrative and Comparative Biology 44:498–509. Anthony, R., J. Estes, M. Ricca, A. Miles, and E. Forsman. 2008. “Bald Ea­ gles and Sea Otters in the Aleutian Archipelago: Indirect Effects of Trophic Cascades,” Ecol­ogy 89:2725–2735. Antoine, D. 2005. “Bridging Ocean Color Observations of the 1980s and 2000s in Search of Long-­Term Trends,” Journal of Geophysical Research 110:1029. Appeltans, W., S. T. Ahyong, G. Anderson, M. V. Angel, T. Artois, N. Bailly, R. Bamber, et al. 2012. “The Magnitude of Global Marine Species Diversity,” Current Biology 2: 2189–2202. Arkema, K. K., G. Guannel, G. Verutes, S. A. Wood, A. Guerry, M. Ruckelshaus, P. Kareiva, M. Lacayo, and J. M. Silver. 2013. “Coastal Habitats Shield ­People and Property from Sea-­Level Rise and Storms,” Nature Climate Change 3:913–918. Armitage, D. W., K. L. Gallagher, N. D. Youngblut, D. H. Buckley, and S. H. Zinder. 2012. “Millimeter-­Scale Patterns of Phyloge­ne­tic and Trait Diversity in a Salt Marsh Microbial Mat,” Frontiers in Microbiology 3:293. Arnaud-­Haond, S., C. M. Duarte, E. Díaz-­Almela, N. Marbà, T. Sintes, and E. A. Serrão. 2012. “Implications of Extreme Life Span in Clonal Organisms: Millenary Clones in Meadows of the Threatened Seagrass Posidonia Oceanica,” PLoS ONE 7:e30454. Aronson, R. B., J. F. Bruno, W. F. Precht, P. W. Glynn, C. D. Harvell, L. Kaufman, C. S. Rogers, E. A. Shinn, and J. F. Valentine. 2003. “­Causes of Coral Reef Degradation,” Science 302:1502–1504. Aronson, R. B., I. G. Macintyre, W. F. Precht, T. J. T. Murdoch, and C. M. Wapnick. 2002. “The Expanding Scale of Species Turnover Events on Coral Reefs in Belize,” Ecological Monographs 72:233–249. Aronson, R. B., and W. F. Precht. 2001. “White-­Band Disease and the Changing Face of Ca­rib­bean Coral Reefs,” Hydrobiologia 460:25–38. Aronson, R. B., W. Precht, M. Toscano, and K. H. Koltes. 2002. “The 1998 Bleaching Event and Its Aftermath on a Coral Reef in Belize,” Marine Biology 141:435–447. Aronson, R. B., K. E. Smith, S. C. Vos, J. B. McClintock, M. O. Amsler, P.-­O. Moksnes, D. S. Ellis, et al. 2015. “No Barrier to Emergence of Bathyal King Crabs on the Antarctic Shelf,” Proceedings of the National Acad­emy of Sciences 112:12997–13002. Arrhenius, O. 1921. “Species and Area,” Journal of Ecol­ogy 9:95–99. Arrigo, K. 2005. “Marine Microorganisms and Global Nutrient Cycles,” Nature 437:349–355. Arrigo, K., D. Robinson, D. Worthen, R. Dunbar, G. DiTullio, M. VanWoert, and M. Lizotte. 1999. “Phytoplankton Community Structure and the Drawdown of Nutrients and CO2 in the Southern Ocean,” Science 283:365–367. Atkins, R. L., J. N. Griffin, C. Angelini, M. I. O’Connor, and B. R. Silliman. 2015. “Consumer-­Plant Interaction Strength: Importance of Body Size, Density and Metabolic Biomass,” Oikos 124:1274–1281. Atwood, T. B., R. M. Connolly, E. G. Ritchie, C. E. Lovelock, M. R. Heithaus, G. C. Hays, J. W. Fourqurean, and P. I. Macreadie. 2015. “Predators Help Protect Carbon Stocks in Blue Carbon Ecosystems,” Nature Climate Change 5:1038–1045. Audzijonyte, A., E. Fulton, M. Haddon, F. Helidoniotis, A. J. Hobday, A. Kuparinen, J. Morrongiello, A. D. Smith, J. Upston, and R. S. Waples. 2016. “Trends and Management Implications of Human-­Influenced Life-­History Changes in Marine Ectotherms,” Fish and Fisheries 17:1005–1028.

Azam, F., T. Fenchel, J. Field, J. Gray, L. Meyer-­Reil, and F. Thingstad. 1983. “The Ecological Role of Water-­Column Microbes in the Sea,” Marine Ecol­ogy Pro­gress Series, Oldendorf 10:257–263. Baden, S., C. Boström, S. Tobiasson, H. Arponen, and P.-­O. Moksnes. 2010. “Relative Importance of Trophic Interactions and Nutrient Enrichment in Seagrass Ecosystems: A Broad-­Scale Field Experiment in the Baltic-­Skagerrak Area,” Limnology and Oceanography 55:1435. Baden, S., A. Emanuelsson, L. Pihl, C. Svensson, and P. Åberg. 2012. “Shift in Seagrass Food Web Structure over De­cades Is Linked to Overfishing,” Marine Ecol­ogy Pro­gress Series 451:61–73. Baird, D., and R. E. Ulanowicz. 1989. “The Seasonal Dynamics of the Chesapeake Bay Ecosystem,” Ecological Monographs, 59:329–364. Baker, S. M. 1909. “On the ­Causes of the Zoning of Brown Seaweeds on the Seashore,” New Phytologist 8:196–202. Bakun, A. 2006. “Wasp-­Waist Populations and Marine Ecosystem Dynamics: Navigating the ‘Predator Pit’ Topographies,” Pro­gress in Oceanography 68:271–288. Balmford, A., P. Gravestock, N. Hockley, C. J. McClean, and C. M. Roberts. 2004. “The Worldwide Costs of Marine Protected Areas,” Proceedings of the National Acad­e my of Sciences 101: 9694–9697. Balvanera, P., C. Kremen, and M. Martínez-­R amos. 2005. “Applying Community Structure Analy­sis to Ecosystem Function: Examples from Pollination and Carbon Storage,” Ecological Applications 15:360–375. Balvanera, P., A. B. Pfisterer, N. Buchmann, J.-­S. He, T. Nakashizuka, D. Raffaelli, and B. Schmid. 2006. “Quantifying the Evidence for Biodiversity Effects on Ecosystem Functioning and Ser­vices,” Ecol­ogy Letters 9:1146–1156. Ban, S. S., N. A. J. Graham, and S. R. Connolly. 2014. “Evidence for Multiple Stressor Interactions and Effects on Coral Reefs,” Global Change Biology 20:681–697. Barber, P. H., S. R. Palumbi, M. V. Erdmann, and M. K. Moosa. 2000. “A Marine Wallace’s Line?” Nature 406:692–693. Barkai, A., and C. McQuaid. 1988. “Predator-­Prey Role Reversal in a Marine Benthic Ecosystem,” Science 242:62. Barnes, C., D. Maxwell, D. C. Reuman, and S. Jennings. 2010. “Global Patterns in Predator-­Prey Size Relationships Reveal Size De­pen­ dency of Trophic Transfer Efficiency,” Ecol­ogy 91:222–232. Barnes, R. D. 1980. Invertebrate Zoology. Philadelphia: Saunders College. Barnosky, A. D. 2008. “Megafauna Biomass Tradeoff as a Driver of Quaternary and ­Future Extinctions,” Proceedings of the National Acad­emy of Sciences 105(Suppl 1):11543–11548. Barnosky, A. D., N. Matzke, S. Tomiya, G. O. U. Wogan, B. Swartz, T. B. Quental, C. Marshall, et al. 2011. “Has the Earth’s Sixth Mass Extinction Already Arrived?” Nature 470:51–57. Bar-­On, Y. M., R. Phillips, and R. Milo. 2018. “The Biomass Distribution on Earth,” Proceedings of the National Acad­emy of Sciences 115:6506–6511. Baross, J. A., and J. W. Deming. 1983. “Growth of ‘Black Smoker’ Bacteria at Temperatures of at Least 250°C,” Nature 303:423–426. Barton, A. D., A. J. Pershing, E. Litchman, N. R. Rec­ord, K. F. Edwards, Z. V. Finkel, T. Kiørboe, and B. A. Ward. 2013. “The Biogeography of Marine Plankton Traits,” Ecol­ogy Letters 16:522–534. Bascompte, J., C. J. Melián, and E. Sala. 2005. “Interaction Strength Combinations and the Overfishing of a Marine Food Web,” Proceedings of the National Acad­emy of Sciences 102:5443–5447. Bates, A. E., N. S. Barrett, and R. D. Stuart-­Smith. 2014. “Resilience and Signatures of Tropicalization in Protected Reef Fish Communities.” Nature Climate Change 4:62–67.

Literature Cited

Baum, D. A., S. D. Smith, and S. S. S. Donovan. 2005. “Evolution. The Tree-­Thinking Challenge,” Science 310:979–980. Baum, J. K., and R. A. Myers. 2004. “Shifting Baselines and the Decline of Pelagic Sharks in the Gulf of Mexico,” Ecol­ogy Letters 7:135–145. Baum, J. K., R. A. Myers, D. G. Kehler, B. Worm, S. J. Harley, and P. A. Doherty. 2003. “Collapse and Conservation of Shark Populations in the Northwest Atlantic,” Science 299:389–392. Baum, J. K., and B. Worm. 2009. “Cascading Top-­Down Effects of Changing Oceanic Predator Abundances,” Journal of Animal Ecol­ogy 78:699–714. Baynham-­Herd, Z., T. Amano, W. J. Sutherland, and P. F. Donald. 2018. “Governance Explains Variation in National Responses to the Biodiversity Crisis,” Environmental Conservation 49:1–12. Beaugrand, G., M. Edwards, and L. Legendre. 2010. “Marine Biodiversity, Ecosystem Functioning, and Carbon Cycles,” Proceedings of the National Acad­emy of Sciences 107:10120–10124. Beaugrand, G., M. Edwards, V. Raybaud, E. Goberville, and R. R. Kirby. 2015. “­Future Vulnerability of Marine Biodiversity Compared with Con­temporary and Past Changes,” Nature Climate Change 5:695–701. Beaugrand, G., and R. R. Kirby. 2018. “How Do Marine Pelagic Species Respond to Climate Change? Theories and Observations,” Annual Review of Marine Science 10:169–197. Beaugrand, G., P. C. Reid, F. Ibanez, J. A. Lindley, and M. Edwards. 2002. “Reorganization of North Atlantic Marine Copepod Biodiversity and Climate,” Science 296(5573):1692–1694. Behrenfeld, M. J. 2010. “Abandoning Sverdrup’s Critical Depth Hypothesis on Phytoplankton Blooms,” Ecol­ogy 91:977–989. Behrenfeld, M. J., R. T. O’Malley, D. A. Siegel, C. R. McClain, J. L. Sarmiento, G. C. Feldman, A. J. Milligan, P. G. Falkowski, R. M. Letelier, and E. S. Boss. 2006. “Climate-­Driven Trends in Con­temporary Ocean Productivity,” Nature 444:752–755. Beja-­Pereira, A., G. Luikart, P. R. ­England, D. G. Bradley, O. C. Jann, G. Bertorelle, A. T. Chamberlain, et al. 2003. “Gene-­Culture Coevolution between ­Cattle Milk Protein Genes and ­Human Lactase Genes,” Nature Ge­ne­tics 35:311–313. Belanger, C. L., D. Jablonski, K. Roy, S. K. Berke, A. Z. Krug, and J. W. Valentine. 2012. “Global Environmental Predictors of Benthic Marine Biogeographic Structure,” Proceedings of the National Acad­emy of Sciences 109:14046–14051. Belgrano, A., A. P. Allen, B. J. Enquist, and J. F. Gillooly. 2002. “Allometric Scaling of Maximum Population Density: A Common Rule for Marine Phytoplankton and Terrestrial Plants,” Ecol­ogy Letters 5:611–613. Bellwood, D. R., C. H. R. Goatley, and O. Bellwood. 2017. “The Evolution of Fishes and Corals on Reefs: Form, Function and Interdependence,” Biological Reviews 92:878–901. Bellwood, D. R., T. P. Hughes, C. Folke, and M. Nyström. 2004. “Confronting the Coral Reef Crisis,” Nature 429:827–833. Bellwood, D. R., T. P. Hughes, and A. S. Hoey. 2006. “Sleeping Functional Group Drives Coral-­Reef Recovery,” Current Biology 16:2434–2439. Belmaker, J., and W. Jetz. 2015. “Relative Roles of Ecological and Energetic Constraints, Diversification Rates and Region History on Global Species Richness Gradients,” Ecol­ogy Letters 18:563–571. Benincà, E., J. Huisman, R. Heerkloss, K. D. Jöhnk, P. Branco, E. H. Van Nes, M. Scheffer, and S. P. Ellner. 2008. “Chaos in a Long-­ Term Experiment with a Plankton Community,” Nature 451:822–825. Benoit-­Bird, K. J., and M. A. Mc­Manus. 2012. “Bottom-­Up Regulation of a Pelagic Community through Spatial Aggregations,” Biology Letters 8:813–816.

395

Bergquist, D. C., F. M. Williams, and C. R. Fisher. 2000. “Longevity Rec­ ord for Deep-­Sea Invertebrate,” Nature 403:499–500. Berkes, F., T. P. Hughes, R. S. Steneck, J. A. Wilson, R. Bellwood, B. Crona, C. Folke, et al. 2006. “Globalization, Roving Bandits, and Marine Resources,” Science 312:1472–1473. Berlow, E. L., J. A. Dunne, N. D. Martinez, P. B. Stark, R. J. Williams, and U. Brose. 2009. “­Simple Prediction of Interaction Strengths in Complex Food Webs,” Proceedings of the National Acad­emy of Sciences 106:187–191. Bertness, M. 1984. “Habitat and Community Modification by an Introduced Herbivorous Snail,” Ecol­ogy 65:370–381. Bertness, M., and P. Ewanchuk. 2002. “Latitudinal and Climate-­Driven Variation in the Strength and Nature of Biological Interactions in New ­England Salt Marshes,” Oecologia 132:392–401. Bertness, M., S. Garrity, and S. Levings. 1981. “Predation Pressure and Gastropod Foraging: A Tropical-­Temperate Comparison,” Evolution 35:995–1007. Bertness, M., and S. Hacker. 1994. “Physical Stress and Positive Associations among Marsh Plants,” American Naturalist 144:363–372. Bertness, M., G. Leonard, J. Levine, P. Schmidt, and A. Ingraham. 1999. “Testing the Relative Contribution of Positive and Negative Interactions in Rocky Intertidal Communities,” Ecol­ogy 80:2711–2726. Bertness, M. D., C. P. Brisson, T. C. Coverdale, M. C. Bevil, S. M. Crotty, and E. R. Suglia. 2014. “Experimental Predator Removal ­Causes Rapid Salt Marsh Die-­Off,” Ecol­ogy Letters 17:830–835. Bertness, M. D., J. F. Bruno, B. R. Silliman, and J. J. Stachowicz, eds. 2014. Marine Community Ecol­ogy and Conservation. Sunderland, MA: Sinauer Associates. Bertness, M. D., and R. M. Callaway. 1994. “Positive Interactions in Communities,” Trends in Ecol­ogy & Evolution 9:191–193. Bertness, M. D., and A. M. Ellison. 1987. “Determinants of Pattern in a New ­England Salt Marsh Plant Community,” Ecological Monographs 57:129–147. Bertness, M. D., and B. R. Silliman. 2014. “Salt Marsh Communities.” Pages 251–270 in M. D. Bertness, J. F. Bruno, B. R. Silliman, and J. J. Stachowicz, editors, Marine Community Ecol­ogy and Conservation. Sunderland, MA: Sinauer Associates. Bettencourt, L. M. A., J. Lobo, D. Helbing, C. Kuhnert, and G. B. West. 2007. “Growth, Innovation, Scaling, and the Pace of Life in Cities,” Proceedings of the National Acad­emy of Sciences 104:7301–7306. Billett, D. S. M., R. S. Lampitt, A. L. Rice, and R. F. C. Mantoura. 1983. “Seasonal Sedimentation of Phytoplankton to the Deep-­Sea Benthos,” Nature 302:520–522. Bilton, D. T., J. Paula, and J. D. D. Bishop. 2002. “Dispersal, Ge­ne­tic Differentiation and Speciation in Estuarine Organisms,” Estuarine Coastal and Shelf Science 55:937–952. Birand, A., A. Vose, and S. Gavrilets. 2012. “Patterns of Species Ranges, Speciation, and Extinction,” American Naturalist 179:1–21. Bird, C. E., I. Fernandez-­Silva, D. J. Skillings, and R. J. Toonen. 2012. “Sympatric Speciation in the Post ‘Modern Synthesis’ Era of Evolutionary Biology,” Evolutionary Biology 39:158–180. Birkeland, C. 1997. “Symbiosis, Fisheries and Economic Development on Coral Reefs,” Trends in Ecol­ogy & Evolution 12:364–367. Bishop, M. J., M. Mayer-­Pinto, L. Airoldi, L. B. Firth, R. L. Morris, L. H. L. Loke, S. J. Hawkins, et al. 2017. “Effects of Ocean Sprawl on Ecological Connectivity: Impacts and Solutions,” Journal of Experimental Marine Biology and Ecol­ogy 492:7–30. Blackburn, T. M., P. Cassey, R. P. Duncan, K. L. Evans, and K. J. Gaston. 2004. “Avian Extinction and Mammalian Introductions on Oceanic Islands,” Science 305:1955–1958.

396

Ocean Ecology

Blanchard, J. L., S. Jennings, R. Holmes, J. Harle, G. Merino, J. I. Allen, J. Holt, N. K. Dulvy, and M. Barange. 2012. “Potential Consequences of Climate Change for Primary Production and Fish Production in Large Marine Ecosystems,” Philosophical Transactions of the Royal Society B-­Biological Sciences 367:2979–2989. Block, B. A., I. D. Jonsen, S. J. Jorgensen, A. J. Winship, S. A. Shaffer, S. J. Bograd, E. L. Hazen, et al. 2011. “Tracking Apex Marine Predator Movements in a Dynamic Ocean,” Nature 475:86–90. Blueweiss, L., H. Fox, V. Kudzma, D. Nakashima, and R. Peters. 1978. “Relationships between Body Size and Some Life History Par­ ameters,” Oecologia 37:257–272. Bode, M., L. Bode, and P. R. Armsworth. 2011. “Dif­fer­ent Dispersal Abilities Allow Reef Fish to Coexist,” Proceedings of the National Acad­emy of Sciences 108:16317–16321. Boden, T. A., R. J. Andres, and G. Sun Marland. “Global, Regional, and National Fossil-Fuel CO2 Emissions (1751–2014) (V. 2017),” United States. https://doi.org/10.3334/CDIAC/00001_V2017. https://www.osti.gov/servlets/purl/1389331. Boden, T.A., G. Marland, and R.J. Andres. 2010. “Global, Regional, and National Fossil-­Fuel CO2 Emissions.” Carbon Dioxide Information Analy­sis Center, Oak Ridge National Laboratory, US Department of Energy, Oak Ridge, TN. doi/10.3334/ CDIAC/00001_V2010. Bohonak, A. J. 1999. “Dispersal, Gene Flow, and Population Structure,” Quarterly Review of Biology 74:21–45. Bolser, R., and M. Hay. 1996. “Are Tropical Plants Better Defended? Palatability and Defenses of Temperate vs. Tropical Seaweeds,” Ecol­ogy 77:2269–2286. Bonan, G. B., and S. C. Doney. 2018. “Climate, Ecosystems, and Planetary ­Futures: The Challenge to Predict Life in Earth System Models,” Science 359:eaam8328. Bongaarts, J. 2009. “­Human Population Growth and the Demographic Transition,” Philosophical Transactions of the Royal Society B-Biological Sciences 364:2985–2990. Bonsdorff, E., E. M. Blomqvist, J. Mattila, and A. Norkko. 1997. “Coastal Eutrophication: ­Causes, Consequences and Perspectives in the Archipelago Areas of the Northern Baltic Sea,” Estuarine Coastal and Shelf Science 44:63–72. Borer, E. T., B. S. Halpern, and E. W. Seabloom. 2006. “Asymmetry in Community Regulation: Effects of Predators and Productivity,” Ecol­ogy 87:2813–2820. Borer, E. T., W. S. Harpole, P. B. Adler, E. M. Lind, J. L. Orrock, E. W. Seabloom, and M. D. Smith. 2014. “Finding Generality in Ecol­ ogy: A Model for Globally Distributed Experiments,” Methods in Ecol­ogy and Evolution 5:65–73. Borer, E. T., E. W. Seabloom, J. B. Shurin, K. E. Anderson, C. A. Blanchette, B. Broitman, S. D. Cooper, and B. S. Halpern. 2005. “What Determines the Strength of a Trophic Cascade?” Ecol­ ogy 86:528–537. Boström, C., S. J. Pittman, C. Simenstad, and R. T. Kneib. 2011. “Seascape Ecol­ogy of Coastal Biogenic Habitats: Advances, Gaps, and Challenges,” Marine Ecol­ogy Pro­gress Series, Oldendorf 427:191–217. Boto, L. 2010. “Horizontal Gene Transfer in Evolution: Facts and Challenges,” Proceedings of the Royal Society B-Biological Sciences 277:819–827. Botsford, L. W. 1997. “The Management of Fisheries and Marine Ecosystems,” Science 277:509–515. Bouillon, S., A. V. Borges, E. Castañeda-­Moya, K. Diele, T. Dittmar, N. C. Duke, E. Kristensen, et al. 2008. “Mangrove Production and Carbon Sinks: A Revision of Global Bud­get Estimates,” Global Biogeochemical Cycles 22, doi/10.1029/2007GB003052.

Bowen, B. W., L. A. Rocha, R. J. Toonen, and S. A. Karl. 2013. “The Origins of Tropical Marine Biodiversity,” Trends in Ecol­ogy & Evolution 28:359–366. Boyce, D. G., M. R. Lewis, and B. Worm. 2010. “Global Phytoplankton Decline over the Past ­Century,” Nature 466:591–596. Boyd, P. W., and M. Bressac. 2016. “Developing a Test-­Bed for Robust Research Governance of Geoengineering: The Contribution of Ocean Iron Biogeochemistry,” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 374:20150299. Boyd, P. W., T. Jickells, C. S. Law, S. Blain, E. A. Boyle, K. O. Buesseler, K. H. Coale, et al. 2007. “Mesoscale Iron Enrichment Experiments 1993–2005: Synthesis and ­Future Directions,” Science 315:612–617. Bracken, M. E. S., and N. H. N. Low. 2012. “Realistic Losses of Rare Species Disproportionately Impact Higher Trophic Levels,” Ecol­ ogy Letters 15:461–467. Branch, G. M., and C. N. Steffani. 2004. “Can We Predict the Effects of Alien Species? A Case-­History of the Invasion of South Africa by Mytilus Galloprovincialis (Lamarck),” Journal of Experimental Marine Biology and Ecol­ogy 300:189–215. Branch, T. A., R. Watson, E. A. Fulton, S. Jennings, C. R. McGilliard, G. T. Pablico, D. Ricard, and S. R. Tracey. 2010. “The Trophic Fingerprint of Marine Fisheries,” Nature 468:431–435. Brandl, S. J., and D. R. Bellwood. 2014. “Individual-­Based Analyses Reveal ­Limited Functional Overlap in a Coral Reef Fish Community,” Journal of Animal Ecol­ogy 83:661–670. Brawley, S. H., J. A. Coyer, A. M. H. Blakeslee, G. Hoarau, L. E. Johnson, J. E. Byers, W. T. Stam, and J. L. Olsen. 2009. “Historical Invasions of the Intertidal Zone of Atlantic North Amer­i­ca Associated with Distinctive Patterns of Trade and Emigration,” Proceedings of the National Acad­emy of Sciences 106: 8239–8244. Breitbart, M. 2012. “Marine Viruses: ‘Truth or Dare,’ ” Annual Review of Marine Science 4:425–448. Breitbart, M., P. Salamon, B. Andresen, J. M. Mahaffy, A. M. Segall, D. Mead, F. Azam, and F. Rohwer. 2002. “Genomic Analy­sis of Uncultured Marine Viral Communities,” Proceedings of the National Acad­emy of Sciences 99:14250–14255. Breitburg, D., L. A. Levin, A. Oschlies, M. Grégoire, F. P. Chavez, D. J. Conley, V. Garçon, et al. 2018. “Declining Oxygen in the Global Ocean and Coastal ­Waters,” Science 359:eaam7240. Bremner, J., S. Rogers, and C. Frid. 2006. “Methods for Describing Ecological Functioning of Marine Benthic Assemblages Using Biological Traits Analy­sis (BTA),” Ecological Indicators 6:609–622. Briffa, M., K. de la Haye, and P. L. Munday. 2012. “High CO2 and Marine Animal Behaviour: Potential Mechanisms and Ecological Consequences,” Marine Pollution Bulletin 64:1519–1528. Briggs, J. C. 1992. “The Marine East Indies: Centre of Origin?” Global Ecol­ogy and Biogeography Letters 2:149–156. Briggs, J. C. 1996. “Tropical Diversity and Conservation,” Conservation Biology 10:713–718. Briggs, J. C. 2007. “Marine Longitudinal Biodiversity: ­Causes and Conservation,” Diversity and Distributions 13:544–555. Briggs, J. C., and B. W. Bowen. 2013. “Marine Shelf Habitat: Biogeography and Evolution,” Journal of Biogeography 40:1023–1035. Brodie, J., C. J. Williamson, D. A. Smale, N. A. Kamenos, N. Mieszkowska, R. Santos, M. Cunliffe, et al. 2014. “The ­Future of the Northeast Atlantic Benthic Flora in a High CO2 World,” Ecol­ ogy and Evolution 4:2787–2798. Brose, U., T. Jonsson, E. L. Berlow, P. Warren, C. Banasek-­Richter, L.-­F. Bersier, J. L. Blanchard, et al. 2006. “Consumer-­Resource

Literature Cited

Body-­Size Relationships in Natu­ral Food Webs,” Ecol­ogy 87:2411–2417. Browman, H. I., R. Law, and C. T. Marshall. 2008. “The Role of Fisheries-­ Induced Evolution,” Science 320:47–50. Brown, C. J., E. A. Fulton, A. J. Hobday, R. J. Matear, H. P. Possingham, C. Bulman, V. Christensen, R. E. Forrest, P. C. Gehrke, and N. A. Gribble. 2010. “Effects of Climate-­Driven Primary Production Change on Marine Food Webs: Implications for Fisheries and Conservation,” Global Change Biology 16:1194–1212. Brown, J. H. 1995. Macroecol­ogy. Chicago: University of Chicago Press. Brown, J. H., J. R. Burger, W. R. Burnside, M. Chang, A. D. Davidson, T. S. Fristoe, M. J. Hamilton, et al. 2014. “Macroecol­ogy Meets Macroeconomics: Resource Scarcity and Global Sustainability,” Ecological Engineering 65:24–32. Brown, J. H., W. R. Burnside, A. D. Davidson, J. P. DeLong, W. C. Dunn, M. J. Hamilton, N. Mercado-­Silva, et al. 2011. “Energetic Limits to Economic Growth,” BioScience 61:19–26. Brown, J. H., J. Gillooly, A. Allen, V. Savage, and G. West. 2004. “­Toward a Metabolic Theory of Ecol­ogy,” Ecol­ogy 85:1771–1789. Brown, J. H., and A. Kodric-­Brown. 1977. “Turnover Rates in Insular Biogeography: Effect of Immigration on Extinction,” Ecol­ogy 58:445–449. Bruce, A. J. 1978. “The Evolution and Zoogeography of Shallow-­Water Tropical Shrimps,” Informative Series, Department of Scientific and Industrial Research, New Zealand 137:337–355. Bruno, J. 2000. “Facilitation of Cobble Beach Plant Communities through Habitat Modification by Spartina Alterniflora,” Ecol­ogy 81:1179–1192. Bruno, J., and B. Cardinale. 2008. “Cascading Effects of Predator Richness,” Frontiers in Ecol­ogy and Environment 6:539–546. Bruno, J. F., L. A. Carr, and M. I. O’Connor. 2015. “Exploring the Role of Temperature in the Ocean through Metabolic Scaling,” Ecol­ogy 96:3126–3140. Bruno, J. F., I. M. Côté, and L. T. Toth. 2018. “Climate Change, Coral Loss, and the Curious Case of the Parrotfish Paradigm: Why ­Don’t Marine Protected Areas Improve Reef Resilience?” Annual Review of Marine Science 11:307–334. Bruno, J. F., and E. R. Selig. 2007. “Regional Decline of Coral Cover in the Indo-­Pacific: Timing, Extent, and Subregional Comparisons,” PLoS ONE 2:e711–8. Bruno, J. F., J. J. Stachowicz, and M. D. Bertness. 2003. “Inclusion of Facilitation into Ecological Theory,” Trends in Ecol­ogy & Evolution 18:119–125. Bruno, J. F., H. Sweatman, W. F. Precht, E. R. Selig, and V. G. W. Schutte. 2009. “Assessing Evidence of Phase Shifts from Coral to Macroalgal Dominance on Coral Reefs,” Ecol­ogy 90:1478–1484. Bulleri, F., and M. G. Chapman. 2010. “The Introduction of Coastal Infrastructure as a Driver of Change in Marine Environments,” Journal of Applied Ecol­ogy 47:26–35. Burge, C. A., C. M. Eakin, C. S. Friedman, B. Froelich, P. K. Hershberger, E. E. Hofmann, L. E. Petes, et al. 2014. “Climate Change Influences on Marine Infectious Diseases: Implications for Management and Society,” Annual Review of Marine Science 6:249–277. Burger, J. R., C. D. Allen, J. H. Brown, W. R. Burnside, A. D. Davidson, T. S. Fristoe, M. J. Hamilton, et al. 2012. “The Macroecol­ogy of Sustainability,” PLoS Biology 10:e1001345. Burger, O., J. P. DeLong, and M. J. Hamilton. 2011. “Industrial Energy Use and the ­Human Life History,” Scientific Reports 1:56. Burkepile, D., and M. Hay. 2006. “Herbivore vs. Nutrient Control of Marine Primary Producers: Context-­Dependent Effects,” Ecol­ogy 87:3128–3139.

397

Burkepile, D. E., and M. E. Hay. 2008. “Herbivore Species Richness and Feeding Complementarity Affect Community Structure and Function on a Coral Reef,” Proceedings of the National Acad­emy of Sciences 105:16201–16206. Burnham, K. P., and D. R. Anderson. 2002. Model Se­lection and Multimodel Inference: A Practical Information-­Theoretic Approach. Second edition. New York: Springer. Burton, R. S., and M. W. Feldman. 1981. “Population Ge­ne­tics of Tigriopus Californicus. II. Differentiation among Neighboring Populations,” Evolution 35:1192–1205. Butman, C., J. Carlton, and S. Palumbi. 1995. “Whaling Effects on Deep-­ Sea Biodiversity,” Conservation Biology 9:462–464. Byers, J. E., and J. M. Pringle. 2006. “­Going against the Flow: Retention, Range Limits and Invasions in Advective Environments,” Marine Ecol­ogy Pro­gress Series 313:27–41. Byrnes, J., J. J. Stachowicz, K. M. Hultgren, A. Randall Hughes, S. V. Olyarnik, and C. S. Thornber. 2005. “Predator Diversity Strengthens Trophic Cascades in Kelp Forests by Modifying Herbivore Be­hav­ior,” Ecol­ogy Letters 9:61–71. Byrnes, J. E., P. L. Reynolds, J. J. Stachowicz, and D. Lusseau. 2007. “Invasions and Extinctions Reshape Coastal Marine Food Webs,” PLoS ONE 2:e295. Cadotte, M. W., B. J. Cardinale, and T. H. Oakley. 2008. “Evolutionary History and the Effect of Biodiversity on Plant Productivity,” Proceedings of the National Acad­emy of Sciences 105: 17012–17017. Cadotte, M. W., and C. M. Tucker. 2017. “Should Environmental Filtering Be Abandoned?” Trends in Ecol­ogy & Evolution 32:429–437. Calbet, A., and M. Landry. 2004. “Phytoplankton Growth, Microzooplankton Grazing, and Carbon Cycling in Marine Systems,” Limnology and Oceanography 49:51–57. Cannicci, S., D. Burrows, S. Fratini, T. J. Smith, J. Offenberg, and F. Dahdouh-­Guebas. 2008. “Faunal Impact on Vegetation Structure and Ecosystem Function in Mangrove Forests: A Review,” Aquatic Botany 89:186–200. Cardinale, B. J., J. E. Duffy, A. Gonzalez, D. U. Hooper, C. Perrings, P. Venail, A. Narwani, et al. 2012. “Biodiversity Loss and Its Impact on Humanity,” Nature 486:59–67. Cardinale, B. J., A. Gonzalez, G. R. H. Allington, and M. Loreau. 2018. “Is Local Biodiversity Declining or Not? A Summary of the Debate over Analy­sis of Species Richness Time Trends,” Biological Conservation 219:175–183. Cardinale, B. J., K. L. Matulich, D. U. Hooper, J. E. Byrnes, E. Duffy, L. Gamfeldt, P. Balvanera, M. I. O’Connor, and A. Gonzalez. 2011. “The Functional Role of Producer Diversity in Ecosystems,” American Journal of Botany 98:572–592. Cardinale, B. J., D. S. Srivastava, J. E. Duffy, J. P. Wright, A. L. Downing, M. Sankaran, and C. Jouseau. 2006. “Effects of Biodiversity on the Functioning of Trophic Groups and Ecosystems,” Nature 443:989–992. Carlson, S. M., and T. R. Seamons. 2008. “A Review of Quantitative Ge­ ne­tic Components of Fitness in Salmonids: Implications for Adaptation to ­Future Change,” Evolutionary Applications 1:222–238. Carlton, J., J. Geller, M. Reaka-­Kudla, and E. Norse. 1999. “Historical Extinctions in the Sea,” Annual Review of Ecol­ogy and Systematics 30:515–538. Carlton, J. T., and A. N. Cohen. 2003. “Episodic Global Dispersal in Shallow ­Water Marine Organisms: The Case History of the Eu­ro­ pean Shore Crabs Carcinus Maenas and C. Aestuarii,” Journal of Biogeography 30:1809–1820.

398

Ocean Ecology

Carlton, J. T., J. K. Tompson, L. E. Schemel, and F. H. Nichols. 1990. “Remarkable Invasion of San Francisco Bay (California, USA), by the Asian Clam Potamocorbula Amurensis. I. Introduction and Dispersal,” Marine Ecol­ogy Pro­gress Series 66:81–94. Caro, A. U., S. A. Navarrete, and J. C. Castilla. 2010. “Ecological Convergence in a Rocky Intertidal Shore Metacommunity Despite High Spatial Variability in Recruitment Regimes,” Proceedings of the National Acad­emy of Sciences 107:18528–18532. Caron, D. A., P. D. Countway, A. C. Jones, D. Y. Kim, and A. Schnetzer. 2012. “Marine Protistan Diversity,” Annual Review of Marine Science 4:467–493. Caron, D. A., and D. A. Hutchins. 2013. “The Effects of Changing Climate on Microzooplankton Grazing and Community Structure: ­Drivers, Predictions and Knowledge Gaps,” Journal of Plankton Research 35:235–252. Carpenter, K. E., M. Abrar, G. Aeby, R. B. Aronson, S. Banks, A. Bruckner, A. Chiriboga, et al. 2008. “One-­Third of Reef-­Building Corals Face Elevated Extinction Risk from Climate Change and Local Impacts,” Science 321:560–563. Carpenter, R. 1986. “Partitioning Herbivory and Its Effects on Coral Reef Algal Communities,” Ecological Monographs 56:345–363. Carpenter, R. 1988. “Mass Mortality of a Ca­rib­bean Sea Urchin: Immediate Effects on Community Metabolism and Other Herbivores,” Proceedings of the National Acad­emy of Sciences 85:511. Carpenter, R. C., 1990. “Mass Mortality of Diadema antillarum. II: Effects on Population Densities and Grazing Intensity of Parrotfishes and Surgeonfishes,” Marine Biology 104:79–86. Carpenter, S., J. Kitchell, J. Hodgson, and P. Cochran. 1987. “Regulation of Lake Primary Productivity by Food Web Structure,” Ecol­ogy 68:1863–1876. Carr, M., T. Anderson, and M. Hixon. 2002. “Biodiversity, Population Regulation, and the Stability of Coral-­Reef Fish Communities,” Proceedings of the National Acad­emy of Sciences 99:11241–11245. Carson, R. T. 2010. “The Environmental Kuznets Curve: Seeking Empirical Regularity and Theoretical Structure,” Review of Environmental Economics and Policy 4:3–23. Casey, J. M., A. H. Baird, S. J. Brandl, M. O. Hoogenboom, J. R. Rizzari, A. J. Frisch, C. E. Mirbach, and S. R. Connolly. 2016. “A Test of Trophic Cascade Theory: Fish and Benthic Assemblages across a Predator Density Gradient on Coral Reefs,” Oecologia 183:161–175. Casini, M., J. Lövgren, J. Hjelm, M. Cardinale, J.-­C. Molinero, and G. Kornilovs. 2008. “Multi-­Level Trophic Cascades in a Heavi­ly Exploited Open Marine Ecosystem,” Proceedings of the Royal Society B-Biological Sciences 275:1793–1801. Caspari, R., and M. H. Wolpoff. 2013. “The Pro­cess of Modern ­Human Origins: The Evolutionary and Demographic Changes Giving Rise to Modern ­Humans.” Pages 355–391 in F. H. Smith and J. C. M. Ahern, editors, The Origins of Modern ­Humans: Biology Reconsidered. Hoboken, NJ: John Wiley & Sons. Caswell, H., and J. E. Cohen. 1991. “Disturbance, Interspecific Interaction and Diversity in Metapopulations,” Biological Journal of the Linnean Society 42:193–218. Cavanaugh, C. M., S. L. Gardiner, M. L. Jones, H. W. Jannasch, and J. B. Waterbury. 1981. “Prokaryotic Cells in the Hydrothermal Vent Tube Worm Riftia pachyptila Jones: Pos­si­ble Chemoautotrophic Symbionts,” Science 213:340–342. Cavanaugh, K. C., J. R. Kellner, A. J. Forde, D. S. Gruner, J. D. Parker, W. Rodriguez, and I. C. Feller. 2013. “Poleward Expansion of Mangroves Is a Threshold Response to Decreased Frequency of Extreme Cold Events,” Proceedings of the National Acad­emy of Sciences 111:723–727.

Cavender-­Bares, J., K. H. Kozak, P. V. A. Fine, and S. W. Kembel. 2009. “The Merging of Community Ecol­ogy and Phyloge­ne­tic Biology,” Ecol­ogy Letters 12:693–715. Cebrián, J. 1999. “Patterns in the Fate of Production in Plant Communities,” American Naturalist 154:449–468. Cebrián, J. 2002. “Variability and Control of Carbon Consumption, Export, and Accumulation in Marine Communities,” Limnology and Oceanography 47:11–22. Cebrián, J., 2004. “Role of First-­Order Consumers in Ecosystem Carbon Flow,” Ecol­ogy Letters 7:232–240. Cebrián, J., and J. Lartigue. 2004. “Patterns of Herbivory and Decomposition in Aquatic and Terrestrial Ecosystems,” Ecological Monographs 74:237–259. Cebrián, J., J. B. Shurin, E. T. Borer, B. J. Cardinale, J. T. Ngai, M. D. Smith, W. F. Fagan, and A. Hector. 2009. “Producer Nutritional Quality Controls Ecosystem Trophic Structure,” PLoS ONE 4:e4929. Cebrián, J., M. Williams, J. McClelland, and I. Valiela. 1998. “The Dependence of Heterotrophic Consumption and C Accumulation on Autotrophic Nutrient Content in Ecosystems,” Ecol­ogy Letters 1:165–170. Cermeño, P., C. de Vargas, F. Abrantes, and P. G. Falkowski. 2010. “Phytoplankton Biogeography and Community Stability in the Ocean,” PLoS ONE 5:e10037. Cermeño, P., and P. G. Falkowski. 2009. “Controls on Diatom Biogeography in the Ocean,” Science 325:1539–1541. Chapin, F. III, B. Walker, R. Hobbs, D. Hooper, J. Lawton, O. Sala, and D. Tilman. 1997. “Biotic Control over the Functioning of Ecosystems,” Science 277:500–504. Chapin, F. S., E. S. Zavaleta, V. T. Eviner, R. L. Naylor, P. M. Vitousek, H. L. Reynolds, D. U. Hooper, S. Lavorel, O. E. Sala, S. E. Hobbie, M. C. Mack, and S. Díaz. 2000. “Consequences of Changing Biodiversity,” Nature 405:234–242. Chapman, M. G., and D. J. Blockley. 2009. “Engineering Novel Habitats on Urban Infrastructure to Increase Intertidal Biodiversity,” Oecologia 161:625–635. Chassot, E., S. Bonhomme, N. K. Dulvy, F. Mélin, R. Watson, D. Gascuel, and O. Le Pape. 2010. “Global Marine Primary Production Constrains Fisheries Catches,” Ecol­ogy Letters 13:495–505. Checkley, D. M., S. Raman, G. L. Maillet, and K. M. Mason. 1988. “Winter Storm Effects on the Spawning and Larval Drift of a Pelagic Fish,” Nature 335:346–348. Chesson, P. 2000. “Mechanisms of Maintenance of Species Diversity,” Annual Review of Ecol­ogy and Systematics 31:343–366. Cheung, W. W. L., V. W. Y. Lam, J. L. Sarmiento, K. Kearney, R. Watson, D. Zeller, and D. Pauly. 2009. “Large-­Scale Re­distribution of Maximum Fisheries Catch Potential in the Global Ocean ­under Climate Change,” Global Change Biology 16:24–35. Cheung, W. W. L., J. L. Sarmiento, J. Dunne, T. L. Frölicher, V. W. Y. Lam, M. L. Deng Palomares, R. Watson, and D. Pauly. 2012. “Shrinking of Fishes Exacerbates Impacts of Global Ocean Changes on Marine Ecosystems,” Nature Climate Change 3:254–258. Cheung, W. W. L., R. Watson, and D. Pauly. 2013. “Signature of Ocean Warming in Global Fisheries Catch,” Nature 497:365–368. Chisholm, S. W., R. J. Olson, E. R. Zettler, and R. Goericke. 1988. “A Novel Free-­Living Prochlorophyte Abundant in the Oceanic Euphotic Zone,” Nature 334:340–343. Chittaro, P. M. 2002. “Species-­Area Relationships for Coral Reef Fish Assemblages of St. Croix, US Virgin Islands,” Marine Ecol­ogy Pro­ gress Series 233:253–261. Chopra, A., and C. H. Lineweaver. 2008. “The Major Elemental Abundance Differences between Life, the Oceans and the Sun,” Proceedings of the 8th Australian Space Science Conference 49–55.

Literature Cited

Christensen, V., and D. Pauly. 1992. “ECOPATH II—­A Software for Balancing Steady-­State Ecosystem Models and Calculating Network Characteristics* 1,” Ecological Modelling 61:169–185. Christensen, V., and C. Walters. 2004. “Ecopath with Ecosim: Methods, Capabilities and Limitations* 1,” Ecological Modelling 172:109–139. Cinner, J. E., C. Huchery, M. A. MacNeil, N. A. J. Graham, T. R. McClanahan, J. Maina, E. Maire, et al. 2016. “Bright Spots among the World’s Coral Reefs,” Nature 535:416–419. Civitello, D. J., J. Cohen, H. Fatima, N. T. Halstead, J. Liriano, T. A. McMahon, C. N. Ortega, et al. 2015. “Biodiversity Inhibits Parasites: Broad Evidence for the Dilution Effect,” Proceedings of the National Acad­emy of Sciences 112:8667–8671. Clapham, M. 1926. “The Ecol­ogy of Rock Pools,” Irish Naturalists’ Journal 1:48–51. Clark, M. 2001. “Are Deepwater Fisheries Sustainable?—­The Example of Orange Roughy (Hoplostethus atlanticus) in New Zealand,” Fisheries Research 51:123–135. Clark, T. D., G. D. Raby, D. G. Roche, S. A. Binning, B. Speers-­Roesch, F. Jutfelt, and J. Sundin. 2020. “Ocean Acidification Does Not Impair the Behaviour of Coral Reef Fishes,” Nature 577:370–375. Clarke, A., and K. Gaston. 2006. “Climate, Energy and Diversity,” Proceedings of the Royal Society B-­Biological Sciences 273:2257. Clausen, K. K., D. Krause-­Jensen, B. Olesen, and N. Marba. 2014. “Seasonality of Eelgrass Biomass across Gradients in Temperature and Latitude,” Marine Ecol­ogy Pro­gress Series 506:71–85. Clavero, M., and E. García-­Berthou. 2005. “Invasive Species Are a Leading Cause of Animal Extinctions,” Trends in Ecol­ogy & Evolution 20:110. Clements, C. S., and M. E. Hay. 2019. “Biodiversity Enhances Coral Growth, Tissue Survivorship and Suppression of Macroalgae,” Nature Ecol­ogy & Evolution 3:178–182. Clements, F. E. 1916. Plant Succession: An Analy­sis of the Development of Vegetation. Washington, DC: Car­ne­gie Institution. Clements, K. D., and J. H. Choat. 2018. “Nutritional Ecol­ogy of Parrotfishes (Scarinae, Labridae).” Pages 42–68 in A. S. Hoey and R. M. Bonaldo, editors, Biology of Parrotfishes. Boca Raton, FL: CRC Press. Clements, K. D., D. Raubenheimer, and J. H. Choat. 2009. “Nutritional Ecol­ogy of Marine Herbivorous Fishes: Ten Years On,” Functional Ecol­ogy 23:79–92. Cloern, J. E., and A. D. Jassby. 2012. “­Drivers of Change in Estuarine-­ Coastal Ecosystems: Discoveries from Four De­cades of Study in San Francisco Bay,” Reviews of Geophysics 50:RG4001. Cohen, A. N., and J. T. Carlton. 1998. “Accelerating Invasion Rate in a Highly Invaded Estuary,” Science 279:555–558. Cohen, J. E., T. Jonsson, and S. R. Carpenter. 2003. “Ecological Community Description Using the Food Web, Species Abundance, and Body Size,” Proceedings of the National Acad­emy of Sciences 100:1781–1786. Coleman, R., A. Underwood, L. Benedetti-­Cecchi, P. Åberg, F. Arenas, J. Arrontes, J. Castro, R. Hartnoll, S. Jenkins, and J. Paula. 2006. “A Continental Scale Evaluation of the Role of Limpet Grazing on Rocky Shores,” Oecologia 147:556–564. Coley, P. D., and J. A. Barone. 1996. “Herbivory and Plant Defenses in Tropical Forests,” Annual Review of Ecol­ogy and Systematics 27:305–335. Collie, J. 2004. “Regime Shifts: Can Ecological Theory Illuminate the Mechanisms?” Pro­gress in Oceanography 60:281–302. Colman, J. 1933. “The Nature of the Intertidal Zonation of Plants and Animals,” Journal of the Marine Biological Association of the UK 18:435–476.

399

Colwell, R. K. 2009. “Biodiversity: Concepts, Patterns, and Mea­sure­ ment.” Pages 257–263 in S. Levin, editor, The Prince­ton Guide to Ecol­ogy. Prince­ton, NJ: Prince­ton University Press. Commito, J., and W. Ambrose Jr. 1985. “Multiple Trophic Levels in Soft-­ Bottom Communities,” Marine Ecol­ogy Pro­gress Series 26:289–293. Connell, J. H. 1961. “The Influence of Interspecific Competition and Other ­Factors on the Distribution of the Barnacle Chthamalus Stellatus,” Ecol­ogy 42:710–723. Connell, J. H. 1978. “Diversity in Tropical Rain Forests and Coral Reefs,” Science 199:1302–1310. Connolly, S. R., and J. Roughgarden. 1998. “A Latitudinal Gradient in Northeast Pacific Intertidal Community Structure: Evidence for an Oceanographically Based Synthesis of Marine Community Theory,” The American Naturalist 151(4):311–326. Connolly, S. R., and J. Roughgarden. 1999. “Theory of Marine Communities: Competition, Predation, and Recruitment-­Dependent Interaction Strength,” Ecological Monographs 69:277–296. Conover, D. O., and S. B. Munch. 2002. “Sustaining Fisheries Yields over Evolutionary Time Scales,” Science 297:94–96. Cooley, S. R., N. Lucey, H. Kite-­Powell, and S. C. Doney. 2012. “Nutrition and Income from Molluscs ­Today Imply Vulnerability to Ocean Acidification Tomorrow,” Fish and Fisheries 13:182–215. Costanza, R. 1989. “What Is Ecological Economics?” Ecological Economics 1:1–7. Costanza, R., R. d’Arge, R. de Groot, S. Farber, M. Grasso, B. Hannon, K. Limburg, et al. 1997. “The Value of the World’s Ecosystem Ser­ vices and Natu­ral Capital,” Nature 387:253–260. Costanza, R., O. Pérez-­Maqueo, M. L. Martinez, P. Sutton, S. J. Anderson, and K. Mulder. 2008. “The Value of Coastal Wetlands for Hurricane Protection,” AMBIO 37:241–248. Costello, C., S. Gaines, and J. Lynham. 2008. “Can Catch Shares Prevent Fisheries Collapse?” Science 321:1678–1681. Costello, C., D. Ovando, T. Clavelle, C. K. Strauss, R. Hilborn, M. C. Melnychuk, T. A. Branch, et al. 2016. “Global Fishery Prospects ­under Contrasting Management Regimes,” Proceedings of the National Acad­emy of Sciences 113:5125–5129. Costello, M. J., S. Wilson, and B. Houlding. 2012. “Predicting Total Global Species Richness Using Rates of Species Description and Estimates of Taxonomic Effort,” Systematic Biology 61:871. Courties, C., A. Vaquer, M. Troussellier, and J. Lautier. 1994. “Smallest Eukaryotic Organism,” Nature 370:255. Cowen, R. K. 2000. “Connectivity of Marine Populations: Open or Closed?” Science 287:857–859. Cowen, R. K. 2006. “Scaling of Connectivity in Marine Populations,” Science 311:522–527. Cowen, R. K., and S. Sponaugle. 2009. “Larval Dispersal and Marine Population Connectivity,” Annual Review of Marine Science 1:443–466. Cowman, P. F., and D. R. Bellwood. 2013. “Vicariance across Major Marine Biogeographic Barriers: Temporal Concordance and the Relative Intensity of Hard versus Soft Barriers,” Proceedings of the Royal Society B-­Biological Sciences 280:20131541. Craft, C., J. Clough, J. Ehman, S. Joye, R. Park, S. Pennings, H. Guo, and M. Machmuller. 2009. “Forecasting the Effects of Accelerated Sea-­Level Rise on Tidal Marsh Ecosystem Ser­vices,” Frontiers in Ecol­ogy and the Environment 7:73–78. Crain, C., B. Silliman, S. Bertness, and M. Bertness. 2004. “Physical and Biotic ­Drivers of Plant Distribution across Estuarine Salinity Gradients,” Ecol­ogy 85:2539–2549.

