WORLD ATLAS OF | BIODIVERSITY 21st CENTURY 3 QQ BRIAN GROOMBRIDGE AND MARTIN D. JENKINS | UNEP WCMC World Atlas of Biodiversity addresses the remark- able growth in concern at all levels for living things and the environment and the increased appreciation of the links between the state of ecosystems and the state of humankind. Building on a wealth of re- search and analysis by the conservation community worldwide, this book provides a comprehensive and accessible view of key global issues in biodiver- sity. It outlines some of the broad ecological relationships between humans and the rest of the material world and summarizes information on the health of the planet. Opening with an outline of some fundamental aspects of material cycles and energy flow in the biosphere, the book goes on to discuss the expansion of this diversity through geo- logical time and the pattern of its distribution over the surface of the Earth, and analyzes trends in the condition of the main ecosystem types and the species integral to them. Digitized by the Internet Archive in 2010 with funding from UNEP-WCMC, Cambridge http://www.archive.org/details/worldatlasofbiod02groo World Atlas of Biodiversity Published in association with UNEP-WCMC by the University of California Press University of California Press Berkeley and Los Angeles, California University of California Press, Ltd. London, England © 2002 UNEP World Conservation Monitoring Centre UNEP-WCMC 219 Huntingdon Road Cambridge CB3 ODL, UK Tel: +44 (0) 1223 277 314 Fax: +44 (0) 1223 277 136 E-mail: infol@unep-wcmc.org Website: www.unep-wcmc.org World Atlas of Biodiversity: Earth's Living Resources in the 21st Century is a revised and updated edition of Global Biodiversity: Earth's Living Resources in the 21st Century No part of this book may be reproduced by any means or transmitted into a machine language without the written permission of the publisher. The contents of this volume do not necessarily reflect the views or policies of UNEP-WCMC, contributory organizations, editors or publishers. The designations employed and the presentations do not imply the expression of any opinion whatsoever on the part of UNEP-WCMC or contributory organizations, editors or publishers concerning the legal status of any country, territory, city or area or its authority, or concerning the delimitation of its frontiers or boundaries or the designation of its name or allegiances. Cloth edition ISBN 0-520-23668-8 Cataloging-in-publication data is on file with the Library of Congress Citation Groombridge B. and Jenkins M.D. (2002) World Atlas of Biodiversity. Prepared by the UNEP World Conservation Monitoring Centre. University of California Press, Berkeley, USA. Moy VK UNEP WCMC World Atlas of Biodiversity Earth's Living Resources in the 21st Century Brian Groombridge & Martin D. Jenkins UNIVERSITY OF CALIFORNIA PRESS BERKELEY Los ANGELES LONDON World Atlas of Biodiversity Prepared by UNEP World Conservation Monitoring Centre 219 Huntingdon Road Cambridge CB3 ODL, UK Tel: +44 (0) 1223 277 314 Fax: +44 (0) 1223 277 136 E-mail: infoldunep-wemc.org Website: www.unep-wemc.org Director Mark Collins Authors Brian Groombridge Martin D. Jenkins Additional contributors Adrian C. Newton (Project manager) Rachel Cook Neil Cox Victoria Gaillard Edmund Green Janina Jakubowska Thomas Kaissl Valerie Kapos Charlotte Lusty Anna Morton Mark Spalding Christoph Zockler Production of maps Simon Blyth with the assistance of Igor Lysenko Corinna Ravilious Jonathan Rhind Layout Yves Messer UNEP WCMC The UNEP World Conservation Monitoring Centre is the biodiversity information and assessment arm of the United Nations Environment Programme, the world’s foremost intergovernmental environmental organization. UNEP-WCMC aims to help decision-makers recognize the value of biodiversity to people everywhere, and to apply this knowledge to all that they do. The Centre's challenge is to transform complex data into policy-relevant information, to build tools and systems for analysis and integration, and to support the needs of nations and the international community as they engage in joint programs of action. A Banson production 27 Devonshire Road Cambridge CB1 2BH, UK Color separations Swaingrove Printed in the UK Acknowledgments First and foremost we would like to express our deepest thanks to the Aventis Foundation, without whose generous funding the research and production work for this book could not have been undertaken. Preparation of the book was also generously supported by the Department of Environment, Food and Rural Affairs (DEFRA) of the UK Government. The Owen Family Trust is also acknowledged for financial support to the first edition of this text. We also acknowledge with thanks the generous assistance extended by the following, listed approximately in the same sequence as the chapters in which their material appears: Christopher Field and George Merchant, Department of Global Ecology, Carnegie Institution of Washington for use of data from a global model of net primary production. Robert Lesslie of the Department of Geography, Australian National University, Canberra, for allowing us to use data resulting