400

Ocean Ecology

Crain, C. M., K. Kroeker, and B. S. Halpern. 2008. “Interactive and Cumulative Effects of Multiple ­Human Stressors in Marine Systems,” Ecol­ogy Letters 11:1304–1315. Cronin, T. W., and R. B. Forward. 1986. “Vertical Migration Cycles of Crab Larvae and Their Role in Larval Dispersal,” Bulletin of Marine Science 39(2):192–201. Crossland, C. J., B. G. Hatcher, and S. V. Smith. 1991. “Role of Coral Reefs in Global Ocean Production,” Coral Reefs 10:55–64. Crouse, D. T., L. B. Crowder, and H. Caswell. 1987. “A Stage-­Based Population Model for Loggerhead Sea Turtles and Implications for Conservation,” Ecol­ogy 68:1412–1423. Crowder, L. B., D. T. Crouse, S. S. Heppell, and T. H. Martin. 1994. “Predicting the Impact of Turtle Excluder Devices on Loggerhead Sea Turtle Populations,” Ecological Applications 4:437–445. Crutzen, P. J. 2002. “Geology of Mankind,” Nature 415:23. Currie, D. J. 1991. “Energy and Large-­Scale Patterns of Animal-­and Plant-­Species Richness,” American Naturalist 137:27–49. Currin, C. A., S. Y. Newell, and H. W. Paerl. 1995. “The Role of Standing Dead Spartina alterniflora and Benthic Microalgae in Salt Marsh Food Webs: Considerations Based on Multiple Stable Isotope Analy­sis,” Marine Ecol­ogy Pro­gress Series 121:99–116. Dafforn, K. A., T. M. Glasby, L. Airoldi, N. K. Rivero, M. Mayer-­Pinto, and E. L. Johnston. 2015. “Marine Urbanization: An Ecological Framework for Designing Multifunctional Artificial Structures,” Frontiers in Ecol­ogy and the Environment 13:82–90. Dafforn, K. A., T. M. Glasby, and E. L. Johnston. 2009. “Links between Estuarine Condition and Spatial Distributions of Marine Invaders,” Diversity and Distributions 15:807–821. Daleo, P., E. Fanjul, A. M. Casariego, B. R. Silliman, M. D. Bertness, and O. Iribarne. 2007. “Ecosystem Engineers Activate Mycorrhizal Mutualism in Salt Marshes,” Ecol­ogy Letters 10:902–908. Damuth, J. 1981. “Population Density and Body Size in Mammals,” Nature 290:699–700. Danovaro, R., C. Gambi, A. Dell’anno, C. Corinaldesi, S. Fraschetti, A. Vanreusel, M. Vincx, and A. J. Gooday. 2008. “Exponential Decline of Deep-­Sea Ecosystem Functioning Linked to Benthic Biodiversity Loss,” Current Biology 18:1–8. Darimont, C., S. Carlson, M. Kinnison, P. Paquet, T. Reimchen, and C. Wilmers. 2009. “­Human Predators Outpace Other Agents of Trait Change in the Wild,” Proceedings of the National Acad­emy of Sciences 106:952–954. Darimont, C. T., C. H. Fox, H. M. Bryan, and T. E. Reimchen. 2015. “The Unique Ecol­ogy of ­Human Predators,” Science 349:858–860. Darling, E. S., L. Alvarez-­Filip, T. A. Oliver, T. R. McClanahan, and I. M. Côté. 2012. “Evaluating Life-­History Strategies of Reef Corals from Species Traits,” Ecol­ogy Letters 15:1378–1386. Darwin, C. 1859. On the Origin of Species by Means of Natu­ral Se­lection, or the Preservation of Favoured Races in the Strug­gle for Life. New York: D. Appleton and Com­pany. Das, S., and J. R. Vincent. 2009. “Mangroves Protected Villages and Reduced Death Toll during Indian Super Cyclone,” Proceedings of the National Acad­emy of Sciences 106:7357–7360. Daskalov, G. 2002. “Overfishing Drives a Trophic Cascade in the Black Sea,” Marine Ecol­ogy Pro­gress Series 225:53–63. Daskalov, G. M., A. N. Grishin, S. Rodionov, and V. Mihneva. 2007. “Trophic Cascades Triggered by Overfishing Reveal Pos­si­ble Mechanisms of Ecosystem Regime Shifts,” Proceedings of the National Acad­emy of Sciences 104:10518–10523. Davenport, A. C., and T. W. Anderson. 2007. “Positive Indirect Effects of Reef Fishes on Kelp Per­for­mance: The Importance of Mesograzers,” Ecol­ogy 88:1548–1561. Davidson, A. D., A. G. Boyer, H. Kim, S. Pompa-­Mansilla, M. J. Hamilton, D. P. Costa, G. Ceballos, and J. H. Brown. 2012. “­Drivers and

Hotspots of Extinction Risk in Marine Mammals,” Proceedings of the National Acad­emy of Sciences 109:3395–3400. Dawkins, R. 1976. The Selfish Gene. Oxford, UK: Oxford University Press. Dawkins, R. 1982. The Extended Phenotype. Oxford, UK: Oxford University Press. Dayton, P. 1971. “Competition, Disturbance, and Community Organ­ ization: The Provision and Subsequent Utilization of Space in a Rocky Intertidal Community,” Ecological Monographs 41:351–389. Deason, E., and T. Smayda. 1982. “Ctenophore-­Zooplankton-­ Phytoplankton Interactions in Narragansett Bay, Rhode Island, USA, during 1972–1977,” Journal of Plankton Research 4:203. De’ath, G., K. E. Fabricius, H. Sweatman, and M. Puotinen. 2012. “The 27-­Year Decline of Coral Cover on the ­Great Barrier Reef and Its ­Causes,” Proceedings of the National Acad­emy of Sciences 109:17995–17999. De Baar, H. J., J. T. De Jong, D. C. Bakker, B. M. Löscher, C. Veth, U. Bathmann, and V. Smetacek. 1995. “Importance of Iron for Plankton Blooms and Carbon Dioxide Drawdown in the Southern Ocean,” Nature 373:412–415. Decho, A. W. 1990. “Microbial Exopolymer Secretions in Ocean Environments: Their Role(s) in Food Webs and Marine Pro­cesses,” Oceanography and Marine Biology: Annual Review 28:73–153. Dee, L. E., S. J. Miller, L. E. Peavey, D. Bradley, R. R. Gentry, R. Startz, S. D. Gaines, and S. E. Lester. 2016. “Functional Diversity of Catch Mitigates Negative Effects of Temperature Variability on Fisheries Yields,” Proceedings of the Royal Society B-­Biological Sciences 283:20161435. de la Haye, K. L., J. I. Spicer, S. Widdicombe, and M. Briffa. 2012. “Reduced pH Sea ­Water Disrupts Chemo-­responsive Behaviour in an Intertidal Crustacean,” Journal of Experimental Marine Biology and Ecol­ogy 412:134–140. Dell, A. I., S. Pawar, and V. M. Savage. 2014. “Temperature Dependence of Trophic Interactions Are Driven by Asymmetry of Species Responses and Foraging Strategy,” Journal of Animal Ecol­ogy 83:70–84. Den Hartog, C. 1970. The Sea-­Grasses of the World. Amsterdam: North-­Holland. Deuser, W. G., E. H. Ross, and R. F. Anderson. 1981. “Seasonality in the Supply of Sediment to the Deep Sargasso Sea and Implications for the Rapid Transfer of M ­ atter to the Deep Ocean,” Deep Sea Research Part A: Oceanographic Research Papers 28:495–505. Deutsch, C., A. Ferrel, B. Seibel, H. O. Pörtner, and R. B. Huey. 2015. “Climate Change Tightens a Metabolic Constraint on Marine Habitats,” Science 348:1132–1135. Deutsch, C., J. L. Penn, and B. Seibel. 2020. “Metabolic Trait Diversity Shapes Marine Biogeography,” Nature 585:557–562. de Vargas, C., S. Audic, N. Henry, J. Decelle, F. Mahé, R. Logares, E. Lara, et al. 2015. “Eukaryotic Plankton Diversity in the Sunlit Ocean,” Science 348:1261605. Devine, J. A., K. D. Baker, and R. L. Haedrich. 2006. “Fisheries: Deep-­ Sea Fishes Qualify as Endangered,” Nature 439:29. de Wit, R., and T. Bouvier. 2006. “ ‘Every­thing Is Everywhere, but, the Environment Selects’; What Did Baas Becking and Beijerinck ­Really Say?” Environmental Microbiology 8:755–758. Diamond, J. M. 1975. “Assembly of Species Communities.” Pages 342– 344 in J. M. Diamond and M. L. Cody, editors, Ecol­ogy and Evolution of Communities. Boston: Harvard University Press. Diana, J. S. 2009. “Aquaculture Production and Biodiversity Conservation,” BioScience 59:27–38. Diaz, R. J., and R. Rosenberg. 2008. “Spreading Dead Zones and Consequences for Marine Ecosystems,” Science 321:926–929.

Literature Cited

Dieckmann, U., and M. Doebeli. 1999. “On the Origin of Species by Sympatric Speciation,” Nature 400:354–357. Dieng, H. B., A. Cazenave, B. Meyssignac, and M. Ablain. 2017. “New Estimate of the Current Rate of Sea Level Rise from a Sea Level Bud­get Approach,” Geophysical Research Letters 44:3744–3751. Dixon, G. B., S. W. Davies, G. A. Aglyamova, and E. Meyer. 2015. “Genomic Determinants of Coral Heat Tolerance across Latitudes,” Science 348:1460–1462. Dixson, D. L., D. Abrego, and M. E. Hay. 2014. “Chemically Mediated Be­ hav­ior of Recruiting Corals and Fishes: A Tipping Point That May Limit Reef Recovery,” Science 345:892–897. Dixson, D. L., G. P. Jones, P. L. Munday, S. Planes, M. S. Pratchett, M. Srinivasan, C. Syms, and S. R. Thorrold. 2008. “Coral Reef Fish Smell Leaves to Find Island Homes,” Proceedings of the Royal Society B-­Biological Sciences 275:2831–2839. Doak, D. F., D. Bigger, E. K. Harding, M. A. Marvier, R. E. O’Malley, and D. Thomson. 1998. “The Statistical Inevitability of Stability-­ Diversity Relationships in Community Ecol­ogy,” American Naturalist 151:264–276. Dobson, A., D. Lodge, J. Alder, G. S. Cumming, J. Keymer, J. McGlade, H. Mooney, J. A. Rusak, O. Sala, V. Wolters, D. Wall, R. Winfree, and M. A. Xenopoulos. 2006. “Habitat Loss, Trophic Collapse, and the Decline of Ecosystem Ser­vices,” Ecol­ogy 87:1915–1924. Dodd, L. F., J. H. Grabowski, M. F. Piehler, I. Westfield, and J. B. Ries. 2015. “Ocean Acidification Impairs Crab Foraging Be­hav­ior,” Proceedings of the Royal Society B-­Biological Sciences 282:20150333. Donato, D. C., J. B. Kauffman, D. Murdiyarso, S. Kurnianto, M. Stidham, and M. Kanninen. 2011. “Mangroves among the Most Carbon-­ Rich Forests in the Tropics,” Nature Geoscience 4:293–297. Doney, S., V. Fabry, R. Feely, and J. Kleypas. 2009. “Ocean Acidification: The Other CO2 Prob­lem,” Annual Review of Marine Science 1:169–192. Doney, S. C., M. Ruckelshaus, J. E. Duffy, J. P. Barry, F. Chan, C. A. En­ glish, H. M. Galindo, J. M. Grebmeier, A. B. Hollowed, and N. Knowlton. 2011. “Climate Change Impacts on Marine Ecosystems,” Annual Review of Marine Science 4:11–37. Donner, S. D., and D. Potere. 2007. “The Inequity of the Global Threat to Coral Reefs,” BioScience 57:214–215. Doonan, I. J., D. Fu, and M. R. Dunn. 2015. “Harvest Control Rules for a Sustainable Orange Roughy Fishery,” Deep Sea Research Part I: Oceanographic Research Papers 98:53–61. Dorenbosch, M., M. G. G. Grol, I. Nagelkerken, and G. van der Velde. 2006. “Seagrass Beds and Mangroves as Potential Nurseries for the Threatened Indo-­Pacific Humphead Wrasse, Cheilinus undulatus, and Ca­rib­bean Rainbow Parrotfish, Scarus guacamaia,” Biological Conservation 129:277–282. Dornelas, M., S. R. Connolly, and T. P. Hughes. 2006. “Coral Reef Diversity Refutes the Neutral Theory of Biodiversity,” Nature 440:80–82. Dornelas, M., N. J. Gotelli, B. McGill, H. Shimadzu, F. Moyes, C. Sievers, and A. E. Magurran. 2014. “Assemblage Time Series Reveal Biodiversity Change but Not Systematic Loss,” Science 344:296–299. Dornelas, M., J. S. Madin, A. H. Baird, and S. R. Connolly. 2017. “Allometric Growth in Reef-­Building Corals,” Proceedings of the Royal Society B-­Biological Sciences 284:20170053. Doropoulos, C., S. Ward, G. Diaz-­Pulido, O. Hoegh-­Guldberg, and P. J. Mumby. 2012. “Ocean Acidification Reduces Coral Recruitment by Disrupting Intimate Larval-­Algal Settlement Interactions,” Ecol­ogy Letters 15:338–346. Douglass, J. G., J. E. Duffy, and J. F. Bruno. 2008. “Herbivore and Predator Diversity Interactively Affect Ecosystem Properties in an Experimental Marine Community,” Ecol­ogy Letters 11:598–608.

401

Douvere, F. 2008. “The Importance of Marine Spatial Planning in Advancing Ecosystem-­Based Sea Use Management,” Marine Policy 32:762–771. Duarte, C. 1991. “Allometric Scaling of Seagrass Form and Productivity,” Marine Ecol­ogy Pro­gress Series, Oldendorf 77:289–300. Duarte, C., and J. Cebrián. 1996. “The Fate of Marine Autotrophic Production,” Limnology and Oceanography 41(8):1758–1766. Duarte, C., and C. Chiscano. 1999. “Seagrass Biomass and Production: A Reassessment,” Aquatic Botany 65:159–174. Duarte, C. M. 1995. “Submerged Aquatic Vegetation in Relation to Dif­ fer­ent Nutrient Regimes,” Ophelia 41:87–112. Duarte, C. M., S. Agusti, E. Barbier, G. L. Britten, J. C. Castilla, J.-­P. Gattuso, R. W. Fulweiler, et al. 2020. “Rebuilding Marine Life,” Nature 580:39–51. Duarte, C. M., H. Kennedy, N. Marbà, and I. Hendriks. 2013. “Assessing the Capacity of Seagrass Meadows for Carbon Burial: Current Limitations and ­Future Strategies,” Ocean and Coastal Management 83:32–38. Duarte, C. M., N. Marbà, and M. Holmer. 2007. “Ecol­ogy: Rapid Domestication of Marine Species,” Science 316:382–383. Ducklow, H. W., and S. C. Doney. 2013. “What Is the Metabolic State of the Oligotrophic Ocean? A Debate,” Annual Review of Marine Science 5:525–533. Ducklow, H. W., D. K. Steinberg, and K. O. Buesseler. 2001. “Upper Ocean Carbon Export and the Biological Pump,” Oceanography 14:50–58. Dudgeon, S. R., R. B. Aronson, J. F. Bruno, and W. F. Precht. 2010. “Phase Shifts and Stable States on Coral Reefs,” Marine Ecol­ogy Pro­ gress Series 413:201–216. Duffy, J. E. 2002. “Biodiversity and Ecosystem Function: The Consumer Connection,” Oikos 99:201–219. Duffy, J. E. 2003. “Biodiversity Loss, Trophic Skew and Ecosystem Functioning,” Ecol­ogy Letters 6:680–687. Duffy, J. E., B. J. Cardinale, K. E. France, P. B. McIntyre, E. Thébault, and M. Loreau. 2007. “The Functional Role of Biodiversity in Ecosystems: Incorporating Trophic Complexity,” Ecol­ogy Letters 10:522–538. Duffy, J. E., C. M. Godwin, and B. J. Cardinale. 2017. “Biodiversity Effects in the Wild Are Common and as Strong as Key ­Drivers of Productivity,” Nature 549:261–264. Duffy, J. E., and M. E. Hay. 2000. “Strong Impacts of Grazing Amphipods on the Organ­ization of a Benthic Community,” Ecological Monographs 70:237–263. Duffy, J. E., J. S. Lefcheck, R. D. Stuart-­Smith, S. A. Navarrete, and G. J. Edgar. 2016. “Biodiversity Enhances Reef Fish Biomass and Re­sis­tance to Climate Change,” Proceedings of the National Acad­emy of Sciences 113:6230–6235. Duffy, J. E., P.-­O. Moksnes, and A. R. Hughes. 2014. “Ecol­ogy of Seagrass Communities.” Pages 271–297 in M. D. Bertness, J. F. Bruno, B. R. Silliman, and J. J. Stachowicz, editors, Marine Community Ecol­ogy and Conservation. Sunderland, MA: Sinauer Associates. Duffy, J. E., P. L. Reynolds, C. Boström, J. A. Coyer, M. Cusson, S. Donadi, J. G. Douglass, et al. 2015. “Biodiversity Mediates Top-­Down Control in Eelgrass Ecosystems: A Global Comparative-­ Experimental Approach,” Ecol­ogy Letters 18:696–705. Duffy, J. E., J. P. Richardson, and E. A. Canuel. 2003. “Grazer Diversity Effects on Ecosystem Functioning in Seagrass Beds,” Ecol­ogy Letters 6:637–645. Duffy, J. E., J. P. Richardson, and K. France. 2005. “Ecosystem Consequences of Diversity Depend on Food Chain Length in Estuarine Vegetation,” Ecol­ogy Letters 8:301–309. Duffy, J. E., and Stachowicz, J. J. 2006. “Why Biodiversity Is Impor­tant to Oceanography: Potential Roles of Ge­ne­tic, Species, and

402

Ocean Ecology

­ rophic Diversity in Pelagic Ecosystem Pro­cesses,” Marine T Ecol­ogy Pro­gress Series 311:179–189. Dugdale, R. C., and J. J. Goering. 1967. “Uptake of New and Regenerated Forms of Nitrogen in Primary Productivity,” Limnology and Oceanography 12:196–206. Duggins, D. 1980. “Kelp Beds and Sea Otters: An Experimental Approach,” Ecol­ogy 61:447–453. Duggins, D. O., C. A. Simenstad, and J. A. Estes. 1989. “Magnification of Secondary Production by Kelp Detritus in Coastal Marine Ecosystems,” Science 245:170–173. Dulvy, N. K., R. P. Freckleton, and N. V. C. Polunin. 2004. “Coral Reef Cascades and the Indirect Effects of Predator Removal by Exploitation,” Ecol­ogy Letters 7:410–416. Dulvy, N. K., J. K. Pinnegar, and J. D. Reynolds. 2009. “Holocene Extinctions in the Sea.” Pages 129–150 in S. T. Turvey, editor, Holocene Extinctions. Oxford, UK: Oxford University Press. Dulvy, N. K., Y. Sadovy, and J. D. Reynolds. 2003. “Extinction Vulnerability in Marine Populations,” Fish and Fisheries 4:25–64. Dungan, M. L., T. E. Miller, and D. A. Thomson. 1982. “Catastrophic Decline of a Top Carnivore in the Gulf of California Rocky Intertidal Zone,” Science 216:989–991. Eakin, C., J. Morgan, S. Heron, T. Smith, and G. Liu. 2010. “Ca­rib­bean Corals in Crisis: Rec­ord Thermal Stress, Bleaching, and Mortality in 2005,” PLoS ONE 11:e13969. Eckert, G. L. 2003. “Effects of the Planktonic Period on Marine Population Fluctuations,” Ecol­ogy 84:372–383. Edgar, G. J., T. J. Alexander, J. S. Lefcheck, A. E. Bates, S. J. Kininmonth, R. J. Thomson, J. E. Duffy, M. J. Costello, and R. D. Stuart-­ Smith. 2017. “Abundance and Local-­Scale Pro­cesses Contribute to Multi-­phyla Gradients in Global Marine Diversity,” Science Advances 3:e1700419. Edgar, G. J., R. D. Stuart-­Smith, T. J. Willis, S. Kininmonth, S. C. Baker, S. Banks, N. S. Barrett, et al. 2014. “Global Conservation Outcomes Depend on Marine Protected Areas with Five Key Features,” Nature 506:216–220. Edmunds, P. J., and R. C. Carpenter. 2001. “Recovery of Diadema antillarum Reduces Macroalgal Cover and Increases Abundance of Juvenile Corals on a Ca­rib­bean Reef,” Proceedings of the National Acad­emy of Sciences 98:5067–5071. Edwards, C. B., A. M. Friedlander, A. G. Green, M. J. Hardt, E. Sala, H. P. Sweatman, I. D. Williams, B. Zgliczynski, S. A. Sandin, and J. E. Smith. 2014. “Global Assessment of the Status of Coral Reef Herbivorous Fishes: Evidence for Fishing Effects,” Proceedings of the Royal Society B-­Biological Sciences 281:20131835. Edwards, K. F., R. J. Best, and J. J. Stachowicz. 2010. “Prey Diversity Is Associated with Weaker Consumer Effects in a Meta-­analysis of Benthic Marine Experiments,” Ecol­ogy Letters 13:194–201. Edwards, M., and A. J. Richardson. 2004. “Impact of Climate Change on Marine Pelagic Phenology and Trophic Mismatch,” Nature 430:881–884. Ehrlich, P. R., and D. Kennedy. 2005. “Millennium Assessment of ­Human Be­hav­ior,” Science 309:562–563. Einbinder, S., A. Perelberg, O. Ben-­Shaprut, M. H. Foucart, and N. Shashar. 2006. “Effects of Artificial Reefs on Fish Grazing in Their Vicinity: Evidence from Algae Pre­sen­ta­tion Experiments,” Marine Environmental Research 61:110–119. Elahi, R., M. I. O’Connor, J. E. K. Byrnes, J. Dunic, B. K. Eriksson, M. J. S. Hensel, and P. J. Kearns. 2015. “Recent Trends in Local-­Scale Marine Biodiversity Reflect Community Structure and ­Human Impacts,” Current Biology 25:1–6. Eldredge, N., J. N. Thompson, P. M. Brakefield, S. Gavrilets, D. Jablonski, J. B. C. Jackson, R. E. Lenski, B. S. Lieberman, M. A. McPeek,

and W. Miller. 2005. “The Dynamics of Evolutionary Stasis,” Paleobiology 31:133–145. Elliott, J. M., and W. Davison. 1975. “Energy Equivalents of Oxygen Consumption in Animal Energetics,” Oecologia 19:195–201. Ellis, E. C. 2015. “Ecol­ogy in an Anthropogenic Biosphere,” Ecological Monographs 85:287–331. Elton, C. S. 1927. Animal Ecol­ogy. New York: Macmillan. Emmerson, M. C., and D. Raffaelli. 2004. “Predator-­Prey Body Size, Interaction Strength and the Stability of a Real Food Web,” Journal of Animal Ecol­ogy 73:399–409. Enquist, B. J., J. H. Brown, and G. B. West. 1998. “Allometric Scaling of Plant Energetics and Population Density,” Nature 395:163–165. Enquist, B. J., G. B. West, E. L. Charnov, and J. H. Brown. 1999. “Allometric Scaling of Production and Life-­History Variation in Vascular Plants,” Nature 401:907–912. Eriksson, B. K., L. Ljunggren, G. Johansson, J. Mattila, A. Rubach, S. R. Berg, and M. Snickars. 2009. “Declines in Predatory Fish Promote Bloom-­Forming Macroalgae,” Ecological Applications 19:1975–1988. Eriksson, B. K., K. Sieben, J. Eklöf, L. Ljunggren, J. Olsson, M. Casini, and U. Bergström. 2011. “Effects of Altered Offshore Food Webs on Coastal Ecosystems Emphasize the Need for Cross-­Ecosystem Management,” AMBIO: Journal of the ­Human Environment 40:786–797. Erlandson, J. M., and T. C. Rick. 2010. “Archaeology Meets Marine Ecol­ ogy: The Antiquity of Maritime Cultures and ­Human Impacts on Marine Fisheries and Ecosystems,” Annual Review of Marine Science 2:231–251. Estes, J., and J. Palmisano. 1974. “Sea Otters: Their Role in Structuring Nearshore Communities,” Science 185:1058. Estes, J., M. Tinker, T. Williams, and D. Doak. 1998. “Killer Whale Predation on Sea Otters Linking Oceanic and Nearshore Ecosystems,” Science 282:473–476. Estes, J. A., D. O. Duggins, and G. B. Rathbun. 1989. “The Ecol­ogy of Extinctions in Kelp Forest Communities,” Conservation Biology 3:252–264. Estes, J. A., M. Heithaus, D. J. McCauley, D. B. Rasher, and B. Worm. 2016. “Megafaunal Impacts on Structure and Function of Ocean Ecosystems,” Annual Review of Environment and Resources 41:83–116. Estes, J. A., J. Terborgh, J. S. Brashares, M. E. Power, J. Berger, W. J. Bond, S. R. Carpenter, et al. 2011. “Trophic Downgrading of Planet Earth,” Science 333:301–306. Etter, R., and J. Grassle. 1992. “Patterns of Species Diversity in the Deep Sea as a Function of Sediment Particle Size Diversity,” Nature 360(6404):576–578. Evans, K. L., P. H. Warren, and K. J. Gaston. 2005. “Species-­Energy Relationships at the Macroecological Scale: A Review of the Mechanisms,” Biological Reviews 80:1–25. Fabricius, K. E. 2005. “Effects of Terrestrial Runoff on the Ecol­ogy of Corals and Coral Reefs: Review and Synthesis,” Marine Pollution Bulletin 50:125–146. Fabricius, K. E., G. De’ath, S. Noonan, and S. Uthicke. 2014. “Ecological Effects of Ocean Acidification and Habitat Complexity on Reef-­Associated Macroinvertebrate Communities,” Proceedings of the Royal Society B-­Biological Sciences 281(1775):20132479. Fabricius, K. E., C. Langdon, S. Uthicke, C. Humphrey, S. Noonan, G. De’ath, R. Okazaki, N. Muehllehner, M. S. Glas, and J. M. Lough. 2011. “Losers and Winners in Coral Reefs Acclimatized to Elevated Carbon Dioxide Concentrations,” Nature Climate Change 1:165–169.

Literature Cited

Falkowski, P., R. Barber, and V. Smetacek. 1998. “Biogeochemical Controls and Feedbacks on Ocean Primary Production,” Science 281:200–207. Falkowski, P. G., T. Fenchel, and E. F. Delong. 2008. “The Microbial Engines That Drive Earth’s Biogeochemical Cycles,” Science 320:1034–1039. FAO. 2009. The State of World Fisheries and Aquaculture. Rome: UN Food and Agriculture Organ­ization. FAO. 2016. The State of World Fisheries and Aquaculture. Rome: UN Food and Agriculture Organ­ization. Farina, J. M., B. R. Silliman, and M. D. Bertness. 2009. “Can Conservation Biologists Rely on Established Community Structure Rules to Manage Novel Systems? . . . ​Not in Salt Marshes,” Ecological Applications 19:413–422. Fauchald, K., and P. A. Jumars. 1979. “The Diet of Worms: A Study of Polychaete Feeding Guilds,” Oceanography and Marine Biology 17:193–284. Felbeck, H. 1981. “Chemoautotrophic Potential of the Hydrothermal Vent Tube Worm, Riftia pachyptila Jones (Vestimentifera),” Science 213:336–338. Feller, I. 1995. “Effects of Nutrient Enrichment on Growth and Herbivory of Dwarf Red Mangrove (Rhizophora Mangle),” Ecological Monographs 65:477–505. Fenchel, T. 1974. “Intrinsic Rate of Natu­ral Increase: The Relationship with Body Size,” Oecologia 14:317–326. Fernández, M., A. Astorga, S. A. Navarrete, C. Valdovinos, and P. A. Marquet. 2009. “Deconstructing Latitudinal Species Richness Patterns in the Ocean: Does Larval Development Hold the Clue?” Ecol­ogy Letters 12:601–611. Ferrario, F., L. Iveša, A. Jaklin, S. Perkol-­Finkel, and L. Airoldi. 2016. “The Overlooked Role of Biotic ­Factors in Controlling the Ecological Per­for­mance of Artificial Marine Habitats,” Journal of Applied Ecol­ogy 53:16–24. Feulner, G., C. Hallmann, and H. Kienert. 2015. “Snowball Cooling ­after Algal Rise,” Nature Geoscience 8:659–662. Feyrer, F., B. Herbold, S. A. Matern, and P. B. Moyle. 2003. “Dietary Shifts in a Stressed Fish Assemblage: Consequences of a Bivalve Invasion in the San Francisco Estuary,” Environmental Biology of Fishes 67:277–288. Field, C., M. Behrenfeld, J. Randerson, and P. Falkowski. 1998. “Primary Production of the Biosphere: Integrating Terrestrial and Oceanic Components,” Science 281:237–240. Filbee-­Dexter, K., and R. E. Scheibling. 2014. “Sea Urchin Barrens as Alternative Stable States of Collapsed Kelp Ecosystems,” Marine Ecol­ogy Pro­gress Series 495:1–25. Finkel, Z. V., J. Beardall, K. J. Flynn, A. Quigg, T. A. V. Rees, and J. A. Raven. 2010. “Phytoplankton in a Changing World: Cell Size and Elemental Stoichiometry,” Journal of Plankton Research 32:119–137. Finnegan, S., S. C. Anderson, P. G. Harnik, C. Simpson, D. P. Tittensor, J. E. Byrnes, Z. V. Finkel, et al. 2015. “Paleontological Baselines for Evaluating Extinction Risk in the Modern Oceans,” Science 348:567–570. Firth, L. B., A. M. Knights, D. Bridger, A. J. Evans, N. Mieszkowska, P. J. Moore, N. E. O’Connor, E. V. Sheehan, R. C. Thompson, and S. J. Hawkins. 2016. “Ocean Sprawl: Challenges and Opportunities for Biodiversity Management in a Changing World.” Pages 201–278 in R. N. Gibson, R. J. A. Atkinson, and J. D. M. Gordon, editors, Oceanography and Marine Biology. Boca Raton, FL: CRC Press. Fisher, C. R., I. R. MacDonald, R. Sassen, C. M. Young, S. A. Macko, S. Hourdez, R. S. Carney, S. Joye, and E. McMullin.

403

2000. “Methane Ice Worms: Hesiocaeca methanicola Colonizing Fossil Fuel Reserves,” Naturwissenschaften 87:184–187. Fisher, J., K. Frank, and W. Leggett. 2010. “Global Variation in Marine Fish Body Size and Its Role in Biodiversity-­Ecosystem Functioning,” Marine Ecol­ogy Pro­gress Series 405:1–13. Flores-­Moreno, H., P. B. Reich, E. M. Lind, L. L. ­Sullivan, E. W. Seabloom, L. Yahdjian, A. S. Macdougall, et al. 2016. “Climate Modifies Response of Non-­native and Native Species Richness to Nutrient Enrichment,” Philosophical Transactions of the Royal Society B-­Biological Sciences 371:20150273. Flynn, D., N. Mirotchnick, M. Jain, M. Palmer, and S. Naeem. 2011. “Functional and Phyloge­ne­tic Diversity as Predictors of Biodiversity-­Ecosystem-­Function Relationships,” Ecol­ogy 92:1573–1581. Fodrie, F. J., K. L. Heck, S. P. Powers, W. M. Graham, and K. L. Robinson. 2009. “Climate-­Related, Decadal-­Scale Assemblage Changes of Seagrass-­Associated Fishes in the Northern Gulf of Mexico,” Global Change Biology, doi/10.1111/j.1365/2486.2009.01889.x. Fodrie, F. J., L. A. Levin, and A. J. Lucas. 2009. “Use of Population Fitness to Evaluate the Nursery Function of Juvenile Habitats,” Marine Ecol­ogy Pro­gress Series 385:39–49. Foley, M. M., B. S. Halpern, F. Micheli, M. H. Armsby, M. R. Caldwell, C. M. Crain, E. Prahler, et al. 2010. “Guiding Ecological Princi­ples for Marine Spatial Planning,” Marine Policy 34:955–966. Foley, N. S., T. M. van Rensburg, and C. W. Armstrong. 2011. “The Rise and Fall of the Irish Orange Roughy Fishery: An Economic Analy­sis,” Marine Policy 35:756–763. Follows, M. J., S. Dutkiewicz, S. Grant, and S. W. Chisholm. 2007. “Emergent Biogeography of Microbial Communities in a Model Ocean,” Science 315:1843–1846. Fonseca, V. G., G. R. Carvalho, W. Sung, H. F. Johnson, D. M. Power, S. P. Neill, M. Packer, et al. 2010. “Second-­Generation Environmental Sequencing Unmasks Marine Metazoan Biodiversity,” Nature Communications 1:1–8. Forster, D., M. Dunthorn, F. Mahé, J. R. Dolan, S. Audic, D. Bass, L. Bittner, et al. 2016. “Benthic Protists: The Under-­Charted Majority,” FEMS Microbiology Ecol­ogy 92. Forster, J., A. G. Hirst, and G. Woodward. 2011. “Growth and Development Rates Have Dif­fer­ent Thermal Responses,” American Naturalist 178:668–678. Fourqurean, J. W., C. M. Duarte, H. Kennedy, N. Marbà, M. Holmer, M. A. Mateo, E. T. Apostolaki, et al. 2012. “Seagrass Ecosystems as a Globally Significant Carbon Stock,” Nature Geoscience 5:1–5. Fox, H. E., J. L. Holtzman, K. M. Haisfield, C. G. McNally, G. A. Cid, M. B. Mascia, J. E. Parks, and R. S. Pomeroy. 2014. “How Are Our MPAs ­Doing? Challenges in Assessing Global Patterns in Marine Protected Area Per­for­mance,” Coastal Management 42:207–226. Fox, J. W. 2013. “The Intermediate Disturbance Hypothesis Should Be Abandoned,” Trends in Ecol­ogy & Evolution 28:86–92. Frank, K., B. Petrie, and N. Shackell. 2007. “The Ups and Downs of Trophic Control in Continental Shelf Ecosystems,” Trends in Ecol­ ogy & Evolution 22:236–242. Frank, K. T., B. Petrie, J. S. Choi, and W. C. Leggett. 2005. “Trophic Cascades in a Formerly Cod-­Dominated Ecosystem,” Science 308:1621–1623. Frank, K. T., and N. L. Shackell. 2001. “Area-­Dependent Patterns of Finfish Diversity in a Large Marine Ecosystem,” Canadian Journal of Fisheries and Aquatic Sciences 58:1703–1707.

404

Ocean Ecology

Freestone, A. L., R. W. Osman, G. M. Ruiz, and M. E. Torchin. 2011. “Stronger Predation in the Tropics Shapes Species Richness Patterns in Marine Communities,” Ecol­ogy 92:983–993. Freytag, J. K., P. R. Girguis, D. C. Bergquist, J. P. Andras, J. J. Childress, and C. R. Fisher. 2001. “A Paradox Resolved: Sulfide Acquisition by Roots of Seep Tubeworms Sustains Net Chemoautotrophy,” Proceedings of the National Acad­emy of Sciences 98:13408–13413. Friedlander, A., and E. DeMartini. 2002. “Contrasts in Density, Size, and Biomass of Reef Fishes between the Northwestern and the Main Hawaiian Islands: The Effects of Fishing Down Apex Predators,” Marine Ecol­ogy Pro­gress Series 230:253–264. Friedlander, A. M., S. A. Sandin, E. E. DeMartini, and E. Sala. 2010. “Spatial Patterns of the Structure of Reef Fish Assemblages at a Pristine Atoll in the Central Pacific,” Marine Ecol­ogy Pro­gress Series, Oldendorf 410:219–231. Frisk, M. G., T. J. Miller, R. J. Latour, and S. J. D. Martell. 2011. “Assessing Biomass Gains from Marsh Restoration in Delaware Bay Using Ecopath with Ecosim,” Ecological Modelling 222:190–200. Froese, R., and D. Pauly, eds. 2019. FishBase (website). www​.­fishbase​.­org. Frumkin, H., G. N. Bratman, S. J. Breslow, B. Cochran, P. H. Kahn Jr., J. J. Lawler, P. S. Levin, et al. 2017. “Nature Contact and ­Human Health: A Research Agenda,” Environmental Health Perspectives 125: 075001. Fuhrman, J. A., J. A. Steele, I. Hewson, M. S. Schwalbach, M. V. Brown, J. L. Green, and J. H. Brown. 2008. “A Latitudinal Diversity Gradient in Planktonic Marine Bacteria,” Proceedings of the National Acad­emy of Sciences 105:7774–7778. Fulton, E. A. 2010. “Approaches to End-­to-­End Ecosystem Models,” Journal of Marine Systems 81:171–183. Gage, J. D., and P. A. Tyler. 1991. Deep-­Sea Biology: A Natu­ral History of Organisms at the Deep-­Sea Floor. Cambridge, UK: Cambridge University Press. Gaichas, S. K. 2008. “A Context for Ecosystem-­Based Fishery Management: Developing Concepts of Ecosystems and Sustainability,” Marine Policy 32:393–401. Gaines, S. D., B. Gaylord, L. R. Gerber, A. Hastings, and B. P. Kinlan. 2007. “Connecting Places: The Ecological Consequences of Dispersal in the Sea,” Oceanography 20(3):90–99. Gaines, S., and J. Lubchenco. 1982. “A Unified Approach to Marine Plant-­ Herbivore Interactions. II. Biogeography,” Annual Review of Ecol­ogy and Systematics 13:111–138. Gallagher, E. D., P. A. Jumars, and D. D. Trueblood. 1983. “Facilitation of Soft-­Bottom Benthic Succession by Tube Builders,” Ecol­ogy 64:1200–1216. Gamfeldt, L., H. Hillebrand, and P. R. Jonsson. 2005. “Species Richness Changes across Two Trophic Levels Si­mul­ta­neously Affect Prey and Consumer Biomass,” Ecol­ogy Letters 8:696–703. Gamfeldt, L., J. S. Lefcheck, J. E. K. Byrnes, B. J. Cardinale, J. E. Duffy, and J. N. Griffin. 2014. “Marine Biodiversity and Ecosystem Functioning: What’s Known and What’s Next?” Oikos 124:252–265. García Molinos, J., B. S. Halpern, D. S. Schoeman, C. J. Brown, W. Kiessling, P. J. Moore, J. M. Pandolfi, E. S. Poloczanska, A. J. Richardson, and M. T. Burrows. 2016. “Climate Velocity and the ­Future Global Re­distribution of Marine Biodiversity,” Nature Climate Change 6:83–88. Gardner, T. A., I. M. Côté, J. A. Gill, A. Grant, and A. R. Watkinson. 2003. “Long-­Term Region-­Wide Declines in Ca­rib­bean Corals,” Science 301:958–960. Gaston, K. J. 1996. “The Multiple Forms of the Interspecific Abundance-­ Distribution Relationship,” Oikos 76:211–220.

Gause, G. F. 1934. “Experimental Analy­sis of Vito Volterra’s Mathematical Theory of the Strug­gle for Existence,” Science 79:16–17. Gedan, K. B., M. L. Kirwan, E. Wolanski, E. B. Barbier, and B. R. Silliman. 2011. “The Pre­sent and ­Future Role of Coastal Wetland Vegetation in Protecting Shorelines: Answering Recent Challenges to the Paradigm,” Climatic Change 106:7–29. Geider, R. J., E. H. Delucia, P. G. Falkowski, A. C. Finzi, J. P. Grime, J. Grace, T. M. Kana, et al. 2001. “Primary Productivity of Planet Earth: Biological Determinants and Physical Constraints in Terrestrial and Aquatic Habitats,” Global Change Biology 7:849–882. Gentry, R. R., H. E. Froehlich, D. Grimm, P. Kareiva, M. Parke, M. Rust, S. D. Gaines, and B. S. Halpern. 2017. “Mapping the Global Potential for Marine Aquaculture,” Nature Ecol­ogy & Evolution 9: 1317–1324 Gerber, L. R., and S. S. Heppell. 2004. “The Use of Demographic Sensitivity Analy­sis in Marine Species Conservation Planning,” Biological Conservation 120:121–128. Gerber, L. R., S. S. Heppell, F. Ballantyne, and E. Sala. 2005. “The Role of Dispersal and Demography in Determining the Efficacy of Marine Reserves,” Canadian Journal of Fisheries and Aquatic Sciences 62:863–871. Gerlach, S. A., A. E. Hahn, and M. Schrage. 1985. “Size Spectra of Benthic Biomass and Metabolism,” Marine Ecol­ogy Pro­gress Series 26:161–173. Gignoux-­Wolfsohn, S. A., and S. V. Vollmer. 2015. “Identification of Candidate Coral Pathogens on White Band Disease-­Infected Staghorn Coral,” PLOS ONE 10:e0134416. Gilbert, B., T. D. Tunney, K. S. McCann, J. P. DeLong, D. A. Vasseur, V. Savage, J. B. Shurin, et al. 2014. “A Bioenergetic Framework for the Temperature Dependence of Trophic Interaction,” Ecol­ogy Letters 17:902–914. Gillooly, J. F., A. P. Allen, G. B. West, and J. H. Brown. 2005. “The Rate of DNA Evolution: Effects of Body Size and Temperature on the Molecular Clock,” Proceedings of the National Acad­emy of Sciences 102:140–145. Gillooly, J. F., E. L. Charnov, G. B. West, V. M. Savage, and J. H. Brown. 2002. “Effects of Size and Temperature on Developmental Time,” Nature 417:70–73. Gleason, H. A. 1917. “The Structure and Development of the Plant Association,” Bulletin of the Torrey Botanical Club 44:463–481. Gleason, H. A. 1926. “The Individualistic Concept of the Plant Association,” Bulletin of the Torrey Botanical Club 53:7–26. Glibert, P. M., R. Azanza, M. Burford, K. Furuya, E. Abal, A. Al-­Azri, F. Al-­Yamani, et al. 2008. “Ocean Urea Fertilization for Carbon Credits Poses High Ecological Risks,” Marine Pollution Bulletin 56:1049–1056. Gluckman, P. D., F. M. Low, T. Buklijas, M. A. Hanson, and A. S. Beedle. 2011. “How Evolutionary Princi­ples Improve the Understanding of ­Human Health and Disease,” Evolutionary Applications 4:249–263. Glynn, P. W., and L. D’Croz. 1990. “Experimental Evidence for High Temperature Stress as the Cause of El Niño-­Coincident Coral Mortality,” Coral Reefs 8:181–191. Glynn, P. W., and G. M. Wellington. 1983. Corals and Coral Reefs of the Galápagos Islands. Oakland: University of California Press. Goldbogen, J. A., D. E. Cade, J. Calambokidis, A. S. Friedlaender, J. Potvin, P. S. Segre, and A. J. Werth. 2017. “How Baleen Whales Feed: The Biomechanics of Engulfment and Filtration,” Annual Review of Marine Science 9:367–386. Goldbogen, J. A., A. S. Friedlaender, J. Calambokidis, M. F. McKenna, M. Simon, and D. P. Nowacek. 2013. “Integrative Approaches to

Literature Cited

the Study of Baleen Whale Diving Be­hav­ior, Feeding Per­for­ mance, and Foraging Ecol­ogy,” BioScience 63:90–100. Gonzalez, A., B. J. Cardinale, G. R. H. Allington, J. Byrnes, K. A. Endsley, D. G. Brown, D. U. Hooper, F. Isbell, M. I. O’Connor, and M. Loreau. 2016. “Estimating Local Biodiversity Change: A Critique of Papers Claiming No Net Loss of Local Diversity,” Ecol­ogy 97:1949–1960. Gooday, A. J. 1988. “A Response by Benthic Foraminifera to the Deposition of Phytodetritus in the Deep Sea,” Nature 332:70–73. Gooday, A. J. 2002. “Biological Responses to Seasonally Varying Fluxes of Organic ­Matter to the Ocean Floor: A Review,” Journal of Oceanography 58:305–322. Goreau, T. F. 1959. “The Physiology of Skeleton Formation in Corals. I. A Method for Measuring the Rate of Calcium Deposition by Corals under Different Conditions,” The Biological Bulletin 116(1):59–75. Gotelli, N. J. 2008. A Primer of Ecol­ogy, 4th edition. Sunderland, MA: Sinauer Associates. Grace, J. B., T. M. Anderson, E. W. Seabloom, E. T. Borer, P. B. Adler, W. S. Harpole, Y. Hautier, et al. 2016. “Integrative Modelling Reveals Mechanisms Linking Productivity and Plant Species Richness,” Nature 529:390–393. Graham, N., N. K. Dulvy, S. Jennings, and N. Polunin. 2005. “Size-­Spectra as Indicators of the Effects of Fishing on Coral Reef Fish Assemblages,” Coral Reefs 24:118–124. Grassle, J., and N. Maciolek. 1992. “Deep-­Sea Species Richness: Regional and Local Diversity Estimates from Quantitative Bottom Samples,” American Naturalist 139(2):313–341. Graves, J. E., and J. R. McDowell. 1995. “Inter-­ocean Ge­ne­tic Divergence of Istiophorid Billfishes,” Marine Biology 122:193–203. Green, A. L., and D. R. Bellwood. 2009. “Monitoring Functional Groups of Herbivorous Reef Fishes as Indicators of Coral Reef Resilience,” IUCN Resilience Science Group, Working Paper, Series 7. Griffin, J. N., J. E. K. Byrnes, and B. J. Cardinale. 2013. “Effects of Predator Richness on Prey Suppression: A Meta-­Analysis,” Ecol­ogy 94:2180–2187. Griffiths, C. L., P. A. R. Hockey, C. Van Erkom Schurink, and P. J. Le Roux. 1992. “Marine Invasive Aliens on South African Shores: Implications for Community Structure and Trophic Functioning,” South African Journal of Marine Science 12:713–722. Grosholz, E. 2002. “Ecological and Evolutionary Consequences of Coastal Invasions,” Trends in Ecol­ogy & Evolution 17:22–27. Grosholz, E. D., and G. M. Ruiz. 1996. “Predicting the Impact of Introduced Marine Species: Lessons from the Multiple Invasions of the Eu­ro­pean Green Crab Carcinus maenas,” Biological Conservation 78:59–66. Gross, K., B. J. Cardinale, J. W. Fox, A. Gonzalez, M. Loreau, H. W. Polley, P. B. Reich, and J. van Ruijven. 2014. “Species Richness and the Temporal Stability of Biomass Production: A New Analy­sis of Recent Biodiversity Experiments,” American Naturalist 183:1–12. Guisande, C., and R. Harris. 1995. “Effect of Total Organic Content of Eggs on Hatching Success and Naupliar Survival in the Copepod Calanus helgolandicus,” Limnology and Oceanography 40:476–482. Guisande, C., J. Sanchez, I. Maneiro, and A. Miranda. 1996. “Trade-­Off between Offspring Number and Offspring Size in the Marine Copepod Euterpina acutifrons at Dif­fer­ent Food Concentrations,” Marine Ecol­ogy Pro­gress Series, Oldendorf 143:37–44. Guo, H., Y. Zhang, Z. Lan, and S. C. Pennings. 2013. “Biotic Interactions Mediate the Expansion of Black Mangrove (Avicennia germi-

405

nans) into Salt Marshes u­ nder Climate Change,” Global Change Biology 19:2765–2774. Gurevitch, J., J. A. Morrison, and L. V. Hedges. 2000. “The Interaction between Competition and Predation: A Meta-­analysis of Field Experiments,” American Naturalist 155:435–453. Haapkylä, J., R. K. F. Unsworth, M. Flavell, D. G. Bourne, B. Schaffelke, and B. L. Willis. 2011. “Seasonal Rainfall and Runoff Promote Coral Disease on an Inshore Reef,” PLoS ONE 6:e16893. Hachich, N. F., M. B. Bonsall, E. M. Arraut, D. R. Barneche, T. M. Lewinsohn, and S. R. Floeter. 2015. “Island Biogeography: Patterns of Marine Shallow-­Water Organisms in the Atlantic Ocean,” Journal of Biogeography 42:1871–1882. Haddock, S. H., M. A. Moline, and J. F. Case. 2009. “Bioluminescence in the Sea,” Annual Review of Marine Science 2:443–493. Hagen, E. M., K. E. McCluney, K. A. Wyant, C. U. Soykan, A. C. Keller, K. C. Luttermoser, E. J. Holmes, J. C. Moore, and J. L. Sabo. 2012. “A Meta-­analysis of the Effects of Detritus on Primary Producers and Consumers in Marine, Freshwater, and Terrestrial Ecosystems,” Oikos 121:1507–1515. Haidt, J. 2012. The Righ­teous Mind: Why Good ­People Are Divided by Politics and Religion. New York: Pantheon Books. Hairston, N. G., F. E. Smith, and L. B. Slobodkin. 1960. “Community Structure, Population Control, and Competition,” American Naturalist 94:421–425. Hallegraeff, G. M. 2010. “Ocean Climate Change, Phytoplankton Community Responses, and Harmful Algal Blooms: A Formidable Predictive Challenge,” Journal of Phycology 46:220–235. Hall-­Spencer, J. M., R. Rodolfo-­Metalpa, S. Martin, E. Ransome, M. Fine, S. M. Turner, S. J. Rowley, D. Tedesco, and M.-­C. Buia. 2008. “Volcanic Carbon Dioxide Vents Show Ecosystem Effects of Ocean Acidification,” Nature 454:96–99. Halpern, B. S., K. A. Selkoe, C. V. Kappel, F. Micheli, J. F. Bruno, K. S. Casey, R. Fujita, et al. 2008. “A Global Map of ­Human Impact on Marine Ecosystems,” Science 319:948–952. Hamner, W. M., M. S. Jones, and J. H. Carleton. 1988. “Zooplankton, Planktivorous Fish, and ­Water Currents on a Windward Reef Face: ­Great Barrier Reef, Australia,” Bulletin of Marine Science 42:459–479. Hansen, B., P. Bjornsen, and P. Hansen. 1994. “The Size Ratio between Planktonic Predators and Their Prey,” Limnology and Oceanography 39:395–403. Hansen, J., M. Sato, R. Ruedy, K. Lo, D. W. Lea, and M. Medina-Elizade. 2006. “Global Temperature Change,” Proceedings of the National Academy of Sciences 103(39):14288–14293. Hanski, I. 1998. “Metapopulation Dynamics,” Nature 396:41–49. Harari, Y. N. 2014. Sapiens: A Brief History of Humankind. New York: HarperCollins. Hardin, G. 1960. “The Competitive Exclusion Princi­ple,” Science 131:1292–1297. Hardin, G. 1968. “The Tragedy of the Commons,” Science 162:1243–1248. Hardy, A. C. 1924. “The Herring in Relation to Its Animate Environment,” Fisheries Investigations 2:53. Harfoot, M. B. J., T. Newbold, D. P. Tittensor, S. Emmott, J. Hutton, V. Lyutsarev, M. J. Smith, J. P. W. Scharlemann, and D. W. Purves. 2014. “Emergent Global Patterns of Ecosystem Structure and Function from a Mechanistic General Ecosystem Model,” PLoS Biology 12:e1001841. Harley, C. D. G., A. R. Hughes, K. M. Hultgren, B. G. Miner, C. J. B. Sorte, C. S. Thornber, L. F. Rodriguez, L. Tomanek, and S. L. Williams. 2006. “The Impacts of Climate Change in Coastal Marine Systems,” Ecol­ogy Letters 9:228–241.

406

Ocean Ecology

Harnik, P. G., H. K. Lotze, S. C. Anderson, Z. V. Finkel, S. Finnegan, D. R. Lindberg, L. H. Liow, et al. 2012. “Extinctions in Ancient and Modern Seas,” Trends in Ecol­ogy & Evolution 27:608–617. Harrison, P. L., R. C. Babcock, G. D. Bull, J. K. Oliver, C. C. Wallace, and B. L. Willis. 1984. “Mass Spawning in Tropical Reef Corals,” Science 223:1186–1189. Harrison, S., and H. Cornell. 2008. “­Toward a Better Understanding of the Regional ­Causes of Local Community Richness,” Ecol­ogy Letters 11:969–979. Harvell, C. D., C. E. Mitchell, J. R. Ward, S. Altizer, A. P. Dobson, R. S. Ostfeld, and M. D. Samuel. 2002. “Climate Warming and Disease Risks for Terrestrial and Marine Biota,” Science 296:2158–2162. Harvey, P. H., R. K. Colwell, J. W. Silvertown, and R. M. May. 1983. “Null Models in Ecol­ogy,” Annual Review of Ecol­ogy and Systematics 14:189–211. Hastings, A., C. L. Hom, S. Ellner, and P. Turchin. 1993. “Chaos in Ecol­ ogy: Is ­Mother Nature a Strange Attractor?” Annual Review of Ecol­ogy and Systematics 24:1–33. Hata, T., J. S. Madin, V. R. Cumbo, M. Denny, J. Figueiredo, S. Harii, C. J. Thomas, and A. H. Baird. 2017. “Coral Larvae Are Poor Swimmers and Require Fine-­Scale Reef Structure to ­Settle,” Scientific Reports 7:932. Hawkins, B. A., F. S. Albuquerque, M. B. Araújo, J. Beck, L. M. Bini, F. J. Cabrero-­Sañudo, I. Castro-­Parga, et al. 2007. “A Global Evaluation of Metabolic Theory as an Explanation for Terrestrial Species Richness Gradients,” Ecol­ogy 88:1877–1888. Hay, M. 1981. “Herbivory, Algal Distribution, and the Maintenance of Between-­Habitat Diversity on a Tropical Fringing Reef,” American Naturalist 118:520–540. Hay, M. 1986. “Associational Plant Defenses and the Maintenance of Species Diversity: Turning Competitors into Accomplices,” American Naturalist 128:617–641. Hay, M., J. Duffy, and W. Fenical. 1990. “Host-­Plant Specialization Decreases Predation on a Marine Amphipod: An Herbivore in Plant’s Clothing,” Ecol­ogy 71:733–743. Hay, M., J. Pawlik, J. Duffy, and W. Fenical. 1989. “Seaweed-­Herbivore-­ Predator Interactions: Host-­Plant Specialization Reduces Predation on Small Herbivores,” Oecologia 81:418–427. Hay, M., and P. Taylor. 1985. “Competition between Herbivorous Fishes and Urchins on Ca­rib­bean Reefs,” Oecologia 65:591–598. Hay, M. E. 1991. “Marine-­Terrestrial Contrasts in the Ecol­ogy of Plant Chemical Defenses against Herbivores,” Trends in Ecol­ogy & Evolution 6:362–365. Hay, M. E. 1992. “The Role of Seaweed Chemical Defenses in the Evolution of Feeding Specialization and in the Mediation of Complex Interactions.” Pages 93–118 in V. J. Paul, editor, Ecological Roles of Marine Natu­ral Products. Ithaca, NY: Cornell University Press. Hay, M. E. 1996. “Marine Chemical Ecol­ogy: What’s Known and What’s Next?” Journal of Experimental Marine Biology and Ecol­ogy 200:103–134. Hay, M. E., and W. Fenical. 1988. “Marine Plant-­Herbivore Interactions: The Ecol­ogy of Chemical Defense,” Annual Review of Ecol­ogy and Systematics 19:111–145. Hay, M. E., and W. Fenical. 1996. “Chemical Ecol­ogy and Marine Biodiversity: Insights and Products from the Sea,” Oceanography 9:10–20. Hay, M. E., Q. E. Kappel, and W. Fenical. 1994. “Synergisms in Plant Defenses against Herbivores: Interactions of Chemistry, Calcification, and Plant Quality,” Ecol­ogy 75:1714–1726. Hay, M. E., V. J. Paul, S. M. Lewis, K. Gustafson, J. Tucker, and R. N. Trindell. 1988. “Can Tropical Seaweeds Reduce Herbivory by

Growing at Night? Diel Patterns of Growth, Nitrogen Content, Herbivory, and Chemical versus Morphological Defenses,” Oecologia 75:233–245. He, Q., and B. R. Silliman. 2016. “Consumer Control as a Common Driver of Coastal Vegetation Worldwide,” Ecological Monographs 86:278–294. Heaney, L. R. 2000. “Dynamic Disequilibrium: A Long-­Term, Large-­ Scale Perspective on the Equilibrium Model of Island Biogeography,” Global Ecol­ogy and Biogeography 9:59–74. Heath, D. D., J. W. Heath, C. A. Bryden, R. M. Johnson, and C. W. Fox. 2003. “Rapid Evolution of Egg Size in Captive Salmon,” Science 299:1738–1740. Hechinger, R. F., K. D. Lafferty, A. P. Dobson, J. H. Brown, and A. M. Kuris. 2011. “A Common Scaling Rule for Abundance, Energetics, and Production of Parasitic and Free-­Living Species,” Science 333:445–448. Heck, K., G. Hays, and R. Orth. 2003. “Critical Evaluation of the Nursery Role Hypothesis for Seagrass Meadows,” Marine Ecol­ogy Pro­ gress Series 253:123–136. Heck, K. Jr., J. Pennock, J. Valentine, L. Coen, and S. Sklenar. 2000. “Effects of Nutrient Enrichment and Small Predator Density on Seagrass Ecosystems: An Experimental Assessment,” Limnology and Oceanography 45:1041–1057. Heck, K. L. Jr., F. J. Fodrie, S. Madsen, C. J. Baillie, and D. A. Byron. 2015. “Seagrass Consumption by Native and a Tropically Associated Fish Species: Potential Impacts of the Tropicalization of the Northern Gulf of Mexico,” Marine Ecol­ogy Pro­gress Series, Oldendorf 520:165–173. Heinsohn, G. E., J. Wake, H. Marsh, and A. V. Spain. 1977. “The Dugong (Dugong dugon) (Muller) in the Seagrass System,” Aquaculture 12:235–248. Heisler, J., P. M. Glibert, J. Burkholder, and D. M. Anderson. 2008. “Eutrophication and Harmful Algal Blooms: A Scientific Consensus,” Harmful Algae 8:3–13. Hellberg, M. E. 2009. “Gene Flow and Isolation among Populations of Marine Animals,” Annual Review of Ecol­ogy Evolution and Systematics 40:291–310. Helmus, M. R., D. L. Mahler, and J. B. Losos. 2015. “Island Biogeography of the Anthropocene,” Nature 513:543–546. Helmuth, B. 2009. “From Cells to Coastlines: How Can We Use Physiology to Forecast the Impacts of Climate Change?” Journal of Experimental Biology 212:753–760. Hemminga, M. A., and C. M. Duarte. 2000. Seagrass Ecol­ogy. Cambridge, UK: Cambridge University Press. Heppell, S., H. Caswell, and L. Crowder. 2000. “Life Histories and Elasticity Patterns: Perturbation Analy­sis for Species with Minimal Demographic Data,” Ecol­ogy 81:654–665. Herman, P.M.J., J. J. Middelburg, J. Van De Koppel, and C. H. R. Heip. 1999. “Ecol­ogy of Estuarine Macrobenthos,” Advances in Ecological Research 29:195–240. Hessler, R. R., and H. L. Sanders. 1967. “Faunal Diversity in the Deep-­ Sea,” Deep Sea Research 14:65–78. Hewson, I., J. B. Button, B. M. Gudenkauf, B. Miner, A. L. Newton, J. K. Gaydos, J. Wynne, et al. 2014. “Densovirus Associated with Sea-­Star Wasting Disease and Mass Mortality,” Proceedings of the National Acad­emy of Sciences 111:17278–17283. Heyward, A. J., and A. P. Negri. 1999. “Natu­ral Inducers for Coral Larval Metamorphosis,” Coral Reefs 18:273–279. Hiddink, J. G., S. Jennings, M. Sciberras, C. L. Szostek, K. M. Hughes, N. Ellis, A. D. Rijnsdorp, R. A. McConnaughey, T. Mazor, R. Hilborn, J. S. Collie, C. R. Pitcher, R. O. Amoroso, A. M. Parma, P. Suuronen, and M. J. Kaiser. 2017. “Global Analy­sis of Deple-

Literature Cited

tion and Recovery of Seabed Biota ­after Bottom Trawling Disturbance,” Proceedings of the National Acad­emy of Sciences 114:8301–8306. Hilborn, R. 2006. “Faith-­Based Fisheries,” Fisheries 31:554–555. Hilborn, R. 2011. “­Future Directions in Ecosystem Based Fisheries Management: A Personal Perspective,” Fisheries Research 108:235–239. Hilborn, R., and U. Hilborn. 2012. Overfishing: What Every­one Needs to Know. Oxford, UK: Oxford University Press. Hilborn, R., and E. Litzinger. 2009. “­Causes of Decline and Potential for Recovery of Atlantic Cod Populations,” Open Fish Science Journal 2:32–38. Hillebrand, H. 2004a. “Strength, Slope and Variability of Marine Latitudinal Gradients,” Marine Ecol­ogy Pro­gress Series 273:251–267. Hillebrand, H. 2004b. “On the Generality of the Latitudinal Diversity Gradient,” American Naturalist 163:192–211. Hillebrand, H., and B. J. Cardinale. 2004. “Consumer Effects Decline with Prey Diversity,” Ecol­ogy Letters 7:192–201. Hixon, M. A., and M. H. Carr. 1997. “Synergistic Predation, Density Dependence, and Population Regulation in Marine Fish,” Science 277:946–949. Hobbs, R. J., S. Arico, J. Aronson, J. S. Baron, P. Bridgewater, V. A. Cramer, P. R. Epstein, et al. 2006. “Novel Ecosystems: Theoretical and Management Aspects of the New Ecological World Order,” Global Ecol­ogy and Biogeography 15:1–7. Hobbs, R. J., E. Higgs, and J. A. Harris. 2009. “Novel Ecosystems: Implications for Conservation and Restoration,” Trends in Ecol­ogy & Evolution 24:599–605. Hochberg, R., S. Atherton, and A. Kieneke. 2014. “Marine Gastrotricha of ­Little Cayman Island with the Description of One New Species and an Initial Assessment of Meiofaunal Diversity,” Marine Biodiversity 44(1):89–113. Hock, K., N. H. Wolff, J. C. Ortiz, S. A. Condie, K. R. N. Anthony, P. G. Blackwell, and P. J. Mumby. 2017. “Connectivity and Systemic Resilience of the ­Great Barrier Reef,” PLoS Biology 15:e2003355. Holdredge, C., M. D. Bertness, and A. H. Altieri. 2009. “Role of Crab Herbivory in Die-­Off of New ­England Salt Marshes,” Conservation Biology 23:672–679. Holdredge, C., M. D. Bertness, N. C. Herrmann, and K. B. Gedan. 2010. “Fiddler Crab Control of Cordgrass Primary Production in Sandy Sediments,” Marine Ecol­ogy Pro­gress Series 399:253–259. Holland, H. D. 2006. “The Oxygenation of the Atmosphere and Oceans,” Philosophical Transactions of the Royal Society B-­Biological Sciences 361:903–915. Holling, C. S. 1973. “Resilience and Stability of Ecological Systems,” Annual Review of Ecol­ogy and Systematics 4:1–23. Holm, P. 2012. “World War II and the ‘­Great Acceleration’ of North Atlantic Fisheries,” Global Environment 5:66–91. Holt, R. D. 2009. “Predation and Community Organ­ization.” Pages 274– 281 in S. Levin, editor, The Prince­ton Guide to Ecol­ogy. Prince­ ton, NJ: Prince­ton University Press. Honda, K., Y. Nakamura, M. Nakaoka, W. H. Uy, and M. D. Fortes. 2013. “Habitat Use by Fishes in Coral Reefs, Seagrass Beds and Mangrove Habitats in the Philippines,” PLoS ONE 8:e65735. Hood, R. R., E. A. Laws, R. A. Armstrong, N. R. Bates, C. W. Brown, C. A. Carlson, F. Chai, et al. 2006. “Pelagic Functional Group Modeling: Pro­gress, Challenges and Prospects,” Deep Sea Research Part II: Topical Studies in Oceanography 53:459–512. Hooper, D. U., E. C. Adair, B. J. Cardinale, J. E. K. Byrnes, B. A. Hungate, K. L. Matulich, A. Gonzalez, J. E. Duffy, L. Gamfeldt, and M. I.

407

O’Connor. 2012. “A Global Synthesis Reveals Biodiversity Loss as a Major Driver of Ecosystem Change,” Nature 486:105–108. Howarth, R., D. Swaney, G. Billen, J. Garnier, B. Hong, C. Humborg, P. Johnes, C. M. Mörth, and R. Marino. 2012. “Nitrogen Fluxes from the Landscape Are Controlled by Net Anthropogenic Nitrogen Inputs and by Climate,” Frontiers in Ecol­ogy and the Environment 10(1):37–43. Howarth, R. W., and R. Marino. 2006. “Nitrogen as the Limiting Nutrient for Eutrophication in Coastal Marine Ecosystems: Evolving Views over Three De­cades,” Limnology and Oceanography 51:364–376. Hsieh, C.-­H., C. S. Reiss, J. R. Hunter, J. R. Beddington, R. M. May, and G. Sugihara. 2006. “Fishing Elevates Variability in the Abundance of Exploited Species,” Nature 443:859–862. Huang, D., and Roy, K., 2015. “The F ­ uture of Evolutionary Diversity in Reef Corals,” Philosophical Transactions of the Royal Society B-­Biological Sciences, 370(1662):20140010. Hubbell, S. P. 1979. “Tree Dispersion, Abundance, and Diversity in a Tropical Dry Forest,” Science 203:1299–1309. Hubbell, S. P. 1997. “A Unified Theory of Biogeography and Relative Species Abundance and Its Application to Tropical Rain Forests and Coral Reefs,” Coral Reefs 16:9–21. Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography. Prince­ton, NJ: Prince­ton University Press. Hughes, A. R., K. J. Bando, L. F. Rodriguez, and S. L. Williams. 2004. “Relative Effects of Grazers and Nutrients on Seagrasses: A Meta-­ analysis Approach,” Marine Ecol­ogy Pro­gress Series 282:87–99. Hughes, A. R., J. E. Byrnes, and J. J. Stachowicz. 2007. “Reciprocal Relationships and Potential Feedbacks between Biodiversity and Disturbance,” Ecol­ogy Letters 10:849–864. Hughes, A. R., and J. J. Stachowicz. 2004. “Ge­ne­tic Diversity Enhances the Re­sis­tance of a Seagrass Ecosystem to Disturbance,” Proceedings of the National Acad­emy of Sciences 101:8998–9002. Hughes, B. B., R. Eby, E. Van Dyke, M. T. Tinker, C. I. Marks, K. S. Johnson, and K. Wasson. 2013. “Recovery of a Top Predator Mediates Negative Eutrophic Effects on Seagrass,” Proceedings of the National Acad­emy of Sciences 110:15313–15318. Hughes, T., M. Rodrigues, D. Bellwood, D. Ceccarelli, O. Hoeghguldberg, L. McCook, N. Moltschaniwskyj, M. S. Pratchett, R. Steneck, and B. Willis. 2007. “Phase Shifts, Herbivory, and the Resilience of Coral Reefs to Climate Change,” Current Biology 17:360–365. Hughes, T. P. 1994. “Catastrophes, Phase Shifts, and Large-­Scale Degradation of a Ca­rib­bean Coral Reef,” Science 265:1547–1551. Hughes, T. P., M. L. Barnes, D. R. Bellwood, J. E. Cinner, G. S. Cumming, J. B. C. Jackson, J. Kleypas, et al. 2017. “Coral Reefs in the Anthropocene,” Nature 546:82–90. Hughes, T. P., S. Carpenter, J. Rockström, M. Scheffer, and B. Walker. 2013. “Multiscale Regime Shifts and Planetary Bound­aries,” Trends in Ecol­ogy & Evolution 28:389–395. Hughes, T. P., J. T. Kerry, M. Álvarez-­Noriega, J. G. Álvarez-­Romero, K. D. Anderson, A. H. Baird, R. C. Babcock, et al. 2017. “Global Warming and Recurrent Mass Bleaching of Corals,” Nature 543:373–377. Hughes, T. P., and J. E. Tanner. 2000. “Recruitment Failure, Life Histories, and Long-­Term Decline of Ca­rib­bean Corals,” Ecol­ogy 81:2250–2263. Huisman, J., and F. J. Weissing. 1999. “Biodiversity of Plankton by Species Oscillations and Chaos,” Nature 402:407–410. Hunt, G., T. M. Cronin, and K. Roy. 2005. “Species-­Energy Relationship in the Deep Sea: A Test Using the Quaternary Fossil Rec­ord,” Ecol­ogy Letters 8:739–747.

408

Ocean Ecology

Huston, M. 1979. “A General Hypothesis of Species Diversity,” American Naturalist 113:81. Huston, M. A. 1994. Biological Diversity: The Coexistence of Species. Cambridge, UK: Cambridge University Press. Hutchings, J. A., and D. J. Fraser. 2008. “The Nature of Fisheries-­and Farming-­Induced Evolution,” Molecular Ecol­ogy 17:294–313. Hutchinson, G. 1961. “The Paradox of the Plankton,” American Naturalist 95:137–145. Hutchinson, G. E. 1957. “Concluding Remarks. Population Studies: Animal Ecol­ogy and Demography,” Cold Spring Harbor Symposium on Quantitative Biology 22:415–457. Hutchinson, G. E. 1959. “Homage to Santa Rosalia or Why Are ­There So Many Kinds of Animals?” American Naturalist 93:145–159. Huxley, J. S. 1932. Prob­lems of Relative Growth. London: Methuen and Com­pany. Idjadi, J. A., S. C. Lee, J. F. Bruno, W. F. Precht, L. Allen-­Requa, and P. J. Edmunds. 2006. “Rapid Phase-­Shift Reversal on a Jamaican Coral Reef,” Coral Reefs 25:209–211. Igulu, M. M., I. Nagelkerken, G. van der Velde, and Y. D. Mgaya. 2013. “Mangrove Fish Production Is Largely Fuelled by External Food Sources: A Stable Isotope Analy­sis of Fishes at the Individual, Species, and Community Levels from Across the Globe,” Ecosystems 16:1336–1352. Ingram, T., and J. B. Shurin. 2009. “Trait-­Based Assembly and Phyloge­ne­ tic Structure in Northeast Pacific Rockfish Assemblages,” Ecol­ ogy 90:2444–2453. Irigoien, X. 2005. “Phytoplankton Blooms: A ‘Loophole’ in Microzooplankton Grazing Impact?” Journal of Plankton Research 27:313–321. Irigoien, X., J. Huisman, and R. Harris. 2004. “Global Biodiversity Patterns of Marine Phytoplankton and Zooplankton,” Nature 429:863–867. Irwin, A. J., Z. V. Finkel, F. E. Muller-­Karger, and L. Troccoli Ghinaglia. 2015. “Phytoplankton Adapt to Changing Ocean Environments,” Proceedings of the National Acad­emy of Sciences 112:5762–5766. Jablonski, D., and R. A. Lutz. 1983. “Larval Ecol­ogy of Marine Benthic Invertebrates: Paleobiological Implications,” Biological Reviews 58:21–89. Jablonski, D., and K. Roy. 2003. “Geo­graph­i­cal Range and Speciation in Fossil and Living Molluscs,” Proceedings of the Royal Society B-­Biological Sciences 270:401–406. Jablonski, D., K. Roy, and J. W. Valentine. 2006. “Out of the Tropics: Evolutionary Dynamics of the Latitudinal Diversity Gradient,” Science 314:102–106. Jablonski, D., and J. J. Sepkoski. 1996. “Paleobiology, Community Ecol­ ogy, and Scales of Ecological Pattern,” Ecol­ogy 77:1367–1378. Jackson, J. 1992. “Pleistocene Perspectives on Coral Reef Community Structure,” American Zoologist 32:719. Jackson, J., M. Donovan, K. Cramer, and V. Lam. 2014. Status and Trends of Ca­rib­bean Coral Reefs: 1970–2012. Gland, Switzerland: IUCN, Global Coral Reef Monitoring Network. Jackson, J. B., M. X. Kirby, W. H. Berger, K. A. Bjorndal, L. W. Botsford, B. J. Bourque, R. H. Bradbury, et al. 2001. “Historical Overfishing and the Recent Collapse of Coastal Ecosystems,” Science 293:629–637. Jackson, J. B. C. 1997. “Reefs since Columbus,” Coral Reefs 16:23–32. Jenkins, S., P. Moore, M. Burrows, D. Garbary, S. Hawkins, A. Ingolfsson, K. Sebens, P. Snelgrove, D. Wethey, and S. Woodin. 2008. “Comparative Ecol­ogy of North Atlantic Shores: Do Differences in Players ­Matter for Pro­cess?” Ecol­ogy 89:3–23.

Jennings, S., and J. L. Blanchard. 2004. “Fish Abundance with No Fishing: Predictions Based on Macroecological Theory,” Journal of Animal Ecol­ogy 73:632–642. Jennings, S., and K. Collingridge. 2015. “Predicting Consumer Biomass, Size-­Structure, Production, Catch Potential, Responses to Fishing and Associated Uncertainties in the World’s Marine Ecosystems,” PLoS ONE 10:e0133794. Jennings, S., S. Greenstreet, and J. Reynolds. 1999. “Structural Change in an Exploited Fish Community: A Consequence of Differential Fishing Effects on Species with Contrasting Life Histories,” Journal of Animal Ecol­ogy 68:617–627. Jennings, S., and M. J. Kaiser. 1998. “The Effects of Fishing on Marine Ecosystems,” Advances in Marine Biology 34:201–212. Jennings, S., and S. Mackinson. 2003. “Abundance–­Body Mass Relationships in Size-­Structured Food Webs,” Ecol­ogy Letters 6:971–974. Jennings, S., F. Mélin, J. L. Blanchard, R. M. Forster, N. K. Dulvy, and R. W. Wilson. 2008. “Global-­Scale Predictions of Community and Ecosystem Properties from ­Simple Ecological Theory,” Proceedings of the Royal Society B-­Biological Sciences 275:1375–1383. Jennings, S., J. Pinnegar, N. Polunin, and K. Warr. 2002. “Linking Size-­ Based and Trophic Analyses of Benthic Community Structure,” Marine Ecol­ogy Pro­gress Series 226:77–85. Jennings, S., J. K. Pinnegar, N. V. C. Polunin, and T. W. Boon. 2001. “Weak Cross-­Species Relationships between Body Size and Trophic Level Belie Power­ful Size-­Based Trophic Structuring in Fish Communities,” Journal of Animal Ecol­ogy 70:934–944. Jennings, S., J. Reynolds, and S. Mills. 1998. “Life History Correlates of Responses to Fisheries Exploitation,” Proceedings of the Royal Society B-­Biological Sciences. Jennings, S., G. D. Stentiford, A. M. Leocadio, K. R. Jeffery, J. D. Metcalfe, I. Katsiadaki, N. A. Auchterlonie, et al. 2016. “Aquatic Food Security: Insights into Challenges and Solutions from an Analy­ sis of Interactions between Fisheries, Aquaculture, Food Safety, ­Human Health, Fish and ­Human Welfare, Economy and Environment,” Fish and Fisheries 17:893–938. Jennings, S., K. J. Warr, and S. Mackinson. 2002. “Use of Size-­Based Production and Stable Isotope Analyses to Predict Trophic Transfer Efficiencies and Predator-­Prey Body Mass Ratios in Food Webs,” Marine Ecol­ogy Pro­gress Series, Oldendorf 240:11–20. Jetz, W., C. Carbone, J. Fulford, and J. H. Brown. 2004. “The Scaling of Animal Space Use,” Science 306:266–268. Jiang, L., and Z. Pu. 2009. “Dif­fer­ent Effects of Species Diversity on Temporal Stability in Single-­Trophic and Multitrophic Communities,” American Naturalist 174:651–659. Jiao, N., G. J. Herndl, D. A. Hansell, R. Benner, G. Kattner, S. W. Wilhelm, D. L. Kirchman, et al. 2010. “Microbial Production of Recalcitrant Dissolved Organic ­Matter: Long-­Term Carbon Storage in the Global Ocean,” Nature Reviews Microbiology 8:593–599. Johannesson, K., M. Panova, P. Kemppainen, C. Andre, E. Rolán-­Alvarez, and R. K. Butlin. 2010. “Repeated Evolution of Reproductive Isolation in a Marine Snail: Unveiling Mechanisms of Speciation,” Philosophical Transactions of the Royal Society B-­Biological Sciences 365:1735–1747. Johansson, J. O. R., and I. Lewis R. R. 1992. “Recent Improvements of ­Water Quality and Biological Indicators in Hillsborough Bay, a Highly Impacted Subdivision of Tampa Bay, Florida,” Science of the Total Environment (Suppl) 1992:1199–1215. Johnson, D. W., and M. A. Hixon. 2011. “Sexual and Lifetime Se­lection on Body Size in a Marine Fish: The Importance of Life-­History Trade-­Offs,” Journal of Evolutionary Biology 24:1653–1663.

Literature Cited

Johnson, Z. I., E. R. Zinser, A. Coe, N. P. McNulty, E. M. S. Woodward, and S. W. Chisholm. 2006. “Niche Partitioning among Prochlorococcus Ecotypes along Ocean-­Scale Environmental Gradients,” Science 311:1737–1740. Jones, A. G., G. I. Moore, C. Kvarnemo, D. Walker, and J. C. Avise. 2003. “Sympatric Speciation as a Consequence of Male Pregnancy in Sea­horses,” Proceedings of the National Acad­emy of Sciences 100:6598–6603. Jones, H. P., and O. J. Schmitz. 2009. “Rapid Recovery of Damaged Ecosystems,” PLoS ONE 4:e5653. Jones, M. C., and W. W. L. Cheung. 2015. “Multi-­model Ensemble Projections of Climate Change Effects on Global Marine Biodiversity,” ICES Journal of Marine Science 72:741–752. Jordà, G., N. Marbà, and C. M. Duarte. 2012. “Mediterranean Seagrass Vulnerable to Regional Climate Warming,” Nature Climate Change 2:1–4. Jørgensen, C., K. Enberg, E. S. Dunlop, R. Arlinghaus, D. S. Boukal, K. Brander, B. Ernande, et al. 2007. “Ecol­ogy: Managing Evolving Fish Stocks,” Science 318:1247–1248. Jorgensen, S. J., C. A. Reeb, T. K. Chapple, S. Anderson, C. Perle, S. R. Van Sommeran, C. Fritz-­Cope, A. C. Brown, A. P. Klimley, and B. A. Block. 2009. “Philopatry and Migration of Pacific White Sharks,” Proceedings of the Royal Society B-­Biological Sciences 277:679–688. Jumars, P. A., K. M. Dorgan, and S. M. Lindsay. 2015. “Diet of Worms Emended: An Update of Polychaete Feeding Guilds,” Annual Review of Marine Science 7:497–520. Kaldy, J. E. 2014. “Effect of Temperature and Nutrient Manipulations on Eelgrass Zostera marina L. from the Pacific Northwest, USA,” Journal of Experimental Marine Biology and Ecol­ogy 453:108–115. Karlson, R., H. Cornell, and T. Hughes. 2004. “Coral Communities Are Regionally Enriched along an Oceanic Biodiversity Gradient,” Nature 429:867–870. Kashtan, N., S. E. Roggensack, S. Rodrigue, J. W. Thompson, S. J. Biller, A. Coe, H. Ding, et al. 2014. “Single-­Cell Genomics Reveals Hundreds of Coexisting Subpopulations in Wild Prochlorococcus,” Science 344:416–420. Kasting, J. F., and J. L. Siefert. 2002. “Life and the Evolution of Earth’s Atmosphere,” Science 296:1066–1068. Kathiresan, K., and B. L. Bingham. 2001. “Biology of Mangroves and Mangrove Ecosystems,” Advances in Marine Biology 40:81–251. Kattge, J., S. Díaz, S. Lavorel, I. C. Prentice, P. Leadley, G. Bönisch, E. Garnier, et al. 2011. “TRY—­A Global Database of Plant Traits,” Global Change Biology 17:2905–2935. Keddy, P. A. 1992. “Assembly and Response Rules: Two Goals for Predictive Community Ecol­ogy,” Journal of Vegetation Science 3:157–164. Keesing, F., L. K. Belden, P. Daszak, A. Dobson, C. D. Harvell, R. D. Holt, P. Hudson, et al. 2010. “Impacts of Biodiversity on the Emergence and Transmission of Infectious Diseases,” Nature 468:647–652. Keith, D. A., J. P. Rodriguez, K. M. Rodríguez-­Clark, E. Nicholson, K. Aapala, A. Alonso, M. Asmussen, et al. 2013. “Scientific Foundations for an IUCN Red List of Ecosystems,” PLoS ONE 8:e62111. Keith, S. A., A. H. Baird, T. P. Hughes, J. S. Madin, and S. R. Connolly. 2013. “Faunal Breaks and Species Composition of Indo-­Pacific Corals: The Role of Plate Tectonics, Environment and Habitat Distribution,” Proceedings of the Royal Society B-­Biological Sciences 280:20130818. Kelleway, J. J., N. Saintilan, P. I. Macreadie, C. G. Skilbeck, A. Zawadzki, and P. J. Ralph. 2015. “Seventy Years of Continuous Encroachment Substantially Increases ‘Blue Carbon’ Capacity as Man-

409

groves Replace Intertidal Salt Marshes,” Global Change Biology 22:1097–1109. Kelly, L. W., K. L. Barott, E. Dinsdale, A. M. Friedlander, B. Nosrat, D. Obura, E. Sala, et al. 2011. “Black Reefs: Iron-­Induced Phase Shifts on Coral Reefs,” ISME Journal 6:638–649. Kelly, R., and S. Palumbi. 2010. “Ge­ne­tic Structure among 50 Species of the Northeastern Pacific Rocky Intertidal Community,” PLoS ONE 5:e8594. Keough, M. J., G. P. Quinn, and A. King. 1993. “Correlations between ­Human Collecting and Intertidal Mollusc Populations on Rocky Shores,” Conservation Biology 7:378–390. Kimmerer, W. 2006. “Response of Anchovies Dampens Effects of the Invasive Bivalve Corbula Amurensis on the San Francisco Estuary Foodweb,” Marine Ecol­ogy Pro­gress Series 324:207–218. Kimmerer, W. J., E. Gartside, and J. J. Orsi. 1994. “Predation by an Introduced Clam as the Likely Cause of Substantial Declines in Zooplankton of San Francisco Bay,” Marine Ecol­ogy Pro­gress Series 113:81–93. King, S. A., and J. N. Lester. 1995. “The Value of Salt Marsh as a Sea Defence,” Marine Pollution Bulletin 30:180–189. Kingsford, M. J., J. M. Leis, and A. Shanks. 2002. “Sensory Environments, Larval Abilities and Local Self-­Recruitment,” Bulletin of Marine Science 70:309–340. Kinlan, B., and S. Gaines. 2003. “Propagule Dispersal in Marine and Terrestrial Environments: A Community Perspective,” Ecol­ogy 84:2007–2020. Kinsey, D. W., and P. J. Davies. 1979. “Effects of Elevated Nitrogen and Phosphorus on Coral Reef Growth,” Limnology and Oceanography 24:935–940. Kirwan, M. L., and J. P. Megonigal. 2013. “Tidal Wetland Stability in the Face of ­Human Impacts and Sea-­Level Rise,” Nature 504:53–60. Kirwan, M. L., A. B. Murray, J. P. Donnelly, and D. R. Corbett. 2011. “Rapid Wetland Expansion during Eu­ro­pean Settlement and Its Implication for Marsh Survival ­under Modern Sediment Delivery Rates,” Geology 39:507–510. Kleiber, M. 1932. “Body Size and Metabolism,” Hilgardia 6:315–353. Knies, J., P. Cabedo-­Sanz, S. T. ­Belt, S. Baranwal, S. Fietz, and A. Rosell-­ Melé. 2014. “The Emergence of Modern Sea Ice Cover in the Arctic Ocean,” Nature Communications 5:5608. Knoll, A. H. 2003. Life on a Young Planet: The First Three Billion Years of Evolution on Earth. Prince­ton, NJ: Prince­ton University Press. Knowlton, N. 1993. “Sibling Species in the Sea,” Annual Review of Ecol­ogy and Systematics 24:189–216. Knowlton, N. 2004. “Multiple ‘Stable’ States and the Conservation of Marine Ecosystems,” Pro­gress in Oceanography 60:387–396. Knowlton, N., and J. B. C. Jackson. 1994. “New Taxonomy and Niche Partitioning on Coral Reefs: Jack of All Trades or Master of Some?” Trends in Ecol­ogy & Evolution 9:7–9. Koch, M., G. Bowes, C. Ross, and X.-­H. Zhang. 2012. “Climate Change and Ocean Acidification Effects on Seagrasses and Marine Macroalgae,” Global Change Biology 19:103–132. Kolar, C. S., and D. M. Lodge. 2002. “Ecological Predictions and Risk Assessment for Alien Fishes in North Amer­i­ca,” Science 298:1233–1236. Kollars, N. M., A. K. Henry, M. A. Whalen, K. E. Boyer, M. Cusson, J. S. Eklöf, C. M. Hereu, et al. 2017. “Meta-­analysis of Reciprocal Linkages between Temperate Seagrasses and Waterfowl with Implications for Conservation,” Frontiers in Plant Science 8:269. Kopp, R. E., A. C. Kemp, K. Bittermann, B. P. Horton, J. P. Donnelly, W. R. Gehrels, C. C. Hay, J. X. Mitrovica, E. D. Morrow, and S. Rahmstorf. 2016. “Temperature-­Driven Global Sea-­Level

410

Ocean Ecology

Variability in the Common Era,” Proceedings of the National Acad­emy of Sciences 113:E1434–­E1441. Kordas, R. L., C. D. G. Harley, and M. I. O’Connor. 2011. “Community Ecol­ogy in a Warming World: The Influence of Temperature on Interspecific Interactions in Marine Systems,” Journal of Experimental Marine Biology and Ecol­ogy 400:218–226. Korpinen, S., T. Honkanen, O. Vesakoski, A. Hemmi, R. Koivikko, J. Loponen, and V. Jormalainen. 2007. “Macroalgal Communities Face the Challenge of Changing Biotic Interactions: Review with Focus on the Baltic Sea,” AMBIO 36:203–211. Koslow, J., G. W. Boehlert, J. D. M. Gordon, R. L. Haedrich, P. Lorance, and N. Parin. 2000. “Continental Slope and Deep-­Sea Fisheries: Implications for a Fragile Ecosystem,” ICES Journal of Marine Science 57:548–557. Kraft, N. J. B., P. B. Adler, O. Godoy, E. C. James, S. Fuller, and J. M. Levine. 2015. “Community Assembly, Coexistence and the Environmental Filtering Meta­phor,” Functional Ecol­ogy 29:592–599. Krausmann, F., K.-­H. Erb, S. Gingrich, H. Haberl, A. Bondeau, V. Gaube, C. Lauk, C. Plutzar, and T. D. Searchinger. 2013. “Global ­Human Appropriation of Net Primary Production Doubled in the 20th ­Century,” Proceedings of the National Acad­emy of Sciences 110:10324–10329. Kristensen, R. M. 2002. “An Introduction to Loricifera, Cycliophora, and Micrognathozoa,” Integrative and Comparative Biology 42:641–651. Kritzer, J. P., M.-­B. DeLucia, E. Greene, C. Shumway, M. F. Topolski, J. Thomas-­Blate, L. A. Chiarella, K. B. Davy, and K. Smith. 2016. “The Importance of Benthic Habitats for Coastal Fisheries,” BioScience 66:274–284. Kroeker, K. J., R. L. Kordas, R. N. Crim, and G. G. Singh. 2010. “Meta-­ analysis Reveals Negative Yet Variable Effects of Ocean Acidification on Marine Organisms,” Ecol­ogy Letters 13:1419–1434. Kroeker, K. J., F. Micheli, and M. C. Gambi. 2012. “Ocean Acidification ­Causes Ecosystem Shifts via Altered Competitive Interactions,” Nature Climate Change 2:1–4. Krug, A. Z., D. Jablonski, J. W. Valentine, and K. Roy. 2009. “Generation of Earth’s First-­Order Biodiversity Pattern,” Astrobiology 9:113–124. Kruk, C., and A. M. Segura. 2012. “The Habitat Template of Phytoplankton Morphology-­Based Functional Groups,” Hydrobiologia 698:191–202. Krylov, A. A., I. A. Andreeva, C. Vogt, J. Backman, V. V. Krupskaya, G. E. Grikurov, K. Moran, and H. Shoji. 2008. “A Shift in Heavy and Clay Mineral Provenance Indicates a ­Middle Miocene Onset of a Perennial Sea Ice Cover in the Arctic Ocean.” Paleoceanography 23:PA1S06. Kuffner, I. B., A. J. Andersson, P. L. Jokiel, K. S. Rod­gers, and F. T. Mackenzie. 2007. “Decreased Abundance of Crustose Coralline Algae Due to Ocean Acidification,” Nature Geoscience 1:114–117. Kurlansky, M. 2001. Cod: A Biography of the Fish That Changed the World. New York: Penguin Books. Ladner, J. T., and S. R. Palumbi. 2012. “Extensive Sympatry, Cryptic Diversity and Introgression throughout the Geographic Distribution of Two Coral Species Complexes,” Molecular Ecol­ogy 21:2224–2238. Lafferty, K. D., A. P. Dobson, and A. M. Kuris. 2006. “Parasites Dominate Food Web Links,” Proceedings of the National Acad­emy of Sciences 103:11211–11216. Lafferty, K. D., and T. H. Suchanek. 2016. “Revisiting Paine’s 1966 Sea Star Removal Experiment, the Most-­Cited Empirical Article in the American Naturalist,” American Naturalist 188:365–378.

Lam, V. W. Y., W. W. L. Cheung, G. Reygondeau, and U. R. Sumaila. 2016. “Projected Change in Global Fisheries Revenues ­under Climate Change,” Scientific Reports 6:32607. Lamb, J. B., J. A. van de ­Water, D. G. Bourne, C. Altier, M. Y. Hein, E. A. Fiorenza, N. Abu, J. Jompa, and C. D. Harvell. 2017. “Seagrass Ecosystems Reduce Exposure to Bacterial Pathogens of ­Humans, Fishes, and Invertebrates,” Science 355:731–733. Lamb, J. B., B. L. Willis, E. A. Fiorenza, C. S. Couch, R. Howard, D. N. Rader, J. D. True, et al. 2018. “Plastic Waste Associated with Disease on Coral Reefs,” Science 359:460–462. Lampert, W. 1989. “The Adaptive Significance of Diel Vertical Migration of Zooplankton,” Functional Ecol­ogy 3:21–27. Landry, M. R., and R. P. Hassett. 1982. “Estimating the Grazing Impact of Marine Micro-­zooplankton,” Marine Biology 67:283–288. Lane, N. 2015. The Vital Question: Energy, Evolution, and the Origins of Complex Life. New York: W. W. Norton and Com­pany. Larkin, P. A. 1977. “An Epitaph for the Concept of Maximum Sustained Yield,” Transactions of the American Fisheries Society 106:1–11. Larson, A. 2009. “Adaptation.” Pages 93–108 in S. Levin, editor, The Prince­ton Guide to Ecol­ogy. Prince­ton, NJ: Prince­ton University Press. Lau, K. V., K. Maher, D. Altiner, B. M. Kelley, L. R. Kump, D. J. Lehrmann, J. C. Silva-­Tamayo, K. L. Weaver, M. Yu, and J. L. Payne. 2016. “Marine Anoxia and Delayed Earth System Recovery ­after the End-­Permian Extinction,” Proceedings of the National Acad­emy of Sciences 113:2360–2365. Lavery, T. J., B. Roudnew, P. Gill, J. Seymour, L. Seuront, G. Johnson, J. G. Mitchell, and V. Smetacek. 2010. “Iron Defecation by Sperm Whales Stimulates Carbon Export in the Southern Ocean,” Proceedings of the Royal Society B-­Biological Sciences 277:3527–3531. Lavorel, S., and E. Garnier. 2002. “Predicting Changes in Community Composition and Ecosystem Functioning from Plant Traits: Revisiting the Holy Grail,” Functional Ecol­ogy 16:545–556. Law, R. 2000. “Fishing, Se­lection, and Phenotypic Evolution,” ICES Journal of Marine Science 57:659–668. Laws, E. A. 2013. “Evaluation of In Situ Phytoplankton Growth Rates: A Synthesis of Data from Varied Approaches,” Annual Review of Marine Science 5:247–268. Lawton, J. 1999. “Are Th ­ ere General Laws in Ecol­ogy?” Oikos 84:177–192. Lee, S. Y. 1995. “Mangrove Outwelling: A Review,” Hydrobiologia 295:203–212. Lee, S. Y. 1998. “Ecological Role of Grapsid Crabs in Mangrove Ecosystems: A Review,” Marine and Freshwater Research 49:335–343. Lefcheck, J. S., J. E. K. Byrnes, F. Isbell, L. Gamfeldt, J. N. Griffin, N. Eisenhauer, M. J. S. Hensel, A. Hector, B. J. Cardinale, and J. E. Duffy. 2015. “Biodiversity Enhances Ecosystem Multifunctionality across Trophic Levels and Habitats,” Nature Communications 6:6936. Lefcheck, J. S., B. B. Hughes, A. J. Johnson, B. W. Pfirrmann, D. B. Rasher, A. R. Smyth, B. L. Williams, M. W. Beck, and R. J. Orth. 2019. “Are Coastal Habitats Impor­tant Nurseries? A Meta-­analysis,” Conservation Letters (2019):e12645. Lefcheck, J. S., R. J. Orth, W. C. Dennison, D. J. Wilcox, R. R. Murphy, J. Keisman, C. Gurbisz, et al. 2018. “Long-­Term Nutrient Reductions Lead to the Unpre­ce­dented Recovery of a Temperate Coastal Region,” Proceedings of the National Acad­emy of Sciences 115:3658–3662. Lefkovitch, L. P. 1965. “The Study of Population Growth in Organisms Grouped by Stages,” Biometrics 21:1–18. Leggett, W. C., and E. Deblois. 1994. “Recruitment in Marine Fishes: Is It Regulated by Starvation and Predation in the Egg and Larval Stages?” Netherlands Journal of Sea Research 32:119–134.

Literature Cited

Leibold, M. A. 1989. “Resource Edibility and the Effects of Predators and Productivity on the Outcome of Trophic Interactions,” American Naturalist 134:922–949. Leibold, M. A. 1996. “A Graphical Model of Keystone Predators in Food Webs: Trophic Regulation of Abundance, Incidence, and Diversity Patterns in Communities,” American Naturalist 147(5):784–812. Leibold, M. A., M. Holyoak, N. Mouquet, P. Amarasekare, J. M. Chase, M. F. Hoopes, R. D. Holt, et al. 2004. “The Metacommunity Concept: A Framework for Multi-­scale Community Ecol­ogy,” Ecol­ogy Letters 7:601–613. Leibold, M. A., and M. A. McPeek. 2006. “Coexistence of the Niche and Neutral Perspectives in Community Ecol­ogy,” Ecol­ogy 87:1399–1410. Leigh, E., R. Paine, J. Quinn, and T. Suchanek. 1987. “Wave Energy and Intertidal Productivity,” Proceedings of the National Acad­emy of Sciences 84:1314. Lenihan, H. S. 1999. “Physical-­Biological Coupling on Oyster Reefs: How Habitat Structure Influences Individual Per­for­mance,” Ecological Monographs 69:251–275. Lenihan, H. S., and F. Micheli. 2001. “Soft-­Sediment Communities.” Pages 253–287 in M. D. Bertness, S. Gaines, and M. E. Hay, editors, Marine Community Ecol­ogy. Sunderland, MA: Sinauer Associates. Leonard, G., J. Levine, P. Schmidt, and M. Bertness. 1998. “Flow-­Driven Variation in Intertidal Community Structure in a Maine Estuary,” Ecol­ogy 79:1395–1411. Le Quéré, C., S. P. Harrison, I. C. Prentice, E. T. Buitenhuis, O. Aumont, L. Bopp, H. Claustre, et al. 2005. “Ecosystem Dynamics Based on Plankton Functional Types for Global Ocean Biogeochemistry Models,” Global Change Biology 11:2016–2040. Leray, M., J. T. Boehm, S. C. Mills, and C. P. Meyer. 2011. “Moorea BIOCODE Barcode Library as a Tool for Understanding Predator-­ Prey Interactions: Insights into the Diet of Common Predatory Coral Reef Fishes,” Coral Reefs 31:383–388. Leray, M., and N. Knowlton. 2015. “DNA Barcoding and Metabarcoding of Standardized Samples Reveal Patterns of Marine Benthic Diversity,” Proceedings of the National Acad­emy of Sciences 112:2076–2081. Leslie, P. H. 1945. “On the Use of Matrices in Certain Population Mathe­ matics,” Biometrika 33:183–212. Lessios, H. A. 1988. “Mass Mortality of Diadema antillarum in the Ca­rib­ be­an: What Have We Learned?” Annual Review of Ecol­ogy and Systematics 19:371–393. Lessios, H. A. 2008. “The ­Great American Schism: Divergence of Marine Organisms ­after the Rise of the Central American Isthmus,” Annual Review of Ecol­ogy Evolution and Systematics 39:63–91. Lessios, H. A. 2016. “The ­Great Diadema antillarum Die-­Off: 30 Years ­Later,” Annual Review of Marine Science 8:267–283. Lessios, H. A., D. R. Robertson, and J. D. Cubit. 1984. “Spread of Diadema Mass Mortality through the Ca­rib­bean,” Science 226:335–337. Lester, S. E., B. S. Halpern, K. Grorud-­Colvert, J. Lubchenco, B. I. Ruttenberg, S. D. Gaines, S. Airamé, and R. R. Warner. 2009. “Biological Effects within No-­Take Marine Reserves: A Global Synthesis,” Marine Ecol­ogy Pro­gress Series 384:33–46. Lester, S. E., B. I. Ruttenberg, S. D. Gaines, and B. P. Kinlan. 2007. “The Relationship between Dispersal Ability and Geographic Range Size,” Ecol­ogy Letters 10:745–758. Levin, L., R. Etter, M. Rex, A. Gooday, and C. Smith. 2001. “Environmental Influences on Regional Deep-­Sea Species Diversity,” Annual Review of Ecol­ogy and Systematics 32:51–93.

411

Levin, L. A. 2005. “Ecol­ogy of Cold Seep Sediments: Interactions of Fauna with Flow, Chemistry and Microbes,” Oceanography and Marine Biology: An Annual Review 43:1–46. Levin, L. A. 2006. “Recent Pro­gress in Understanding Larval Dispersal: New Directions and Digressions,” Integrative and Comparative Biology 46:282–297. Levin, L. A., and M. Sibuet. 2012. “Understanding Continental Margin Biodiversity: A New Imperative,” Annual Review of Marine Science 4:79–112. Levin, P. S., I. Kaplan, R. Grober-­Dunsmore, P. M. Chittaro, S. Oyamada, K. Andrews, and M. Mangel. 2009. “A Framework for Assessing the Biodiversity and Fishery Aspects of Marine Reserves,” Journal of Applied Ecol­ogy 46:735–742. Levin, S. 1992. “The Prob­lem of Pattern and Scale in Ecol­ogy: The Robert H. MacArthur Award Lecture,” Ecol­ogy 73:1943–1967. Levin, S. 1998. “Ecosystems and the Biosphere as Complex Adaptive Systems,” Ecosystems 1:431–436. Levine, J. M., J. S. Brewer, and M. D. Bertness. 1998. “Nutrients, Competition and Plant Zonation in a New ­England Salt Marsh,” Journal of Ecol­ogy 86:285–292. Levins, R. 1970. “Extinction,” Lectures on Mathe­matics in the Life Sciences 2:75–107. Levitus, S., J. I. Antonov, T. P. Boyer, O. K. Baranova, H. E. Garcia, R. A. Locarnini, A. V. Mishonov, et al. 2012. “World Ocean Heat Content and Thermosteric Sea Level Change (0–2000 m), 1955–2010,” Geophysical Research Letters 39. Lewis, J. R. 1964. The Ecol­ogy of Rocky Shores. London: En­glish Universities Press. Lewis, S. M. 1986. “The Role of Herbivorous Fishes in the Organ­ization of a Ca­rib­bean Reef Community,” Ecological Monographs 56:183–200. Li, H., and R. Durbin. 2011. “Inference of ­Human Population History from Individual Whole-­Genome Sequences,” Nature 475:493–496. Li, W. K. W. 2002. “Macroecological Patterns of Phytoplankton in the Northwestern North Atlantic Ocean,” Nature 419: 154–157. Lima-­Mendez, G., K. Faust, N. Henry, J. Decelle, S. Colin, F. Carcillo, S. Chaffron, et al. 2015. “Determinants of Community Structure in the Global Plankton Interactome,” Science 348:1262073. Lin, D., L. Hanscom, A. Murthy, A. Galli, M. Evans, E. Neill, M. Mancini, et al. 2018. “Ecological Footprint Accounting for Countries: Updates and Results of the National Footprint Accounts, 2012–2018,” Resources 7:58. Lindeman, R. L. 1942. “The Trophic-­Dynamic Aspect of Ecol­ogy,” Ecol­ ogy 23:399–417. Linville, R. G., S. N. Luoma, L. Cutter, and G. A. Cutter. 2002. “Increased Selenium Threat as a Result of Invasion of the Exotic Bivalve Potamocorbula amurensis into the San Francisco Bay-­Delta,” Aquatic Toxicology 57:51–64. Litchman, E., P. de Tezanos Pinto, K. F. Edwards, C. A. Klausmeier, C. T. Kremer, and M. K. Thomas. 2015. “Global Biogeochemical Impacts of Phytoplankton: A Trait-­Based Perspective,” Journal of Ecol­ogy 103:1384–1396. Litchman, E., and C. A. Klausmeier. 2008. “Trait-­Based Community Ecol­ogy of Phytoplankton,” Annual Review of Ecol­ogy Evolution and Systematics 39:615–639. Litchman, E., M. D. Ohman, and T. Kiorboe. 2013. “Trait-­Based Approaches to Zooplankton Communities,” Journal of Plankton Research 35:473–484. Littler, M., P. Taylor, and D. Littler. 1986. “Plant Defense Associations in the Marine Environment,” Coral Reefs 5:63–71.

412

Ocean Ecology

Littler, M. M., and D. S. Littler. 1980. “The Evolution of Thallus Form and Survival Strategies in Benthic Marine Macroalgae: Field and Laboratory Tests of a Functional Form Model,” American Naturalist 116:25–44. Locey, K. J., and J. T. Lennon. 2016. “Scaling Laws Predict Global Microbial Diversity,” Proceedings of the National Acad­emy of Sciences 113:5970–5975. Lochte, K., and C. M. Turley. 1988. “Bacteria and Cyanobacteria Associated with Phytodetritus in the Deep Sea,” Nature 333:67–69. Loeuille, N., and M. Loreau. 2005. “Evolutionary Emergence of Size-­ Structured Food Webs,” Proceedings of the National Acad­emy of Sciences 102:5761–5766. Logares, R., S. Audic, D. Bass, L. Bittner, C. Boutte, R. Christen, J.-­M. Claverie, et al. 2014. “Patterns of Rare and Abundant Marine Microbial Eukaryotes,” Current Biology 24:813–821. Loh, T.-­L., S. E. McMurray, T. P. Henkel, J. Vicente, and J. R. Pawlik. 2015. “Indirect Effects of Overfishing on Ca­rib­bean Reefs: Sponges Overgrow Reef-­Building Corals,” PeerJ 3:e901. Lohrer, A. M., S. F. Thrush, and M. M. Gibbs. 2004. “Bioturbators Enhance Ecosystem Function through Complex Biogeochemical Interactions,” Nature 431:1092–1095. Lomolino, M. V. 2000. “Ecol­ogy’s Most General, Yet Protean Pattern: The Species-­Area Relationship,” Journal of Biogeography 27:17–26. Long, J. D., G. W. Smalley, T. Barsby, J. T. Anderson, and M. E. Hay. 2007. “Chemical Cues Induce Consumer-­Specific Defenses in a Bloom-­Forming Marine Phytoplankton,” Proceedings of the National Acad­emy of Sciences 104:10512–10517. Longhurst, A. 2007. “Doubt and Certainty in Fishery Science: Are We ­Really Headed for a Global Collapse of Stocks?” Fisheries Research 86:1–5. Longhurst, A. 2010. Mismanagement of Marine Fisheries. Cambridge, UK: Cambridge University Press. Longhurst, A., S. Sathyendranath, T. Platt, and C. Caverhill. 1995. “An Estimate of Global Primary Production in the Ocean from Satellite Radiometer Data,” Journal of Plankton Research 17:1245–1271. Longhurst, A. R. 2001. Ecological Geography of the Sea. New York: Academic Press. Longhurst, A. R. 2007. Ecological Geography of the Sea. Second edition. New York: Academic Press. Longhurst, A. R., A. W. Bedo, and W. G. Harrison. 1990. “Vertical Flux of Respiratory Carbon by Oceanic Diel Mi­grant Biota,” Deep Sea Research 37:685–694. Longhurst, A. R., and W. G. Harrison. 1988. “Vertical Nitrogen Flux from the Oceanic Photic Zone by Diel Mi­grant Zooplankton and Nekton,” Deep Sea Research 35:881–889. Longhurst, A. R., and W. G. Harrison. 1989. “The Biological Pump: Profiles of Plankton Production and Consumption in the Upper Ocean,” Pro­gress in Oceanography 22:47–123. Longo, G. O., M. E. Hay, C. E. L. Ferreira, and S. R. Floeter. 2018. “Trophic Interactions across 61 Degrees of Latitude in the Western Atlantic,” Global Ecol­ogy and Biogeography 18:107–117. Lonsdale, P. 1977. “Clustering of Suspension-­Feeding Macrobenthos near Abyssal Hydrothermal Vents at Oceanic Spreading Centers,” Deep Sea Research 24:857–863. Lopez, G. R., and J. S. Levinton. 1987. “Ecol­ogy of Deposit-­Feeding Animals in Marine Sediments,” Quarterly Review of Biology 62:235–260. López-­Urrutia, Á., E. San Martin, R. P. Harris, and X. Irigoien. 2006. “Scaling the Metabolic Balance of the Oceans,” Proceedings of the National Acad­emy of Sciences 103:8739–8744. Loreau, M. 2000. “Biodiversity and Ecosystem Functioning: Recent Theoretical Advances,” Oikos 91:3–17.

Loreau, M. 2010. From Populations to Ecosystems: Theoretical Foundations for a New Ecological Synthesis. Prince­ton, NJ: Prince­ton University Press. Loreau, M., S. Naeem, P. Inchausti, J. Bengtsson, J. P. Grime, A. Hector, D. U. Hooper, et al. 2001. “Biodiversity and Ecosystem Functioning: Current Knowledge and ­Future Challenges,” Science 294:804–808. Losos, J., and R. Ricklefs. 2009. “Adaptation and Diversification on Islands,” Nature 457:830–836. Losos, J. B. 2011. “Seeing the Forest for the Trees: The Limitations of Phylogenies in Comparative Biology,” American Naturalist 177:709–727. Losos, J. B., and R. E. Glor. 2003. “Phyloge­ne­tic Comparative Methods and the Geography of Speciation,” Trends in Ecol­ogy & Evolution 18:220–227. Lotze, H. K., H. S. Lenihan, B. J. Bourque, R. H. Bradbury, R. G. Cooke, M. C. Kay, S. M. Kidwell, M. X. Kirby, C. H. Peterson, and J. B. C. Jackson. 2006. “Depletion, Degradation, and Recovery Potential of Estuaries and Coastal Seas,” Science 312:1806–1809. Lovelock, C. E., D. R. Cahoon, D. A. Friess, G. R. Guntenspergen, K. W. Krauss, R. Reef, K. Rogers, et al. 2015. “The Vulnerability of Indo-­Pacific Mangrove Forests to Sea-­Level Rise,” Nature 526:559–563. Lovelock, C. E., I. C. Feller, M. C. Ball, J. Ellis, and B. Sorrell. 2007. “Testing the Growth Rate vs. Geochemical Hypothesis for Latitudinal Variation in Plant Nutrients,” Ecol­ogy Letters 10:1154–1163. Lovelock, J. 1995. The Ages of Gaia: A Biography of Our Living Earth. New York: Norton. Lovelock, J. E. 1972. “Gaia as Seen through the Atmosphere,” Atmospheric Environment 6:579–580. Lovelock, J. E. 1979. Gaia: A New Look at Life on Earth. Oxford, UK: Oxford University Press. Lovelock, J. E., and L. Margulis. 1974. “Atmospheric Homeostasis by and for the Biosphere: The Gaia Hypothesis,” Tellus 26:2–10. Lubchenco, J. 1978. “Plant Species Diversity in a Marine Intertidal Community: Importance of Herbivore Food Preference and Algal Competitive Abilities,” American Naturalist 112:23–39. Lubchenco, J., E. B. Cerny-­Chipman, J. N. Reimer, and S. A. Levin. 2016. “The Right Incentives Enable Ocean Sustainability Successes and Provide Hope for the ­Future,” Proceedings of the National Acad­emy of Sciences 51:14507–14514. Lubchenco, J., and S. Gaines. 1981. “A Unified Approach to Marine Plant-­ Herbivore Interactions. I. Populations and Communities,” Annual Review of Ecol­ogy and Systematics 12:405–437. Lubchenco, J., and B. A. Menge. 1978. “Community Development and Per­sis­tence in a Low Rocky Intertidal Zone,” Ecological Monographs 48:67–94. Lubchenco, J., B. Menge, S. Garrity, P. Lubchenco, L. Ashkenas, S. Gaines, R. Emlet, J. Lucas, and S. Strauss. 1984. “Structure, Per­sis­tence, and Role of Consumers in a Tropical Rocky Intertidal Community (Taboguilla Island, Bay of Panama),” Journal of Experimental Marine Biology and Ecol­ogy 78:23–73. Luck, G. W., S. Lavorel, S. McIntyre, and K. Lumb. 2012. “Improving the Application of Vertebrate Trait-­Based Frameworks to the Study of Ecosystem Ser­vices,” Journal of Animal Ecol­ogy 81:1065–1076. Luiz, O. J., A. P. Allen, D. R. Robertson, S. R. Floeter, M. Kulbicki, L. Vigliola, R. Becheler, and J. S. Madin. 2013. “Adult and Larval Traits as Determinants of Geographic Range Size among Tropical Reef Fishes,” Proceedings of the National Acad­emy of Sciences 110:16498–16502.

Literature Cited

Lüthi, D., M. Le Floch, B. Bereiter, T. Blunier, J.-­M. Barnola, U. Siegenthaler, D. Raynaud, et al. 2008. “High-­Resolution Carbon Dioxide Concentration Rec­ord 650,000–800,000 Years before Pre­sent,” Nature 453:379–382. MacArthur, R. 1955. “Fluctuations of Animal Populations and a Mea­sure of Community Stability,” Ecol­ogy 36:533–536. MacArthur, R. 1958. “Population Ecol­ogy of Some Warblers of Northeastern Coniferous Forests,” Ecol­ogy 39:599–619. MacArthur, R., and R. Levins. 1967. “The Limiting Similarity, Convergence, and Divergence of Coexisting Species,” American Naturalist 101:377–385. MacArthur, R., and E. Wilson. 1963. “An Equilibrium Theory of Insular Zoogeography,” Evolution 17:373–387. MacArthur, R., and E. Wilson. 1967. The Equilibrium Theory of Island Biogeography. Prince­ton, NJ: Prince­ton University Press. MacArthur, R. H. 1972. Geo­graph­ic­ al Ecol­ogy: Patterns in the Distribution of Species. Prince­ton, NJ: Prince­ton University Press. MacAvoy, S. E., C. R. Fisher, R. S. Carney, and S. A. Macko. 2005. “Nutritional Associations among Fauna at Hydrocarbon Seep Communities in the Gulf of Mexico,” Marine Ecol­ogy Pro­gress Series 292:51–60. Macdonald K. S., III, R. Ríos, and J. E. Duffy. 2006. “Biodiversity, Host Specificity, and Dominance by Eusocial Species among Sponge-­Dwelling Alpheid Shrimp on the Belize Barrier Reef,” Diversity and Distributions 12:165–178. Mace, G. M., B. Reyers, R. Alkemade, R. Biggs, F. S. Chapin III, S. E. Cornell, S. Díaz, et al. 2014. “Approaches to Defining a Planetary Boundary for Biodiversity,” Global Environmental Change 28:289–297. Machovsky-­Capuska, G. E., A. M. Se­nior, S. J. Simpson, and D. Raubenheimer. 2016. “The Multidimensional Nutritional Niche,” Trends in Ecol­ogy & Evolution 31:355–365. MacIntyre, H. L., R. J. Geider, and D. C. Miller. 1996. “Microphytobenthos: The Ecological Role of the ‘Secret Garden’ of Unvegetated, Shallow-­Water Marine Habitats. I. Distribution, Abundance and Primary Production,” Estuaries 19:186–201. Mac Nally, R., C. Albano, and E. Fleishman. 2014. “A Scrutiny of the Evidence for Pressure-­Induced State Shifts in Estuarine and Nearshore Ecosystems,” Austral Ecol­ogy 39:898–906. MacNeil, M. A., N. A. J. Graham, J. E. Cinner, S. K. Wilson, I. D. Williams, J. Maina, S. Newman, et al. 2015. “Recovery Potential of the World’s Coral Reef Fishes,” Nature 520:341–344. Madin, E. M. P., S. D. Gaines, and R. R. Warner. 2010. “Field Evidence for Pervasive Indirect Effects of Fishing on Prey Foraging Be­hav­ ior,” Ecol­ogy 91:3563–3571. Madin, E. M. P., J. S. Madin, and D. J. Booth. 2011. “Landscape of Fear Vis­i­ble from Space,” Scientific Reports 1:1–4. Madin, J. S., K. D. Anderson, M. H. Andreasen, T. C. L. Bridge, S. D. Cairns, S. R. Connolly, E. S. Darling, et al. 2016. “The Coral Trait Database, a Curated Database of Trait Information for Coral Species from the Global Oceans,” Scientific Data 3:1–22. Madin, J. S., M. O. Hoogenboom, S. R. Connolly, E. S. Darling, D. S. Falster, D. Huang, S. A. Keith, et al. 2016. “A Trait-­Based Approach to Advance Coral Reef Science,” Trends in Ecol­ogy & Evolution 31:419–428. Magurran, A. E., and B. J. McGill, eds. 2011. Biological Diversity: Frontiers in Mea­sure­ment and Assessment. Oxford, UK: Oxford University Press. Mahon, R., P. McConney, and R. N. Roy. 2008. “Governing Fisheries as Complex Adaptive Systems,” Marine Policy 32:104–112. Mangel, M., P. Levin, and A. Patil. 2006. “Using Life History and Per­sis­ tence Criteria to Prioritize Habitats for Management and Conservation,” Ecological Applications 16:797–806.

413

Mann, K. 1988. “Production and Use of Detritus in Vari­ous Freshwater, Estuarine, and Coastal Marine Ecosystems,” Limnology and Oceanography 33:910–930. Mann, K. H. 1982. Ecology of Coastal Waters: A Systems Approach, Vol. 8. Berkeley: University of California Press. Mann, K. H., and J. Lazier. 1996. Dynamics of Marine Ecosystems: Biological-­Physical Interaction in the Oceans. Malden, MA: Blackwell Publishing. Marañón, E. 2015. “Cell Size as a Key Determinant of Phytoplankton Metabolism and Community Structure,” Annual Review of Marine Science 7:241–264. Marba, N., and C. M. Duarte. 1998. “Rhizome Elongation and Seagrass Clonal Growth,” Marine Ecol­ogy Pro­gress Series 174:269–280. Margalef, R. 1963. “On Certain Unifying Princi­ples in Ecol­ogy,” American Naturalist 47:357–374. Margalef, R. 1978. “Life-­Forms of Phytoplankton as Survival Alternatives in an Unstable Environment,” Oceanologica Acta 1:493–509. Marquet, P. A., S. A. Navarrete, and J. C. Castilla. 1990. “Scaling Population Density to Body Size in Rocky Intertidal Communities,” Science 250:1125–1127. Marquet, P. A., S. A. Navarrete, and J. C. Castilla. 1995. “Body Size, Population Density, and the Energetic Equivalence Rule,” Journal of Animal Ecol­ogy 64:325. Marsh, H., G. De’ath, N. Gribble, and B. Lane. 2005. “Historical Marine Population Estimates: Triggers or Targets for Conservation? The Dugong Case Study,” Ecological Applications 15:481–492. Marshall, D. J., P. J. Krug, E. K. Kupriyanova, M. Byrne, and R. B. Emlet. 2012. “The Biogeography of Marine Invertebrate Life Histories,” Annual Review of Ecol­ogy Evolution and Systematics 43:97–114. Martin, D. L., Y. Chiari, E. Boone, T. D. Sherman, C. Ross, S. Wyllie-­ Echeverria, J. K. Gaydos, and A. A. Boettcher. 2016. “Functional, Phyloge­ne­tic and Host-­Geographic Signatures of Labyrinthula spp. Provide for Putative Species Delimitation and a Global-­Scale View of Seagrass Wasting Disease,” Estuaries and Coasts 39(5):1–20. Martin, J. H., K. H. Coale, K. S. Johnson, and S. E. Fitzwater. 1994. “Testing the Iron Hypothesis in Ecosystems of the Equatorial Pacific Ocean,” Nature 371:123–129. Martin, J. H., and S. E. Fitzwater. 1988. “Iron Deficiency Limits Phytoplankton Growth in the North-­East Pacific Subarctic,” Nature 331:341–343. Martin, J. H., and R. Michael Gordon. 1988. “Northeast Pacific Iron Distributions in Relation to Phytoplankton Productivity,” Deep Sea Research Part A: Oceanographic Research Papers 35:177–196. Martiny, A. C., C. T. A. Pham, F. W. Primeau, J. A. Vrugt, J. K. Moore, S. A. Levin, and M. W. Lomas. 2013. “Strong Latitudinal Patterns in the Elemental Ratios of Marine Plankton and Organic ­Matter,” Nature Geoscience 6:279–283. Martiny, J. B. H., B. J. M. Bohannan, J. H. Brown, R. K. Colwell, J. A. Fuhrman, J. L. Green, M. C. Horner-­Devine, et al. 2006. “Microbial Biogeography: Putting Microorganisms on the Map,” Nature Reviews Microbiology 4:102–112. Massana, R. 2011. “Eukaryotic Picoplankton in Surface Oceans,” Annual Review of Microbiology 65:91–110. Mate, B. R., V. Y. Ilyashenko, A. L. Bradford, V. V. Vertyankin, G. A. Tsidulko, V. V. Rozhnov, and L. M. Irvine. 2015. “Critically Endangered Western Gray Whales Migrate to the Eastern North Pacific,” Biology Letters 11:20150071. May, R. M. 1973. “Stability and Complexity in Model Ecosystems,” Monographs in Population Biology 6:1–235.

414

Ocean Ecology

May, R. M. 1974. “Biological Populations with Nonoverlapping Generations: Stable Points, Stable Cycles, and Chaos,” Science 186:645–647. May, R. M. 1988. “How Many Species Are ­There on Earth?” Science 241:1441–1449. Mayhew, P. J., M. A. Bell, T. G. Benton, and A. J. McGowan. 2012. “Biodiversity Tracks Temperature over Time,” Proceedings of the National Acad­emy of Sciences 109:15141–15145. Mayr, E. 1942. Systematics and the Origin of Species, from the Viewpoint of a Zoologist. Cambridge, MA: Harvard University Press. Mayr, E. 1954. “Geographic Speciation in Tropical Echinoids,” Evolution 8:1–18. Mayr, E. 1963. Animal Species and Evolution. Cambridge, MA: Belknap Press. McCann, K. S. 2000. “The Diversity-­Stability Debate,” Nature 405:228–233. McCauley, D. J., M. L. Pinsky, S. R. Palumbi, J. A. Estes, F. H. Joyce, and R. R. Warner. 2015. “Marine Defaunation: Animal Loss in the Global Ocean,” Science 347:1255641. McClain, C. R., A. P. Allen, D. P. Tittensor, and M. A. Rex. 2012. “Energetics of Life on the Deep Seafloor,” Proceedings of the National Acad­emy of Sciences 109:15366–15371. McClanahan, T. R. 2018. “Community Biomass and Life History Benchmarks for Coral Reef Fisheries,” Fish and Fisheries 19:471–488. McClanahan, T. R., S. D. Donner, J. A. Maynard, M. A. MacNeil, N. A. J. Graham, J. Maina, A. C. Baker, et el. 2012. “Prioritizing Key Resilience Indicators to Support Coral Reef Management in a Changing Climate,” PLoS ONE 7:e42884. McClanahan, T. R., and J. O. Omukotu. 2011. “Comparison of Modern and Historical Fish Catches (AD 750–1400) to Inform Goals for Marine Protected Areas and Sustainable Fisheries,” Conservation Biology 25:945–955. McClenachan, L. 2009. “Historical Declines of Goliath Grouper Populations in South Florida, USA,” Endangered Species Research 7:175–181. McDevitt-­Irwin, J. M., J. C. Iacarella, and J. K. Baum. 2016. “Reassessing the Nursery Role of Seagrass Habitats from Temperate to Tropical Regions: A Meta-­analysis,” Marine Ecol­ogy Pro­gress Series 557:133–143. McGill, B. J., B. J. Enquist, E. Weiher, and M. Westoby. 2006. “Rebuilding Community Ecol­ogy from Functional Traits,” Trends in Ecol­ ogy & Evolution 21:178–185. McGlathery, K., K. Sundback, and I. Anderson. 2007. “Eutrophication in Shallow Coastal Bays and Lagoons: The Role of Plants in the Coastal Filter,” Marine Ecol­ogy Pro­gress Series 348:1–18. McGowan, J., and P. Walker. 1985. “Dominance and Diversity Maintenance in an Oceanic Ecosystem,” Ecological Monographs 55:103–118. McGurk, M. D. 1986. “Natu­ral Mortality of Marine Pelagic Fish Eggs and Larvae: Role of Spatial Patchiness,” Marine Ecol­ogy Pro­gress Series 34:227–242. McKeon, C. S., M. X. Weber, S. E. Alter, N. E. Seavy, E. D. Crandall, D. J. Barshis, E. D. Fechter-­Leggett, and K. L. L. Oleson. 2015. “Melting Barriers to Faunal Exchange across Ocean Basins,” Global Change Biology 22:465–473. McKinney, M. L. 2006. “Urbanization as a Major Cause of Biotic Homogenization,” Biological Conservation 127:247–260. McKinney, M. L., and J. L. Lockwood. 1999. “Biotic Homogenization: A Few Winners Replacing Many Losers in the Next Mass Extinction,” Trends in Ecol­ogy & Evolution 14:450–453. McMahon, K. W., S. R. Thorrold, L. A. Houghton, and M. L. Berumen. 2015. “Tracing Carbon Flow through Coral Reef Food Webs Using a Compound-­Specific Stable Isotope Approach,” Oecologia 180:809–821.

McNeill, J. R. 2001. Something New ­under the Sun: An Environmental History of the Twentieth-­Century World. New York: W. W. Norton & Com­pany. Meadows, D. H., D. L. Meadows, J. Randers, and W. W. Behrens III. 1972. The Limits to Growth. New York: Universe Books. Mellin, C., M. A. MacNeil, A. J. Cheal, M. J. Emslie, and M. J. Caley. 2016. “Marine Protected Areas Increase Resilience among Coral Reef Communities,” Ecol­ogy Letters 19:629–637. Menge, B. 1976. “Organ­ization of the New ­England Rocky Intertidal Community: Role of Predation, Competition, and Environmental Heterogeneity,” Ecological Monographs 46:355–393. Menge, B., C. Blanchette, P. Raimondi, T. Freidenburg, S. Gaines, J. Lubchenco, D. Lohse, G. Hudson, M. Foley, and J. Pamplin. 2004. “Species Interaction Strength: Testing Model Predictions along an Upwelling Gradient,” Ecological Monographs 74:663–684. Menge, B., J. Lubchenco, and L. Ashkenas. 1985. “Diversity, Heterogeneity and Consumer Pressure in a Tropical Rocky Intertidal Community,” Oecologia 65:394–405. Menge, B., J. Lubchenco, and L. Ashkenas. 1986. “Experimental Separation of Effects of Consumers on Sessile Prey in the Low Zone of a Rocky Shore in the Bay of Panama: Direct and Indirect Consequences of Food Web Complexity,” Journal of Experimental Marine Biology and Ecol­ogy 100:225–269. Menge, B., and J. Sutherland. 1976. “Species Diversity Gradients: Synthesis of the Roles of Predation, Competition, and Temporal Heterogeneity,” American Naturalist 110:351. Menge, B., and J. Sutherland. 1987. “Community Regulation: Variation in Disturbance, Competition, and Predation in Relation to Environmental Stress and Recruitment,” American Naturalist 130:730–757. Menge, B. A. 1995. “Indirect Effects in Marine Rocky Intertidal Interaction Webs: Patterns and Importance,” Ecological Monographs 65:21–74. Menge, B. A., E. L. Berlow, C. A. Blanchette, S. A. Navarrete, and S. B. Yamada. 1994. “The Keystone Species Concept: Variation in Interaction Strength in a Rocky Intertidal Habitat,” Ecological Monographs 64:249–286. Menge, B. A., and J. Lubchenco. 1981. “Community Organ­ization in Temperate and Tropical Rocky Intertidal Habitats: Prey Refuges in Relation to Consumer Pressure Gradients,” Ecological Monographs 51:429–450. Menge, B. A., J. Lubchenco, M. E. S. Bracken, F. Chan, M. M. Foley, T. L. Freidenburg, S. D. Gaines, et al. 2003. “Coastal Oceanography Sets the Pace of Rocky Intertidal Community Dynamics,” Proceedings of the National Acad­emy of Sciences 100:12229–12234. Merrie, A., D. C. Dunn, M. Metian, A. M. Boustany, Y. Takei, A. O. Elferink, Y. Ota, V. Christensen, P. N. Halpin, and H. Österblom. 2014. “An Ocean of Surprises—­Trends in ­Human Use, Unexpected Dynamics and Governance Challenges in Areas beyond National Jurisdiction,” Global Environmental Change 27:19–31. Mertz, D. B., D. A. Cawthon, and T. Park. 1976. “An Experimental Analy­ sis of Competitive Indeterminacy in Tribolium,” Proceedings of the National Acad­emy of Sciences 73:1368–1372. Meyer, J. L., E. T. Schultz, and G. S. Helfman. 1983. “Fish Schools: An Asset to Corals,” Science 220:1047–1049. Michaels, A. F., and M. W. Silver. 1988. “Primary Production, Sinking Fluxes and the Microbial Food Web,” Deep Sea Research Part A: Oceanographic Research Papers 35:473–490. Micheli, F. 1999. “Eutrophication, Fisheries, and Consumer-­Resource Dynamics in Marine Pelagic Ecosystems,” Science 285:1396–1398.

Literature Cited

Micheli, F., and B. S. Halpern. 2005. “Low Functional Redundancy in Coastal Marine Assemblages,” Ecol­ogy Letters 8:391–400. Micheli, F., P. J. Mumby, D. R. Brumbaugh, K. Broad, C. P. Dahlgren, A. R. Harborne, K. E. Holmes, C. V. Kappel, S. Y. Litvin, and J. N. Sanchirico. 2014. “High Vulnerability of Ecosystem Function and Ser­vices to Diversity Loss in Ca­rib­bean Coral Reefs,” Biological Conservation 171:186–194. Micheli, F., and C. Peterson. 1999. “Estuarine Vegetated Habitats as Corridors for Predator Movements,” Conservation Biology 13:869–881. Micheli, F., C. H. Peterson, L. S. Mullineaux, C. R. Fisher, Susan W. Mills, G. Sancho, G. Johnson, and H. S. Lenihan. 2002. “Predation Structures Communities at Deep-­Sea Hydrothermal Vents,” Ecological Monographs 72:365–382. Middelburg, J. J. 2011. “Chemoautotrophy in the Ocean,” Geophysical Research Letters 38. Millennium Ecosystem Assessment. 2006a. “Coastal Systems.” Page 37 in Marine and Coastal Ecosystems and ­Human Well-­Being: Synthesis. Nairobi: UN Environment Programme. Millennium Ecosystem Assessment. 2006b. Marine and Coastal Ecosystems and ­Human Well-­Being: Synthesis. Nairobi: UN Environment Programme. Miller, A. W., and G. M. Ruiz. 2014. “Arctic Shipping and Marine Invaders,” Nature Climate Change 4:413–416. Miller, D. D., U. R. Sumaila, D. Copeland, D. Zeller, B. Soyer, T. Nikaki, G. Leloudas, S. T. Fjellberg, R. Singleton, and D. Pauly. 2016. “Cutting a Lifeline to Maritime Crime: Marine Insurance and IUU Fishing,” Frontiers in Ecol­ogy and the Environment 14(7):357–362. Miller, T. E., J. H. Burns, P. Munguia, E. L. Walters, J. M. Kneitel, P. M. Richards, N. Mouquet, and H. L. Buckley. 2005. “A Critical Review of Twenty Years’ Use of the Resource-­R atio Theory,” American Naturalist 165:439–448. Miller, T. R., A. Wiek, D. Sarewitz, J. Robinson, L. Olsson, D. Kriebel, and D. Loorbach. 2014. “The ­Future of Sustainability Science: A Solutions-­Oriented Research Agenda,” Sustainability Science 9:239–246. Mittelbach, G. G., D. W. Schemske, H. V. Cornell, A. P. Allen, J. M. Brown, M. B. Bush, S. P. Harrison, et al. 2007. “Evolution and the Latitudinal Diversity Gradient: Speciation, Extinction and Biogeography,” Ecol­ogy Letters 10:315–331. Mokany, K., S. Ferrier, S. R. Connolly, P. K. Dunstan, E. A. Fulton, M. B. Harfoot, T. D. Harwood, et al. 2015. “Integrating Modelling of Biodiversity Composition and Ecosystem Function,” Oikos 125:10–19. Molnar, J. L., R. L. Gamboa, C. Revenga, and M. D. Spalding. 2008. “Assessing the Global Threat of Invasive Species to Marine Biodiversity,” Frontiers in Ecol­ogy and the Environment 6:485–492. Moncreiff, C. A., and M. J. ­Sullivan. 2001. “Trophic Importance of Epiphytic Algae in Subtropical Seagrass Beds: Evidence from Multiple Stable Isotope Analyses,” Marine Ecol­ogy Pro­gress Series 215:93–106. Montagna, P. 1995. “Rates of Metazoan Meiofaunal Microbivory: A Review,” Vie et Milieu 45:1–9. Moore, C. M., M. M. Mills, K. R. Arrigo, I. Berman-­Frank, L. Bopp, P. W. Boyd, E. D. Galbraith, et al. 2013. “Pro­cesses and Patterns of Oceanic Nutrient Limitation,” Nature Geoscience 6:701–710. Moore, J. C., E. L. Berlow, D. C. Coleman, P. C. Ruiter, Q. Dong, A. Hastings, N. C. Johnson, et al. 2004. “Detritus, Trophic Dynamics and Biodiversity,” Ecol­ogy Letters 7:584–600. Moore, K. A., and J. C. Jarvis. 2008. “Environmental ­Factors Affecting Recent Summertime Eelgrass Diebacks in the Lower Chesa-

415

peake Bay: Implications for Long-­Term Per­sis­tence,” Journal of Coastal Research 55:135–147. Mora, C., O. Aburto-­Oropeza, A. A. Bocos, P. M. Ayotte, S. Banks, A. G. Bauman, M. Beger, S. Bessudo, D. J. Booth, and E. Brokovich. 2011. “Global ­Human Footprint on the Linkage between Biodiversity and Ecosystem Functioning in Reef Fishes,” PLoS Biology 9:e1000606. Mora, C., P. M. Chittaro, P. F. Sale, J. P. Kritzer, and S. A. Ludsin. 2003. “Patterns and Pro­cesses in Reef Fish Diversity,” Nature 421:933–936. Mora, C., R. Danovaro, and M. Loreau. 2014. “Alternative Hypotheses to Explain Why Biodiversity-­Ecosystem Functioning Relationships Are Concave-­Up in Some Natu­ral Ecosystems but Concave-­ Down in Manipulative Experiments,” Scientific Reports 4. Mora, C., D. P. Tittensor, S. Adl, A. G. B. Simpson, and B. Worm. 2011. “How Many Species Are ­There on Earth and in the Ocean?” PLoS Biology 9:e1001127. Mora, C., C.-­L. Wei, A. Rollo, T. Amaro, A. R. Baco, D. Billett, L. Bopp, et al. 2013. “Biotic and ­Human Vulnerability to Projected Changes in Ocean Biogeochemistry over the 21st ­Century,” PLoS Biology 11:e1001682. Moran, N. A., J. P. McCutcheon, and A. Nakabachi. 2008. “Genomics and Evolution of Heritable Bacterial Symbionts,” Annual Review of Ge­ne­tics 42:165–190. Morán, X. A. G., L. Alonso-­Sáez, E. Nogueira, H. W. Ducklow, N. González, Á. López-­Urrutia, L. Díaz-­Pérez, A. Calvo-­Díaz, N. Arandia-­Gorostidi, and T. M. Huete-­Stauffer. 2015. “More, Smaller Bacteria in Response to Ocean’s Warming?” Proceedings of the Royal Society B-­Biological Sciences 282:20150371. Morán, X. A. G., Á. López-­Urrutia, A. Calvo-­Díaz, and W. K. W. Li. 2010. “Increasing Importance of Small Phytoplankton in a Warmer Ocean,” Global Change Biology 16:1137–1144. Morato, T., R. Watson, T. Pitcher, and D. Pauly. 2006. “Fishing Down the Deep,” Fish and Fisheries 7:24–34. Morgan, S., and J. Christy. 1995. “Adaptive Significance of the Timing of Larval Release by Crabs,” American Naturalist 145:457–479. Morgan Ernest, S. M., B. J. Enquist, J. H. Brown, E. L. Charnov, J. F. Gillooly, V. M. Savage, E. P. White, et al. 2003. “Thermodynamic and Metabolic Effects on the Scaling of Production and Population Energy Use,” Ecol­ogy Letters 6:990–995. Morse, A., K. Iwao, M. Baba, and K. Shimoike. 1996. “An Ancient Chemosensory Mechanism Brings New Life to Coral Reefs,” Biological Bulletin 191(2):149–154. Moses, M. E., and J. H. Brown. 2003. “Allometry of ­Human Fertility and Energy Use,” Ecol­ogy Letters 6:295–300. Mouquet, N., P. Munguia, J. M. Kneitel, and T. E. Miller. 2003. “Community Assembly Time and the Relationship between Local and Regional Species Richness,” Oikos 103:618–626. Mousseau, T., and D. Roff. 1987. “Natu­ral Se­lection and the Heritability of Fitness Components,” Heredity 59:181–197. Moy, F. E., and H. Christie. 2012. “Large-­Scale Shift from Sugar Kelp (Saccharina latissima) to Ephemeral Algae along the South and West Coast of Norway,” Marine Biology Research 8:309–321. Muehlstein, L., D. Porter, and F. Short. 1991. “Labyrinthula zosterae sp. nov., the Causative Agent of Wasting Disease of Eelgrass, Zostera marina,” Mycologia 83(2):180–191. Muller-­Parker, G., C. F. D’elia, and C. B. Cook. 2015. “Interactions between Corals and Their Symbiotic Algae.” Pages 99–116 in Charles Birkeland, editor, Coral Reefs in the Anthropocene. Dordrecht, Germany: Springer. Mumby, P. J., C. P. Dahlgren, A. R. Harborne, C. V. Kappel, F. Micheli, D. R. Brumbaugh, K. E. Holmes, et al. 2006. “Fishing, Trophic

416

Ocean Ecology

Cascades, and the Pro­cess of Grazing on Coral Reefs,” Science 311:98–101. Mumby, P. J., A. J. Edwards, J. E. Arias-­González, K. C. Lindeman, P. G. Blackwell, A. Gall, M. I. Gorczynska, et al. 2004. “Mangroves Enhance the Biomass of Coral Reef Fish Communities in the Ca­rib­bean,” Nature 427:533–536. Mumby, P. J., A. R. Harborne, J. Williams, C. V. Kappel, D. R. Brumbaugh, F. Micheli, K. E. Holmes, C. P. Dahlgren, C. B. Paris, and P. G. Blackwell. 2007. “Trophic Cascade Facilitates Coral Recruitment in a Marine Reserve,” Proceedings of the National Acad­emy of Sciences 104:8362–8367. Munday, P. L., L. van Herwerden, and C. L. Dudgeon. 2004. “Evidence for Sympatric Speciation by Host Shift in the Sea,” Current Biology 14:1498–1504. Murdoch, W. W. 1969. “Switching in General Predators: Experiments on Predator Specificity and Stability of Prey Populations,” Ecological Monographs 39:335–354. Muscatine, L. 1990. “The Role of Symbiotic Algae in Carbon and Energy Flux in Reef Corals,” Coral Reefs 25:1–29. Mutshinda, C. M., Z. V. Finkel, C. E. Widdicombe, and A. J. Irwin. 2016. “Ecological Equivalence of Species within Phytoplankton Functional Groups,” Functional Ecol­ogy 30:1714–1722. Myers, A. A. 1996. “Species and Generic Gamma-­Scale Diversity in Shallow-­Water Marine Amphipoda with Par­tic­ul­ ar Reference to the Mediterranean,” Journal of the Marine Biological Association of the UK 76:195–202. Myers, R. A., J. K. Baum, T. D. Shepherd, S. P. Powers, and C. H. Peterson. 2007. “Cascading Effects of the Loss of Apex Predatory Sharks from a Coastal Ocean,” Science 315:1846–1850. Myers, R. A., and B. Worm. 2003. “Rapid Worldwide Depletion of Predatory Fish Communities,” Nature 423:280–283. Nagelkerken, I., and S. D. Connell. 2015. “Global Alteration of Ocean Ecosystem Functioning Due to Increasing ­Human CO2 Emissions,” Proceedings of the National Acad­emy of Sciences 112:13272–13277. Nagelkerken, I., M. G. G. Grol, and P. J. Mumby. 2012. “Effects of Marine Reserves versus Nursery Habitat Availability on Structure of Reef Fish Communities,” PLoS ONE 7:e36906. Nagelkerken, I., B. D. Russell, B. M. Gillanders, and S. D. Connell. 2015. “Ocean Acidification Alters Fish Populations Indirectly through Habitat Modification,” Nature Climate Change 6:89–93. Naidoo, G. 2009. “Differential Effects of Nitrogen and Phosphorus Enrichment on Growth of Dwarf Avicennia marina Mangroves,” Aquatic Botany 90:184–190. Narayan, S., M. W. Beck, P. Wilson, C. J. Thomas, A. Guerrero, C. C. Shepard, B. G. Reguero, G. Franco, J. C. Ingram, and D. Trespalacios. 2017. “The Value of Coastal Wetlands for Flood Damage Reduction in the Northeastern USA,” Scientific Reports, 1–12. Naylor, R. L., R. J. Goldburg, J. H. Primavera, N. Kautsky, M. C. Beveridge, J. Clay, C. Folke, J. Lubchenco, H. Mooney, and M. Troell. 2000. “Effect of Aquaculture on World Fish Supplies,” Nature 405:1017–1024. Naylor, R. L., S. L. Williams, and D. R. Strong. 2001. “Ecol­ogy. Aquaculture—­A Gateway for Exotic Species,” Science 294:1655–1656. Nei, M., and S. Kumar. 2000. Molecular Evolution and Phyloge­ne­tics. New York: Oxford University Press. Neigel, J. E. 2003. “Species-­Area Relationships and Marine Conservation,” Ecological Applications 13(Suppl):S138–­S145. Nekola, J. C., C. D. Allen, J. H. Brown, J. R. Burger, A. D. Davidson, T. S. Fristoe, M. J. Hamilton, et al. 2013. “The Malthusian-­

Darwinian Dynamic and the Trajectory of Civilization,” Trends in Ecol­ogy & Evolution 28:127–130. Neori, A., T. Chopin, M. Troell, A. H. Buschmann, G. P. Kraemer, C. Halling, M. Shpigel, and C. Yarish. 2004. “Integrated Aquaculture: Rationale, Evolution and State of the Art Emphasizing Seaweed Biofiltration in Modern Mariculture,” Aquaculture 231:361–391. Neubauer, P., O. P. Jensen, J. A. Hutchings, and J. K. Baum. 2013. “Resilience and Recovery of Overexploited Marine Populations,” Science 340:347–349. Newell, R. 1988. “Ecological Changes in Chesapeake Bay: Are They the Result of Overharvesting the American Oyster, Crassostrea virginica?” Pages 29–31 in M. P. Lynch and E. C. Krome, editors, Understanding the Estuary. Gloucester Point, VA: Chesapeake Research Consortium. Newman, M. J. H., G. A. Paredes, E. Sala, and J. B. C. Jackson. 2006. “Structure of Ca­rib­bean Coral Reef Communities across a Large Gradient of Fish Biomass,” Ecol­ogy Letters 9:1216–1227. Newman, S. P., E. H. Meesters, C. S. Dryden, S. M. Williams, C. Sanchez, P. J. Mumby, and N. V. C. Polunin. 2015. “Reef Flattening Effects on Total Richness and Species Responses in the Ca­rib­ bean,” Journal of Animal Ecol­ogy 84:1678–1689. Newton, K., I. Cote, G. Pilling, S. Jennings, and N. Dulvy. 2007. “Current and ­Future Sustainability of Island Coral Reef Fisheries,” Current Biology 17:655–658. Nichols, F. H., J. K. Thompson, and L. E. Schemel. 1990. “Remarkable Invasion of San Francisco Bay (California, USA) by the Asian Clam Potamocorbula amurensis. II. Displacement of a Former Community,” Marine Ecol­ogy Pro­gress Series 66:95–101. Nickols, K. J., J. W. White, J. L. Largier, and B. Gaylord. 2015. “Marine Population Connectivity: Reconciling Large-­Scale Dispersal and High Self-­Retention,” American Naturalist 185:196–211. Nixon, S. W. 1982. “Nutrient Dynamics, Primary Production and Fisheries Yields of Lagoons,” Oceanologica Acta (Special Issue):357–372. Nixon, S. W., and B. A. Buckley. 2002. “ ‘A Strikingly Rich Zone’—­ Nutrient Enrichment and Secondary Production in Coastal Marine Ecosystems,” Estuaries 25:782–796. Norberg, J. 2004. “Biodiversity and Ecosystem Functioning: A Complex Adaptive Systems Approach,” Limnology and Oceanography 49:1269–1277. Norberg, J., D. P. Swaney, J. Dushoff, J. Lin, R. Casagrandi, and S. A. Levin. 2001. “Phenotypic Diversity and Ecosystem Functioning in Changing Environments: A Theoretical Framework,” Proceedings of the National Acad­emy of Sciences 98:11376–11381. Nordlund, L. M., E. W. Koch, E. B. Barbier, and J. C. Creed. 2016. “Seagrass Ecosystem Ser­vices and Their Variability across Genera and Geo­graph­i­cal Regions,” PLoS ONE 11:e0163091–23. Normile, D., 2009. “Round and Round: A Guide to the Carbon Cycle,” Science 325:1642–1643. Norse, E. A. 1993. Global Marine Biological Diversity: A Strategy for Building Conservation into Decision Making. Washington, DC: Island Press. Norse, E. A., S. Brooke, W. W. L. Cheung, M. R. Clark, I. Ekeland, R. Froese, K. M. Gjerde, et al. 2012. “Sustainability of Deep-­Sea Fisheries,” Marine Policy 36:307–320. Nürnberg, S., and M. Aberhan. 2013. “Habitat Breadth and Geographic Range Predict Diversity Dynamics in Marine Mesozoic Bivalves,” Paleobiology 39:360–372. O’Brien, K. R., M. Waycott, P. Maxwell, G. A. Kendrick, J. W. Udy, A. J. P. Ferguson, K. Kilminster, et al. 2018. “Seagrass Ecosystem Trajectory Depends on the Relative Timescales of Re­sis­tance, Recovery and Disturbance,” Marine Pollution Bulletin 134:166–176.

Literature Cited

O’Connor, M. I. 2009. “Warming Strengthens an Herbivore-­Plant Interaction,” Ecol­ogy 90:388–398. O’Connor, M. I., J. F. Bruno, S. D. Gaines, B. S. Halpern, S. E. Lester, B. P. Kinlan, and J. M. Weiss. 2007. “Temperature Control of Larval Dispersal and the Implications for Marine Ecol­ogy, Evolution, and Conservation,” Proceedings of the National Acad­emy of Sciences 104:1266–1271. O’Connor, M. I., B. Gilbert, and C. J. Brown. 2011. “Theoretical Predictions for How Temperature Affects the Dynamics of Interacting Herbivores and Plants,” American Naturalist 178:626–638. O’Connor, M. I., A. Gonzalez, J. Byrnes, B. J. Cardinale, J. E. Duffy, L. Gamfeldt, J. N. Griffin, et al. 2017. “A General Biodiversity-­ Function Relationship Is Mediated by Trophic Level,” Oikos 126:18–31. O’Connor, M. I., M. F. Piehler, D. M. Leech, A. Anton, and J. F. Bruno. 2009. “Warming and Resource Availability Shift Food Web Structure and Metabolism,” PLoS Biology 7:e1000178. O’Dea, A., H. A. Lessios, A. G. Coates, R. I. Eytan, S. A. Restrepo-­ Moreno, A. L. Cione, L. S. Collins, et al. 2016. “Formation of the Isthmus of Panama,” Science Advances 2:e1600883. Odling-­Smee, F. J., K. N. Laland, and M. W. Feldman. 1996. “Niche Construction,” American Naturalist 147:641–648. Odum, E. P. 1969. “The Strategy of Ecosystem Development,” Science 164:262–270. Odum, E. P., H. T. Odum, and J. Andrews. 1971. Fundamentals of Ecol­ogy. Third edition. Philadelphia, PA: W. B. Saunders Co. Odum, H., and E. Odum. 1955. “Trophic Structure and Productivity of a Windward Coral Reef Community on Eniwetok Atoll,” Ecological Monographs 25:291–320. Odum, H. T. 1957. “Trophic Structure and Productivity of Silver Springs, Florida,” Ecological Monographs 27(1):55–112. Odum, W., and E. J. Heald. 1975. “The Detritus Based Food Web on an Estuarine Mangrove Community,” Estuarine Research 1:265–286. Organisation for Economic Co-­Operation and Development (OECD). 2017. The Ocean Economy in 2030. London: IWA Publishing. Offre, P., A. Spang, and C. Schleper. 2013. “Archaea in Biogeochemical Cycles,” Annual Review of Microbiology 67:437–457. Ogden, J. C., R. A. Brown, and N. Salesky. 1973. “Grazing by the Echinoid Diadema antillarum philippi: Formation of Halos around West Indian Patch Reefs,” Science 182:715–717. O’Gorman, E. J., and M. C. Emmerson. 2009. “Perturbations to Trophic Interactions and the Stability of Complex Food Webs,” Proceedings of the National Acad­emy of Sciences 106:13393–13398. Olden, J., N. Poff, and K. Bestgen. 2008. “Trait Synergisms and the Rarity, Extirpation, and Extinction Risk of Desert Fishes,” Ecol­ogy 89:847–856. O’Neill, R. V. 2001. “Is It Time to Bury the Ecosystem Concept? (with Full Military Honors, of Course!),” Ecol­ogy 82:3275–3284. O’Neill, R. V., D. L. Deangelis, R. B. Waide, and G. E. Allen. 1986. A Hierarchical Concept of Ecosystems. Monographs in Population Biology 23. Prince­ton, NJ: Prince­ton University Press. Oreskes, N., and E. M. Conway. 2011. Merchants of Doubt: How a Handful of Scientists Obscured the Truth on Issues from Tobacco Smoke to Global Warming. New York: Bloomsbury Press. Organ, C., C. L. Nunn, Z. Machanda, and R. W. Wrangham. 2011. “Phyloge­ne­tic Rate Shifts in Feeding Time during the Evolution of Homo,” Proceedings of the National Acad­emy of Sciences 108:14555–14559. Orr, J. C., V. J. Fabry, O. Aumont, L. Bopp, S. C. Doney, R. A. Feely, A. Gnanadesikan, et al. 2005. “Anthropogenic Ocean Acidifica-

417

tion over the Twenty-­First ­Century and Its Impact on Calcifying Organisms,” Nature 437:681–686. Orth, R., K. Heck, and J. Van Montfrans. 1984. “Faunal Communities in Seagrass Beds: A Review of the Influence of Plant Structure and Prey Characteristics on Predator-­Prey Relationships,” Estuaries and Coasts 7:339–350. Orth, R., and J. Van Montfrans. 1984. “Epiphyte-­Seagrass Relationships with an Emphasis on the Role of Micrograzing: A Review,” Aquatic Botany 18:43–69. Östman, Ö., J. Eklöf, B. K. Eriksson, J. Olsson, P.-­O. Moksnes, and U. Bergström. 2016. “Top-­Down Control as Impor­tant as Nutrient Enrichment for Eutrophication Effects in North Atlantic Coastal Ecosystems,” Journal of Applied Ecol­ogy 53:1138–1147. Ostrom, E. 2009. “A General Framework for Analyzing Sustainability of Social-­Ecological Systems,” Science 325:419–422. Overland, J. E., D. B. Percival, and H. O. Mofjeld. 2006. “Regime Shifts and Red Noise in the North Pacific,” Deep Sea Research Part I: Oceanographic Research Papers 53:582–588. Pacala, S., and R. Socolow. 2004. “Stabilization Wedges: Solving the Climate Prob­lem for the Next 50 Years with Current Technologies,” Science 305:968–972. Padilla, D., and B. Allen. 2000. “Paradigm Lost: Reconsidering Functional Form and Group Hypotheses in Marine Ecol­ogy,” Journal of Experimental Marine Biology and Ecol­ogy 250:207–221. Paerl, H. W., and J. Huisman. 2008. “CLIMATE: Blooms Like It Hot.” Science 320:57–58. Paine, R. 1966. “Food Web Complexity and Species Diversity,” American Naturalist 100:65–75. Paine, R. 1974. “Intertidal Community Structure,” Oecologia 15:93–120. Paine, R. 1980. “Food Webs: Linkage, Interaction Strength and Community Infrastructure,” Journal of Animal Ecol­ogy 49:667–685. Paine, R. T. 1992. “Food-­Web Analy­sis through Field Mea­sure­ment of per Capita Interaction Strength,” Nature 355:73–75. Paine, R. T. 1994. Marine Rocky Shores and Community Ecol­ogy: An Experimentalist’s Perspective. Volume 4. Oldendorf/Luhe, Germany: Ecol­ogy Institute. Palmer, A. R. 1979. “Fish Predation and the Evolution of Gastropod Shell Sculpture: Experimental and Geographic Evidence,” Evolution 33:697–713. Palumbi, S. 1994. “Ge­ne­tic Divergence, Reproductive Isolation, and Marine Speciation,” Annual Review of Ecol­ogy and Systematics 25:547–572. Palumbi, S. R. 2001. “­Humans as the World’s Greatest Evolutionary Force,” Science 293:1786–1790. Palumbi, S. R. 2009. “Speciation and the Evolution of Gamete Recognition Genes: Pattern and Pro­cess,” Heredity 102:66–76. Pandolfi, J. M., R. H. Bradbury, E. Sala, T. P. Hughes, K. A. Bjorndal, R. G. Cooke, D. McArdle, et al. 2003. “Global Trajectories of the Long-­ Term Decline of Coral Reef Ecosystems,” Science 301:955–958. Park, T. 1948. “Experimental Studies of Interspecies Competition. 1. Competition between Populations of the Flour Beetles, Tribolium confusum Duval and Tribolium castaneum Herbst,” Ecological Monographs 18:265–307. Parrish, F. A. 2009. “Do Monk Seals Exert Top-­Down Pressure in Subphotic Ecosystems?” Marine Mammal Science 25:91–106. Parsons, T. R., M. Takabashi, and B. Hargrave. 1984. Biological Oceanographic Pro­cesses. Oxford, UK: Pergamon Press. Pastor, J., B. Koeck, P. Astruch, and P. Lenfant. 2013. “Coastal Man-­Made Habitats: Potential Nurseries for an Exploited Fish Species, Diplodus sargus (Linnaeus, 1758),” Fisheries Research 148:74–80.

418

Ocean Ecology

Patrick, W. S., and J. S. Link. 2015. “Myths That Continue to Impede Pro­ gress in Ecosystem-­Based Fisheries Management,” Fisheries 40:155–160. Pauly, D., 1980. “On the Interrelationships between Natu­ral Mortality, Growth Par­ameters, and Mean Environmental Temperature in 175 Fish Stocks,” ICES Journal of Marine Science 39:175–192. Pauly, D. 1995. “Anecdotes and the Shifting Baseline Syndrome of Fisheries,” Trends in Ecol­ogy & Evolution 10:430. Pauly, D., and V. Christensen. 1995. “Primary Production Required to Sustain Global Fisheries,” Nature 374:255–257. Pauly, D., V. Christensen, J. Dalsgaard, R. Froese, and F. Torres. 1998. “Fishing Down Marine Food Webs,” Science 279:860–863. Pauly, D., V. Christensen, S. Guénette, T. J. Pitcher, U. R. Sumaila, C. J. Walters, R. Watson, and D. Zeller. 2002. “­Towards Sustainability in World Fisheries,” Nature 418:689–695. Pavoine, S., and M. B. Bonsall. 2011. “Mea­sur­ing Biodiversity to Explain Community Assembly: A Unified Approach,” Biological Reviews 86:792–812. Pawlik, J. R., D. E. Burkepile, and R. V. Thurber. 2016. “A Vicious Circle? Altered Carbon and Nutrient Cycling May Explain the Low Resilience of Ca­rib­bean Coral Reefs,” BioScience 66:470–476. Payne, J. L., A. M. Bush, N. A. Heim, M. L. Knope, and D. J. McCauley. 2016. “Ecological Selectivity of the Emerging Mass Extinction in the Oceans,” Science 353:1284–1286. Pearson, R. G., and T. P. Dawson. 2003. “Predicting the Impacts of Climate Change on the Distribution of Species: Are Bioclimate Envelope Models Useful?” Global Ecol­ogy and Biogeography 12:361–371. Pearson, T. H., and R. Rosenberg. 1978. “Macrobenthic Succession in Relation to Organic Enrichment and Pollution of the Marine Environment,” Oceanography and Marine Biology: An Annual Review 16:229–311. Pendleton, L., D. C. Donato, B. C. Murray, S. Crooks, W. A. Jenkins, S. Sifleet, C. Craft, et al. 2012. “Estimating Global ‘Blue Carbon’ Emissions from Conversion and Degradation of Vegetated Coastal Ecosystems,” PloS One 7(9):e43542. Pennings, S., E. Siska, and M. Bertness. 2001. “Latitudinal Differences in Plant Palatability in Atlantic Coast Salt Marshes,” Ecol­ogy 82:1344–1359. Pennings, S. C., and R. M. Callaway. 1992. “Salt Marsh Plant Zonation: The Relative Importance of Competition and Physical ­Factors,” Ecol­ogy 73:681–690. Pennings, S. C., C.-­K . Ho, C. S. Salgado, K. Wieski, N. Davé, A. E. Kunza, and E. L. Wason. 2009. “Latitudinal Variation in Herbivore Pressure in Atlantic Coast Salt Marshes,” Ecol­ogy 90:183–195. Pepin, P., and R. A. Myers. 1991. “Significance of Egg and Larval Size to Recruitment Variability of Temperate Marine Fish,” Canadian Journal of Fisheries and Aquatic Sciences 48:1820–1828. Pereira, H. M., P. W. Leadley, V. Proença, R. Alkemade, J. P. W. Scharlemann, J. F. Fernandez-­Manjarrés, M. B. Araújo, et el. 2010. “Scenarios for Global Biodiversity in the 21st ­Century,” Science 330:1496–1501. Pereyra, R. T., L. Bergström, L. Kautsky, and K. Johannesson. 2009. “Rapid Speciation in a Newly Opened Postglacial Marine Environment, the Baltic Sea,” BMC Evolutionary Biology 9:70. Pergent, G., J. Romero, C. Pergent-­Martini, M. A. Mateo, and C. F. Boudouresque. 1994. “Primary Production, Stocks and Fluxes in the Mediterranean Seagrass Posidonia oceanica,” Marine Ecol­ ogy Pro­gress Series 106:139–146. Perry, C. T., G. N. Murphy, P. S. Kench, S. G. Smithers, E. N. Edinger, R. S. Steneck, and P. J. Mumby. 2013. “Caribbean-­Wide De-

cline in Carbonate Production Threatens Coral Reef Growth,” Nature Communications 4:1402. Pershing, A. J., M. A. Alexander, C. M. Hernandez, L. A. Kerr, A. Le Bris, K. E. Mills, J. A. Nye, et al. 2015. “Slow Adaptation in the Face of Rapid Warming Leads to Collapse of the Gulf of Maine Cod Fishery,” Science 350:809–812. Pershing, A. J., L. B. Christensen, N. R. Rec­ord, G. D. Sherwood, and P. B. Stetson. 2010. “The Impact of Whaling on the Ocean Carbon Cycle: Why Bigger Was Better,” PLoS ONE 5:e12444. Pershing, A. J., K. E. Mills, N. R. Rec­ord, K. Stamieszkin, K. V. Wurtzell, C. J. Byron, D. Fitzpatrick, W. J. Golet, and E. Koob. 2014. “Evaluating Trophic Cascades as ­Drivers of Regime Shifts in Dif­fer­ent Ocean Ecosystems,” Philosophical Transactions of the Royal Society B-­Biological Sciences 370:20130265. Petchey, O., and K. Gaston. 2002. “Functional Diversity (Fd), Species Richness and Community Composition,” Ecol­ogy Letters 5:402–411. Petchey, O. L., A. P. Beckerman, J. O. Riede, and P. H. Warren. 2008. “Size, Foraging, and Food Web Structure,” Proceedings of the National Acad­emy of Sciences 105:4191–4196. Petersen, C. G. J. 1918. “The Sea Bottom and Its Production of Fish-­ Food,” Report of the Danish Biological Station 25:1–66. Petersen, J. K., J. W. Hansen, M. B. Laursen, P. Clausen, J. Carstensen, and D. J. Conley. 2008. “Regime Shift in a Coastal Marine Ecosystem,” Ecological Applications 18:497–510. Petersen, J. M., F. U. Zielinski, T. Pape, R. Seifert, C. Moraru, R. Amann, S. Hourdez, et al. 2011. “Hydrogen Is an Energy Source for Hydrothermal Vent Symbioses,” Nature 476:176–180. Peterson, B. J., and Fry, B. 1987. “Stable Isotopes in Ecosystem Studies,” Annual Review of Ecol­ogy and Systematics 18:293–320. Peterson, C. 1984. “Does a Rigorous Criterion for Environmental Identity Preclude the Existence of Multiple Stable Points?” American Naturalist 124:127–133. Peterson, C. H. 1979. “Predation, Competitive Exclusion, and Diversity in the Soft-­Sediment Benthic Communities of Estuaries and Lagoons.” Pages 233–264 in R. J. Livingston, editor, Ecological Pro­cesses in Coastal and Marine Systems. Boston: Springer. Peterson, C. H., and S. V. Andre. 1980. “An Experimental Analy­sis of Interspecific Competition among Marine Filter Feeders in a Soft-­Sediment Environment,” Ecol­ogy 61:129–139. Pianka, E. R. 1966. “Latitudinal Gradients in Species Diversity: A Review of Concepts,” American Naturalist 100:33–46. Pietsch, T. W. 2005. “Dimorphism, Parasitism, and Sex Revisited: Modes of Reproduction among Deep-­Sea Ceratioid Anglerfishes (Teleostei: Lophiiformes),” Ichthyological Research 52:207–236. Pimm, S. L., C. N. Jenkins, R. Abell, T. M. Brooks, J. L. Gittleman, L. N. Joppa, P. H. Raven, C. M. Roberts, and J. O. Sexton. 2014. “The Biodiversity of Species and Their Rates of Extinction, Distribution, and Protection,” Science 344:1246752. Pineda, J. 1991. “Predictable Upwelling and the Shoreward Transport of Planktonic Larvae by Internal Tidal Bores,” Science 253:548–549. Pinker, S. 2011. The Better Angels of Our Nature: The Decline of Vio­lence in History and Its C ­ auses. New York: Penguin UK. Pinker, S. 2018. Enlightenment Now: The Case for Reason, Science, Humanism, and Pro­gress. New York: Penguin. Pittman, S., J. Christensen, C. Caldow, C. Menza, and M. Monaco. 2007. “Predictive Mapping of Fish Species Richness across Shallow-­ Water Seascapes in the Ca­rib­bean,” Ecological Modelling 204:9–21.

Literature Cited

Pittman, S. J., B. M. Costa, and T. A. Battista. 2009. “Using Lidar Bathymetry and Boosted Regression Trees to Predict the Diversity and Abundance of Fish and Corals,” Journal of Coastal Research 10053:27–38. Platt, T., C. Fuentes-­Yaco, and K. T. Frank. 2003. “Spring Algal Bloom and Larval Fish Survival,” Nature 423:398–399. Poff, N., J. Olden, N. Vieira, D. Finn, M. Simmons, and B. Kondratieff. 2006. “Functional Trait Niches of North American Lotic Insects,” Journal of the North American Benthological Society 25:730–755. Polidoro, B. A., K. E. Carpenter, L. Collins, N. C. Duke, A. M. Ellison, J. C. Ellison, E. J. Farnsworth, et al. 2010. “The Loss of Species: Mangrove Extinction Risk and Geographic Areas of Global Concern,” PLoS ONE 5:e10095. Pollock, F. J., J. B. Lamb, S. N. Field, S. F. Heron, B. Schaffelke, G. Shedrawi, D. G. Bourne, and B. L. Willis. 2014. “Sediment and Turbidity Associated with Offshore Dredging Increase Coral Disease Prevalence on Nearby Reefs,” PLoS ONE 9:e102498–9. Poloczanska, E. S., C. J. Brown, and W. J. Sydeman. 2013. “Global Imprint of Climate Change on Marine Life,” Nature Climate Change 3:919–925. Polovina, J. J. 1984. “Model of a Coral Reef Ecosystem I. The ECOPATH Model and Its Application to French Frigate Shoals,” Coral Reefs 3:1–11. Pomeroy, L. 1974. “The Ocean’s Food Web, a Changing Paradigm,” Bio Science 24:499–504. Pomeroy, L., C. D’Elia, and L. Schaffner. 2006. “Limits to Top-­Down Control of Phytoplankton by Oysters in Chesapeake Bay,” Marine Ecol­ogy Pro­gress Series 325:301–309. Ponsard, J., M.-­A . Cambon-­Bonavita, M. Zbinden, G. Lepoint, A. Joassin, L. Corbari, B. Shillito, L. Durand, V. Cueff-­Gauchard, and P. Compère. 2012. “Inorganic Carbon Fixation by Chemosynthetic Ectosymbionts and Nutritional Transfers to the Hydrothermal Vent Host-­Shrimp Rimicaris exoculata,” ISME Journal 7:96–109. Poore, A. G. B., A. H. Campbell, R. A. Coleman, G. J. Edgar, V. Jormalainen, P. L. Reynolds, E. E. Sotka, et al. 2012. “Global Patterns in the Impact of Marine Herbivores on Benthic Primary Producers,” Ecol­ogy Letters 15:912–922. Pörtner, H.-­O., and A. P. Farrell. 2008. “Physiology and Climate Change,” Science 322:690–692. Posey, M. H., and W. G. Ambrose. 1994. “Effects of Proximity to an Offshore Hard-­Bottom Reef on Infaunal Abundances,” Marine Biology 118:745–753. Potter, I. C., J. R. Tweedley, M. Elliott, and A. K. Whitfield. 2015. “The Ways in Which Fish Use Estuaries: A Refinement and Expansion of the Guild Approach,” Fish and Fisheries 16:230–239. Power, M., D. Tilman, J. Estes, B. Menge, W. Bond, L. Mills, G. Daily, J. Castilla, J. Lubchenco, and R. Paine. 1996. “Challenges in the Quest for Keystones,” BioScience 46:609–620. Power, M. E. 1992. “Top-­Down and Bottom-­Up Forces in Food Webs: Do Plants Have Primacy?” Ecol­ogy 73:733–746. Pratchett, M. S. 2005. “Dietary Overlap among Coral-­Feeding Butterflyfishes (Chaetodontidae) at Lizard Island, Northern ­Great Barrier Reef,” Marine Biology 148:373–382. Preisser, E. L., D. I. Bolnick, and M. F. Benard. 2005. “Scared to Death? The Effects of Intimidation and Consumption in Predator-­Prey Interactions,” Ecol­ogy 86:501–509. Prowe, A. E. F., M. Pahlow, S. Dutkiewicz, M. Follows, and A. Oschlies. 2012. “Top-­Down Control of Marine Phytoplankton Diversity in a Global Ecosystem Model,” Pro­gress in Oceanography 101:1–13.

419

Ptacnik, R., A. G. Solimini, T. Andersen, T. Tamminen, P. Brettum, L. Lepisto, E. Willen, and S. Rekolainen. 2008. “Diversity Predicts Stability and Resource Use Efficiency in Natu­ral Phytoplankton Communities,” Proceedings of the National Acad­emy of Sciences 105:5134–5138. Pulkkinen, K., L.-­R . Suomalainen, A. F. Read, D. Ebert, P. Rintamaki, and E. T. Valtonen. 2009. “Intensive Fish Farming and the Evolution of Pathogen Virulence: The Case of Columnaris Disease in Finland,” Proceedings of the Royal Society B-­Biological Sciences 277:593–600. Quenouille, B., N. Hubert, E. Bermingham, and S. Planes. 2011. “Speciation in Tropical Seas: Allopatry Followed by Range Change,” Molecular Phyloge­ne­tics and Evolution 58:546–552. Quigg, A., Z. Finkel, A. Irwin, Y. Rosenthal, T. Ho, J. Reinfelder, O. Schofield, F. Morel, and P. Falkowski. 2003. “The Evolutionary Inheritance of Elemental Stoichiometry in Marine Phytoplankton,” Nature 425:291–294. Rabalais, N. N., R. E. Turner, and B. Gupta. 2007. “Sediments Tell the History of Eutrophication and Hypoxia in the Northern Gulf of Mexico,” Ecological Applications 17:S129–­S143. Ramirez-­Llodra, E., T. M. Shank, and C. R. German. 2007. “Biodiversity and Biogeography of Hydrothermal Vent Species: Thirty Years of Discovery and Investigations,” Oceanography 20:30–41. Ramirez-­Llodra, E., P. A. Tyler, M. C. Baker, O. A. Bergstad, M. R. Clark, E. Escobar, L. A. Levin, et al. 2011. “Man and the Last ­Great Wilderness: ­Human Impact on the Deep Sea,” PLoS ONE 6:e22588. Rasher, D. B., A. S. Hoey, and M. E. Hay. 2013. “Consumer Diversity Interacts with Prey Defenses to Drive Ecosystem Function,” Ecol­ ogy 94:1347–1358. Rasmussen, E. 1977. “The Wasting Disease of Eelgrass (Zostera marina) and Its Effects on Environmental ­Factors and Fauna.” Pages 1–52 in C. P. McRoy and C. Helfferich, editors, Seagrass Ecosystems: A Scientific Perspective. New York: Dekker. Raven, J. A. 2009. “Contributions of Anoxygenic and Oxygenic Phototrophy and Chemolithotrophy to Carbon and Oxygen Fluxes in Aquatic Environments,” Aquatic Microbial Ecol­ogy 56:177–192. Raymundo, L. J., A. R. Halford, A. P. Maypa, and A. M. Kerr. 2009. “Functionally Diverse Reef-­Fish Communities Ameliorate Coral Disease,” Proceedings of the National Acad­emy of Sciences 106:17067–17070. Rec­ord, N. R., A. J. Pershing, and F. Maps. 2012. “First Princi­ples of Copepod Development Help Explain Global Marine Diversity Patterns,” Oecologia 170:289–295. Redfield, A. C. 1934. “On the Proportions of Organic Derivations in Sea ­Water and Their Relation to the Composition of Plankton.” Pages 176–192 in James Johnstone Memorial Volume. Liverpool, UK: University Press of Liverpool. Redfield, A. C. 1972. “Development of a New ­England Salt Marsh,” Ecological Monographs 42:201–237. Reecht, Y., M.-­J. Rochet, V. M. Trenkel, S. Jennings, and J. K. Pinnegar. 2013. “Use of Morphological Characteristics to Define Functional Groups of Predatory Fishes in the Celtic Sea,” Journal of Fish Biology 83:355–377. Reichman, O. 2007. “Predator Diversity and Trophic Interactions,” Ecol­ ogy 88:2415–2426. Reid, P. C., D. G. Johns, M. Edwards, M. Starr, and P. Snoeijs. 2007. “A Biological Consequence of Reducing Arctic Ice Cover: Arrival of the Pacific Diatom Neodenticula seminae in the North Atlantic for the First Time in 800 000 Years,” Global Change Biology 13:1910–1921.

420

Ocean Ecology

Reid, W. V., D. Chen, L. Goldfarb, H. Hackmann, Y. T. Lee, K. Mokhele, E. Ostrom, et al. 2010. “Earth System Science for Global Sustainability: ­Grand Challenge,” Science 330:916–917. Reise, K. 2002. “Sediment Mediated Species Interactions in Coastal ­Waters,” Journal of Sea Research 48:127–141. Remane, A. 1934. “Die Brackwasserfauna,” Zoologischer Anzeiger 7 (Suppl.):34–74. Reuman, D. C., C. Mulder, C. Banašek-­Richter, M. F. C. Blandenier, A. M. Breure, H. Den Hollander, J. M. Kneitel, D. Raffaelli, G. Woodward, and J. E. Cohen. 2009. “Allometry of Body Size and Abundance in 166 Food Webs,” Advances in Ecological Research 41:1–44. Reusch, T., C. Boström, W. T. Stam, and J. L. Olsen. 1999. “An Ancient Eelgrass Clone in the Baltic,” Marine Ecol­ogy Pro­gress Series 183:301–304. Reusch, T. B. H., A. Ehlers, A. Hämmerli, and B. Worm. 2005. “Ecosystem Recovery ­after Climatic Extremes Enhanced by Genotypic Diversity,” Proceedings of the National Acad­emy of Sciences 102:2826–2831. Rex, M. 1981. “Community Structure in the Deep-­Sea Benthos,” Annual Review of Ecol­ogy and Systematics 12:331–353. Rex, M., C. Stuart, R. Hessler, J. Allen, H. Sanders, and G. Wilson. 1993. “Global-­Scale Latitudinal Patterns of Species Diversity in the Deep-­Sea Benthos,” Nature 365:636–639. Reynolds, L. K., M. Waycott, K. J. McGlathery, R. J. Orth, and J. C. Zieman. 2012. “Eelgrass Restoration by Seed Maintains Ge­ne­tic Diversity: Case Study from a Coastal Bay System,” Marine Ecol­ogy Pro­gress Series 448:223–233. Reynolds, P. L., and J. F. Bruno. 2013. “Multiple Predator Species Alter Prey Be­hav­ior, Population Growth, and a Trophic Cascade in a Model Estuarine Food Web,” Ecological Monographs 83:119–132. Reynolds, P. L., J. P. Richardson, and J. E. Duffy. 2014. “Field Experimental Evidence That Grazers Mediate Transition between Microalgal and Seagrass Dominance,” Limnology and Oceanography 59:1053–1064. Reynolds, P. L., and E. E. Sotka. 2011. “Non-­consumptive Predator Effects Indirectly Influence Marine Plant Biomass and Palatability,” Journal of Ecol­ogy 99:1272–1281. Reynolds, P. L., J. J. Stachowicz, K. Hovel, C. Boström, K. Boyer, M. Cusson, J. S. Eklöf, et al. 2018. “Latitude, Temperature and Habitat Complexity Predict Predation Pressure in Eelgrass Beds across the Northern Hemi­sphere,” Ecol­ogy 99:29–35. Rhoads, D., and D. Young. 1970. “The Influence of Deposit-­Feeding Organisms on Sediment Stability and Community Trophic Structure,” Journal of Marine Research 28:150–178. Ricciardi, A., and E. Bourget. 1999. “Global Patterns of Macroinvertebrate Biomass in Marine Intertidal Communities,” Marine Ecol­ogy Pro­gress Series 185:21–35. Richards, D. R., and D. A. Friess. 2016. “Rates and ­Drivers of Mangrove Deforestation in Southeast Asia, 2000–2012,” Proceedings of the National Acad­emy of Sciences 113:344–349. Richardson, A. J., A. Bakun, G. C. Hays, and M. J. Gibbons. 2009. “The Jellyfish Joyride: ­Causes, Consequences and Management Responses to a More Gelatinous ­Future,” Trends in Ecol­ogy & Evolution 24:312–322. Ricklefs, R. E. 1987. “Community Diversity: Relative Roles of Local and Regional Pro­cesses,” Science 235:167–171. Ricklefs, R. E. 2004. “A Comprehensive Framework for Global Patterns in Biodiversity,” Ecol­ogy Letters 7:1–15. Ricklefs, R. E. 2008. “Disintegration of the Ecological Community,” American Naturalist 172:741–750.

Ricklefs, R. E., and D. Schluter. 1993. Species Diversity in Ecological Communities: Historical and Geo­graph­i­cal Perspectives. Chicago: University of Chicago Press. Riginos, C., Y. M. Buckley, S. P. Blomberg, and E. A. Treml. 2014. “Dispersal Capacity Predicts Both Population Ge­ne­tic Structure and Species Richness in Reef Fishes,” American Naturalist 184:52–64. Riley, G. A. 1946. “­Factors Controlling Phytoplankton Populations on Georges Bank,” Journal of Marine Research 6:54–73. Rivero-­Calle, S., A. Gnanadesikan, C. E. Del Castillo, W. M. Balch, and S. D. Guikema. 2015. “Multidecadal Increase in North Atlantic Coccolithophores and the Potential Role of Rising CO2,” Science 350:1533–1537. Roberts, C. 2002. “Deep Impact: The Rising Toll of Fishing in the Deep Sea,” Trends in Ecol­ogy & Evolution 17:242–245. Rocha, L. A., C. R. Rocha, D. R. Robertson, and B. W. Bowen. 2008. “Comparative Phylogeography of Atlantic Reef Fishes Indicates Both Origin and Accumulation of Diversity in the Ca­rib­ bean,” BMC Evolutionary Biology 8:157. Rockström, J., W. Steffen, K. Noone, A. Persson, F. S. Chapin, E. F. Lambin, T. M. Lenton, et al. 2009. “A Safe Operating Space for Humanity,” Nature 461:472–475. Rodriguez, J., J. Tintoré, J. T. Allen, J. M. Blanco, and D. Gomis. 2001. “Mesoscale Vertical Motion and the Size Structure of Phytoplankton in the Ocean,” Nature 410:360–363. Roff, G., and P. J. Mumby. 2012. “Global Disparity in the Resilience of Coral Reefs,” Trends in Ecol­ogy & Evolution 27:404–413. Rohde, K. 1992. “Latitudinal Gradients in Species Diversity: The Search for the Primary Cause,” Oikos 65(3):514–527. Rohr, J. R., D. J. Civitello, F. W. Halliday, P. J. Hudson, K. D. Lafferty, C. L. Wood, and E. A. Mordecai. 2020. “­Towards Common Ground in the Biodiversity-­Disease Debate,” Nature Ecol­ogy & Evolution 4:24–33. Röhr, M. E., M. Holmer, J. K. Baum, M. Björk, K. Boyer, D. Chin, L. Chalifour, et al. 2018. “Blue Carbon Storage Capacity of Temperate Eelgrass (Zostera marina) Meadows,” Global Biogeochemical Cycles 32:1457–1475. Rolland, J., F. L. Condamine, F. Jiguet, and H. Morlon. 2014. “Faster Speciation and Reduced Extinction in the Tropics Contribute to the Mammalian Latitudinal Diversity Gradient,” PLoS Biology 12:e1001775. Roman, J., J. A. Estes, L. Morissette, C. Smith, D. Costa, J. McCarthy, J. B. Nation, S. Nicol, A. Pershing, and V. Smetacek. 2014. “Whales as Marine Ecosystem Engineers,” Frontiers in Ecol­ogy and the Environment 12:377–385. Roman, J., and S. R. Palumbi. 2003. “Whales before Whaling in the North Atlantic,” Science 301:508–510. Rosenzweig, M. L., and Z. Abramsky. 1993. “How Are Diversity and Productivity Related?” Pages 52–65 in R. E. Ricklefs and D. Schluter, editors, Species Diversity in Ecological Communities: Historical and Geo­graph­i­cal Perspectives. Chicago: University of Chicago Press. Roughgarden, J., S. Gaines, and H. Possingham. 1988. “Recruitment Dynamics in Complex Life Cycles,” Science 241:1460–1466. Roy, K., D. Jablonski, J. W. Valentine, and G. Rosenberg. 1998. “Marine Latitudinal Diversity Gradients: Tests of Causal Hypotheses,” Proceedings of the National Acad­emy of Sciences 95:3699–3702. Ruddiman, W. F. 2013. “The Anthropocene,” Annual Review of Earth and Planetary Sciences 41:45–68. Ruesink, J. L., B. E. Feist, C. J. Harvey, J. S. Hong, A. C. Trimble, and L. M. Wisehart. 2006. “Changes in Productivity Associated with Four Introduced Species: Ecosystem Transformation of a

Literature Cited

‘Pristine’ Estuary,” Marine Ecol­ogy Pro­gress Series 311:2013–2215. Ruesink, J. L., H. S. Lenihan, A. C. Trimble, K. W. Heiman, F. Micheli, J. E. Byers, and M. C. Kay. 2005. “Introduction of Non-­native Oysters: Ecosystem Effects and Restoration Implications,” Annual Review of Ecol­ogy Evolution and Systematics 36:643–689. Ruff, S. E., J. F. Biddle, A. P. Teske, K. Knittel, A. Boetius, and A. Ramette. 2015. “Global Dispersion and Local Diversification of the Methane Seep Microbiome,” Proceedings of the National Acad­ emy of Sciences 112:4015–4020. Ruiz, G., P. Fofonoff, A. Hines, and E. Grosholz. 1999. “Non-­indigenous Species as Stressors in Estuarine and Marine Communities: Assessing Invasion Impacts and Interactions,” Limnology and Oceanography 44:950–972. Ruiz, G. M., P. W. Fofonoff, J. T. Carlton, M. J. Wonham, and A. H. Hines. 2000. “Invasion of Coastal Marine Communities in North Amer­i­ca: Apparent Patterns, Pro­cesses, and Biases,” Annual Review of Ecol­ogy and Systematics 31:481–531. Rundle, H. D., and P. Nosil. 2005. “Ecological Speciation,” Ecol­ogy Letters 8:336–352. Russell, B. D., S. D. Connell, S. Uthicke, N. Muehllehner, K. E. Fabricius, and J. M. Hall-­Spencer. 2013. “­Future Seagrass Beds: Can Increased Productivity Lead to Increased Carbon Storage?” Marine Pollution Bulletin 73:463–469. Russell, R., A. D. Guerry, P. Balvanera, R. K. Gould, X. Basurto, K. M. A. Chan, S. Klain, J. Levine, and J. Tam. 2013. “­Humans and Nature: How Knowing and Experiencing Nature Affect Well-­ Being,” Annual Review of Environment and Resources 38:473–502. Rutherford, S., S. D’Hondt, and W. Prell. 1999. “Environmental Controls on the Geographic Distribution of Zooplankton Diversity,” Nature 400:749–753. Sadovy, Y. 2005. “Trou­ble on the Reef: The Imperative for Managing Vulnerable and Valuable Fisheries,” Fish and Fisheries 6:167–185. Saintilan, N., N. C. Wilson, K. Rogers, A. Rajkaran, and K. W. Krauss. 2013. “Mangrove Expansion and Salt Marsh Decline at Mangrove Poleward Limits,” Global Change Biology 20:147–157. Sala, E. and S. Giakoumi. 2018. “No-­take Marine Reserves Are the Most Effective Protected Areas in the Ocean,” ICES Journal of Marine Science 75:1166–1168. Sala, E., and M. H. Graham. 2002. “Community-­Wide Distribution of Predator-­Prey Interaction Strength in Kelp Forests,” Proceedings of the National Acad­emy of Sciences 99:3678–3683. Sale, P. F. 1977. “Maintenance of High Diversity in Coral Reef Fish Communities,” American Naturalist 111:337–359. Salo, T., and C. Gustafsson. 2016. “The Effect of Ge­ne­tic Diversity on Ecosystem Functioning in Vegetated Coastal Ecosystems,” Ecosystems 19:1429–1444. Salomon, A. K., S. K. Gaichas, O. P. Jensen, V. N. Agostini, N. A. Sloan, J. Rice, T. R. McClanahan, M. H. Ruckelshaus, P. S. Levin, N. K. Dulvy, and E. A. Babcock. 2011. “Bridging the Divide between Fisheries and Marine Conservation Science,” Bulletin of Marine Science 87:251–274. Sanders, H., and R. Hessler. 1969. “Ecol­ogy of the Deep-­Sea Benthos,” Science 163:1419. Sandin, S. A., J. E. Smith, E. E. Demartini, E. A. Dinsdale, S. D. Donner, A. M. Friedlander, T. Konotchick, et al. 2008. “Baselines and Degradation of Coral Reefs in the Northern Line Islands,” PLoS ONE 3:e1548. Sandom, C., S. Faurby, B. Sandel, and J. C. Svenning. 2014. “Global Late Quaternary Megafauna Extinctions Linked to ­Humans, Not

421

Climate Change,” Proceedings of the Royal Society B-­Biological Sciences 281:20133254. Sanford, E. 1999. “Regulation of Keystone Predation by Small Changes in Ocean Temperature,” Science 283:2095. Sanford, E. 2014. “The Biogeography of Marine Communities.” Pages 131–163 in M. D. Bertness, J. F. Bruno, B. R. Silliman, and J. J. Stachowicz, editors, Marine Community Ecol­ogy and Conservation. Sunderland, MA: Sinauer Associates. Sanz-­Aguilar, A., J. M. Igual, G. Tavecchia, M. Genovart, and D. Oro. 2016. “When Immigration Mask Threats: The Rescue Effect of a Scopoli’s Shearwater Colony in the Western Mediterranean as a Case Study,” Biological Conservation 198:33–36. Sathirathai, S., and E. B. Barbier. 2001. “Valuing Mangrove Conservation in Southern Thailand,” Con­temporary Economic Policy 19:109–122. Savage, V. M., J. F. Gillooly, J. H. Brown, G. B. West, and E. L. Charnov. 2004. “Effects of Body Size and Temperature on Population Growth,” American Naturalist 163:429–441. Savage, V. M., J. F. Gillooly, W. H. Woodruff, G. B. West, A. P. Allen, B. J. Enquist, and J. H. Brown. 2004. “The Predominance of Quarter-­Power Scaling in Biology,” Functional Ecol­ogy 18:257–282. Sax, D., and S. Gaines. 2008. “Species Invasions and Extinction: The ­Future of Native Biodiversity on Islands,” Proceedings of the National Acad­emy of Sciences 105:11490. Sax, D., and S. D. Gaines. 2003. “Species Diversity: From Global Decreases to Local Increases,” Trends in Ecol­ogy & Evolution 18:561–566. Scheffer, M. 2004. “Mechanisms for Marine Regime Shifts: Can We Use Lakes as Microcosms for Oceans?” Pro­gress in Oceanography 60:303–319. Scheffer, M., S. Carpenter, J. A. Foley, C. Folke, and B. Walker. 2001. “Catastrophic Shifts in Ecosystems,” Nature 413:591–596. Scheffer, M., and S. R. Carpenter. 2003. “Catastrophic Regime Shifts in Ecosystems: Linking Theory to Observation,” Trends in Ecol­ ogy & Evolution 18:648–656. Scheffer, M., S. Rinaldi, J. Huisman, and F. J. Weissing. 2003. “Why Plankton Communities Have No Equilibrium: Solutions to the Paradox,” Hydrobiologia 491:9–18. Schemske, D. W., G. G. Mittelbach, H. V. Cornell, J. M. Sobel, and K. Roy. 2009. “Is ­There a Latitudinal Gradient in the Importance of Biotic Interactions?” Annual Review of Ecol­ogy Evolution and Systematics 40:245–269. Schindler, D. E., R. Hilborn, B. Chasco, C. P. Boatright, T. P. Quinn, L. A. Rogers, and M. S. Webster. 2010. “Population Diversity and the Portfolio Effect in an Exploited Species,” Nature 465:609–612. Schindler, D. W. 1978. “­Factors Regulating Phytoplankton Production and Standing Crop in the World’s Freshwaters,” Limnology and Oceanography 23:478–486. Schluter, D. 1996. “Ecological Causes of Adaptive Radiation,” The American Naturalist 148:S40–S64. Schmidt, A. L., M. Coll, and H. K. Lotze. 2017. “Regional-­Scale Differences in Eutrophication Effects on Eelgrass-­Associated (Zostera marina) Macrofauna,” Estuaries and Coasts 40:1096–1112. Schmitz, O. J. 2007. “Predator Diversity and Trophic Interactions,” Ecology 88(10):2415–2426. Schmoker, C., S. Hernandez-­Leon, and A. Calbet. 2013. “Microzooplankton Grazing in the Oceans: Impacts, Data Variability, Knowledge Gaps and ­Future Directions,” Journal of Plankton Research 35:691–706.

422

Ocean Ecology

Schramski, J. R., D. K. Gattie, and J. H. Brown. 2015. “­Human Domination of the Biosphere: Rapid Discharge of the Earth-­Space Battery Foretells the ­Future of Humankind,” Proceedings of the National Acad­emy of Sciences 112:9511–9517. Schultze, E. D., and H. A. Mooney. 1994. Biodiversity and Ecosystem Function. New York: Springer. Schwartz, J. H., and I. Tattersall. 2010. “Fossil Evidence for the Origin of Homo sapiens,” Yearbook of Physical Anthropology 53:94–121. Schwinghamer, P. 1981. “Characteristic Size Distributions of Integral Benthic Communities,” Canadian Journal of Fisheries and Aquatic Sciences 38:1255–1263. Seehausen, O. 2006. “African Cichlid Fish: A Model System in Adaptive Radiation Research,” Proceedings of the Royal Society B-­ Biological Sciences 273:1987–1998. Self, R., and P. Jumars. 1988. “Cross-­Phyletic Patterns of Particle Se­ lection by Deposit Feeders,” Journal of Marine Research 46:119–143. Sepkoski, J. J. Jr. 1981. “A ­Factor Analytic Description of the Phanerozoic Marine Fossil Rec­ord,” Paleobiology 7:36–53. Seto, K. C., B. Güneralp, and L. R. Hutyra. 2012. “Global Forecasts of Urban Expansion to 2030 and Direct Impacts on Biodiversity and Carbon Pools,” Proceedings of the National Acad­emy of Sciences 109:16083–16088. Shakun, J. D., P. U. Clark, F. He, S. A. Marcott, A. C. Mix, Z. Liu, B. Otto-­ Bliesner, A. Schmittner, and E. Bard. 2012. “Global Warming Preceded by Increasing Carbon Dioxide Concentrations during the Last Deglaciation,” Nature 484:49–54. Shanks, A. 1983. “Surface Slicks Associated with Tidally Forced Internal Waves May Transport Pelagic Larvae of Benthic Invertebrates and Fishes Shoreward,” Marine Ecol­ogy Pro­gress Series, Oldendorf 13:311–315. Sheldon, R. W., A. Prakash, and W. H. Sutcliffe. 1972. “The Size Distribution of Particles in the Ocean,” Limnology and Oceanography 17:327–340. Shen, S. Z., J. L. Crowley, Y. Wang, S. A. Bowring, D. H. Erwin, P. M. Sadler, C. Q. Cao, et al. 2011. “Calibrating the End-­Permian Mass Extinction,” Science 334:1367–1372. Sheppard, C., A. Harris, and A. Sheppard. 2008. “Archipelago-­Wide Coral Recovery Patterns since 1998 in the Chagos Archipelago, Central Indian Ocean,” Marine Ecol­ogy Pro­gress Series, Oldendorf 362:109–117. Sherman, K. 1991. “The Large Marine Ecosystem Concept: Research and Management Strategy for Living Marine Resources,” Ecological Applications 1(4):349–360. Shi, D., Y. Xu, B. M. Hopkinson, and F. M. M. Morel. 2010. “Effect of Ocean Acidification on Iron Availability to Marine Phytoplankton,” Science 327:676–679. Shinen, J. L., and S. A. Navarrete. 2014. “Lottery Coexistence on Rocky Shores: Weak Niche Differentiation or Equal Competitors Engaged in Neutral Dynamics?” American Naturalist 183:342–362. Shiomoto, A., K. Tadokoro, K. Nagasawa, and Y. Ishida. 1997. “Trophic Relations in the Subarctic North Pacific Ecosystem: Pos­si­ble Feeding Effect from Pink Salmon,” Marine Ecol­ogy Pro­gress Series 150:75–85. Shmida, A., and M. V. Wilson. 1985. “Biological Determinants of Species Diversity,” Journal of Biogeography 12:1–20. Short, F., T. Carruthers, W. Dennison, and M. Waycott. 2007. “Global Seagrass Distribution and Diversity: A Bioregional Model,” Journal of Experimental Marine Biology and Ecol­ogy 350(1–2):3–20.

Short, F., and H. Neckles. 1999. “The Effects of Global Climate Change on Seagrasses,” Aquatic Botany 63:169–196. Short, F. T., L. Muehlstein, and D. Porter. 1987. “Eelgrass Wasting Disease: Cause and Recurrence of a Marine Epidemic,” Biological Bulletin 173:557–562. Short, F. T., B. Polidoro, S. R. Livingstone, K. E. Carpenter, S. Bandeira, J. S. Bujang, H. P. Calumpong, et al. 2011. “Extinction Risk Assessment of the World’s Seagrass Species,” Biological Conservation 144:1961–1971. Short, S. M. 2012. “The Ecol­ogy of Viruses That Infect Eukaryotic Algae,” Environmental Microbiology 14:2253–2271. Shurin, J., E. Borer, E. Seabloom, K. Anderson, C. Blanchette, B. Broitman, S. Cooper, and B. Halpern. 2002. “A Cross-­Ecosystem Comparison of the Strength of Trophic Cascades,” Ecol­ogy Letters 5:785–791. Shurin, J. B., D. S. Gruner, and H. Hillebrand. 2005. “All Wet or Dried Up? Real Differences between Aquatic and Terrestrial Food Webs,” Proceedings of the Royal Society B-­Biological Sciences 273:1–9. Siepielski, A. M., K.-­L. Hung, E. E. Bein, and M. A. McPeek. 2010. “Experimental Evidence for Neutral Community Dynamics Governing an Insect Assemblage,” Ecol­ogy 91:847–857. Sieracki, M. E., P. G. Verity, and D. K. Stoecker. 1993. “Plankton Community Response to Sequential Silicate and Nitrate Depletion during the 1989 North Atlantic Spring Bloom,” Deep Sea Research II 40:213–225. Sih, A., P. Crowley, M. McPeek, J. Petranka, and K. Strohmeier. 1985. “Predation, Competition, and Prey Communities: A Review of Field Experiments,” Annual Review of Ecol­ogy and Systematics 16:269–311. Silliman, B. R. 2005. “Drought, Snails, and Large-­Scale Die-­Off of Southern U.S. Salt Marshes,” Science 310:1803–1806. Silliman, B. R., and M. D. Bertness. 2002. “A Trophic Cascade Regulates Salt Marsh Primary Production,” Proceedings of the National Acad­emy of Sciences 99:10500–10505. Silliman, B. R., and S. Y. Newell. 2003. “Fungal Farming in a Snail,” Proceedings of the National Acad­emy of Sciences 100:15643–15648. Silliman, B. R., E. Schrack, Q. He, R. Cope, A. Santoni, T. van der Heide, R. Jacobi, M. Jacobi, and J. Van De Koppel. 2015. “Facilitation Shifts Paradigms and Can Amplify Coastal Restoration Efforts,” Proceedings of the National Acad­emy of Sciences 112:14295–14300. Simpson, S. D., M. Meekan, J. Montgomery, R. McCauley, and A. Jeffs. 2005. “Homeward Sound,” Science 308:221. Simpson, S. D., S. Jennings, M. P. Johnson, J. L. Blanchard, P.-­J. Schön, D. W. Sims, and M. J. Genner. 2011. “Continental Shelf-­Wide Response of a Fish Assemblage to Rapid Warming of the Sea,” Current Biology 21:1565–1570. Sirot, C., S. Villéger, D. Mouillot, A. M. Darnaude, J. Ramos Miranda, D. Flores Hernandez, and J. Panfili. 2015. “Combinations of Biological Attributes Predict Temporal Dynamics of Fish Species in Response to Environmental Changes,” Ecological Indicators 48:147–156. Siska, E., S. Pennings, T. Buck, and M. Hanisak. 2002. “Latitudinal Variation in Palatability of Salt-­Marsh Plants: Which Traits Are Responsible?” Ecol­ogy 83:3369–3381. Slater, G. J., J. A. Goldbogen, and N. D. Pyenson. 2017. “In­de­pen­dent Evolution of Baleen Whale Gigantism Linked to Plio-­ Pleistocene Ocean Dynamics,” Proceedings of the Royal Society B-­Biological Sciences 284:20170546. Smalley, A. E. 1960. “Energy Flow of a Salt Marsh Grasshopper Population,” Ecol­ogy 41:672–677.

Literature Cited

Smetacek, V. 1999. “Diatoms and the Ocean Carbon Cycle,” Protist 150:25–32. Smetacek, V., and U. Passow. 1990. “Spring Bloom Initiation and Sverdrup’s Critical-­Depth Model,” Limnology and Oceanography 35:228–234. Smil, V. 1999. “Detonator of the Population Explosion,” Nature 400:415. Smith, C. R. 1992. “Whale Falls: Chemosynthesis on the Deep Seafloor,” Oceanus 35:74–78. Smith, J., C. Smith, and C. Hunter. 2001. “An Experimental Analy­sis of the Effects of Herbivory and Nutrient Enrichment on Benthic Community Dynamics on a Hawaiian Reef,” Coral Reefs 19:332–342. Smith, M. D., A. Oglend, A. J. Kirkpatrick, F. Asche, L. S. Bennear, J. K. Craig, and J. M. Nance. 2017. “Seafood Prices Reveal Impacts of a Major Ecological Disturbance,” Proceedings of the National Acad­emy of Sciences 114:1512–1517. Smith, T. J. 1987. “Seed Predation in Relation to Tree Dominance and Distribution in Mangrove Forests,” Ecol­ogy 68:266–273. Sogin, M. L., H. G. Morrison, J. A. Huber, D. M. Welch, S. M. Huse, P. R. Neal, J. M. Arrieta, and G. J. Herndl. 2006. “Microbial Diversity in the Deep Sea and the Underexplored ‘Rare Biosphere.’ ” Proceedings of the National Acad­emy of Sciences 103:12115–12120. Solan, M., B. J. Cardinale, A. L. Downing, K. A. M. Engelhardt, J. L. Ruesink, and D. S. Srivastava. 2004. “Extinction and Ecosystem Function in the Marine Benthos,” Science 306:1177–1180. Sole, R. V., and J. X. M. Montoya. 2001. “Complexity and Fragility in Ecological Networks,” Proceedings of the Royal Society B-­Biological Sciences 268:2039–2045. Sommer, U. 1994. “The Impact of Light Intensity and Daylength on Silicate and Nitrate Competition among Marine Phytoplankton,” Limnology and Oceanography 39:1680–1688. Sommer, U. 1995. “An Experimental Test of the Intermediate Disturbance Hypothesis Using Cultures of Marine Phytoplankton,” Limnology and Oceanography 40:1271–1277. Southward, A. J. 1960. “On Changes of Sea Temperature in the En­glish Channel,” Journal of the Marine Biological Association of the UK 39:449–458. Spalding, M. 2010. World Atlas of Mangroves. Milton Park, UK: Routledge. Spalding, M., H. Fox, G. Allen, N. Davidson, Z. Ferdana, M. Finlayson, B. Halpern, M. Jorge, A. Lombana, and S. Lourie. 2007. “Marine Ecoregions of the World: A Bioregionalization of Coastal and Shelf Areas,” BioScience 57:573–583. Springer, A. M., J. A. Estes, G. B. van Vliet, T. M. Williams, D. F. Doak, E. M. Danner, K. A. Forney, and B. Pfister. 2003. “Sequential Megafaunal Collapse in the North Pacific Ocean: An Ongoing Legacy of Industrial Whaling?” Proceedings of the National Acad­emy of Sciences 100:12223–12228. Srinivasan, U. T., W. W. L. Cheung, R. Watson, and U. R. Sumaila. 2010. “Food Security Implications of Global Marine Catch Losses Due to Overfishing,” Journal of Bioeconomics 12:183–200. Srivastava, D. S. 1999. “Using Local-­Regional Richness Plots to Test for Species Saturation: Pitfalls and Potentials,” Journal of Animal Ecol­ogy 68:1–16. Srivastava, D. S., M. W. Cadotte, A. A. M. MacDonald, R. G. Marushia, and N. Mirotchnick. 2012. “Phyloge­ne­tic Diversity and the Functioning of Ecosystems,” Ecol­ogy Letters 15:637–648. Stachowicz, J., H. Fried, R. Osman, and R. Whitlatch. 2002. “Biodiversity, Invasion Re­sis­tance, and Marine Ecosystem Function: Reconciling Pattern and Pro­cess,” Ecol­ogy 83:2575–2590.

423

Stachowicz, J., R. Whitlatch, and R. Osman. 1999. “Species Diversity and Invasion Re­sis­tance in a Marine Ecosystem,” Science 286:1577–1579. Stachowicz, J. J., J. F. Bruno, and J. E. Duffy. 2007. “Understanding the Effects of Marine Biodiversity on Communities and Ecosystems,” Annual Review of Ecol­ogy Evolution and Systematics 38:739–766. Stachowicz, J. J., J. R. Terwin, R. B. Whitlatch, and R. W. Osman. 2002. “Linking Climate Change and Biological Invasions: Ocean Warming Facilitates Nonindigenous Species Invasions,” Proceedings of the National Acad­emy of Sciences 99: 15497–15500. Stanley, S. 1968. “Post-­Paleozoic Adaptive Radiation of Infaunal Bivalve Molluscs: A Consequence of Mantle Fusion and Siphon Formation,” Journal of Paleontology 42:214–229. Stebbins, G. L. 1974. Flowering Plants: Evolution above the Species Level. Cambridge, MA: Belknap Press. Steele, J. 1991. “Marine Functional Diversity,” BioScience 41:470–474. Steele, J. 2004. “Regime Shifts in the Ocean: Reconciling Observations and Theory,” Pro­gress in Oceanography 60:135–141. Steele, J. H. 1985. “A Comparison of Terrestrial and Marine Ecological Systems,” Nature 313:355–358. Steffen, W., W. Broadgate, L. Deutsch, O. Gaffney, and C. Ludwig. 2015. “The Trajectory of the Anthropocene: The ­Great Acceleration,” Anthropocene Review 2:81–98. Steffen, W., J. Crutzen, and J. R. McNeill. 2007. “The Anthropocene: Are ­Humans Now Overwhelming the ­Great Forces of Nature?” AMBIO: Journal of the ­Human Environment 36:614–621. Steinberg, D. K., and M. R. Landry. 2017. “Zooplankton and the Ocean Carbon Cycle,” Annual Review of Marine Science 9:413–444. Steinberg, D. K., B. A. Van Mooy, K. O. Buesseler, P. W. Boyd, T. Kobari, and D. M. Karl. 2008. “Bacterial vs. Zooplankton Control of Sinking Particle Flux in the Ocean’s Twilight Zone,” Limnology and Oceanography 53:1327–1338. Steneck, R. S., D. R. Bellwood, and M. E. Hay. 2017. “Herbivory in the Marine Realm,” Current Biology 27:R1–­R6. Steneck, R. S., and M. N. Dethier. 1994. “A Functional Group Approach to the Structure of Algal-­Dominated Communities,” Oikos 69:476–498. Steneck, R. S., M. H. Graham, B. J. Bourque, D. Corbett, J. M. Erlandson, J. A. Estes, and M. J. Tegner. 2002. “Kelp Forest Ecosystems: Biodiversity, Stability, Resilience and ­Future,” Environmental Conservation 29:24. Steneck, R. S., P. J. Mumby, C. MacDonald, D. B. Rasher, and G. Stoyle. 2018. “Attenuating Effects of Ecosystem Management on Coral Reefs,” Science Advances 4:eaao5493. Steneck, R. S., and L. Watling. 1982. “Feeding Capabilities and Limitation of Herbivorous Molluscs: A Functional Group Approach,” Marine Biology 68:299–319. Stergiou, K. 2002. “Overfishing, Tropicalization of Fish Stocks, Uncertainty and Ecosystem Management: Resharpening Ockham’s Razor,” Fisheries Research 55:1–9. Stern, N. H. 2007. The Economics of Climate Change: The Stern Review. Cambridge, UK: Cambridge University Press. Sterner, R. W., and J. J. Elser. 2002. Ecological Stoichiometry: The Biology of Ele­ments from Molecules to the Biosphere. Prince­ton, NJ: Prince­ ton University Press. Stevens, G. C. 1989. “The Latitudinal Gradient in Geo­graph­ic­ al Range: How So Many Species Coexist in the Tropics,” American Naturalist 133:240–256.

424

Ocean Ecology

Stevenson, C., L. S. Katz, F. Micheli, B. Block, K. W. Heiman, C. Perle, K. Weng, R. Dunbar, and J. Witting. 2007. “High Apex Predator Biomass on Remote Pacific Islands,” Coral Reefs 26:47–51. Stibor, H., O. Vadstein, S. Diehl, A. Gelzleichter, T. Hansen, F. Hantzsche, A. Katechakis, et al. 2004. “Copepods Act as a Switch between Alternative Trophic Cascades in Marine Pelagic Food Webs,” Ecol­ogy Letters 7:321–328. Stocker, T. F., D. Qin, G. K. Plattner, L. V. Alexander, S. K. Allen, N. L. Bindoff, F. M. Breon, et al. 2013. “Technical Summary.” Pages 33–115 in T. F. Stocker, D. Qin, G. K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex, and P. M. Midgley, editors, Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press. Stommel, H. 1963. “Va­ri­e­ties of Oceanographic Experience: The Ocean Can Be Investigated as a Hydrodynamical Phenomenon as Well as Explored Geo­graph­i­cally,” Science 139:572–576. Stomp, M., J. Huisman, F. De Jongh, A. Veraart, D. Gerla, M. Rijkeboer, B. Ibelings, U. Wollenzien, and L. Stal. 2004. “Adaptive Divergence in Pigment Composition Promotes Phytoplankton Biodiversity,” Nature 432:104–107. Strain, E. M. A., R. J. Thomson, F. Micheli, F. P. Mancuso, and L. Airoldi. 2014. “Identifying the Interacting Roles of Stressors in Driving the Global Loss of Canopy-­Forming to Mat-­Forming Algae in Marine Ecosystems,” Global Change Biology 20:3300–3312. Stramma, L., S. Schmidtko, L. A. Levin, and G. C. Johnson. 2010. “Ocean Oxygen Minima Expansions and Their Biological Impacts,” Deep-­Sea Research Part I 57:587–595. Strathmann, R. 1985. “Feeding and Nonfeeding Larval Development and Life-­History Evolution in Marine Invertebrates,” Annual Review of Ecol­ogy and Systematics 16:339–361. Strathmann, R. 1990. “Why Life Histories Evolve Differently in the Sea I,” American Zoologist 30:197–207. Striebel, M., S. Behl, S. Diehl, and H. Stibor. 2009. “Spectral Niche Complementarity and Carbon Dynamics in Pelagic Ecosystems,” American Naturalist 174:141–147. Strong, A., S. Chisholm, C. Miller, and J. Cullen. 2009. “Ocean Fertilization: Time to Move On,” Nature 461:347–348. Stuart-­Smith, R. D., A. E. Bates, J. S. Lefcheck, J. E. Duffy, S. C. Baker, R. J. Thomson, J. F. Stuart-­Smith, et al. 2013. “Integrating Abundance and Functional Traits Reveals New Global Hotspots of Fish Diversity,” Nature 501(7468): 539–542. Stuart-­Smith, R. D., G. J. Edgar, N. S. Barrett, S. J. Kininmonth, and A. E. Bates. 2015. “Thermal Biases and Vulnerability to Warming in the World’s Marine Fauna,” Nature 528:88–92. Suchanek, T. H. 1979. “The Mytilus californianus Community: Studies on the Composition, Structure, Organ­ization, and Dynamics of a Mussel Bed,” PhD thesis, University of Washington. Suchanek, T. H., S. L. Williams, J. C. Ogden, D. K. Hubbard, and I. P. Gill. 1985. “Utilization of Shallow-­Water Seagrass Detritus by Carribbean Deep-­Sea Macrofauna: Δ13C Evidence,” Deep Sea Research Part A: Oceanographic Research Papers 32:201–214. Sul, W. J., T. A. Oliver, H. W. Ducklow, L. A. Amaral-­Zettler, and M. L. Sogin. 2013. “Marine Bacteria Exhibit a Bipolar Distribution,” Proceedings of the National Acad­emy of Sciences 110:2342–2347. ­Sullivan, B. K., T. D. Sherman, V. S. Damare, O. Lilje, and F. H. Gleason. 2013. “Potential Roles of Labyrinthula spp. in Global Seagrass Population Declines,” Fungal Ecol­ogy 6:328–338.

Sumaila, U. R., A. S. Khan, A. J. Dyck, R. Watson, G. Munro, P. Tydemers, and D. Pauly. 2010. “A Bottom-­Up Re-­estimation of Global Fisheries Subsidies,” Journal of Bioeconomics 12:201–225. Sunagawa, S., L. P. Coelho, S. Chaffron, J. R. Kultima, K. Labadie, G. Salazar, B. Djahanschiri, et al. 2015. “Structure and Function of the Global Ocean Microbiome,” Science 348:1261359. Sutherland, J. 1974. “Multiple Stable Points in Natu­ral Communities,” American Naturalist 108:859–873. Sutherland, W. J., R. Aveling, T. M. Brooks, M. Clout, L. V. Dicks, L. Fellman, E. Fleishman, et al. 2014. “A Horizon Scan of Global Conservation Issues for 2014,” Trends in Ecol­ogy & Evolution 29:15–22. Sutherland, W. J., S. H. M. Butchart, B. Connor, C. Culshaw, L. V. Dicks, J. Dinsdale, H. Doran, et al. 2018. “A 2018 Horizon Scan of Emerging Issues for Global Conservation and Biological Diversity,” Trends in Ecol­ogy & Evolution 33:47–58. Suttle, C. A. 2007. “Marine Viruses—­Major Players in the Global Ecosystem,” Nature Reviews Microbiology 5:801–812. Sverdrup, H. U. 1953. “On Conditions for the Vernal Blooming of Phytoplankton,” Journal du Conseil 18:287–295. Swartz, W., E. Sala, S. Tracey, R. Watson, and D. Pauly. 2010. “The Spatial Expansion and Ecological Footprint of Fisheries (1950 to Pre­ sent),” PLoS ONE 5:e15143. Sweetman, A. K., A. R. Thurber, C. R. Smith, L. A. Levin, C. Mora, C.-­L. Wei, A. J. Gooday, et al. 2017. “Major Impacts of Climate Change on Deep-­Sea Benthic Ecosystems,” Elementa: Science of the Anthropocene 5:1–23. Swenson, W., D. S. Wilson, and R. Elias. 2000. “Artificial Ecosystem Se­ lection,” Proceedings of the National Acad­emy of Sciences 97:9110–9114. Szathmary, E., and J. Maynard Smith. 1995. “The Major Evolutionary Transitions,” Nature 374:227–232. Tamburello, N., I. M. Côté, and N. K. Dulvy. 2015. “Energy and the Scaling of Animal Space Use,” American Naturalist 186:196–211. Tansley, A. G. 1935. “The Use and Abuse of Vegetational Concepts and Terms,” Ecol­ogy 16:284–307. Taylor, M. S., and M. E. Hellberg. 2003. “Ge­ne­tic Evidence for Local Retention of Pelagic Larvae in a Ca­rib­bean Reef Fish,” Science 299:107–109. Taylor, M. S., and M. E. Hellberg. 2005. “Marine Radiations at Small Geographic Scales: Speciation in Neotropical Reef Gobies (Elacatinus),” Evolution 59:374–385. Teal, J. M. 1962. “Energy Flow in the Salt Marsh Ecosystem of Georgia,” Ecol­ogy 43:614–624. Teal, J. M., and J. W. Kanwisher. 1966. “Gas Transport in the Marsh Grass, Spartina alterniflora,” Journal of Experimental Botany 17:355–361. Tegner, M., and P. K. Dayton. 2000. “Ecosystem Effects of Fishing in Kelp Forest Communities,” ICES Journal of Marine Science 57:579–589. Tegner, M. J., and P. K. Dayton. 1987. “El Nino Effects on Southern California Kelp Forest Communities,” Advances in Ecological Research 17:243–279. Teh, L. S. L., L. C. L. Teh, and U. R. Sumaila. 2013. “A Global Estimate of the Number of Coral Reef Fishers,” PLoS ONE 8:e65397–10. Telesh, I. V., and V. V. Khlebovich. 2010. “Principal Pro­cesses within the Estuarine Salinity Gradient: A Review,” Marine Pollution Bulletin 61:149–155. Tenore, K. R. 1983. “Organic Nitrogen and Caloric Content of Detritus. III. Effect on Growth of a Deposit-­Feeding Polychaete, Capitella capitata,” Estuarine Coastal and Shelf Science 17:733–742. Tepolt, C. K. 2014. “Adaptation in Marine Invasion: A Ge­ne­tic Perspective,” Biological Invasions 17:887–903.

Literature Cited

Terborgh, J. W., and J. Faaborg. 1980. “Saturation of Birds Communities in the West Indies,” American Naturalist 116:178–195. Thackeray, S. J., T. H. Sparks, M. Frederiksen, S. Burthe, P. J. Bacon, J. R. Bell, M. S. Botham, et al. 2010. “Trophic Level Asynchrony in Rates of Phenological Change for Marine, Freshwater and Terrestrial Environments,” Global Change Biology 16:3304–3313. Thayer, G. W., K. A. Bjorndal, J. C. Ogden, S. L. Williams, and J. C. Zieman. 1984. “Role of Larger Herbivores in Seagrass Communities,” Estuaries 7:351–376. Thayer, G. W., W. J. Kenworthy, and M. S. Fonseca. 1984. Ecol­ogy of Eelgrass Meadows of the Atlantic Coast: A Community Profile. Washington, DC: US Fish and Wildlife Ser­vice. Thayer, G. W., P. L. Parker, M. W. LaCroix, and B. Fry. 1978. “The Stable Carbon Isotope Ratio of Some Components of an Eelgrass, Zostera marina, Bed,” Oecologia 35:1–12. Thompson, R. C., and T. P. Crowe. 2002. “Rocky Intertidal Communities: Past Environmental Changes, Pre­sent Status and Predictions for the Next 25 Years,” Environmental Conservation 29:1–24. Thorrold, S. R., C. Latkoczy, P. K. Swart, and C. M. Jones. 2001. “Natal Homing in a Marine Fish Metapopulation,” Science 291:297–299. Thorson, G. 1950. “Reproductive and Larval Ecol­ogy of Marine Bottom Invertebrates,” Biological Reviews of the Cambridge Philosophical Society 25:1–45. Thorson, G. 1957. “Bottom Communities (Sublittoral or Shallow Shelf),” Geological Society of Amer­ic­ a Memoirs 67:461–534. Thrush, S. F., and P. K. Dayton. 2002. “Disturbance to Marine Benthic Habitats by Trawling and Dredging: Implications for Marine Biodiversity,” Annual Review of Ecol­ogy and Systematics 33:449–473. Thrush, S. F., J. E. Hewitt, M. Gibbs, C. Lundquist, and A. Norkko. 2006. “Functional Role of Large Organisms in Intertidal Communities: Community Effects and Ecosystem Function,” Ecosystems 9:1029–1040. Thrush, S. F., R. D. Pridmore, J. E. Hewitt, and V. J. Cummings. 1992. “Adult Infauna as Facilitators of Colonization on Intertidal Sandflats,” Journal of Experimental Marine Biology and Ecol­ogy 159:253–265. Tilman, D. 1982. “Resource Competition and Community Structure,” Monographs in Population Biology 17:1–296. Tilman, D. 1999. “The Ecological Consequences of Changes in Biodiversity: A Search for General Princi­ples,” Ecol­ogy 80:1455–1474. Tilman, D., C. L. Lehman, and K. T. Thomson. 1997. “Plant Diversity and Ecosystem Productivity: Theoretical Considerations,” Proceedings of the National Acad­emy of Sciences 94:1857–1861. Tittensor, D. P., C. Mora, W. Jetz, H. K. Lotze, D. Ricard, E. V. Berghe, and B. Worm. 2010. “Global Patterns and Predictors of Marine Biodiversity across Taxa,” Nature 466:1098–1101. Tittensor, D. P., and B. Worm. 2016. “A Neutral-­Metabolic Theory of Latitudinal Biodiversity,” Global Ecol­ogy and Biogeography 25:630–641. Tollefson, J. 2012. “Ocean-­Fertilization Proj­ect Off Canada Sparks Furor,” Nature 490:458. Tomas, F., B. Martínez-­Crego, G. Hernán, and R. Santos. 2015. “Responses of Seagrass to Anthropogenic and Natu­ral Disturbances Do Not Equally Translate to Its Consumers,” Global Change Biology 21:4021–4030. Toth, G. B., and H. Pavia. 2007. “Induced Herbivore Re­sis­tance in Seaweeds: A Meta-­analysis,” Journal of Ecol­ogy 95:425–434. Toth, L. T., R. van Woesik, T. J. T. Murdoch, S. R. Smith, J. C. Ogden, W. F. Precht, and R. B. Aronson. 2014. “Do No-­Take Reserves Benefit

425

Florida’s Corals? 14 Years of Change and Stasis in the Florida Keys National Marine Sanctuary,” Coral Reefs 33:565–577. Trebilco, R., J. K. Baum, A. K. Salomon, and N. K. Dulvy. 2013. “Ecosystem Ecol­ogy: Size-­Based Constraints on the Pyramids of Life,” Trends in Ecol­ogy & Evolution 28:423–431. Trick, C. G., B. D. Bill, W. P. Cochlan, M. L. Wells, V. L. Trainer, and L. D. Pickell. 2010. “Iron Enrichment Stimulates Toxic Diatom Production in High-­Nitrate, Low-­Chlorophyll Areas,” Proceedings of the National Acad­emy of Sciences 107:5887–5892. Trussell, G. C., P. J. Ewanchuk, and M. D. Bertness. 2002. “Field Evidence of Trait-­Mediated Indirect Interactions in a Rocky Intertidal Food Web,” Ecol­ogy Letters 5:241–245. Turner, G. M. 2008. “A Comparison of the Limits to Growth with 30 Years of Real­ity,” Global Environmental Change 18:397–411. Turner, R. E. 1977. “Intertidal Vegetation and Commercial Yields of Penaeid Shrimp,” Transactions of the American Fisheries Society 106:411–416. Twining, B. S., and S. B. Baines. 2013. “The Trace Metal Composition of Marine Phytoplankton,” Annual Review of Marine Science 5:191–215. Ulanowicz, R. E. 1986. Growth and Development: Ecosystems Phenomenology. New York: Springer Verlag. Underwood, A. J., and P. G. Fairweather. 1989. “Supply-­Side Ecol­ogy and Benthic Marine Assemblages,” Trends in Ecol­ogy & Evolution 4:16–20. Unsworth, R. K. F., S. L. Hinder, O. G. Bodger, and L. C. Cullen-­ Unsworth. 2014. “Food Supply Depends on Seagrass Meadows in the Coral Triangle,” Environmental Research Letters 9:094005. Valdivia, A., C. E. Cox, and J. F. Bruno. 2017. “Predatory Fish Depletion and Recovery Potential on Ca­rib­bean Reefs,” Science Advances 3:e1601303. Valentine, J., and J. Duffy. 2006. “The Central Role of Grazing in Seagrass Ecol­ogy.” Pages 463–501 in A. W. D. Larkum, R. J. Orth, and C. M. Duarte, editors, Seagrasses: Biology, Ecol­ogy, and Conservation. New York: Springer. Valentine, J., and K. Heck Jr. 1999. “Seagrass Herbivory: Evidence for the Continued Grazing of Marine Grasses,” Marine Ecol­ogy Pro­ gress Series 176:291–302. Valiela, I. 1995. Marine Ecological Pro­cesses. New York: Springer. Valiela, I., and M. Cole. 2002. “Comparative Evidence That Salt Marshes and Mangroves May Protect Seagrass Meadows from Land-­ Derived Nitrogen Loads,” Ecosystems 5:92–102. Valiela, I., D. Rutecki, and S. Fox. 2004. “Salt Marshes: Biological Controls of Food Webs in a Diminishing Environment,” Journal of Experimental Marine Biology and Ecol­ogy 300:131–159. van der Heide, T., E. H. Van Nes, M. M. van Katwijk, H. Olff, and A. J. P. Smolders. 2011. “Positive Feedbacks in Seagrass Ecosystems—­ Evidence from Large-­Scale Empirical Data,” PLoS ONE 6:e16504. Vandermeer, J. 2006. “Metabolic Theories in Ecol­ogy,” Trends in Ecol­ogy & Evolution 21:136–140. van der Zee, E. M., C. Angelini, L. L. Govers, M. J. A. Christianen, A. H. Altieri, K. J. van der Reijden, B. R. Silliman, et al. 2016. “How Habitat-­Modifying Organisms Structure the Food Web of Two Coastal Ecosystems,” Proceedings of the Royal Society B-­ Biological Sciences 283:20152326. Van Dover, C., C. German, K. Speer, L. Parson, and R. Vrijenhoek. 2002. “Evolution and Biogeography of Deep-­Sea Vent and Seep Invertebrates,” Science 295:1253.

426

Ocean Ecology

Van Houtan, K. S., L. Mcclenachan, and J. N. Kittinger. 2013. “Seafood Menus Reflect Long-­Term Ocean Changes,” Frontiers in Ecol­ ogy and the Environment 11:289–290. van Katwijk, M. M., A. R. Bos, V. N. de Jonge, L. S. A. Hanssen, D. C. R. Hermus, and D. J. de Jong. 2009. “Guidelines for Seagrass Restoration: Importance of Habitat Se­lection and Donor Population, Spreading of Risks, and Ecosystem Engineering Effects,” Marine Pollution Bulletin 58:179–188. van Nes, E., T. Amaro, M. Scheffer, and G. Duineveld. 2007. “Pos­si­ble Mechanisms for a Marine Benthic Regime Shift in the North Sea,” Marine Ecol­ogy Pro­gress Series 330:39–47. van Oppen, M. J., B. J. McDonald, B. Willis, and D. J. Miller. 2001. “The Evolutionary History of the Coral Genus Acropora (Scleractinia, Cnidaria) Based on a Mitochondrial and a Nuclear Marker: Reticulation, Incomplete Lineage Sorting, or Morphological Convergence?” Molecular Biology and Evolution 18:1315–1329. Vaughan, N. E., and T. M. Lenton. 2011. “A Review of Climate Geoengineering Proposals,” Climatic Change 109:745–790. Vellend, M. 2010. “Conceptual Synthesis in Community Ecol­ogy,” Quarterly Review of Biology 85:183–206. Vellend, M. 2016. The Theory of Ecological Communities. Prince­ton, NJ: Prince­ton University Press. Vellend, M., L. Baeten, I. H. Myers-­Smith, S. C. Elmendorf, R. Beauséjour, C. D. Brown, P. De Frenne, K. Verheyen, and S. Wipf. 2013. “Global Meta-­analysis Reveals No Net Change in Local-­ Scale Plant Biodiversity over Time,” Proceedings of the National Acad­emy of Sciences 110:19456. Venter, J. C., K. Remington, J. F. Heidelberg, A. L. Halpern, D. Rusch, J. A. Eisen, D. Wu, et al. 2004. “Environmental Genome Shotgun Sequencing of the Sargasso Sea,” Science 304:66–74. Vergés, A., C. Doropoulos, H. A. Malcolm, M. Skye, M. Garcia-­Pizá, E. M. Marzinelli, A. H. Campbell, et al. 2016. “Long-­Term Empirical Evidence of Ocean Warming Leading to Tropicalization of Fish Communities, Increased Herbivory, and Loss of Kelp,” Proceedings of the National Acad­emy of Sciences 113:13791–13796. Vergés, A., P. D. Steinberg, M. E. Hay, A. G. B. Poore, A. H. Campbell, E. Ballesteros, K. L. Heck, et al. 2014. “The Tropicalization of Temperate Marine Ecosystems: Climate-­Mediated Changes in Herbivory and Community Phase Shifts,” Proceedings of the Royal Society B-­Biological Sciences 281:20140846. Verity, P., and V. Smetacek. 1996. “Organism Life Cycles, Predation, and the Structure of Marine Pelagic Ecosystems,” Marine Ecol­ogy Pro­gress Series 130:277–293. Vermeij, G. 2004. “Ecological Avalanches and the Two Kinds of Extinction,” Evolutionary Ecol­ogy Research 6:315–337. Vermeij, G. J. 1976. “Interoceanic Differences in Vulnerability of Shelled Prey to Crab Predation,” Nature 260:135–136. Vermeij, G. J. 1978. Biogeography and Adaptation: Patterns of Marine Life. Cambridge, MA: Harvard University Press. Vermeij, G. J. 1991. “Anatomy of an Invasion: The Trans-­Arctic Exchange,” Paleobiology 17:281–307. Vermeij, G. J. 2004. Nature: An Economic History. Prince­ton, NJ: Prince­ ton University Press. Vermeij, G. J., G. P. Dietl, and D. G. Reid. 2008. “The Trans-­Atlantic History of Diversity and Body Size in Ecological Guilds,” Ecol­ogy 89:S39–52. Vermeij, G. J., and P. D. Roopnarine. 2008. “The Coming Arctic Invasion,” Science 321:780–781. Vermeij, M. J. A., K. L. Marhaver, C. M. Huijbers, I. Nagelkerken, and S. D. Simpson. 2010. “Coral Larvae Move ­toward Reef Sounds,” PLoS ONE 5:e10660.

Via, S. 2001. “Sympatric Speciation in Animals: The Ugly Duckling Grows Up,” Trends in Ecol­ogy & Evolution 16:381–390. Villéger, S., S. Brosse, M. Mouchet, D. Mouillot, and M. J. Vanni. 2017. “Functional Ecol­ogy of Fish: Current Approaches and ­Future Challenges,” Aquatic Sciences 79:783–801. Villéger, S., N. W. H. Mason, and D. Mouillot. 2008. “New Multidimensional Functional Diversity Indices for a Multifaceted Framework in Functional Ecol­ogy,” Ecol­ogy 89:2290–2301. Villéger, S., J. Ramos Miranda, D. Flores Hernandez, A. Sosa Lopez, and D. Mouillot. 2008. “Stable Trophic Structure across Coastal Nekton Assemblages despite High Species Turnover,” Marine Ecol­ogy Pro­gress Series 364:135–146. Vince, S. W., and A. A. Snow. 1984. “Plant Zonation in an Alaskan Salt Marsh. II. An Experimental Study of the Role of Edaphic Conditions,” Journal of Ecol­ogy 72:669–684. Violle, C., M.-­L. Navas, D. Vile, E. Kazakou, C. Fortunel, I. Hummel, and E. Garnier. 2007. “Let the Concept of Trait Be Functional!” Oikos 116:882–892. Vitousek, P., P. Ehrlich, A. Ehrlich, and P. Matson. 1986. “­Human Appropriation of the Products of Photosynthesis,” BioScience 36:368–373. Vitousek, P. M., H. A. Mooney, J. Lubchenco, and J. M. Melillo. 1997. “­Human Domination of Earth’s Ecosystems,” Science 277:494–499. Vitousek, S., P. L. Barnard, C. H. Fletcher, N. Frazer, L. Erikson, and C. D. Storlazzi. 2017. “Doubling of Coastal Flooding Frequency within De­cades Due to Sea-­Level Rise,” Scientific Reports 7:1399. Waide, R. B., M. R. Willig, C. F. Steiner, G. Mittelbach, L. Gough, S. I. Dodson, G. P. Juday, and R. Parmenter. 1999. “The Relationship between Productivity and Species Richness,” Annual Review of Ecol­ogy and Systematics 30:257–300. Wallace, B. P., S. S. Heppell, R. L. Lewison, S. Kelez, and L. B. Crowder. 2008. “Impacts of Fisheries Bycatch on Loggerhead Turtles Worldwide Inferred from Reproductive Value Analyses,” Journal of Applied Ecol­ogy 45:1076–1085. Wang, W., Z. Yan, S. You, Y. Zhang, L. Chen, and G. Lin. 2011. “Mangroves: Obligate or Facultative Halophytes? A Review,” Trees 25:953–963. Ware, D. M., and R. E. Thomson. 2005. “Bottom-­Up Ecosystem Trophic Dynamics Determine Fish Production in the Northeast Pacific,” Science 308:1280–1284. ­Waters, C. N., J. Zalasiewicz, C. Summerhayes, A. D. Barnosky, C. Poirier, A. Gałuszka, A. Cearreta, et al. 2016. “The Anthropocene Is Functionally and Stratigraphically Distinct from the Holocene,” Science 351:aad2622. Watling, L., and E. A. Norse. 1998. “Disturbance of the Seabed by Mobile Fishing Gear: A Comparison to Forest Clearcutting,” Conservation Biology 12:1180–1197. Watson, R., and D. Pauly. 2001. “Systematic Distortions in World Fisheries Catch Trends,” Nature 414:534–536. Watson, R. A., W. W. Cheung, J. A. Anticamara, R. U. Sumaila, D. Zeller, and D. Pauly. 2013. “Global Marine Yield Halved as Fishing Intensity Redoubles,” Fish and Fisheries, 14:493–503. Watson, S.-­A ., S. Lefevre, M. I. McCormick, P. Domenici, G. E. Nilsson, and P. L. Munday. 2014. “Marine Mollusc Predator-­Escape Behaviour Altered by Near-­Future Carbon Dioxide Levels,” Proceedings of the Royal Society B-­Biological Sciences 281:20132377. Watzin, M. 1985. “Interactions among Temporary and Permanent Meiofauna: Observations on the Feeding and Be­hav­ior of Selected Taxa,” Biological Bulletin 169:397.

Literature Cited

Watzin, M. 1986. “Larval Settlement into Marine Soft-­Sediment Systems: Interactions with the Meiofauna,” Journal of Experimental Marine Biology and Ecol­ogy 98:65–113. Waycott, M., C. M. Duarte, T. J. B. Carruthers, R. J. Orth, W. C. Dennison, S. Olyarnik, A. Calladine, et al. 2009. “Accelerating Loss of Seagrasses across the Globe Threatens Coastal Ecosystems,” Proceedings of the National Acad­emy of Sciences 106:12377–12381. Webb, C. O., D. D. Ackerly, M. A. McPeek, and M. J. Donoghue. 2002. “Phylogenies and Community Ecol­ogy,” Annual Review of Ecol­ ogy and Systematics 33:475–505. Webster, N. S., S. Uthicke, E. S. Botté, F. Flores, and A. P. Negri. 2012. “Ocean Acidification Reduces Induction of Coral Settlement by Crustose Coralline Algae,” Global Change Biology 19:303–315. Weersing, K., and R. Toonen. 2009. “Population Ge­ne­tics, Larval Dispersal, and Connectivity in Marine Systems,” Marine Ecol­ogy Pro­ gress Series 393:1–12. Wei, C.-­L., G. T. Rowe, E. Escobar-­Briones, A. Boetius, T. Soltwedel, M. J. Caley, Y. Soliman, et al. 2010. “Global Patterns and Predictions of Seafloor Biomass Using Random Forests,” PLoS ONE 5:e15323. Weinberger, F. 2007. “Pathogen-­Induced Defense and Innate Immunity in Macroalgae,” Biological Bulletin 213:290–302. Wellington, G. M., and P. W. Glynn. 1983. “Environmental Influences on Skeletal Banding in Eastern Pacific (Panama) Corals,” Coral Reefs 1:215–222. West, G. B., J. H. Brown, and B. J. Enquist. 1997. “A General Model for the Origin of Allometric Scaling Laws in Biology,” Science 276:122–126. West-­Eberhard, M. 1989. “Phenotypic Plasticity and the Origins of Diversity,” Annual Review of Ecol­ogy and Systematics 20:249–278. Whalen, M. A., R. D. Whippo, J. J. Stachowicz, P. H. York, E. Aiello, T. Alcoverro, A. H. Altieri, L. Benedetti-Cecchi, C. Bertolini, M. Bresch, and F. Bulleri. 2020. “Climate Drives the Geography of Marine Consumption by Changing Predator Communities.,” Proceedings of the National Academy of Sciences 117(45):28160–28166. Whitfield, A. K., M. Elliott, A. Basset, S. J. M. Blaber, and R. J. West. 2012. “Paradigms in Estuarine Ecology—­A Review of the Remane Diagram with a Suggested Revised Model for Estuaries,” Estuarine Coastal and Shelf Science 97:78–90. Whittaker, R. H. 1972. “Evolution and Mea­sure­ment of Species Diversity,” Taxon 21:213–251. Whittaker, R. J., K. A. Triantis, and R. J. Ladle. 2008. “A General Dynamic Theory of Oceanic Island Biogeography,” Journal of Biogeography 35:977–994. Whittaker, R. J., K. J. Willis, and R. Field. 2001. “Scale and Species Richness: T ­ owards a General, Hierarchical Theory of Species Diversity,” Journal of Biogeography 28:453–470. Wiens, J. J., D. D. Ackerly, A. P. Allen, B. L. Anacker, L. B. Buckley, H. V. Cornell, E. I. Damschen, et al. 2010. “Niche Conservatism as an Emerging Princi­ple in Ecol­ogy and Conservation Biology,” Ecol­ogy Letters 13:1310–1324. Wiley, E. O. 1988. “Vicariance Biogeography,” Annual Review of Ecol­ogy and Systematics 19:513–542. Wiley, E. O., and B. S. Lieberman. 2011. Phyloge­ne­tics: Theory and Practice of Phyloge­ne­tic Systematics. Second edition. Hoboken, NJ: Wiley. Williams, G. C., and R. M. Nesse. 1991. “The Dawn of Darwinian Medicine,” Quarterly Review of Biology 66:1–22.

427

Williams, J., and S. Jackson. 2007. “Novel Climates, No-­Analog Communities, and Ecological Surprises,” Frontiers in Ecol­ogy and the Environment 5:475–482. Williams, P. J., P. D. Quay, T. K. Westberry, and M. J. Behrenfeld. 2013. “The Oligotrophic Ocean Is Autotrophic,” Annual Review of Marine Science 5:535–549. Williams, S. L., R. Ambo-­R appe, C. Sur, J. M. Abbott, and S. R. Limbong. 2017. “Species Richness Accelerates Marine Ecosystem Restoration in the Coral Triangle,” Proceedings of the National Acad­ emy of Sciences 114:11986–11991. Williamson, P., D. Wallace, C. S. Law, and P. W. Boyd. 2012. “Ocean Fertilization for Geoengineering: A Review of Effectiveness, Environmental Impacts and Emerging Governance,” Pro­cess Safety and Environmental Protection 90:475–488. Wilsey, B. J., L. M. Martin, and H. Wayne Polley. 2005. “Predicting Plant Extinction Based on Species-­Area Curves in Prairie Fragments with High Beta Richness,” Conservation Biology 19:1835–1841. Wilson, W. H. 1990. “Competition and Predation in Marine Soft-­ Sediment Communities,” Annual Review of Ecol­ogy and Systematics 21:221–241. Wing, S., J. Largier, L. Botsford, and J. Quinn. 1995. “Settlement and Transport of Benthic Invertebrates in an Intermittent Upwelling Region,” Limnology and Oceanography 40:316–329. Wing, S. R., and E. S. Wing. 2001. “Prehistoric Fisheries in the Ca­rib­ bean,” Coral Reefs 20:1–8. Witman, J. D., and P. K. Dayton. 2001. “Rocky Subtidal Communities,” Pages 339–366 in M. D. Bertness, S. D. Gaines, and M. E. Hay, editors, Marine Community Ecol­ogy. Sunderland, MA: Sinauer Associates. Witman, J. D., R. J. Etter, and F. Smith. 2004. “The Relationship between Regional and Local Species Diversity in Marine Benthic Communities: A Global Perspective,” Proceedings of the National Acad­emy of Sciences 101:15664–15669. Witman, J. D., and K. Roy, eds. 2009. Marine Macroecol­ogy. Chicago: University of Chicago Press. Wittmann, A. C., and H.-­O. Pörtner. 2013. “Sensitivities of Extant Animal Taxa to Ocean Acidification,” Nature Climate Change 3:995–1001. Woese, C. R., O. Kandler, and M. L. Wheelis. 1990. “­Towards a Natu­ral System of Organisms: Proposal for the Domains Archaea, Bacteria, and Eucarya,” Proceedings of the National Acad­emy of Sciences 87:4576–4579. Wolanski, E., and W. M. Hamner. 1988. “Topographically Controlled Fronts in the Ocean and Their Biological Influence,” Science 241:177–181. Wolff, T. 1979. “Macrofaunal Utilization of Plant Remains in the Deep Sea,” Sarsia 64:117–143. Wood, R. 1998. “The Ecological Evolution of Reefs,” Annual Review of Ecol­ogy and Systematics 29:179–206. Wood, S. A., S. A. Lilley, D. R. Schiel, and J. B. Shurin. 2010. “Organismal Traits Are More Impor­tant than Environment for Species Interactions in the Intertidal Zone,” Ecol­ogy Letters 13:1160–1171. Woodin, S. A. 1976. “Adult-­Larvae Interactions in Dense Infaunal Assemblages: Patterns of Abundance,” Journal of Marine Research 34:1. Woodley, J. D., E. A. Chornesky, P. A. Clifford, J. B. Jackson, L. S. Kaufman, N. Knowlton, J. C. Lang, et al. 1981. “Hurricane Allen’s Impact on Jamaican Coral Reefs,” Science 214:749–755. Woodson, C. B., and S. Y. Litvin. 2015. “Ocean Fronts Drive Marine Fishery Production and Biogeochemical Cycling,” Proceedings of the National Acad­emy of Sciences 112:1710–1715.

428

Ocean Ecology

Woodward, G., B. Ebenman, M. Emmerson, J. Montoya, J. Olesen, A. Valido, and P. Warren. 2005. “Body Size in Ecological Networks,” Trends in Ecol­ogy & Evolution 20:402–409. Woolley, S. N. C., D. P. Tittensor, P. K. Dunstan, G. Guillera-­Arroita, J. J. Lahoz-­Monfort, B. A. Wintle, B. Worm, and T. D. O’Hara. 2016. “Deep-­Sea Diversity Patterns Are ­Shaped by Energy Availability,” Nature 533:393–396. Wootton, J. 1994. “Predicting Direct and Indirect Effects: An Integrated Approach Using Experiments and Path Analy­sis,” Ecol­ogy 75:151–165. Wootton, J. T. 2005. “Field Pa­ram­e­terization and Experimental Test of the Neutral Theory of Biodiversity,” Nature 433:309–312. Wootton, J. T., and M. Emmerson. 2005. “Mea­sure­ment of Interaction Strength in Nature,” Annual Review of Ecol­ogy Evolution and Systematics 36:419–444. Wootton, J. T., C. A. Pfister, and J. D. Forester. 2008. “Dynamic Patterns and Ecological Impacts of Declining Ocean pH in a High-­ Resolution Multi-­year Dataset,” Proceedings of the National Acad­emy of Sciences 105:18848–18853. Worm, B., E. B. Barbier, N. Beaumont, J. E. Duffy, C. Folke, B. S. Halpern, J. B. C. Jackson, et al. 2006. “Impacts of Biodiversity Loss on Ocean Ecosystem Ser­vices,” Science 314:787–790.

Worm, B., R. Hilborn, J. K. Baum, T. A. Branch, J. S. Collie, C. Costello, M. J. Fogarty, et al. 2009. “Rebuilding Global Fisheries,” Science 325:578. Worm, B., and H. Lotze. 2006. “Effects of Eutrophication, Grazing, and Algal Blooms on Rocky Shores,” Limnology and Oceanography 51:569–579. Worm, B., and R. A. Myers. 2003. “Meta-­analysis of Cod-­Shrimp Interactions Reveals Top-­Down Control in Oceanic Food Webs,” Ecol­ogy 84:162–173. Worm, B., and R. T. Paine. 2016. “­Humans as a Hyperkeystone Species,” Trends in Ecol­ogy & Evolution 31:600–607. Wright, I. J., P. B. Reich, M. Westoby, D. D. Ackerly, Z. Baruch, F. Bongers, J. Cavender-­Bares, et al. 2004. “The Worldwide Leaf Economics Spectrum,” Nature 428:821–827. Yachi, S., and M. Loreau. 1999. “Biodiversity and Ecosystem Productivity in a Fluctuating Environment: The Insurance Hypothesis,” Proceedings of the National Acad­emy of Sciences 96:1463–1468. zu Ermgassen, P. S. E., M. D. Spalding, B. Blake, L. D. Coen, B. Dumbauld, S. Geiger, J. H. Grabowski, et al. 2012. “Historical Ecol­ogy with Real Numbers: Past and Pre­sent Extent and Biomass of an Imperilled Estuarine Habitat,” Proceedings of the Royal Society B-­Biological Sciences 279:3393–3400.

Photo Credits Fig. 1.3: (A) NASA; (B) Jialiang Gao; (C) Phillip Capper Fig. 1.4: NASA Fig. 2.1: (A) Reinhardt Møbjerg Kristensen; (B) Ulf Jondelius Fig. 2.2: (A) Emmett Duffy; (B) Zatelmar, Wikimedia; (C) Steve Rupp, National Science Foundation Fig. 2.3: (A) Luke Thompson and Nikki Watson; (B) Octavio Aburto; (C) Maggie Johnson; (D) Christoffer Boström; (E) Emmett Duffy Fig. 2.4: (A–­C) NOAA; (D) John R. Dolan; (E) Tintinnidguy, Wikimedia; (F) Science Source Fig. 2.6: Samsara, Wikimedia Fig. 2.7: (A) wildestanimal, Shutterstock; (B) Karen Osborn; (C) Dante Fenolio; (D) Eric Heupel; (E) Atomic Roderick, Shutterstock; (F) Alexander Semenov Fig. 3.1: Deep Time Maps Fig. 4.13: (A) John James Audubon; (B) Greenpeace; (C) Akira Kuwata/ Nordic Council of Ministers; (D) Richard Aronson and James B. McClintock; (E) Peter Southwood; (F) James St. John Fig. 5.8: H. Zell Fig. 6.4: Espen Rekdal, BluePlanetArchive​.­com Fig. 6.7: Choksawatdikorn Fig. 6.8: (A) sciencephoto​.­com; (B) William Harrigan, Florida Keys National Marine Sanctuary Fig. 6.10: Quintin Muñoz Fig. 6.14: David Starr Jordan Fig. 7.6: (A) Klaus Stiefel; (B) Rickard Zerpe; (C) Connie Chen Fig 8.1: Emmett Duffy Fig. 9.5: (A) Albert Calbet, Marine Zooplankton Ecol­ogy Group, ICM-­CSIC, Barcelona, Spain; (B) John Dolan; (C) Ernst Haeckel; (D) Alex Mustard; (E) Emmett Duffy; (F) Brandon B, Shutterstock Fig. 10.4: NOAA Fisheries/Lisa Conger Fig. 10.10: (A) Emma Kissling; (B, C) NOAA; (D) Neil Bromhall; (E) Robert Vrijenhoek

Fig. 10.13: (A) Chris German, WHOI/NSF, NASA/ROV Jason 2012, © Woods Hole Oceanographic Institution; (B) MARUM, Center for Marine Environmental Sciences, University of Bremen; (C, D) NOAA Fig. 11.3: (A) Trish Hartmann; (B) Joost van Uffelen; (C) Emmett Duffy; (D) SERC/Keira Heggie Fig. 11.9: ­After CCRM, ­Virginia Institute of Marine Science Fig. 11.13: Damsea, Shutterstock Fig. 12.1: lkonya Fig. 12.2: (A) Alex Mustard; (B) NASA Fig. 12.3: Ahmed Amir Fig. 12.6: Robert Steneck Fig. 12.7: (A, C, D, F) G. Allen; (B) R. Hamilton; (E) A. Hoey Fig. 12.13: Bette Willis Fig. B2.2.1: (A) Flip Nicklen/Minden Pictures; (B) Craig Smith/NOAA Fig. B3.1.1: (A) R. T. Pereyra; (B) Emőke Dénes; (C) P. L. Munday; (D) timsimages​.­uk, Shutterstock Fig. B3.2.1: Photo credit: W. B. Miller Fig. B6.1.1: (left to right): NOAA, NOAA, Quintín Muñoz Fig. B6.2.1: NOAA Fig. B6.3.1: National Marine Fisheries Ser­vice Fig. B6.5.1: MDC Seamarc Maldives Fig. B7.1.1: Michael Maggs Fig. B8.3.1: Derek Keats Fig. B9.1.1. Timothy Knepp/US Fish and Wildlife Ser­vice Fig. B10.1.1: (top left) Maxim Yakovlev; (bottom left) Hans Hillewaert; (right) US Department of Health and ­Human Ser­vices Fig. B10.2.1: (A) NOAA; (B) Malcolm Clark, NIWA, New Zealand Fig. B11.1.1: Loren Chipman Fig. B11.2.1: (l to r): Andrew Cohen; Andrew Cohen; US Fish and Wildlife Ser­vice Fig. B12.1.1: (l to r): Ken Clifton; zsispeo; Rich Carey; Philippe Bourjon Fig. B12.2.1: Klaus Stiefel

Index A page number followed by f refers to a figure and a page number followed by t indicates a ­table. abiotic environment: biodiversity and, 2, 378; components of, 3 abundance, scaling with body mass, 166–67, 237, 238f, 240, 241f acidification. See ocean acidification; volcanic vents Acropora: dozens of coexisting species of, 205; white-­band disease of, 185, 362–63, 370, 375 Acropora cervicornis. See staghorn coral Acropora millepora: acidification affecting larval settlement of, 93–94; heritable heat tolerance in, 127–28 Acropora palmata. See elkhorn coral active continental margins, 297 active metabolic rate, 120 adaptation, 126, 128, 135 age classes, 145; in field populations, 147; Leslie matrix and, 147–48; reproductive values of, 146 age-­structured populations, 147–48 Agulhas current, 41 algae: defenses against herbivores, 177–78, 179, 181f, 186; as disparate group, 13, 20, 22; functional classification of, 32, 33f; as most of ocean’s primary producers, 225; seagrass competition with, 319–21, 320f; in sediment, 32; viruses infecting, 25; vulnerability to grazing, 32, 34f. See also crustose coralline algae; ephemeral algae; filamentous (turf ) algae; macroalgae; perennial algae; phytoplankton algal matrices, 354f, 355. See also filamentous (turf) algae Allee effect, 251, 286 allometric equations, 121 allometric exponent, 121, 122f allometric scaling laws, 123–24. See also metabolic scaling allometry, 121; macroecol­ogy and, 134 allopatric speciation, 49, 50 alpha diversity (α), 45 alternative stable states, 208, 249–52; on coral reefs, 363, 365–66, 365f, 366f; in marine fisheries, 250–51, 250f; rapid shifts among, 224; rarity of evidence for, 380; on temperate rocky reefs, 181 ammonium, in seawater, 115, 116, 226 amphipods: direct development of, 18, 140; nonconsumptive effects of predators on, 182; overgrazing stressed kelps, 302; Phronima, 23f; on seagrass blade, 227f; species-­area relationship, 46f anglerfish, 277, 278f anoxic sediment layer, 30 anoxic sediments: coastal, 310, 314, 328, 332, 333; of cold seeps, 283 Antarctic: circumpolar current and, 39; predation of king crab in, 88f, 91, 215 Anthropocene epoch, 1, 2–3, 4; beginning of, 75–76; biodiversity and, 36–37, 101–5, 135; central challenges for, 111; coastal ecosystems in, 337–44; coral reefs in, 369–74, 370f, 372f; earth system status in, 75, 77f; evolution driven by ­human impact in, 135; fossil fuels and, 78; humanity as major force of nature in, 74; hybridization

increasing in, 21; industrial fixation of nitrogen and, 79–80, 80f; mangroves in, 334–35; rocky shores in, 309–10; salt marshes in, 330–31; seagrass meadows in, 322–23, 325; sediment bottoms in, 315–17, 316f; species interactions in, 191–93; technological innovation and, 84. See also ­Great Acceleration Anthropocene ocean: biogeography of, 68–72, 69f, 71f; cautious optimism about, 108–10, 109f; ecosystems in, 257–58; ­human impacts on, 87, 88f, 89; marine organisms in, 135–36; marine populations in, 168; predicted species distributions in, 69–70, 69f antitropical distribution patterns, 60 apparent competition, 173, 173f aquaculture: disease in marine organisms and, 101, 104–5; evolutionary changes caused by, 104–5; global growth of, 340, 341f; hybridization resulting from, 21; integrated, 106; management of, 101; mangrove forests cleared for, 334–35; of oysters, 339; prospects if better planned and managed, 340; of shrimps, 77f, 334, 340 aragonite, 372 Archaea, 19–20, 19f; metabolic characteristics of, 25, 26, 27t; at methane seeps, 283–84; picoplanktonic, 22; primary producers among, 225 Arctic opening, 70–72, 78 Arctic sea ice extent, 78, 79f arrow worms (Chaetognatha), 12, 14f artificial reefs, 344 artificial structures, 105, 343–44 ascidians, ocean warming and nonnative species of, 103 Asian clam (Corbula amurensis), 342–43, 342f assemblage, 197 Atlantic menhaden (Brevoortia tyrannus), 162, 162f Atlantic Ocean: ancient separation from Indian and Pacific Oceans, 39; Ca­rib­bean as biodiversity hotspot in, 60; passive continental margins of, 297; species richness in, 52; trait differences on two sides of, 63–64. See also North Atlantic atmosphere of Earth, with or without life, 2, 2t atolls, 347; Enewetak, 222, 222f, 223, 350; evolution of, 350, 350f ATP, 115, 118 autotrophy, 26, 27t, 118 bacteria: on algal turfs of reefs, 355; chemoautotrophic, 26, 27t, 282–83; consuming phytoplankton primary production, 272; latitudinal diversity gradients of, 286–87; of mesopelagic zone, 264; ocean warming and, 215; picoplanktonic, 22; primary producers among, 225. See also cyanobacteria Bacteria, 19–20, 19f Baltic Sea: collapse of cod population in, 252, 276; as estuary, 298; productivity enhanced by eutrophication of, 341 Banks, Joseph, 260 barnacles: experiments with competition between, 174–75, 174f, 176, 304; life histories of, 139f, 149f; neutral theory vs. determinism and, 205, 206

barrier reef, 350, 350f basal metabolic rate, 120, 121, 123 bathypelagic zone, 263f, 264–65 Becking, Lourens, 286 be­hav­ior: ­human, 108; phenotypic plasticity of, 128; in predator-­prey interactions, 182–84 Belize Barrier Reef, 370–71 benthic biomass, deep-­sea global distribution of, 276, 277f benthic ecosystems: altered by acidification, 94; changed by eutrophication, 258; coastal, 300, 302–3; defined, 21; marine ecol­ogy of, 8, 9; organic enrichment and, 315–16, 316f benthic organisms: functional groups of, 30–36; invertebrates with pelagic larvae, 201; reef fishes with benthic egg guarding, 54–55, 55f benthic substrata, 30. See also sediments benthopelagic animals, 278 Bertness, Mark, 328 beta diversity (β), 45; reduced by ocean warming, 103 biodiversity: acidification and, 94–95; in the Anthropocene, 36–37, 101–5, 135; central importance of, 5; conceptual scheme for pro­cesses in, 44–45, 45f; in coral reef communities, 347, 351, 357–60; in deep sea, 48, 265, 279–80, 280f, 284–88, 285f; defined, 11; diseases exacerbated by decline in, 341, 343; driving ecosystem pro­cesses, 378; ecosystem functioning and, 241–45, 243f; in framework for thinking about ecol­ogy, 3–4, 3f, 4f; functional classification of, 14, 18, 21–28, 30–36; functional consequences of decline in, 102–3; global hotspot of, 48; ­human dependence on, 36–37; ­human impacts on, 36, 72, 258–59; integrative models of, 60; interaction strength in tropical communities and, 68; magnitude of, 11; major patterns in, 44–48; of marine vs. terrestrial animals, 12–13, 13t, 14f; of marine vs. terrestrial primary producers, 13, 15f; phyloge­ne­tic classification of, 14, 18–21, 19f; planetary boundary for, 103; regime shifts promoted by low values of, 252; resource availability and, 284–86, 285f; spatial organ­ization of, 45–46, 46f; stability of ecosystems and, 245–48, 369; vertical and horizontal components of, 232, 232f, 246. See also diversity in communities; functional diversity; species richness biogeochemical pro­cesses, 3, 4, 5, 6; coastal zones and, 303; of coral reef ecosystems, 367–68; microbial biosphere and, 293; in sediment bottoms, 310, 314, 315, 321 biogeography, 38; of Anthropocene ocean, 68–72, 69f, 71f; biodiversity and, 44–48; classification of coastal regions in, 62–63, 62f; classification of pelagic ocean in, 61–62, 61f, 63; community structure and, 208–9; earth history and, 38–40, 39f; fisheries management and, 63; of functional traits, 63–64, 66f; ocean circulation and, 40–44; pressing questions in, 72–73; rocky shore species and, 308; of species interactions, 64, 66–68, 67f. See also dispersal; island biogeography, theory of

432

Index

biological accommodation, 280 biological oceanography, 8–9 biological pump, 273–74, 273f; climate warming and, 26, 290, 291, 292; metazoan heterotrophs and, 28; midwater fishes and, 28, 273; organismal pro­cesses involved in, 237; tropicalizing plankton and, 70, 73 bioluminescence, 23f, 277 biomass production: control of, 228–36; as 10% of animal’s ingested energy, 237–38; as yield in fisheries, 117. See also primary producer biomass biomass pyramids, 239–40, 239f biomes: anthropogenic, 106; of Longhurst’s scheme, 61–62, 63 bioturbation, 314, 315; by crabs in mangrove forests, 333 birds: as bycatch in fisheries, 201; salt marsh habitat for, 328; seabirds rescued by dispersal, 201; waterfowl feeding on seagrass, 321 bivalves, marine: coastal, in MEOW provinces, 63; as concentrators of heavy metals, 343; epifaunal, 34; evolutionary radiation ­after origin of siphon, 56; infaunal, 33, 34, 312, 312f, 313, 314; latitudinal diversity gradient and, 60; ocean acidification and, 135; temperature influence on speciation in, 53–54; at vents and seeps, 281, 283, 284. See also clams; mussels in salt marshes; mussels on intertidal shores; oyster reefs; oysters Black Sea: low-­diversity regime shifts in, 252; pelagic food web of, 250f, 251; trophic cascade in, 251, 274, 275, 276 bleaching of corals, 370, 370f, 371, 372f, 374; affecting dif­fer­ent types differently, 353; flattened reefs in Ca­rib­bean and, 363; risk of extinction and, 102; thermal tolerance and, 121 blooms of algae and phytoplankton: with chemically noxious properties, 257–58, 266, 341; decline of seagrass and, 323f, 324, 325; in freshwater ecosystems, 232; invasive zooplankton in Black Sea and, 275, 275f, 276; spring bloom, 265–66, 278, 279; stimulated by iron, 266. See also eutrophication bluefin tuna, 97–98, 102, 140 blue ling (Molva dipterygia), 289 blue ­whale, 23f, 28, 264f body size: of benthic organisms, 30, 32f; biological pump and, 274; carbon flux to deep sea and, 287; classes of plankton and, 22, 22t; consumer-­ prey interactions and, 182, 183f, 184; declining due to ­human impact, 89, 193; declining with food scarcity in deep sea, 276; ecosystem pro­cesses and, 237–40; extinction vulnerability and, 59, 101–2; ­future development of theory based on, 136; of harvested fish, 95, 96f, 98, 104, 156–57, 157f, 193; home range size and, 167, 167f; interaction strengths and, 189, 190, 190f, 193; of keystone species, 102; larval mortality and, 141f; as master trait, 121; metabolic scaling as function of, 121–24, 122f, 123f, 165–66; of pelagic organisms, 22–24, 24f; sensory functions affected by transitions in, 124; speciation and, 54; species richness and, 210; temperature-­mediated changes in, 90, 215; trophic level of fish and, 144, 144f; vulnerability to exploitation and, 65. See also cell size of phytoplankton body velocity, and temperature effect on predation, 191–92 bottom type, and species composition, 48

bottom-up control, 113f, 229–31, 230f, 231f, 233; in estuarine ecosystems, 302; of fish biomass, 276; of hydrothermal vent communities, 283; as incomplete analy­sis of mangroves, 333; salt marsh ecosystems and, 330 Briggs, John, 60 brittle stars (Ophiuroidea), 286 brooding. See direct development brown web, 235, 236 buoyancy in seawater, 124–25; of bloom-­forming algal cells, 266 butterflyfishes (Chaetodontidae), 202, 202f, 364, 364f bycatch in fisheries, 97, 98, 100, 106; in deep sea, 289; of seabirds, 201; of sea turtles, 151–52, 153 calcareous algae, 32, 33f. See also crustose coralline algae calcareous organisms, and ocean acidification, 93, 94–95, 135–36 calcium carbonate: of corals, 350, 351–52, 351f, 371, 372; ocean acidification and, 92f, 93, 135; as plant defense against herbivores, 178 California halibut (Paralichthys californicus), 155 carbonate precipitation at seeps, 284 carbon content, 113–14 carbon cycle: biological pump and, 273–74, 273f; destabilized by fossil fuel consumption, 78; phytoplankton in, 267; sediment pro­cesses and, 314 carbon dioxide (CO2): atmospheric rise in, 78, 78f; biological pump and, 273, 273f; diffusing into ocean surface, 289; emissions from mangrove clearance, 335; ocean acidification and, 92–95, 92f; phytoplankton species composition and, 237; trend since 1972 predictions, 81f, 82 carbon export to deep ocean: diel vertical migration and, 273; phytoplankton species composition and, 237; ­whales and, 29. See also export of production to deep ocean carbon fixation by photosynthesis, 118 carbon sinks: mangrove production and, 334; salt marshes as, 330 carbon storage in sediments or soil: from mangroves, 335; from seagrasses, 321–22, 322f; from terrestrial and marine ecosystems, 322f; trophic skew and, 338 Ca­rib­bean as biodiversity hotspot, 60 Ca­rib­bean reefs: biodiversity in communities of, 357; bleaching of corals on, 371, 372f; carbonate thickness in, 350; compared to Indo–­West Pacific, 366–67; decline of, 370–71, 370f; diseases with impacts on, 362–63 (see also Diadema antillarum); flattening of, 363; loss of carbonate structure on, 372; non-­stony corals on, 354; overgrown by sponges on overfished reefs, 360–61. See also Jamaican coral reefs carnivores: ecosystem energy pathways and, 113f; nutrition of food chosen by, 182 carpet-­of-­mouths hypothesis, 142–43 carry­ing capacity (K): body mass and, 166; regime shifts in predator-­prey systems and, 251 catch-­share programs, 100–101 cell size of phytoplankton: diversity of, 285, 285f; as fundamental trait, 132–33, 132f, 267; responses to physical forcing and, 267; in warming ocean, 291, 292

centers of accumulation, 60 centers of origin, 60 centers of overlap, 60 central ocean gyres, 40f, 41–42, 41f; oligotrophic systems of, 228, 261, 276; warming-­induced decline of phytoplankton biomass in, 291 Challenger expedition, 48, 261, 279 chaos, 271 chaotic dynamics, 269, 271, 293 chemical cues: amphipod reaction to fish predation and, 182; for larval settlement, 163, 163f; Phaeocystis colony formation and, 179 chemical defenses of plants and algae, 178; of bloom-­forming algae and phytoplankton, 257, 266, 341; of coral reef algae, 355, 359, 360; of host plants of mesograzer species, 179, 181f, 186; induced in marine algae, 179 chemoautotrophs, unicellular, 23, 24f, 26, 27t, 225 chemosynthetic ecosystems, 281–84, 281f chicken bones, 76, 104 chlorophyll: iron required for synthesis of, 117; primary producer biomass and, 9f chlorophyll a, 25, 118, 119 chlorophyll b, 25–26 chlorophyll c, 26 chlorophyll maximum, 61, 263, 335 circumpolar current, 39, 63 clades, 18f, 20 clams: Asian clam invasion, 342–43, 342f; at hydrothermal vents, 281, 283 Clements, Frederic, 196 climate: biogeography and, 38; biological pump and, 274; ocean circulation and, 40–41 climate change: biotic homogenization driven by, 103–4; communities and, 214–15; coral reef conservation and, 374; disease in marine organisms and, 307, 341; ecosystem models and, 254; evolution in response to, 158; during Last Glacial Maximum, 70; novel ecosystems resulting from, 257; species interactions and, 191–92; as threat to economic worldview, 86. See also climate warming climate variation: fishery yield stabilized by diversity and, 247; regime shifts in pelagic systems and, 249 climate warming, 78, 78f; Anthropocene ocean and, 289–92; biogeographic reor­ga­ni­za­tion and, 68–71; communities and, 90–91; coral diseases and, 364; in ecosystem modeling of primary production, 254; eutrophication and, 341; low-­oxygen zones due to, 316; mangroves and, 332, 333, 335; metabolic balance of ocean and, 287; metabolic models and, 135; with most heat absorbed by ocean, 90; salt marsh ecosystems disturbed by, 331; seagrass ecosystems and, 322–23. See also climate change; ocean warming; sea level rise climax community, 197, 220, 253 clownfish (Amphiprion percula), 163 CO2. See carbon dioxide (CO2) coast: cultural eutrophication on, 79; development on, 322, 323, 330, 340; extending from waterline to continental shelf, 296; interactions among habitats on, 335–36; oceanography of, 296–97; vegetation protecting against storms and erosion on, 321, 330, 334 coastal biome, 62 coastal boundary layer, and local retention of larvae, 161

Index

coastal ecosystems, 44, 302–4; in the Anthropocene, 337–44; climate of adjacent continent and, 297; ­human dependence on, 6; less nutritious higher plants in, 231; materials delivered by the land to, 296; restoration of, 323–25, 323f, 324f; sea level rise and, 323; ser­vices to ­humans from, 303–4; summary of, 346; trophic cascades in, 234f, 302. See also mangroves; rocky intertidal communities; salt marshes; seagrasses; sediment bottoms coastal infrastructure: anthropogenic biomes and, 106; global increase in jellyfish and, 105; of urbanized estuaries, 343–44 coastal marine communities: homogenization of, 103; invaded by nonindigenous species, 103, 104f; trophic skew in, 193, 193f. See also coastal ecosystems coastal marine organisms, and island biogeography, 58 coastal regions: with high productivity, 261; MEOW biogeographic realms of, 62–63, 62f; oceanography of, 44 coastal squeeze, 323, 330 coastal upwelling, 44, 62, 261, 297, 298f, 299, 305 coccolithophorid phytoplankton, 135, 136, 267, 267f, 292 cod (Gadus morhua): Baltic Sea collapse of, 252, 276; North Atlantic fisheries for, 64, 65, 252; tropicalization and, 70 coexistence of competing species: differences in dispersal ability and, 202, 202f; ecological drift and, 203; neutral theory and, 205; stable or unstable, 176. See also niche differentiation; niche partitioning cognitive psy­chol­ogy, 381, 382 cold seeps, 281f, 283–84 comb jellies. See ctenophores commerce: globalized, 72; homogenization of ocean driven by, 103; invasive species facilitated by, 339; transport of nonnative species by, 71 communities: accidents of history vs. environmental forcing and, 72; in the Anthropocene, 214–15; climate warming and, 90–91; concepts of, 196–98; defining variables of, 195; definitions of, 197–98; ecosystems and, 211, 213–14; functional structure of, 211, 213; functional traits and, 187–89, 211, 213–14; history and geography in structuring of, 197; number and kinds of species in, 195–96; ocean acidification and, 93–95; ocean warming and composition of, 192; pelagic, 269–71; phyloge­ne­tic structure of, 214; with similar composition in similar habitats, 197. See also diversity in communities; rocky intertidal communities community assembly, 199–200, 200f, 201 community dominant, 173, 175 community dynamics, 198–203. See also dispersal; ecological drift; ecological se­lection; speciation community metabolism, 222 community modules, 173, 173f community organ­ization: functional traits and, 187–89, 211, 213–14; in mangroves, 333; in rocky intertidal communities, 2, 9, 306–8, 308f, 309f; in salt marshes, 328–29; in seagrass meadows, 319–21, 320f; in sediment bottoms, 313–14; of sediment fauna subject to organic enrichment, 315–16, 316f; species interactions and, 2, 9. See also community structure community structure: climate change and, 121, 291; deterministic interactions and, 205; ecosystem

functioning and, 218. See also community organ­ization competition: climate warming causing shift to facilitation from, 192; interspecific, 175–76; in mangrove forests, 333; natu­ral se­lection and, 126; in salt marshes, 328, 329, 330; specialists vs. generalists and, 55; speciation and, 49–50, 51, 53, 55; three pos­si­ble outcomes of, 176; Tilman’s resource-­ratio theory of, 176 competitive exclusion: accelerated by herbivorous snail, 175; chaotic dynamics and, 269, 271; despite long-­distance dispersal, 50; disturbances maintaining higher diversity and, 207, 280; ­limited by larval dispersal, 201; nonequilibrium dynamics and, 270–71; paradox of the plankton and, 198, 269–70; resource partitioning and, 269, 270; on rocky intertidal shores, 174–75, 174f, 306–7; selective predator with effect on, 185. See also niche differentiation competitive exclusion princi­ple, 176, 270 complementarity: diverse prey community and, 246; productivity of diverse species and, 242. See also niche partitioning complex adaptive systems: of biosphere containing humanity, 84, 85, 379; earth system as, 377, 378; ecosystem as, 4, 224–25, 236; in ecosystem model of phytoplankton, 255; unpredictable but resilient, 377 connectivity: ge­ne­tic markers of, 162, 164; geochemical tracers and, 164; larval dispersal and, 159, 169 Connell, Joseph, 174–75, 176, 207, 304, 306 conservation: cautious optimism about, 108–10, 109f; community interactions applied to, 345; of coral reefs, 374, 375; ecosystem approach in, 257; elasticity analyses and, 151–52, 154, 154f; evidence-­based, 381; life histories and, 151–55; local engagement in management and, 106, 110; of mangroves, 231; marine ecoregions classification and, 62, 63; nonconsumptive effects of predators and, 184; of seagrass ecosystems, 231, 322–23, 323f, 325; social-­ cultural institutions and, 106 consumer-­prey interactions: diverse prey assemblages and, 246, 248f; as function of three pro­cesses, 121; stoichiometric mismatch in, 115, 115f; temperature and, 121; warming-­related life history changes and, 91 consumers, major energy pathways of, 112, 113f consumption, in energy bud­get, 118, 120f continental shelves: of active and passive margins, 297; coastal ocean extending to edge of, 296; disproportionate significance of, 296; organic ­matter delivered to, 303; trophic cascades over, 274 continents, history of, 38–40, 39f convergent fronts, 42–43, 42f Cook, James, 260, 347 copepods: biogeographic patterns in functional traits of, 64; biological pump and, 274; calanoid, 268, 272; climate warming and, 68–69, 70, 90; as macroplankton, 22; meiofaunal, 33; as mesoplankton, 22; population growth ­under food-­poor conditions, 146–47, 147f; resilient pattern of species diversity in, 271; sensitivity of life stages to temperature, 166; trade-­off between egg size and number in, 140, 141f coral gobies (Gobiodon), 51, 51f

433

coral larvae: crustose coralline algae and, 93–94, 161; inhibited by degradation of reefs, 365; ocean acidification and, 93–94, 372; from robust source reefs, 374; sensory biology of, 163, 163f; specific species of macroalgae and, 163 coral reef food web, simulated species removals in, 190 coral reefs: abiotic environment of, 351–52; in the Anthropocene, 369–74, 370f, 372f; biodiversity of, 347, 351, 357–60; chemically defended organisms on, 186; conservation and management of, 374, 375; cyanobacteria on, 178; dead zones in, 341; diseases with impacts on, 362–64, 362f; diversity in Atlantic vs. Pacific, 208; ecosystem ser­vices to ­humans from, 371; of Enewetak Atoll, 222, 222f, 223, 350, 367; evolution of, 350, 350f; fisheries on, 253, 368–69, 369f, 373; ­future of, 373–74, 380–81; geomorphology of, 348–51, 349f, 350f; halos of grazed areas surrounding, 177, 177f, 319; healthy vs. degraded algal-­ dominated, 163, 163f, 363, 365–66, 365f, 366f; herbivorous fishes on, 179, 248, 355–56, 356f, 365, 368–69, 373, 375; ­human impacts on, 367; intermediate disturbance hypothesis and, 207, 208f; legacy of Pleistocene ice ages in, 208; mangrove interactions with, 334; near urban centers, 337; ocean acidification and, 372; ocean warming and, 371, 372f; as oligotrophic systems, 218, 222, 228, 351, 352, 368; paradox of low fisheries yields in, 253; paradox of productivity of, 347, 352, 367; plastic pollution on, 364; primary producers of, 222, 354–55, 354f; promising topics for research on, 374–75; regional variation in, 366–67; in state of crisis, 363, 369–71, 370f; summary of, 375–76; transition from kelp forests to, 192; trophic cascades on communities of, 361–62; uncertain role of algal proliferation on, 375. See also Ca­rib­bean reefs; crustose coralline algae; filamentous (turf) algae; G ­ reat Barrier Reef; Jamaican coral reefs; reef communities; reef fishes corals: acidification altering communities of, 94; anatomy and physiology of, 351f, 352; communities supporting better condition of, 110; of deep sea, 278f, 289, 292; as dominant coastal primary producers, 301; double blow to calcification of, 93; facilitation by fire corals, 186; fishes feeding on live corals, 179; as foundation species, 102–3, 301, 352; functional classification of, 353–54, 353f; hermatypic, 347; heterotrophic capacity of, 352; hybridization among, 21; Indo-­Pacific species distributions of, 39–40, 57; life history matrices for, 148–49, 149f, 150–51, 152t; local vs. regional control of diversity and, 212–13, 212f; loop diagrams of life histories, 149f; maintained by herbivore functional diversity, 246, 248; as net producers due to symbiotic algae, 296; neutral theory and, 205, 206f; restoration proj­ects for, 109f; seagrasses reducing disease prevalence in, 336; symbiotic algae of (see zooxanthellae); thermal tolerances of, 121; threatened species of, 102; trait-­based approaches to, 137. See also Acropora; bleaching of corals; coral larvae; holobiont, coral-­algal Coral Triangle, 48, 350; biodiversity in, 357; distributions of species in, 57; diversity of seagrasses in, 318; as evolutionary center of origin, 60. See also Indo–­West Pacific reefs

434

Index

cordgrasses. See Spartina cordgrasses Coriolis force, 40–42, 40f crabs: Eu­ro­pean green crab (Carcinus maenas), 310; king crab (Paralomis birsteini), 88f, 91, 215; in mangrove forests, 333; predatory, using seagrass near oyster reefs, 336; in salt marshes, 329 critical depth, 265 cross-­habitat diversity gradient, 54 crown-­of-­thorns starfish (Acanthaster planci), 361–62, 365, 370f, 374 crustaceans: euphausiid (krill), 22, 264f, 268, 272, 274; mesograzing, 179; zooplanktonic, top-­down control of diatoms by, 233. See also amphipods; copepods; crabs crustose coralline algae, 15f, 88f; acidification and, 93–94, 372; coral larvae settling on, 372; grazing intensity and, 360; on hot tropical shores, 304; overgrown by fleshy seaweeds, 93, 94; on temperate rocky reefs, 181 cryptic species, 156, 205 ctenophores, top-­down control of zooplankton by, 233, 234f, 275, 275f cultural eutrophication, 79 cyanobacteria: avoided by herbivores, 214; on coral reefs, 178, 355; as dominant primary producers of ocean, 268–69; fossils similar to, 311; of harmful algal blooms, 341; origin of chloroplasts and, 19; origin of oxygenic photosynthesis and, 1; in sediment, 32; stromatolites built by, 348. See also Prochlorococcus Cycliophora, 11, 12f cytometric diversity of phytoplankton, 285, 285f damselfishes: Ca­rib­bean (Stegastes partitus), 155, 156f; on coral reefs, 355, 362, 369; species interactions stabilizing populations of, 248 dark reaction, 118 Darwin, Charles, 49–50, 60, 126, 155, 245, 260; coral reefs and, 347, 350, 358, 367 dead zones, low-­oxygen, 110, 257, 316, 340–41 decomposers, and energy pathways, 113f decomposition: nitrate accumulation from, 115–16; quality of primary producers and, 115, 115f, 116f deep scattering layer, 28, 264 deep sea, 276–80; adaptations to life in, 276–77, 278f; biodiversity in, 48, 265, 279–80, 280f, 284–88, 285f; dead zones in, 316, 341; detritus as main food source in, 235, 278–79; episodic food sources in, 279; fisheries of, 288–89, 290, 292; global distribution of benthic biomass in, 276, 277f; ­limited knowledge of, 276; pelagic-­benthic coupling in, 278–79, 279f; scarcity of food in, 276; seagrass detritus in, 321; seasonality of material flux to, 278–79, 279f; volume of, 276 defaunation, 215, 258 democracy, 378 demographic modeling. See matrix population models demographic par­ameters, and metabolic allometry, 166–67 demographic transition, ­human, 168 depensation, 251 deposit-­feeders, 312, 312f, 313, 314, 315; detritivores as, 236; macroinfauna as, 33–34; in salt marshes, 328 depth diversity gradient, 48 deserts: of deep sea, 277, 281; of open ocean, 42, 222, 261, 347, 352, 367. See also oligotrophic systems

determinism: in community structure, 205, 206, 207; ecosystem pro­cesses and, 225 detritivores, 113f, 236; of deep-­sea benthos, 277; increased by nutrient enrichment, 341; ­limited by nitrogen availability, 115; nutrition of food chosen by, 182 detritus, 235–36, 235f; aggregates of, 235; on algal turfs of reefs, 355; in coastal systems dominated by higher plants, 231; consumers of, 233, 235–36 (see also detritivores); declining exponentially with depth, 264, 264f, 278; defined, 233; from epipelagic ecosystem, 264; grazed in seagrass meadows, 319; from kelps of rocky coasts, 181, 232; from mangrove production, 333, 334; morphous vs. amorphous, 235, 235f; pathways of, 235–36, 235f; as primary deep sea food source, 235, 278–79; in reef fishes’ diet, 359; in salt marshes, 325, 328, 330; from ungrazed seagrass, 317, 321 development, 139. See also life histories Diadema antillarum (long-­spined sea urchin), 175, 184–85, 362, 362f, 363, 373, 375 diatoms: biological pump and, 274; conditions favoring, 267, 267f; disfavored by ocean warming, 292; epiphytic on seagrasses, 319; functional differentiation from dinoflagellates, 270; ocean acidification affecting iron use of, 292; responding to day length more than temperature, 192; in sediment, 32; silica cell wall as defense of, 268; Si:N ratios and, 129, 130; species composition associated with nutrient availability and, 286; in spring bloom, 266; top-­down control by crustacean zooplankton, 233; trans-­Arctic invasion of North Atlantic by, 71, 88f; warming of polar w ­ aters and, 26 dilution hypothesis, 364 dinoflagellates: conditions favoring, 267, 267f; functional differentiation from diatoms, 270; of red tides, 266. See also zooxanthellae direct development (brooding), 140, 141–42; of peracarid crustaceans, 18, 140 directional se­lection, 156–57, 157f discrete rate of increase: of population, 146, 148, 152t; of trait or gene, 146 disease in ­humans: declining biodiversity and, 341, 343; seagrasses reducing prevalence of, 336 disease in marine organisms, 184–85; aquaculture as risk for, 101, 104–5; climate change and, 341; coral reefs impacted by, 362–63; coral species threatened by, 102; in seagrass meadows, 320–21, 322–23; seagrass role in reducing prevalence of, 336; transported with nonnative oysters, 339; unanswered questions about, 375; warming favoring the spread of, 91, 102, 192, 322–23 disease organisms, as keystone species, 173, 175, 362, 364 disequilibrium in community ecol­ogy, 207 dispersal: across Arctic Ocean, 198; artificial structures as barrier to, 344; composition of communities and, 45, 45f, 198; distance variation between types of organisms, 158–59, 158f, 164; ge­ne­tic markers of, 162, 164; global microbial diversity and, 286; ­human impact on barriers to, 72; in integrative models of biodiversity, 60; latitudinal diversity gradient and, 60; local community diversity and, 200–201; marine reserves and, 154; in metacommunity, 201; in metapopulations, 138; rescue effect of, 201; species richness and, 209–10,

209f; theory of island biogeography and, 57–58, 57f. See also larval dispersal dispersal ability, 56–57; coexistence of ecologically equivalent species and, 202, 202f; ge­ne­tic structure in reef fishes and, 54–55, 55f, 57; geographic range size and, 167; speciation and, 50, 51, 54–55, 55f, 57 dispersal filter, 200, 200f dissolved organic carbon (DOC), 273f, 354 dissolved organic ­matter (DOM): absorbed by corals, 352; Bacteria and Archaea feeding on, 25; in detritus formation, 235; in pelagic ecosystem pro­cesses, 129f; in sediments, 30, 32 disturbance: coral disease outbreaks and, 363, 364; deep-­sea biodiversity and, 280, 285; diversity in communities and, 207, 245–46; reef community diversity and, 358; to salt marsh ecosystems, 331. See also intermediate disturbance hypothesis diversification rate hypothesis, 53 diversity in communities, 203–11; adaptation to environmental change and, 225; disturbance and, 207, 245–46; local vs. regional control of, 212–13, 212f; metabolic theory and, 210–11; neutral theory of, 204–6; ocean warming and, 214–15; predation affecting level of, 175, 185; productivity and, 368; regional enrichment and, 209–10, 209f. See also biodiversity diversity minimum, estuarine, 299, 299f DNA sequencing: reconstruction of phylogeny from, 20–21. See also metagenomics DOC. See dissolved organic carbon DOM. See dissolved organic ­matter domestication of marine species, 104–5, 158. See also aquaculture domestication of the ocean, 104–5 donor-­controlled systems, 235 downwelling at convergent front, 42, 42f dugongs, 302, 319, 321, 323, 355 dynamic equilibrium model, 207 EAF (ecosystem approach to fisheries), 63, 100 East Indies. See Coral Triangle EBFM (ecosystem-­based fisheries management), 100, 106 echinoderms, 12, 14f; ocean acidification and, 135. See also sea urchins; starfish ecological drift, 45, 45f, 198, 201, 203, 359. See also neutral theory ecological economics, 80, 108 ecological efficiency, 237–38 ecological engineering: in coastal communities dominated by macrophytes, 302; in sediment communities, 314 ecological extinction, 102, 258 ecological opportunity hypothesis, 55 ecological se­lection, 45, 45f, 198, 199–200; in coral reef communities, 357, 359–60; in rocky intertidal communities, 206; sympatric speciation and, 50 ecological speciation, 50, 51 ecological stoichiometry, 115, 115f, 116f ecol­ogy: adapted to human-­dominated world, 6; complexity in, 7–8; framework for thinking about, 3–4, 3f, 4f; ­human sciences essential to, 84 economic growth, 380. See also economics economics: conservation goals and, 110; ecological, 80, 108; energy use and, 77f, 84, 85f; evidence-­based understanding of, 381; ignoring

Index

biophysical constraints, 86; in natu­ral history of ­humans, 86 Ecopath mass balance model, 253 Ecopath with Ecosim (EwE), 253–54 ecoregions classification, 62–63, 62f ecosystem approach to fisheries (EAF), 63, 100 ecosystem-­based fisheries management (EBFM), 100, 106 ecosystem engineering by ­humans, 85, 378 ecosystem functioning, 218 ecosystem modeling: in fisheries management, 252–55, 253f, 254f; ­future of, 259; of global ecosystem, 255–57, 256f ecosystems: in the Anthropocene, 257–58; biodiversity and, 241–48; communities and, 211, 213–14, 224–25; as complex adaptive systems, 4, 224–25, 236; conceptual history of, 219–25; control of biomass and productivity in, 228–36, 258; functional traits and, 128, 129f, 211, 213, 236–37; irreversible transitions in, 380–81; Odum’s engineering meta­phor for, 222, 222f; organismal traits in, 236–37; predictability of major properties of, 293; resilience of, 380–81; risk assessment for, 257; terrestrial, 12, 13t, 293 ecosystem ser­vices to ­humans, 218; coastal, 303–4 ecotypes, 50, 51; of Prochlorococcus, 256f, 270 eelgrass (Zostera marina): beneficial nutrient enrichment in meadows of, 341; coastal sediment conditions and, 251; food web of animals supported by, 187, 187f; ge­ne­tic diversity of, 325; legacy of glaciation and invasion on, 40; length of shoots in, 318; organic carbon stored in sediment by, 322; trophic cascades involving, 302, 324–25, 324f; wasting disease in 1930s and, 175, 321, 322–23; wide range of salinity tolerance in, 318. See also seagrasses eelgrass, Japa­nese (Zostera japonica), 339, 340f effect traits, 21, 215–16 Ekman transport, 41, 44, 349–50 elasticity, 148, 152t; management uses of, 151–52, 154, 154f electrical cir­cuit approach to ecosystems, 222, 223, 223f elemental composition of organisms, 113–15, 114f. See also stoichiometry elkhorn coral (Acropora palmata), 15f; beginning to revive, 381; threatened status of, 88f, 102; white-­band disease of, 185, 362–63, 370, 375 El Niño event of 1982–1983, 371 Elton, Charles, 120, 238 emergent properties: climate change effects on ecosystems and, 257, 293; of ecological networks, 189–91; of ecosystems, 7, 218, 219, 224; in ecosystem simulations, 255; of food webs, 186, 187. See also complex adaptive systems endemicity, 44, 45; within MEOW realms and provinces, 63; modern extinctions and, 59; of remote islands, 58 endosymbiosis, in origin of eukaryotic cell and mitochondrion, 56 end-­Permian mass extinction, 58 end-­to-­end ecosystem models, 254–55, 254f energetic equivalence rule, 166–67, 190; population-­ level interaction strengths and, 190; species richness and, 210; trophic levels and, 240, 241f energy: as central ­human resource, 86; in early ecosystem studies, 222–23, 223f; major pathways of producers and consumers of, 112, 113f;

metabolism and, 117; mostly lost to animal’s respiration and excretion, 237–38; species diversity and, 210–11; species-­energy hypotheses and, 54, 210–11, 285–86; stored in carbon-­carbon bonds, 114 energy bud­get of organism, 118, 120f energy use by ­humans: economic activity and, 77f, 84, 85f; ­human life history traits and, 168. See also fossil fuels Enewetak Atoll: drilling to volcanic basement of, 350; Odum study at, 222, 222f, 223, 367 Enteromorpha, 310 environmental filtering, 200, 200f environmental forcing vs. accidents of history, 72 environmental niche models, 69–70, 69f, 106; climate change and, 158 ephemeral algae: eutrophication and, 258, 310, 324; facilitation by herbivores feeding on, 302; on rocky shores, 307, 310; seagrasses and, 324–25; traits of, 133 epifauna, 30, 34; damaged by fishing machinery, 99; seagrass providing habitat for, 321 epipelagic zone, 263–64, 263f epiphytic algae on seagrasses, 319, 324 equalizing mechanisms, and unstable coexistence, 176 equilibrium. See nonequilibrium pro­cesses Estes, James, 181, 232 estuaries, 298–300; bottom-up control in, 302; circulation patterns of, 298f, 299; dead zones in, 341; defined, 298; diversity gradients in, 299, 299f; dominant primary producers of, 300; fauna of sediment bottom in, 312f; invasive species in, 339, 340f, 342–43, 342f; oyster reefs in, 299, 338; of passive continental margins, 297; primary and secondary production in, 302–3, 303f; reor­ga­nized by nonnative species, 339, 340f; salt marshes in, 326; seagrasses in, 317, 318, 324–25; sea level rise and, 323; trophic cascades in, 276, 324–25; urban centers and, 337, 343–44 estuarine fish species, functional traits predicting demographic changes in, 189 Eukarya, 19, 19f, 20; endosymbiotic origin of, 56; primary producers among, 225 euphausiid crustaceans (krill), 22; biological pump and, 274; deep foraging by ­whales and, 264f; as herbivores of plankton, 268; in pelagic food web, 272 Eu­ro­pean green crab (Carcinus maenas), 310 eurythermal species, 121, 291 eutrophication, 257–58, 340–41; in Black Sea, 275; climate warming and, 341; coastal community composition and, 309–10; as control on seagrass distribution, 251; disturbing salt marsh ecosystems, 331; ephemeral algae and, 310; nitrogen availability and, 309–10, 324; trophic cascades in coastal ecosystems and, 302, 324–25. See also blooms of algae and phytoplankton evidence-­based approach, 381–82 evolution: coastal geography and, 44; human-­induced, 98, 104–5, 135, 158, 168, 169–70; in urbanized environments, 344. See also natu­ral se­lection EwE (Ecopath with Ecosim), 253–54 excretion: by heterotrophs, 116, 118. See also fecal pellets exploitative competition, 126, 173, 173f, 176 export of production to deep ocean, 226, 228; by biological pump, 273–74, 273f; microbial loop and, 267. See also carbon export to deep ocean

435

extended phenotype, 83 extinction: abiotic forcing vs. organismal traits and, 73; body size and, 136; climate warming and, 103; ecological, 102, 258; ecological specialization and, 55; functional redundancy and, 246; ­human impacts and, 36, 58, 59; in integrative models of biodiversity, 60; latitudinal diversity gradient and, 53, 60; of marine species in modern times, 59, 101–2, 102f; predicted in Anthropocene ocean, 69–70, 69f; prevented by rescue effect, 201; solar energy and, 54; temperature in fossil rec­ord and, 124; theory of island biogeography and, 57–58, 57f, 59; two distinct syndromes of, 58–59. See also mass extinctions extinction debt, 103 extremophile Archaea, 19 facilitation, 185–86, 186f, 189; in coastal communities, 302; by foundation species, 185–86, 189, 304; in mangrove forests, 333; in rocky intertidal communities, 307; in salt marsh communities, 329, 330; in sediment communities, 314; stressful environments causing shift to, 192 fecal pellets: of metazoan heterotrophs, 28, 274; of zooplankton, 235, 237, 273, 278 fecundity: age-­specific, 145, 147, 148; larval dispersal and, 142–43; scaling with body mass, 166 fertilization of ocean with iron, 107, 266 filamentous (turf) algae, 178, 179, 222, 354–55, 354f, 356; epiphytic on seagrasses, 319; fishes excreting nutrients absorbed from, 367; keeping pace with herbivory, 360; mass die-­off of Diadema and, 362, 362f fire corals, 186 fisheries, marine: alternative stable states in, 250–51, 250f; biomass production in, 117; bottom-up control of yield in, 230, 231f; coastal, 340; coastal and estuarine productivity of, 302, 303f, 304, 326, 326f; on coral reefs, 253, 368–69, 369f, 373; current state of, 97–98; of deep sea, 288–89, 290, 292; directional se­lection in, 156–57, 157f; diversity enhancing stability of yield in, 246, 247, 247f; economic and nutritional importance of, 6, 95, 368; economic and nutritional losses in, 98, 100; ecosystem impacts of, 98–99; eutrophication enhancing productivity of, 341; exploitation rates in, 97–98, 97f, 100; “fishing down the food web” in, 215; ­future of, 99–101; hope for sustainable ­f uture in, 108, 109–10; ­human evolutionary pressure in, 104–5; ­human impacts on, 95, 96f, 97, 135f, 193, 337; illegal fishing in, 110; mangrove contribution to production in, 334, 336; ocean warming and, 91, 291; precipitous decline in, 89; of rocky subtidal habitats, 309; salt marsh ecosystems and, 330; seagrasses supporting ser­vices to, 321; shifting baseline and, 87; supported by soft sediments, 315; trophic cascades induced by, 361–62; tropicalization and, 70; upwelling systems that sustain high levels of, 228. See also bycatch in fisheries; industrial fishing fisheries management, 97–98; challenges of Anthropocene and, 106, 108; coral reef resilience and, 373–74; in developing tropical nations, 373; ecosystem focus in, 257; energy and ele­ment bud­gets in, 117; food webs used in, 186–87, 187f; ­future of, 99–101; good governance and, 381–82; Hjört-­Cushing match-­mismatch hypothesis and, 169; inflated Chinese catch statistics and, 252;

436

Index

fisheries management (continued) large marine ecosystems in, 63; modeling in, 252–55, 253f, 254f; nonlinear responses and, 99; specific production (P/B) and, 151; weakness of ge­ne­tic approaches to, 165 fishes: artificial structures providing nursery for juveniles of, 344; biomass increasing with species diversity of, 244, 245f; body size and vulnerability to exploitation of, 65; mangroves providing habitat for juveniles of, 333, 336; as nekton, 28; as paraphyletic group, 20; productivity boosted by interactions among habitats of, 336; seagrass providing habitat for juveniles of, 321, 336; size spectra of, 135f, 239; thriving when protected, 381; trait-­based approaches to, 137. See also reef fishes fishes, herbivorous, 35–36; behavioral responses to predators and, 184; on coral reefs, 179, 248, 355–56, 356f, 365, 368–69, 373, 375; feeding on macroalgae, 179, 215, 360; functional groups of, 179; ocean warming and, 192, 215; in rocky intertidal communities, 305; on seagrass meadows, 319, 322 fishes, predatory: acidification and, 94; amphipod reaction to chemical cues from, 182; migrating from protection of coral reefs to seagrass beds, 335; top-­down control of coastal ecosystems by, 325 fish larvae: planktotrophic, seasonality of recruitment, 169; recruitment variability based on size of, 141, 142; sensory biology of, 163, 163f; warming-­related life history changes in, 192 fitness: ecological se­lection and, 199; life history traits and, 127; natu­ral se­lection and, 126; quantitative definition of, 155; reduced by competition, 175; species interactions and, 129 Flavobacterium columnare, 105 fleshy seaweeds: Diadema mass die-­off and, 362, 362f; favored in biotic homogenization, 104; low grazing on reefs and, 360; shifts ­toward dominance by, 94, 103 flowering plants, 13, 30; detritus-­based food web and, 235, 303; as dominant primary producers of estuaries, 300; herbivore effects on, 34f; low nutritional quality of, 115, 131, 301–2; low vulnerability to herbivory, 231, 301–2. See also macrophytes; mangroves; marsh grasses; seagrasses food chain: classical pelagic food chain, 267, 272, 272f, 291; energy and ele­ment bud­gets for, 117; modules of, 173f food supply to ­humans, 81f, 82 food webs, 173f, 186–87; detritus-­based, 235; early example based on eelgrass, 187, 187f; interaction strengths in, 189; ocean warming and, 291; simulations of, 189–91 (see also ecosystem modeling); supporting North Sea herring, 187, 188f; vulnerable to loss of well-­connected species, 189–90 foraminifera: of deep-­sea benthos, 280, 286; planktonic, with faster speciation at high temperatures, 53f, 54 fossil fuels: catalyzing ­Great Acceleration, 78; economic activity and, 84; as existential threat to coral reefs, 370, 371; ­human development supported by, 6, 74–75, 75f, 77–78; nitrogen oxides from combustion of, 79, 80f; trend in extraction since 1972 predictions, 81f, 82; unsustainable

consumption rate of, 74, 86, 378. See also climate change; climate warming; ocean acidification fouling species: characteristic suite of, 344; interaction pathways in seagrass beds and, 320f; nonnative ascidians, 103; nonnative invertebrates in harbors, 343; temperature effect on consumer pressure on, 66, 192 foundation species, 30, 34, 180–81; alternative ecosystem states on temperate rocky reefs and, 181; Anthropocene transition from, 36; artificial structures reducing establishment of, 344; calcareous, ocean acidification and, 93; coastal macrophytes as, 296, 300, 301; corals as, 102–3, 301, 352; declining due to ­human impact, 191, 337, 338; dependent on predator control of herbivores, 302; facilitation by, 185–86, 189, 304; large impacts on communities by, 190–91, 193; management of nutrient loading and, 343; nontrophic interactions and, 189; often declining fastest, 102–3; of salt marshes, 325, 327, 329; of sediment bottoms, 314 fringing reefs, 350, 350f functional biology, 5 functional diversity, 5; in ecosystem modeling, 228; ecosystems stabilized by, 246; extinction risk and, 59, 73; of herbivores, 227, 227f; species richness and, 241–42 functional groups, 128; fates of primary production and, 230–31; of herbivores, 179; of phytoplankton, 131–33, 132f; species diversity and, 242; trophic structure and, 211, 213; of vent communities, 283 functional groups, benthic, 30–36; herbivory and, 131; of macrophytes, 133 functional groups, pelagic, 21–28, 129f; body size and, 22–24, 24f; nutrition and metabolism of, 23, 24f, 25–28, 27t functional redundancy, 246 functional traits: biogeography of, 63–64; bridging levels of organ­ization, 133–34, 187; community organ­ization and, 187–89, 211, 213–14; defined, 21; ecosystem pro­cesses and, 128, 129f, 211, 213, 236–37; effect traits, 21, 215–16; in interaction networks, 189; limiting similarity in, 187–89, 188f; in models of global change, 106; predictions based on study of, 136; responses to environmental forcing and, 189; response traits, 21, 215–16; trade-­offs in, 128, 131, 136–37; trait-­based ecol­ogy and, 129–31. See also functional groups; phyloge­ne­tic conservatism of traits; traits fundamental niche, 129, 199, 199f Gaia hypothesis, 2, 221 gamete recognition proteins, 50 gamma diversity (γ), 45 gas hydrates, 284 GDP (gross domestic product), 77f, 84, 85f General Ecosystem Model, 255–57 ge­ne­tic drift: connectivity among populations and, 164; environmental disturbances and, 201 genotype, 127 geochemical tracers, 164–65, 165f geoengineering, 107 geographic range size. See range size geomorphology, 3; of coral reefs, 348–51, 349f, 350f; ­human transformation of, 6, 7f; of mangroves, 332; of rocky shores, 305, 306f; of salt marshes, 326–27, 327f; of seagrass meadows, 317–18 Georges Bank haddock, 250–51, 250f

geostrophic flow, 40f, 42; trades biome and, 61 ­giant kelp (Macrocystis pyrifera), 13, 15f, 30, 181 glacial isostatic adjustment, 91 glaciations: Antarctic, 39; in Pleistocene ice ages, 39, 39f, 40 Gleason, Henry, 197 Global Fishing Watch, 110 globalization, 103–4 global warming. See climate warming Gorgonian corals, 354 governance, 381–82 gray ­whales, 29, 29f, 159–60 grazers: benthic, 34–36; in coastal ecosystems, 302; harmful algal blooms and, 257–58; micrograzers of phytoplankton, 268; pelagic ecosystems and, 274; pelagic functional types of, 268–69; on spring bloom, 265, 266. See also herbivores grazing: ecosystem pathway of, 113f; eutrophication and, 310; highest on nutrient-­rich microalgae, 131, 230–31; inhibited by acidification, 94; intensified around artificial structures, 105, 344; lowest on higher plants, 131, 231; macroalgal functional groups and, 133, 134f; in marine vs. terrestrial ecosystems, 125–26; in modern vs. historical seagrass ecosystems, 321; overgrazing caused by trophic skew, 338; on phytoplankton, 268, 269; rapid turnover of primary production by, 228, 228f; in salt marshes, 329, 330; stronger at low latitudes, 66, 67f; top-­down control by, 226–27, 228, 228f. See also herbivory ­Great Acceleration, 6, 76–82; decrease in extreme poverty since, 380; indicators of, 76, 77f; industrial-­scale fishing in, 95 ­great auk, 88f, 101 ­Great Barrier Reef: bleaching of corals on, 371, 373, 374; coexistence of ecologically equivalent fish species on, 202, 202f; crown-­of-­thorns starfish on, 361–62; decline of coral cover on, 370f; disease linked to terrestrial runoff and, 363; dozens of coexisting Acropora species on, 205; herbivore abundance and algal proliferation on, 375; herbivores controlling algae on, 360, 361f; microhabitats and diets of herbivorous fishes on, 359; resilient source reefs in, 374; seen from space, 349f, 350; semisynchronous mass spawning of corals on, 143–44; signs of some recovery on, 373; weak top-­down forcing on, 362 ­Great Barrier Reef Marine Park Act, 100 ­Great Oxygenation Event, 1, 2 green algae (Chlorophyceae), 26 green­house effect, 78 green­house gases: climate warming and, 289; increasing since early ­human history, 75, 76f; methane as, 284; reaching equilibrium only ­after de­cades, 95. See also carbon dioxide (CO2) green web, 235, 236 green world hypothesis, 228–29, 229f, 232 gross primary production: defined, 118; highest on coral reefs, 367–68 Gulf Stream, 41, 297 gyre circulation, 40f, 41–42. See also central ocean gyres Haber-­Bosch pro­cess, 79 habitat: conservation strategies based on life histories and, 154–55; deep-­sea heterogeneity of, 280; ­human impact on, 337–38; provided by seagrasses, 321; species richness and age of, 52, 52f

Index

habitat area: speciation and, 50, 52; species richness and, 211; theory of island biogeography and, 57–58, 57f. See also species-­area relation (SAR) habitat complementarity, 335–36 habitat fragmentation: extinction driven by, 59, 72; metapopulations and, 159; speciation and, 357 habitat se­lection, and speciation, 50, 51 Halimeda, 178 Hardin, Garrett, 86 hard substrata, 30; ­human infrastructure used as, 105 Hardy, Alistair, 187, 188f herbivores: ammonium and urea excreted by, 226; of coastal benthic systems, 302; on coral reefs, 355–56, 356f, 360–61, 361f, 373, 375 (see also fishes, herbivorous; sea urchins); ecosystem energy pathways and, 113f; functional diversity of, 227, 227f; functional groups of, 179; grazing pelagic primary production, 268, 269; interactions between plants and, 176–79, 181f; latitudinal trends in impact of, 67–68, 67f; ­limited by nitrogen availability, 115, 115f; in mangrove forests, 333; nutritional quality of primary producers and, 115, 115f, 116f; range expansion due to tropicalization and, 70, 71f; in seagrass meadows, 319–21, 320f; stoichiometry of food choice by, 182; temperature-­ mediated increase in consumption by, 191, 193–94; in transition between coral-­and algal-­dominated states, 360, 362. See also fishes, herbivorous; grazers herbivory: plant defenses against, 177, 178, 268 (see also chemical defenses); plants’ phyloge­ne­ tic conservatism and, 214; plant stoichiometry and, 194; vigorous on coral reefs, 347; warming temperatures and, 192, 193. See also grazing hermatypic corals, 347 Hessler, Robert, 279–80 heterotrophic bacteria, 23, 24f, 25 heterotrophs: metazoan invertebrate plankton, 28; multicellular, 23, 24f, 28; nitrogen from excretion by, 116, 226; unicellular eukaryotic, 23, 24f, 26–27 heterotrophy, 118, 120 higher plants. See flowering plants high-­nitrogen low-­chlorophyll (HNLC) regions, 107, 117, 266, 292 Hilborn, Ray, 100 historical pro­cesses, 72, 197, 198, 207–9. See also biogeography Hjört, Johan, 169 Hjört-­Cushing match-­mismatch hypothesis, 169 holistic vs. reductionist approaches, 7–8 holobiont, coral-­algal, 347, 351f, 352, 367 Holocene epoch, 4, 377 homeothermic megafauna, 23, 24f, 28–30, 29f homeothermic vertebrates, interaction strengths of, 193 homogenization of ocean life, 103–4 Homo sapiens: dependence on the ocean, 6; evolving in response to niche construction, 85; as keystone species, 108, 345, 378; natu­ral and cultural history of, 83–84; origin of, 1; outsized influence on Earth’s ecosystems, 1, 6; as top predator of ocean, 87, 292; ultrasociality of, 84, 378. See also ­human society horizontal gene transfer, 19–20, 21, 25 HSS (Hairston, Smith, and Slobodkin), 228–29 Hubbell, Stephen, 201, 203, 204 ­human behavioral and social ecol­ogy, 108 ­human society: advancement in well-­being of, 74–75, 378, 379; continuing dependence on biological

communities, 378; energy use by, 84, 85f, 86; extinctions driven by, 36, 58, 59; impacts on natu­ral world, 337; trends in socioeconomic development, 77f; tribal and emotional motivations in, 378, 382. See also Homo sapiens Hutchinson, G. E., 130, 195, 198, 357; deep-­sea biodiversity and, 280; on disequilibrium, 207, 271; niche definition of, 128–29. See also paradox of the plankton Huxley, Julian, 121 hybrid end-­to-­end ecosystem models, 254–55 hybridization, 21 hydrothermal vents, 281–83, 281f; global distribution of, 282, 282f; ­whale carcasses and, 29 hyperdiverse tropical forests, 204 Hypnea, facilitated by Sargassum, 185–86, 186f hypoxic regions. See dead zones, low-­oxygen hysteresis, 249, 250f, 251; in Black Sea, 275; in coral reef communities, 365, 365f indirect interactions, 173 Indo–­West Pacific reefs: Ca­rib­bean reefs compared to, 366–67; dilution hypothesis and coral disease in, 364; diversity in, 366–67; soft corals in, 354. See also Coral Triangle Indo–­West Pacific region, 350, 357 induced defenses of plants, 178–79 industrial agriculture, 85 industrial fertilizer, 229 industrial fishing, 72, 88f, 95, 97; government subsidies for, 109; seafloor destruction caused by, 98–99, 289, 292; trawling of seabed in, 292, 295, 315, 316–17, 337 Industrial Revolution, 74, 75, 76, 78, 79, 86 infauna, 30, 33–34, 312, 313–15; increased by nutrient enrichment, 341; larvae of invertebrates of, 313–14; macroinfauna, 32f, 33–34; regime shifts and, 252; seagrass providing habitat for, 321; sediment grain size diversity and, 280. See also meiofauna innovation. See technological innovation instantaneous increase: of population, 146, 153f, 165–66; of trait or gene frequencies, 155 insurance hypothesis, 246, 247 integrated polyculture systems, 106 interaction filter, 200, 200f interaction modules, 173, 173f interaction networks. See networks, ecological interaction strengths, 171, 172f, 173, 193–94; defined, 190; in food webs, 186; organismal traits as predictors of, 189; population-­level, 190, 190f; in rocky shore communities, 190, 190f; temperature-­mediated changes in, 191–92 interactome, 17, 17f interference competition, 126; among infaunal species, 313 intermediate disturbance hypothesis, 207, 208, 208f, 270; mixing of ­water column and, 285. See also disturbance International Convention on the Conservation of Antarctic Marine Living Resources, 100 interspecific competition, 175–76 intertidal zone, 297–98; mangroves and, 331–32; rocky shore community interactions in, 171, 172, 173; salt marshes and, 325, 327 intrinsic growth rate of population (r), 146, 153f, 165–66

437

invasions of nonnative species, 338–39, 339f; of Antarctic shelf by king crab, 88f, 91, 215; Arctic opening and, 71; by Asian clam in San Francisco Bay, 342–43, 342f; in Black Sea, 275, 275f, 276; coastal patterns of, 103, 104f; commercial shipping as driver of, 103; in ­earlier earth history, 338–39; in estuaries, 339, 340f, 342–43, 342f; extinctions driven by, 59; favored by globalization, 103; fisheries-­related, 103; by fleshy algae favored by acidification, 94, 136; increasing diversity in some areas, 72; low-­diversity regime shifts triggered by, 252; ocean warming and, 103; often lower in food chain, 104; by pathogens, 184; predicted in Anthropocene ocean, 69–70, 69f; on rocky shores, 310; trophic skew exacerbated by, 193, 338; by weedy species, 338. See also nonnative species inverted biomass pyramids, 151 iron: Ca­rib­bean vs. Pacific reefs and, 367; fertilization of ocean with, 107, 266; limiting primary production, 117, 266, 367; ocean acidification and, 292; phytoplankton growth and, 114; wind-­borne, 291, 293, 367 island area and age, and species richness, 52, 52f island biogeography, theory of, 57–58, 57f; extinction due to habitat loss and, 59; ­human impacts on diversity and, 72; neutral theory and, 203, 204 Isthmus of Panama: allopatric speciation and, 49; origin of, 39, 297; vicariance across, 57 IUCN Red List, 101, 102f, 136, 257, 371 Jamaican coral reefs, 148–49, 150–51, 152t; beginning to come back, 373, 381; decline of, 363, 365 Japa­nese eelgrass (Zostera japonica), 339, 340f jellyfish, global increase in, 105 kelps: climate warming and acidification and, 338; fishing as cause of overgrazing in, 338; ­giant kelp (Macrocystis pyrifera), 13, 15f, 30, 181; high-­latitude forests of, 300, 338; life histories of, 139f; predator diversity affecting productivity of, 242, 244f; on rocky shores, 181, 301, 304; in trophic cascade, 179, 180f, 181, 232, 249; vulnerable to ­human disturbance, 103; warming-­related transition to coral reefs from, 192; warm-­ temperature stress leading to herbivore outbreak in, 302 key innovations, diversification stimulated by, 56 keystone predation, 173, 173f keystone species, 173; disease organisms as, 173, 175, 184, 307, 362, 364; Homo sapiens as, 108, 345, 378; ­human impacts on, 72, 191; with large impacts on communities, 190–91, 193; often declining fastest, 102; sea star Pisaster ochraceous, 12, 14f, 68, 91, 173, 175, 304, 306–7 killer ­whales: expanded into Hudson Bay, 71–72; trophic cascade beginning with, 180f, 181, 232 king crab (Paralomis birsteini), 88f, 91, 215 Kleiber, Max, 121 krill. See euphausiid crustaceans Kuroshio Current, 41, 297 Labyrinthula zosterae, 320–21, 323 lakes, regime shifts in, 251, 252 landscapes of fear, 177f, 183–84 large-­bodied species: ecological extinction of, 258; ­human impact falling on, 337, 338, 378. See also body size

438

Index

large marine ecosystems (LMEs), 63 larval choice of habitat, sensory cues for, 163, 163f larval dispersal, 158–59; behavioral strategies and, 161, 162, 164; of benthic invertebrates, 201; as bet-­hedging, 142; as a central theme in ecol­ogy, 138; connectivity and, 159; distances of, 158–59; as enduring prob­lem in ecol­ogy, 142–44, 169; habitat complementarity and, 336; hydrodynamic simulation of, 161; methods for studying, 159; oceanography integrated with research on, 169; from rocky shore communities, 305–6; se­lection for high fecundity and, 142–43 larval populations on rocky shores, 305–6, 307; energy of ­water movement and, 308, 309f latitudinal bands, 44 latitudinal diversity gradient, 47, 47f; debate about ­causes of, 72; diversification rate hypothesis and, 53–54; in fossil marine bivalves, 60; habitat area and, 50, 52, 211; interaction of temperature and area in, 211; speciation and, 49, 50, 52, 53, 54; summary of ­factors driving, 73 latitudinal trends: in life history related to temperature, 64, 66f; in plant defenses, 178; in species interactions, 64, 66–68, 67f lecithotrophy, 140 Lefkovitch matrix, 148, 149f, 152t Leibig’s law of the minimum, 226 Leopold, Aldo, 378 Leslie, Patrick H., 147 Leslie matrix, 145f, 147–48 lettuce coral (Agaricia), 151, 370 life histories, 139–42; of clonal invertebrates and plants, 148; conservation and, 151–55, 168; of coral species, 353–54, 353f; in deep sea, 277; evolutionary change and, 155; fish using multiple habitats during, 336; loop diagrams of, 148, 149f; marine protected areas and, 154; metabolic scaling and, 165–66; phenological mismatch and, 192; trade-­offs in, 140–41, 141f. See also phenological mismatch; phenology life history traits, 127–28 life history transitions: elasticity of population growth rate and, 148, 152t; Lefkovitch matrix and, 148 life ­tables, 145, 145f, 146–47, 147f; modeling with ­limited data from, 152, 154f lifetime reproductive output, 145–46; of industrialized ­humans, 168 light, and photosynthesis, 225–26 light reactions, 118 ­limited access privilege programs (LAPPs), 100–101 limiting similarity, 187–89, 188f The Limits to Growth, 80–82, 81f, 86 Lindeman, Raymond, 219, 220f, 222, 223, 330 lionfish, 104 Littorina littorea, 310 local richness, 199 local saturation vs. regional enrichment, 209–10, 209f loggerhead turtles, fishing impacts on, 151–52, 153 logistic population growth, 144, 145 Longhurst, Alan, 61–62, 63, 97, 100, 265, 272 longitudinal diversity gradient, 48, 60 long-­spined sea urchin. See Diadema antillarum loop diagrams, 148, 149f Loricifera, 11, 12f, 33 lottery hypothesis, 202, 205, 359 Lovelock, James, 2 Lubchenco, Jane, 175 Lyell, Charles, 350

MacArthur, Robert, 57, 64, 197, 245, 280, 358, 359 macroalgae, 15f, 30, 32, 33f, 34f; blooms of, 257, 258; coral larvae oriented to specific species of, 163; on coral reefs, 355, 360; defenses against microbial pathogens of, 179; detritus-­based food web and, 235; dominating reefs with high river runoff, 352; functional groups of, 133, 134f; on healthy vs. degraded coral reefs, 163, 163f, 363, 365–66, 365f, 366f; herbivore intimidation by predators and, 182; herbivore pressure on artificial structures and, 105; herbivorous fishes feeding on, 179, 215, 360; high-­latitude forests of, 300; life histories of, 139, 139f; on rocky shores, 301; sponges in mutually beneficial relationship with, 354. See also algae; ephemeral algae; kelps; perennial algae; seaweeds macroecological scaling, of community interactions, 190, 190f macroecol­ogy, 133–35; of open ocean, 284–88, 288f; of populations, 165–68; of trophic interactions, 240, 241f macroepifauna, 34 macroinfauna, 33–34 macrophytes: benthic, 30–32, 33f, 34f, 133; of coastal ecosystems, 296, 300–302, 301t, 303; as foundation species, 301; lake-­water sediment and, 251. See also flowering plants; macroalgae; seaweeds macroplankton, 22, 22t, 28 Magnuson-­Stevens Fishery Conservation and Management Act, 99, 100, 257 management: large strongly-­interacting species and, 191; marine ecoregions classification and, 62–63; population ge­ne­tic markers and, 164. See also fisheries management manatees, 302, 319, 321, 355, 373 mangroves, 331–35; in the Anthropocene, 334–35; basic features of, 331–32; cleared for aquaculture, 340; as coastal primary producers, 296; community organ­ization and key interactions in, 333; conservation of, 231; detritus-­based food web and, 235; ecosystem pro­cesses in, 334; ecosystem ser­vices provided by, 334; estuarine habitats dominated by, 299, 300; expanding into salt marshes, 335; fish production associated with, 334, 336; as foundation species, 102–3; geomorphology and environment of, 332; global distribution of, 317f; historical focus on bottom-up control in, 333; latitudinal distributions compared to salt marsh habitat, 325, 331; lower nutritional quality of, 115, 131, 231; near urban centers, 337; organisms and traits of, 332–33; physiological adaptations of, 332; plant species in, 331, 331f; prop roots of, 332, 333; in shallow, soft sediment environments, 300, 300f, 301; yield in fisheries near to, 326f; zonation of, 332 mantis shrimps, reef-­dwelling, 57 Margalef, Ramon, 220, 267, 270, 272, 285 Margulis, Lynn, 2 marine ecol­ogy: contrasted with biological oceanography, 4f, 8, 9; historically local focus of, 345; origins and motivation for, 95, 117 Marine Ecoregions of the World (MEOW), 62–63, 62f marine mammals: body size and vulnerability to exploitation of, 65; extinction risk among, 59, 136; overharvesting of, 338; trophic cascades and, 274. See also sirenians; ­whales

marine protected areas, 109, 109f, 110, 377; life histories and, 154 marine snow, 235, 278, 279 marine spatial planning, 100, 106 marsh grasses, 131, 296, 299, 300, 300f, 301. See also salt marshes Martin, John, 107, 117 mass balance, 117, 118 mass balance models, 252–55, 253f, 254f; of Black Sea, 275 mass extinctions: defined, 101; ecological opportunity and, 55–56, 56f; during most of earth history, 58–59; not yet happening, 379–80; sixth, 59, 72, 101 matrix population models, 145f, 147–49, 149f, 152t; evolution and, 155; habitat protection and, 154; in sea turtle conservation, 153, 153f maximum sustainable yield (MSY), 97, 97f, 98, 99, 100, 149, 151 May, Robert, 271 Mayr, Ernst, 49 mean temperature of the catch (MTC), 70 megafauna, threatened, 108, 109f meiofauna, 30, 32–33, 32f, 35f, 312, 313, 314 Menge, Bruce, 307, 308f menhaden spawning, 162, 162f MEOW (Marine Ecoregions of the World), 62–63, 62f meroplankton, 139 mesograzers, 179, 181f, 186; mutualistic, 319, 320f; in seagrass meadows, 319, 320f, 321 mesopelagic zone, 263f, 264 mesoplankton, 22, 22t, 28 mesopredators: on coral reefs, 362; trophic skew causing eruption of, 338 metabolic balance of open ocean, 287, 288f metabolic ecol­ogy, 117, 124; balance between photosynthesis and respiration and, 287; climate warming and, 135; dif­fer­ent components of metabolism and, 166; macroecol­ogy and, 134; species diversity and, 210–11; temperature effects on communities and, 216; trophic level and life history in, 240 metabolic rates: of deep-­sea animals, 287; higher in warm places and times, 121; low in deep sea, 277; speciation and, 53–54; specific production (P/B) and, 151 metabolic scaling: of abundance with body mass, 166–67, 237, 238f, 240, 241f; with body mass and temperature, 121–24, 122f, 123f; deep-­sea ecosystems and, 287; herbivore effects on plant biomass and, 194; life history syndromes and, 165–66; of metabolic components with temperature, 166 metabolic scope, 120–21 metabolic theory of ecol­ogy. See metabolic ecol­ogy metabolism, 117; aerobic, 1; of heterotrophs, 118, 120, 120f; of plants vs. herbivores, 125, 125f metacommunities, 201, 345; of coexisting reef fish species, 202; neutral theory of biodiversity and, 204 metagenomics, 16–17, 16f, 19; parasitic plankton and, 26–27, 184; phytoplankton and, 25; of species interactions, 17, 17f metapopulations, 138; conceptual framework for, 159, 159f; ge­ne­tic markers and, 164; geochemical tracers and, 164 metazoan heterotrophs. See multicellular heterotrophs meteorite strike, end-­Cretaceous, 58–59

Index

methane clathrates, 284 methane ice worm, 281f, 284 methane seeps, 281f, 283–84 microalgae: of algal turfs, 376; detritus from, 236; high rate of grazing on, 131, 230–31; nutrient content of, 259; of sediment bottoms, 311, 315. See also phytoplankton microbes: in algal turfs of reefs, 355; aquaculture driving pathogenicity in, 104–5; as detritivores, 236; diversity of, 14, 15f; genomic studies of, 14, 16–17, 16f, 17f; number of species on Earth, 11; pathogenic, 184–85; sediment-­dwelling, 31f, 32; suggested defenses of macroalgae against, 179. See also Archaea; bacteria; disease in marine organisms microbial communities, replacing foundation species, 103, 338 microbial diversity, global controls on, 286–87 microbial loop, 272–73, 272f; defined, 272; phytoplankton size structure and, 267; picoplanktonic eukaryotes in, 22; picoplanktonic production passing through, 132, 269 microbial mats: foundation species yielding dominance to, 103; on sediments, 311 microbiome of ocean, and horizontal gene transfer, 19–20 microplankton, 22, 22t microzooplankton: grazing by, 268, 269, 272–73, 293; ignorance about basic physiology of, 293 mid-­ocean ridges, 282, 282f midwater fishes, 264; biological pump and, 28, 273; Hawaiian monk seal preying on, 274 migration of large marine vertebrates, 159–60, 160f Mismanagement of Marine Fisheries (Longhurst), 100 mitochondrion, endosymbiotic origin of, 56 mixed layer, 263; four main biomes and, 61–62 mixotrophs, 23, 24f modules, sets of species as, 173, 173f mollusks: deep-­sea diversity of, 286; expansion through Arctic into Atlantic, 71; mesograzing, 179; ocean acidification and, 135; range size and speciation in, 52, 53f; vulnerable to ­human disturbance, 103. See also bivalves, marine monophyletic clade, 18f, 20 monsoon, 61 mortality rate, scaling with body mass and temperature, 166 MSX, 339 multicellular heterotrophs, 23, 24f, 28 mussels in salt marshes, 328, 329 mussels on intertidal shores, 189, 198, 304, 307, 308; invasive Mytilus galloprovincialis in Africa, 310; predation on (see Pisaster ochraceous) mutation rates: solar energy and, 54; temperature and, 53, 53f mutualistic mesograzers, 319 NADH, 118 nanoflagellates, 267; controlling bacteria in ­water column, 272 nanoplankton, 22, 22t natu­ral history: fundamental importance to ecol­ogy of, 7, 83, 359, 378; generalities that transcend the details of, 7–8, 187, 214; of h­ umans, 83–84, 86, 106, 378 natu­ral se­lection, 8, 126–27, 155–58; at dif­fer­ent life stages, 151; directional se­lection, 156–57, 157f; ecosystem pro­cesses and, 224; stabilizing se­lection, 155–56, 156f. See also evolution

nekton, 28; trophic diversity in Gulf of Mexico and, 213 nematodes: of deep-­sea benthos, 287; meiofaunal, 33, 35f, 313 net plankton, 22, 272 net primary production (NPP): defined, 118; of dif­fer­ent plant communities, 301t; herbivore biomass correlated with, 230, 230f; ­human appropriation of, 79; low on coral reefs, 368; magnitudes and fates of, 116f; by marine phytoplankton, 225. See also primary production network-­based ecosystem models, 186 networks, ecological, 186–91; loss of strongly-­ interacting species in, 189–91; nontrophic interactions in, 189; in rocky intertidal zone, 189; topologies of, 171, 172f, 186, 189. See also food webs; interaction strengths neutral theory, 203–6; deterministic interactions and, 205, 206; ecological drift and, 203; habitat area and, 211; lottery hypothesis in, 202, 205, 359. See also ecological drift; neutral theory of biodiversity and biogeography neutral theory of biodiversity and biogeography, 58, 203, 204–6; island theory and, 57. See also neutral theory new production, 116, 226, 228 niche: concepts of, 128–29; defined, 130, 199, 199f; dimensions of, 128–30, 130f, 176 niche construction by ­humans, 85 niche differentiation: coexistence of competing species and, 174–75, 176, 198; ecological se­lection and, 200; in phytoplankton, 130, 130f, 198; vulnerability to selective grazing and, 270. See also specialization niche partitioning, 357; efficient pro­cessing of ecosystem resources and, 242; by photopigment diversity, 119; of reef fishes, 358–59, 358f. See also complementarity; resource partitioning nitrate: new production fueled by, 226, 228; reduced in nitrogen cycle, 117; in seawater, 115–16, 226 nitrite: requirement for reduction before biological assimilation, 116; in seawater, 115–16, 226 nitrogen: in animals vs. plants, 115; anthropogenic flux into coastal ­waters, 80f, 257, 309–10, 322, 323f, 324, 324f, 336, 340, 341; atmospheric, from fossil fuel combustion, 79, 80f; availability and fluxes of, 114, 116; biological pump and, 273, 274; growth rate ­limited by, 120; intentional fertilization of sea with, 341; limiting primary producer biomass, 226, 226f; in protein, 114, 115; seawater forms of, 115–16, 226; sediment organisms’ pro­cessing of, 314. See also Haber-­ Bosch pro­cess; nutrient loading nitrogen cycle: changed by ­humans, 378; iron used by microbes in, 117. See also nitrogen nitrogen fixation: by cyanobacteria, 267, 355; industrial, 79–80, 80f; iron limitation of, 117 no-­analog ecosystems, 105. See also novel ecosystems nonallopatric speciation, 50, 51 nonequilibrium pro­cesses: diversity and, 207; in ecosystems, 223–24; historical ­factors and, 208; paradox of plankton diversity and, 270–71 nonlinear environmental forcing: fish populations and, 100; regime shifts and, 249 nonlinear population dynamics, 271 nonlinear responses to stressors: by crown-­of-­thorns starfish, 365; fish populations and, 99, 108 nonlinear systems, 377

439

nonnative species: favored by artificial structures, 105, 344; as fouling invertebrates in harbors, 343; intentionally introduced, 339, 340f; novel ecosystems dominated by, 105; se­lection pressures changed by, 344; Willapa Bay reor­ga­ni­za­tion by, 339, 340f. See also invasions of nonnative species North Atlantic: cod fisheries of, 64, 65, 252; impoverished in coastal marine species, 40; low-­diversity regime shifts in, 252 novel ecosystems, 105; climate change and, 257; species differences in dispersal and, 164; in urbanized environments, 344 NPP. See net primary production (NPP) null models, 203–4. See also neutral theory NutNet group, 345 nutricline, 61 nutrient availability in the ocean, 225, 226, 226f; phytoplankton community composition and, 267, 267f. See also eutrophication; nitrogen; oligotrophic systems nutrient loading: algal blooms generated by, 258; algal dominance of coral reefs and, 365f; collapse of San Francisco Bay fish populations due to, 105; declining coastal ecosystems and, 346; economic value in reduction of, 341; managed to maintain foundation species, 343; phytoplankton blooms generated by, 316; rapid responses to, 126; seagrass tolerance for, 317; synergizing with other stressors, 343; temperature interacting with, 125. See also eutrophication; nitrogen nutritional quality of primary producers, 115, 115f, 116f, 131, 231 Obama, Barack, 379 ocean acidification, 92–95, 92f, 135–36; coral reefs and, 372; kelp forests and, 338; predicted effects of, 290, 292; seagrasses benefiting from, 319 ocean con­vey­or ­belt, 43, 43f ocean currents: circumpolar, 39, 63; wind and, 40, 41, 297 oceanography: biological, 8–9; history of, 260–61; physical, and population models, 169, 170 oceans, history of continents and, 38–40, 39f ocean warming, 78, 79f, 89–91; acidification enhanced by, 92; algae-­herbivore interactions and, 125; coral reefs and, 371, 372f; hotspots of, 70; physiological effects of, 90, 90f; predicted effects of, 290–92; seagrass ecosystems and, 321; species differences in dispersal and, 164; species interactions and, 191–92. See also climate warming; sea level rise ochre sea star. See Pisaster ochraceous Odum, Eugene P., 219, 220, 221t, 222, 253, 330 Odum, Howard T., 219, 222, 223, 330 oligotrophic systems: biological pump and, 237; carbon balance in, 274; of central ocean gyres, 228, 261, 276; coastal macrophytes in, 317, 332; of coral reefs, 218, 222, 228, 351, 352, 368; perennial foundation species in, 110; picoplankton in, 14, 15f, 64, 132, 237, 267; supported by regenerated nitrogen, 228, 267. See also deserts open ocean: in Anthropocene, 289–93; controls on biodiversity in, 284–87, 285f; macroecol­ogy of ecosystem pro­cesses in, 287–88, 288f. See also pelagic ocean optimism: vs. pragmatism, 381; reasons for, 108–10, 109f orange roughy (Hoplostethus atlanticus), 289, 290 organic enrichment of benthos, 315–16, 316f

440

Index

Ostreococcus tauri, 13 Ostrom, Elinor, 87 otoliths, geochemical signatures of, 164–65, 165f “out of the tropics” hypothesis, 60 overfishing, 97, 98, 106; algal blooms generated by, 258; of coastal fisheries, 340; coral reef conservation and, 374, 375; coral reef disease and, 364; of coral reef fishes, 360–61, 365, 368, 369f, 373, 375; destabilizing ecosystems, 248; Jamaican coral reefs and, 363, 365; of large fishes, 338; life history and vulnerability to, 168; marine food webs altered by, 258; regime shifts triggered by, 251, 274; trophic cascades driven by, 274–76 overfishing debt, 98 oxycalorific coefficient, 120 oxygen: declining in warmer ­water, 215, 291, 341; depleted in polluted sediments, 316, 316f; minimum in upper mesopelagic, 264, 291; predicted seawater decrease by 2100, 290; released by photosynthesis, 118 oxygen consumption, 120; in sediments, 314; temperature and, 121 oyster reefs, 300f, 314–15; estuarine habitats dominated by, 299, 338; nearly wiped out, 338, 339; predatory crabs on, 336; restoration proj­ects for, 109f, 345 oysters: aquaculture of, 339; as foundation species, 301; ­human exploitation of, 314, 315f; intentionally introduced, 339, 340f; introduced along with disease-­causing organisms, 339 Pacific Ocean: active continental margins of, 297; separated from Atlantic 2.8 Mya, 39; species richness in, 52 Paine, Robert, 175, 304 paleontology, 197, 209; stasis observed in, 156 Pangaea, 38, 39f paradox of hyperdiverse tropical forests, 204 paradox of the plankton, 130, 198, 207; proposed solutions to, 269–71; reef fish diversity and, 359 parallel communities, 48 parapatric speciation, 50 paraphyletic group, 18f, 20 parasites, 184–85; heterotrophic protists as, 26–27; in microbial loop, 272; scaling of abundance with body size of, 240, 241f; selectively grazing on phytoplankton, 270 parrotfishes (Scaridae), 179, 355, 356f, 359, 360, 373 parsimony, in phyloge­ne­tic reconstruction, 21 particulate organic carbon (POC): declining flux with ocean warming, 291; from mesopelagic microbes and zooplankton, 273, 273f; seasonality of flux to deep seabed, 278 passive continental margins, 297; salt marshes on, 327; sediment plains of, 310 path de­pen­dency, of ecosystems, 224, 249 pathogens, as type of predator, 184. See also disease in marine organisms pelagic armourhead (Pseudopentaceros wheeleri), 289 pelagic-­benthic coupling: deep-­sea, 278–79, 279f; sediment in shallow ­water and, 315 pelagic communities, 269–71 pelagic ecosystems: biological oceanography of, 8–9; defined, 21; functioning of, 272–76; physical forcing of, 261–66; representative organisms of, 23f; top-­down control in, 274–76

pelagic food webs, 272–73, 272f pelagic functional diversity, 21–28; body size and, 22–24, 24f; nutrition and metabolism, 23, 24f, 25–28, 27t pelagic ocean: biogeographic provinces of, 61, 61f, 63; four main biomes of, 61–62. See also open ocean perennial algae, 133, 258, 302, 307, 308, 310 periwinkle (Littorina saxatilis), 51, 51f Perkinsus spp., 339 Petersen, C. G. J., 186 Phaeocystis: in blooms, 266; climate warming and, 26; defensive colony formation of, 268; P. globosa, chemical sense of, 179 phages, 24–25 phase shifts: in complex adaptive systems, 249; between coral-­and algal-­dominated reefs, 365, 365f, 366f, 373; mediated by herbivores, 70. See also alternative stable states; regime shifts phenological mismatch: climate warming and, 90–91; species interactions and, 192 phenology: climate warming and, 68, 69, 292; defined, 68 phenotype, 126, 127, 127f phenotypic plasticity, 128 phosphorus, 114, 115, 119; intentional fertilization of sea with, 341; ­limited availability in the ocean, 226, 226f photosynthesis, 118; on coral reefs, 367–68; fossil fuel built by, 86; herbivore feeding on plankton and, 125, 125f; iron required for, 117; origin of, 1; pigments used in, 25–26, 118, 119, 119f; primary production by, 118, 225–26; by zooxanthellae, 351–52, 371 photosynthetically active radiation (PAR), 118, 119, 211, 225 phototrophs, unicellular, 22, 23, 24f, 25–26, 27t. See also phytoplankton phyla: of marine animals, 12–13, 13t, 14f; of marine autotrophs, 13; of recently discovered invertebrates, 11, 12f; of terrestrial animals, 12, 13t phyloge­ne­tic classification, 14, 18–21 phyloge­ne­tic conservatism of traits, 18; in benthic macrophytes, 32, 34f, 133; community structure and, 214; in functional form models, 133; of marine primary producers, 25; of suites of interrelated functional traits, 131 phyloge­ne­tic reconstruction, 18, 20–21 phyloge­ne­tic structure of communities, 214 phyloge­ne­tic systematics, 18 phyloge­ne­tic trees, 18f, 20 phylogeny, 18 physical oceanography, and population models, 169, 170 phytoplankton, 25–26; bacterial associations with, 25; biogeographic patterns in functional traits of, 64; carbon fixed by, 273; cell size as fundamental trait of, 132–33, 132f, 267; in coastal regions, 300–301; compared to land plants, 230; diverse taxa of, 267, 267f; diversity of, 285, 285f; diversity of photosynthetic pigments in, 118, 119, 119f; ecosystem model based on, 255, 256f; equatorial upwelling and, 42; eukaryotic, 25–26; evolving in response to climate change, 158; functional groups of, 131–33, 132f; functional types of, 266–68, 268f; global and regional patterns of, 9, 9f, 261–62, 294; grazing on, 268, 269; herbivore interactions with, 125–26, 125f; Hutchinson’s

paradox of, 198 (see also paradox of the plankton); induced defenses of, 179; iron requirements of, 117, 266; large-­scale patterns of biomass in, 226; net primary production by, 225; niche differentiation in, 130, 130f; nitrogen assimilated by, 116; nutrient limitation in the world ocean and, 226, 226f; ocean acidification and, 136; ocean warming and biomass of, 291; resource use efficiency and biodiversity of, 244; retained in epipelagic zone, 263; in sediment bottom food webs, 311, 312; species composition of, 237, 285; traits of higher plants compared to, 5; variance in nutrient requirements of, 237; viral influence on dynamics of, 25. See also cell size of phytoplankton; plankton picoplankton, 22, 22t, 25, 267. See also Prochlorococcus pigments, photosynthetic, 25–26, 118, 119, 119f Pinker, Steven, 379 Pisaster ochraceous (ochre sea star): as keystone predator, 12, 14f, 68, 91, 173, 175, 304, 306–7; temperature-­mediated effect on predation by, 191; viral infection of, 307 planetary atmospheres, 2t plankton: consumer size scaling with prey size in, 182, 183f; four biomes affecting productivity of, 61–62; less genet­ically structured than benthic organisms, 55; oceanographic research on communities of, 8–9; size classes of, 22, 22t. See also phytoplankton; zooplankton planktonic larvae, warming-­related life history changes in, 192 planktotrophic larvae, less common in cold regions, 64, 66f planktotrophy, 140–42 plant communities, biomass and NPP of, 301t plant-­herbivore interactions, 176–79; climate change and, 91; phyloge­ne­tically conserved traits and, 214. See also chemical defenses plants: in ecosystem energy pathways, 113f; global database of terrestrial species, 136–37; referring to all primary producers, 225; stoichiometry of, 115, 115f. See also flowering plants plastic pollution, 289, 364 plastids, 25, 26 pneumatophores, 331f, 332 POC. See particulate organic carbon (POC) poikilothermic vertebrates, interaction strengths of, 193 polar biome, 62 polar regions, higher productivity in, 261 politics, 84–85, 106, 108 pollution by ­human activities, 337; declining in prosperous socie­ties, 380; plastic in, 289, 364; of sediment communities, 315–16, 316f; in urbanized estuaries, 344; ­water pollution and coastal restoration, 322, 323f polychaetes, 33–34, 35f; on sediment bottoms, 312, 312f, 314 polygenic traits, 127 polynyas, 43 polyphyletic group, 18f, 20 population ge­ne­tic structure, and connectivity, 162, 164 population growth, ­human, 75, 76f; driven by technological innovation, 83, 84; impact on earth system and, 84; industrial nitrogen fertilization and, 79–80; trend since 1972 predictions, 81f, 82, 379 population growth, quantitative theory of, 144–49

Index

population growth rate: discrete (λ), 146, 148, 152t; elasticity of, 148, 152, 152t, 154f; intrinsic (r), 146, 153f, 165–66 populations: in the Anthropocene, 168; defined, 138; macroecol­ogy of, 165–68; smaller at higher temperatures, 215 portfolio effect, 246, 247 pragmatism, 381 predation: artificial structures and, 344; climate warming and, 191–92; diversity of communities affected by, 175, 185; by keystone species Pisaster ochraceous, 12, 14f, 68, 91, 175; larval dispersal as protection from, 142–43; mass spawning of coral species and, 143–44; mesograzers protected from, 179, 181f, 186; more intense at low latitudes, 66–68, 67f; ocean acidification and, 94, 136; phenological mismatch and, 192; in sediment communities, 313; in vent communities, 283 predator-­prey systems, regime shifts in, 251–52 predators: declining on coral reefs, 356; diversity affecting trophic cascade and, 242, 244f; generalist or specialist, 184; nonconsumptive effects of, 182–84; ocean acidification and, 372; pathogens as, 184; in salt marshes, 329, 330; in seagrass meadows, 319, 320f; traits of, 184; trophic skew caused by decline of, 193. See also top-­down control press perturbations, 365, 376 prey: diversity of, and predator impact, 246, 248f, 308; traits of, 182–84, 183f primary producer biomass: of dif­fer­ent plant communities, 301, 301t; global distribution of, 294; satellite imagery of, 9, 9f, 98, 261–62; virus-­induced reduction of, 25. See also primary productivity primary producers: of coastal ocean, 296; consumers’ profound effect on, 224; on coral reefs, 222, 354–55, 354f; diversity of, 13, 15f; extinctions during most of earth history and, 58–59; major types in aquatic systems, 27t; marine vs. terrestrial, 13, 15f; nutrient influence on competition and dominance in, 194; nutritional quality of dif­fer­ent types of, 115, 115f, 116f, 131, 231; pelagic vs. benthic, 5; stoichiometry of, 115, 115f; as vascular plants and algae, 225. See also algae; flowering plants; macrophytes primary production, 118, 225–26; acidification experiments and, 93; appropriated by marine fisheries, 98; in coastal and estuarine ecosystems, 300–301, 302–3, 303f; on coral reefs, 222; fates for dif­fer­ent types of producers, 116f; ­limited by inorganic nutrients, 225, 226, 226f; ­limited by trace nutrients, 117; of mangrove trees, 334; supported in seagrass habitats, 321. See also gross primary production; net primary production (NPP) primary productivity: defined, 118; in four main biomes, 61–62; in salt marshes, 326, 330. See also primary producer biomass; productivity Prochlorococcus, 14, 15f, 22, 25, 132, 237, 267; ecotypes of, 256f, 270; model reproducing distributions of, 255, 256f productivity: of coastal and estuarine fisheries, 302, 303f, 304, 326, 326f; of coastal vs. oceanic ­waters, 301; control of biomass distribution and, 228–36; of coral reef ecosystems, 367; decreasing in warmer ocean, 290, 291; in estuaries, 299; global distribution of, 261–62; in harvested fish

populations, 157. See also primary productivity; secondary (animal) production protein: metabolism of, 120; nitrogen in, 114, 115 protists: heterotrophic, 26–27; multicellular, 30; phyloge­ne­tic relationships and, 20; size classes of, 22t; traditional kingdom of, 20 Protozoa, as paraphyletic group, 20 pycnocline, 42, 43, 262, 263f; at coastal regions, 44 pyramid of numbers, 238, 239f quantum yield of photosynthesis, 118 rabbitfishes (Siganidae), 355, 359, 360 range size: body size and, 167, 167f; climate-­mediated expansion of, 103; extinction risk and, 59; macroecol­ogy of, 167–68; speciation and, 52 Rapoport’s rule, 287 reaction norm, 128 realized niche, 129, 130, 199, 199f recalcitrant carbon reservoir, 273 red algae, 26 Redfield, Alfred, 115 red-­listing of ecosystems, 257 red seaweed Hypnea, facilitated by Sargassum, 185–86, 186f red tides, 266, 267f reef communities: ancient earth history of, 348; nutrient inputs shifting dominant organisms of, 352; of other than stony corals, 354. See also coral reefs; oyster reefs; temperate reefs reef fishes: biomass increasing with species diversity of, 245f; coexistence of ecologically equivalent species, 202, 205; coral gobies, 51, 51f; dispersal mode affecting ge­ne­tic structure in, 54–55, 55f, 161; diversity of, 358–59, 358f, 366–67, 369; East Indies as center of origin for, 60; ecological se­lection in, 359; feeding in seagrasses or mangroves, 335–36, 367; fisheries of, 253, 368–69, 369f, 373; geographic range size of, 167–68; geological history affecting distributions of, 57; grazing on algal turfs, 355; herbivorous, 179, 248, 355–56, 356f, 365, 368–69, 373, 375; hydrodynamic simulations of larval dispersal in, 161; larval sensory biology and, 163, 163f; microhabitats of, 359; proximity to nursery habitat of, 336; slow life histories of, 368; topographic complexity and, 351. See also fishes, herbivorous Reef Life Survey program, 345 regenerated production, 116, 226, 228 regime shifts, 249; empirical evidence for, 250–51, 250f, 380; mechanisms of, 251–52; trophic cascades and, 274, 275, 276, 293. See also alternative stable states; phase shifts regional enrichment, 209–10, 209f, 212–13, 212f regional richness, 199, 209–10 regional species pool, 198–99, 200, 201; species density of coral reefs and, 357 Remane, Adolf, 299 reproductive value, 146, 148 rescue effect, 201 resource availability, and biodiversity, 284–86, 285f resource partitioning: deep-­sea biodiversity and, 280; by phytoplankton, 270. See also niche partitioning respiration, 120, 120f; on coral reefs, 367–68; herbivore interactions with plankton and, 125, 125f; pelagic profile of, 264 respiration rates, rising with temperature, 214–15 response traits, 21, 215–16

441

Riftia pachyptila, 281f, 283 Riley, Gordon, 265 risk assessment of ecosystems, 257 river deltas, 298. See also estuaries rivers, high productivity around mouths of, 261 rockfish. See Sebastes, limiting similarity in rock wall invertebrate communities, regional enrichment of, 212f, 213 rockweed (Fucus), 51, 51f; as invasive species, 310; trophic cascades on shores dominated by, 302, 325 rocky intertidal communities, 304–10; in the Anthropocene, 309–10; community organ­ization of, 2, 9, 306–8, 308f, 309f; competitive exclusion in, 174–75, 174f, 176; deterministic control of, 206; differing on opposite sides of Atlantic, 197; facilitation in, 307; interaction networks in, 189; interaction strengths in, 190, 190f; neutral theory tested in, 204–6; organisms of, 305–6; per­sis­tent coexistence of Chilean barnacles in, 205; pro­cesses affecting structure of, 198; species interactions in, 173, 174–75, 174f; ­water flow and energy in, 308, 309f; zonation of, 195, 196f, 304. See also rocky shores rocky shores: as dominant coastal habitat in many places, 309; dominant macrophytes of, 181, 301; geomorphology and environment of, 305, 306f; nonconsumptive effects of predators in communities on, 183; surfgrasses on, 317; wave energy on, 305, 306f, 309. See also rocky intertidal communities; rocky subtidal habitats rocky subtidal habitats, 304; alternative ecosystem states on, 181; environmental ­drivers on, 305; kelps and coralline algal crusts on, 181; productive macroalgae on, 309. See also rocky shores Romer, Paul, 381 Ross Sea protected area, 110 salmon: diversity in Alaskan fishery of, 247, 247f; farmed, 104, 105, 158; ocean fertilization experiment and, 107 salps, 22, 23f; biological pump and, 237, 274; as grazers on picoplankton, 268 salt marshes, 325–31; in the Anthropocene, 330–31; basic features of, 325–26; community organ­ ization in, 328–29; cordgrass in trophic cascade of, 234f; detritus-­based food web and, 235; drought stress leading to herbivore outbreak in, 302; ecosystem ser­vices of, 330; expanded by Eu­ro­pean colonization of Amer­i­cas, 330; geomorphology and environment of, 326–27, 327f; global distribution of, 317f; mangroves expanding into, 335; metabolic scaling of grazing snails in, 194; nontrophic interactions in, 189; organisms and traits in, 327–28; overgrazing caused by fishing and, 338; plants growing better ­under salt stress and, 192; plant species characteristic of, 325; restoration of, 345; species interactions in, 328, 329, 330; Teal’s early ecosystem study of, 222–23; top-­down vs. bottom up control in, 328–29, 330; trophic cascades in, 329; zonation of, 327, 327f, 328. See also marsh grasses sampling phenomenon: diverse prey community and, 246; productivity of diverse species and, 242 Sanders, Howard, 279–80 San Francisco Bay estuary: Asian clam invasion in, 342–43, 342f; as most invaded estuary, 339; as novel ecosystem, 105

442

Index

SAR. See species-­area relation Sargassum, 23f; facilitation of seaweed Hypnea via associational defense with, 185–86, 186f; temperature effect on herbivory in, 125 satellite imaging: of global primary producer biomass, 9, 9f, 98, 261–62; oceanographic data derived from, 61; of phytoplankton biomass in warmer ocean, 291 scaling relationships. See metabolic scaling Scotian Shelf, regime shift in, 276 sea bass. See Sebastes, limiting similarity in seagrasses, 15f, 30, 31–32, 36; biological traits of, 318–19; conservation of, 231; detritus-­based food web and, 235; epiphytic algae on, 319, 324; estuarine habitats dominated by, 299, 300; as foundation species, 102–3; ge­ne­tic and species diversity of, 325; global distribution of, 317, 317f; low nutritional quality of, 115, 131, 231; regime shifts involving sediment conditions and, 251; of sediment shores, 301; spatially escaping from herbivores, 177; submerged, as coastal primary producers, 296. See also eelgrass (Zostera marina); eelgrass, Japa­nese (Zostera japonica) seagrass meadows, 300f, 317–25; African, nontrophic interactions in, 189; ancient reef-­like escarpments of, 318, 318f; in the Anthropocene, 322–23, 325; community organ­ization in, 319–21, 320f; diverse herbivores supported by, 319; ecosystem pro­cesses and ser­v ices in, 321–22; epiphytic algae with productivity superior to, 319; geomorphology and environment of, 317–18; large vertebrates historically grazing on, 373; overgrazing caused by fishing and, 338; protected from nitrogen loading by marshes, 336; providing corridors for predatory crabs, 336; reducing disease prevalence in corals and ­humans, 336; restoration of, 323, 323f, 325; in shallow soft sediment environments, 300; supporting diversity and productivity, 317 seagrass wasting disease, 175, 321, 322–23, 341 sea level rise, 78, 79f, 91–92; coastal impact of, 337–38; coastal vegetation protecting against, 303; mangroves threatened by, 335; marshes threatened by, 330 sea otters, 179, 180f, 181, 232, 249, 324 sea slugs: as mesograzers on chemically defended algae, 181f, 186; Phyllaplysia smaragda, 101 sea squirts, favored in biotic homogenization, 104 sea star wasting syndrome, 307, 341 sea surface temperature: geographic distribution of anomalies in, 80f; rise in, 78, 79f sea turtles, 89, 102, 102f; Ca­rib­bean decline in, 323, 373; as coastal herbivores, 302, 319, 321; conservation of, 151–52, 153; grazing on coral reefs, 355; overharvesting of, 338 sea urchins, 12; on Ca­rib­bean reefs, killed by epidemic, 175; grazing on coral reefs, 355, 360, 361f; grazing on seagrass meadows, 319, 321; ocean acidification and, 135–36; in trophic cascade, 179, 180f, 181, 232, 249. See also Diadema antillarum seaweeds, 15f, 30, 32; calcium carbonate in tissues of, 178; canopy-­forming, as foundation species, 344; canopy-­forming, multiple stressors on, 343; chemically defended, mesograzers on, 181f, 186; as coastal primary producers, 296; facilitation between sparid fishes and, 185–86; Halimeda defenses against herbivores, 178; reefs dominated by, 365; on rocky shores, 304, 305. See also fleshy seaweeds; kelps; macroalgae

Sebastes, limiting similarity in, 187–89, 188f secondary metabolites, 178 secondary (animal) production: acidification experiments and, 93; in estuaries and coastal ecosystems, 299, 302–3, 303f; in salt marshes, 326, 328, 330; supported in seagrass habitats, 321. See also productivity sedimentary shores: macrophytes on, 301; non-­native species on artificial structures of, 344; wave energy on, 306f sediment bottoms, 310–17; in the Anthropocene, 315–17, 316f; community organ­ization in, 313–14; ecosystem pro­cesses and ser­vices in, 314–15; environmental ­drivers of, 310–11; ­human impacts on, 314–15; organic enrichment changing community structure of, 315–16, 316f; organisms of (see sediment-­dwelling organisms); parallel communities and, 48; pollution of, 315–16, 316f; properties of sediments in, 310–11, 311f; regime shifts of plants rooted in, 251, 252; seagrasses rooting in, 317, 321; ­water depth and, 311 sediment-­dwelling organisms, 311–13, 312f; body sizes of, 30, 32f; of deep sea, 48; microbes, 31f, 32; sediment grain size diversity and, 280; similar in distant regions, 48. See also infauna; meiofauna sediments, 30; biogeography of predators digging in, 64; classification of grain size in, 310–11; deep-­sea, 136, 244; layering of, 30, 31f; microbial mats on, 311; primary production stored in, 231; ungrazed seagrass buried in, 317; vertical geochemical zonation in, 311, 311f, 335 self-­organization: fisheries management and, 108; of ­humans to behave sustainably, 87 semistable states, 249, 251 sensitive dependence on initial conditions, 224, 271. See also chaos sensory physiology: acidification and, 93, 136; body size and, 124; larval choice of habitat and, 163, 163f. See also chemical cues Sepkoski, John, 54 sexual se­lection: stabilizing se­lection and, 155, 156f; sympatric speciation and, 50, 51 shallow regions, 44 shared derived characters, 20–21 sharks: banned fishing of, 110; on coral reefs, 356; direct development in, 140; inverted biomass pyramid for, 240; migration patterns of white sharks, 160; slow life history of, 102 shelf-­break front, 44, 297; coastal biome and, 62 shifting baseline phenomenon, 87, 89, 258, 379 shrimps: aquaculture of, 77f, 334, 340; host specialization of snapping shrimps, 357, 359; reef-­dwelling mantis shrimps, 57 silicon, 271. See also diatoms silversides (Menidia menidia), 157 sirenians, 302, 319, 321 sixth mass extinction, 59, 72, 101 size spectra, 237–40, 238f Smetacek, Victor, 268 snapping shrimps (Alpheidae), host specialization of, 357, 359 snowball earth, 1 social-­ecological systems, 106; fisheries as, 108; general approach to challenges in, 110; poorest ­people depending on Anthropocene ocean and, 293 sociality of ­humans, 84, 378 social policy, 381–82

sociocultural niche construction, 85 socioeconomic status, and energy use, 84, 85f soft corals (Alcyonaceae), 354 soft substrata. See sediments solar energy: latitudinal gradient of, 40, 41; species richness and, 54 Sommer, Ulrich, 129, 130 source-­sink dynamics for community dispersal, 201 southeast Asia, as evolutionary hot­house, 44 sparid fishes, facilitation between seaweeds and, 185–86 Spartina cordgrasses, 325, 327–28, 327f; facilitative interactions of, 329; salt marsh zonation and, 328 Spartina alterniflora, 66–67, 325, 327, 327f; in anoxic environments, 328; facilitation by, 186, 329; invaded from East to West Coast of USA, 339, 340f; salt marsh zonation and, 328; supporting snail and crab populations, 329; trophic cascade in salt marshes and, 234f Spartina patens, 328 specialization: coexistence of species and, 270; diversity of snapping shrimps and, 357, 359; extinction risk and, 59; speciation and, 55. See also niche differentiation speciation, 44–45, 45f, 49–56; allopatric, 49, 50; body size and, 54; coastal geography and, 44; community dynamics and, 209; composition of communities and, 198; dispersal ability and, 54–55, 55f, 57; ecological, 50, 51; ecological opportunity and, 55–56, 56f; ecological specialization and, 55; extinction rates and, 59; habitat age and, 52, 52f; habitat area and, 50, 52; island biogeography and, 58; latitudinal diversity gradient and, 60; in mixed patterns of biodiversity, 60; parapatric, 50; summary of ­factors leading to, 56; sympatric, 50, 51; temperature and, 53–54, 53f, 124 species, total number on Earth, 11 species-­area relation (SAR), 45–46, 46f; extinction due to habitat loss and, 59, 72; extinctions before ­human population stabilizes and, 379; latitudinal diversity gradient and, 50, 52, 211 species density. See species richness species distribution models, 69–70, 69f species-­energy hypotheses, 54, 210–11; deep-­sea animals and, 285–86 species interactions: in the Anthropocene, 191–93; on artificial structures, 105; biogeography of, 64, 66–68; climate change and, 191–92; community organ­ization and, 2, 9; between competitors, 175–76; ecosystem stability and, 248; functional traits in models of global change and, 106; general considerations about, 171–75; graphical webs of, 171, 172f; latitudinal diversity gradient and, 53; in mangroves, 333; metagenomics of, 17, 17f; between plants and herbivores, 176–79, 181f; between prey and predators, 182–85; realized niche and, 199, 199f; in rocky intertidal communities, 307; in salt marshes, 328, 329; temperature and, 121, 124; three general types of, 171; warming-­related interactions and, 91. See also interaction strengths; networks, ecological species pool, 38, 49 species richness: body size and, 210; centers of origin and, 60; climate change affecting, 103; on coral reefs, 357; defined, 44; dispersal and, 209–10, 209f; equilibrium of ­factors determining, 58; estuarine, 299, 299f; foundation species in community and, 189; functional diversity and,

Index

241–42; habitat area and, 211; of plankton, 285; of rocky shore prey communities, 308; species-­ energy hypothesis and, 54; temperature and, 54, 210; temperature-­related trends in life history and, 64, 66f; theory of island biogeography and, 57–58, 57f; in unified neutral theory of biodiversity, 204. See also biodiversity; diversity in communities specific production (P/B), 151 sperm ­whale, 264 spinner dolphin (Stenella longirostris), 51, 51f spiny lobsters (Panulirus argus): larval dispersal as protection for, 142–43, 143f; seagrass as larval settlement habitat for, 336 sponges (Porifera), 12–13, 14f; on degraded Ca­rib­bean reefs, 354; on overfished Ca­rib­bean reefs, 360–61 spreading centers, 282 spring bloom, 265–66, 278, 279 squids, 28 squirrelfishes (Sargocentron), 359 stabilizing mechanisms, and coexistence, 176 stabilizing se­lection, 155–56, 156f stable age distribution, 148 stage-­structured populations, 148–49, 149f, 152t; choice of habitats to protect and, 154; vulnerability of populations and, 168 staghorn coral (Acropora cervicornis): beginning to revive, 381; threatened status of, 102; white-­band disease of, 185, 362–63, 370, 375 starfish. See brittle stars; crown-­of-­thorns starfish; Pisaster ochraceous (ochre sea star); sun star Stebbins, G. L ­ edyard, 49 Steller’s sea cow, 101, 302, 309 stenothermal species, 121 Stern, Nicholas, 86 stoichiometry, 113–15, 114f; defined, 114; ecological, 115, 115f, 116f; ecosystem pro­cesses and, 237; food choice influenced by, 182; of phytoplankton in warming ocean, 292 Stommel, Henry, 261, 262f storage, 116f, 118; in sediments of coastal systems, 131, 303, 330, 334 stressful environments: facilitation and, 192, 302, 307, 308f, 329, 333; low diversity in, 307; ­water flow and, 308 stressors, multiple: coastal, 341, 343; on coral reefs, 363, 364, 370–71, 375 stromatolites, 311, 348, 348f substrata, 30; epifauna on, 34. See also sediments succession of a community, 196–97, 219–20 sun star (Heliaster kubiniji), 307 surfgrasses (Phyllospadix spp.), 317 surgeonfishes (Acanthuridae), 355, 356f, 359 survivorship, 145, 147 suspension-­feeders, 312, 312f, 313, 314, 315; increased by nutrient enrichment, 341, 352; macroinfauna as, 33–34; in salt marshes, 328 sustainable systems of resource use, 87 Sutherland, John, 307, 308f Sverdrup, Harald, 265–66 sympatric speciation, 50, 51 synapomorphies, 20–21 Syndiniales, 27 syngameons, 21 systems approach in ecol­ogy, 222, 223 tagging of marine animals, 159–60, 160f Tansley, Arthur, 219 Teal, John, 222–23

technological innovation, 75, 76, 76f; fossil fuel use and, 78; Haber-­Bosch pro­cess and, 79; ­human population growth and, 83, 84 tectonic events: long-­term legacies of, 209. See also volcanism tectonic plates: hydrothermal vents and, 282; Indo-­Pacific coral species and, 39–40, 57 temperate reefs: ocean-­warming shift to coral dominance of, 70, 71f; rocky, alternative ecosystem states on, 181 temperature: balance between photosynthesis and respiration and, 287; green­house effect and, 78, 78f; interaction strengths and, 193–94; latitudinal patterns in interactions and, 67–68, 67f; life history and, 64, 66f, 166; metabolic scaling as function of, 122f, 123–24, 123f, 166; metabolism of heterotrophs vs. autotrophs and, 191; of open ocean vs. land, 297; organism’s normal range of, 120–21; oxygen consumption and, 121; of sea surface, 78, 79f, 80f; speciation and, 53–54, 53f; species interactions and, 121, 125; species richness and, 210. See also climate warming; ocean warming terrestrial ecosystems, 12, 13t; connections to marine ecosystems, 293 Tethys Sea, 38, 39, 39f, 348 thermal niche models, 291–92 thermocline, 263f, 264 thermohaline circulation, 43, 43f Thorson, Gunnar, 48, 64 Thorson’s rule, 64 tides, 297 tipping points, 249, 343, 363, 365, 380 top-­down control, 113f, 231–32, 232f; in coastal ecosystems, 302; in pelagic ecosystems, 274–76; in salt marsh ecosystems, 328–29, 330 top predators: abundant on coral reefs, 355–56; cascading effects of decline of, 180–81, 215, 232, 233; extinctions of, 193; feeding at convergent fronts, 42; food webs destabilized by loss of, 190, 232; homeothermic, 28; ­human role in ocean as, 87, 292; intensely harvested, 95; nekton as, 28; strongly reduced in modern times, 102, 258; targeted ­human pressure on, 233; trophic skew caused by loss of, 338 trace ele­ments: limiting primary production, 117; population structure and, 164–65, 165f. See also iron trade. See commerce trades biome, 61 trade winds, 40; monsoon and, 61 tragedy of the commons, 86–87; catch-­share programs and, 101 trait-­based approaches, 131, 136–37 traits: ecosystem functioning and, 224–25; extinction vulnerability and, 59; taxonomic group effects and, 73. See also functional traits; phyloge­ne­tic conservatism of traits transport structures, fractal network of, 122–23, 124 trawling damage to seabed, 292, 295, 315, 316–17, 337 tree of life, 18–21, 19f Trichodesmium, 267, 274 trophic amensalism, 313 trophic cascades, 173, 173f, 179, 180–81, 180f, 232–33; in Black Sea, 251, 274, 275, 276; in coastal ecosystems, 234f, 302; in coral reef communities, 361–62; foundation species depending on, 302; green world hypothesis and, 229; ­human impact

443

on large animals and, 258; increasing with predator diversity, 242, 244f; nonconsumptive effects in, 183–84; overfishing and, 274–76; pelagic, 293; in salt marshes, 329, 331; stronger in marine than in terrestrial systems, 230; uncommon in open ocean, 274, 276 trophic-­dynamic approach of Lindeman, 219, 220f, 222, 330 trophic efficiency, 237–38 trophic levels, 130; alternating between bottom-up and top-­down control, 233, 234f; biomass distributions among, 135, 135f; correlated with body size of individual fish, 144, 144f, 184; diversity-­biomass relationship and, 242–43, 245f; ecosystem concept and, 219; energetic equivalence rule and, 240, 241f; functional groups and, 213; green world hypothesis and, 228–29, 229f; interaction strengths and, 189, 190, 190f. See also bottom-up control; top-­down control trophic skew, 193, 193f, 215, 233, 258, 338 tropicalization, 68, 70, 71f; of seagrass-­associated fish communities, 322 tube worms, 281, 281f, 283, 284 turbulence: composition of phytoplankton and, 267, 267f, 292; recruitment of coral larvae and, 365; sediment grain size and, 311 turf algae. See filamentous (turf) algae turnover. See beta diversity (β) turtle excluder devices, 153 turtles. See sea turtles ultrasocial be­hav­ior, ­human, 84, 378 unicellular chemoautotrophs, 23, 24f, 26, 27t, 225 unicellular eukaryotic heterotrophs, 23, 24f, 26–27 unicellular phototrophs, 22, 23, 24f, 25–26, 27t unicornfish (Naso), 360 unified neutral theory of biodiversity. See neutral theory of biodiversity and biogeography upwelling: around reef islands, 367; climate warming and, 90, 91; coastal, 44, 62, 261, 297, 298f, 299, 305; equatorial, 42, 261–62; nitrate coming from, 116, 117, 226, 228; Pisaster ochraceous feeding on mussels and, 91 urbanized estuaries, 337, 343–44 urea: ocean fertilization proposal and, 107; in seawater, 115, 116, 226 values, 107–8, 110 vascular plants. See flowering plants velvet worms (Onychophora), 12 vents. See hydrothermal vents; volcanic vents Vermeij, Geerat, 58–59 vertebrates: grazing on seagrass meadows, 373; interaction strengths of, 193 vertical migration, diel, 261, 264; carbon export to deep ocean and, 273; between complementary habitats, 335 vicariance biogeography, 56, 57 viruses, 23, 24–25, 24f; in microbial loop, 272, 272f; phytoplankton defenses against, 268 visually foraging poikilotherms, 23, 24f, 28 volcanic vents: acidified ocean w ­ aters at, 94, 135–36, 372; low coral diversity at, 372 volcanism, and end-­Permian mass extinction, 58 Wallace, Alfred Russel, 358 wall of mouths, 142

444

Index

water-­column stratification, 262–65, 263f; four pelagic biomes and, 61; gyre circulation and, 41, 42; ocean warming and, 291, 292; phytoplankton cell size and, 132–33, 285; phytoplankton diversity and composition related to, 285; spring bloom and, 265–66 ­water flow: community interactions on rocky shores and, 308, 309f. See also wave energy ­water pollution. See pollution by ­human activities ­water quality: coral reefs and, 363, 364; oyster reefs and, 338 wave energy: on rocky shores, 305, 306f, 309; sediment grain size and, 311. See also ­water flow weakfish (Cynoscion regalis), geochemical signatures in, 164–65, 165f web of life, 21

weedy species: favored in biotic homogenization, 103–4; replacing foundation species, 36, 103, 338 westerlies biome, 61–62 western boundary currents, 41 wetlands. See coastal ecosystems ­whales, 28–30, 29f; blue ­whale, 23f, 28, 264f; evolved to exploit deep-­water prey, 264; fishing impacts on, 292; gray ­whale, 29, 29f, 159–60; invertebrates feeding on carcasses of, 278f, 279, 280; as keystone species, 102; sperm ­whale, 264; threatened species of, 102, 102f white-­band disease of Acropora, 185, 362–63, 370, 375 Williamson, Oliver, 87 Wilson, Edward, 57 wind: coastal upwelling and, 298f, 299; mixed layer and, 263; ocean currents and, 40, 41, 297; thermohaline circulation and, 43 Wootton, Tim, 205, 206

World3 computer model, 80, 82 worms: meiofaunal nematodes, 33, 35f, 313; nematodes of deep-­sea benthos, 287; on sediment bottoms, 312, 312f, 313–14 xenophyophores, 280 zooplankton: biogeographic patterns in functional traits of, 64; biological pump and, 273–74, 273f; crustacean, top-­down control of diatoms by, 233; ctenophores in top-­down control of, 233, 234f, 275, 275f; fecal pellets of, 235, 237, 273, 278; grazing by microzooplankton, 268, 269, 272–73, 293; ocean warming and, 215, 291–92; selective grazing by, 270. See also plankton zooxanthellae, 222, 301, 351–52, 351f, 354, 367, 368, 371, 375–76