from his global wilderness analysis. BirdLife International, of Cambridge, UK, for allowing use of spatial data on endemic bird areas and on threatened bird species. Gene Carl Feldman, Oceanographer at NASA/Goddard Space Flight Center, Greenbelt, Maryland, for approving use of material from the SeaWiFS Project of NASA/Goddard Space Flight Center and ORBIMAGE. The University of Maryland Global Land Cover Facility, for facilitating use of land cover data. Professor Wilhelm Barthlott of the Botanisches Institut und Botanischer Garten, Rheinischen Friedrich-Wilhelms-Universitat, Bonn, for kindly allowing use of a map showing contours of global plant species diversity. Jonathan Loh, responsible for the WWF Living Planet Report, for kindly approving use of global trend indices from the Living Planet Report 2000. John E.N. Veron, Chief Scientist at the Australian Institute of Marine Sciences, Townsville, Queensland for allowing use of coral generic diversity data. Several biologists associated with IUCN/SSC specialist groups on fishes, mollusks and inland water crustacea, for providing data and expertise on important areas for freshwater biodiversity collated in an earlier publication: Gerald R. Alien (Western Australian Museum]; the late Denton Belk {Texas}; Philippe Bouchet (Laboratoire de Biologie des invertébrés marins et malacologie, Muséum National d'Histoire Naturelle, Paris}; Keith Crandall (Department of Zoology, Brigham Young University]; Neil Curmberlidge (Department of vi Biology, Northern Michigan University}; Olivier Gargominy (Laboratoire de Biologie des invertébrés marins et malacologie, Muséum National d'Histoire Naturelle, Paris); Maurice Kottelat (Cornol, Switzerland}; Sven O. Kullander (Department of Vertebrate Zoology, Swedish Museum of Natural History, Stockholm); Christian Lévéque (ORSTOM, Paris]; R. von Sternberg (Center for Intelligent Systems, State University of New York at Binghamton]; Guy Teugels (Laboratoire Ichthyologie, Musée Royal de l'Afrique Centrale, Tervuren). Ben ten Brink, Jan Bakkes and Jaap van Woerden for facilitating use of material illustrating work on scenarios carried out at the Rijksinstituut voor Volksgezondheid en Milieu [RIVM], Bilthoven, the Netherlands. Christian Nellemann (Norwegian Institute for Nature Research], Hugo Ahlenius (UNEP GRID- Arendal) and the Secretariat of GLOBIO (Global methodology for mapping human impacts on the biosphere], for material applying this approach to scenario development. PHOTOGRAPHS Pages: 6, L. Olesen/UNEP/Still Pictures; 7, L.L. Hock/UNEP/Topham; 14, B. Groombridge; 15, M. Wakabayashi/UNEP/Topham; 37, Giotto Castelli; 38, M. Friedlander/UNEP/Topham; 39, UNEP/Topham; 72, G. Bluhm/UNEP/Topham; 73, K. Kaznaki/UNEP/Topham; 86, M. Schneider/UNEP/Topham; 93, M.R. Andrianavalona/UNEP/Still Pictures; 96, H. Mundell/UNEP/Topham; 98, R. Faria/UNEP/Topham; 99, UNEP/Topham; 103 J. Nuab/UNEP/Topham; 110 top, R. del Rosarion/UNEP/Topham;110 bottom, Mazinsky/UNEP/Topham; 118, E. Green; 143 top, E. Green; bottom, M. Spalding; 144, M. Garcia Blanco/UNEP/Topham; 151, E. Green; 152, D. Nayak/UNEP/Still Pictures; 154, E. Green; 155, D. Seifert/UNEP/Still Pictures; 165, UNEP/Topham; 169, S.W. Ming/UNEP/Still Pictures; 174, F. Colombini/UNEP/Topham; 178, C.K. Au/UNEP/Still Pictures; 185, K. Lohua/UNEP/Still Pictures; 194, C. Petersen/UNEP/Topham; 212, P. Garside/UNEP/Topham; 216, C. Senanunsakul/UNEP/Still Pictures Contents Foreword xi 3.5 Frequency of percent extinction Preface xii per million year period 29 Introduction 1 3.6 Number of family extinctions per geological interval through CuapTer 1: The biosphere 3 the Phanerozoic 29 Maps TABLE 1.1. Physical geography of the Earth 4 3.1 The principal mass extinctions 1.2. Primary production in the in the Phanerozoic fossil biosphere 8 record 30 FIGURE 1.1 Hypsographic curve 5 CHaPTER 4: Humans, food and TABLES biodiversity 33 1.1 Global annual net primary MAPS production 10 4.1. Early human dispersal 34 1.2 Estimated global carbon 4.2 Livestock breeds: numbers budget and biomass totals 11 and status 42 4.3 FAO world diet classes 48 Cuapter 2: The diversity of organisms 13 4.4 Human population density 52 FIGURE 4.5 Terrestrial wilderness 54 2.1 The phylogenetic tree 19 4.6 Vertebrate extinctions since TABLES AD1500 56 2.1 Estimated numbers of 4.7. Threatened mammal species 58 described species, and 4.8 Critically endangered possible global total 18 mammals and birds 62 2.2 Key features of the major 4.9 Threatened bird species density 64 groups of living organisms 20 FIGURES Boxes 4.1. Human population 47 2.1. New species discoveries 16 4.2 Vertebrate extinctions by 2.2 Improving taxonomic period since AD1500 59 knowledge and capacity 17 TABLES 4.1 Top ten food commodities, Cuapter 3: Biodiversity through time 23 ranked by percentage FIGURES contribution to global food 3.1 The four eons of the supply 4) geological timescale 24 4.2 World diet classes 44 3.2 Periods and eras of the 4.3. Examples of diversity in Phanerozoic 26 agricultural systems 45 3.3 Animal family diversity 4.4 Number of individuals and through time 27 biomass, selected organisms 46 3.4 Plant diversity through time 28 4.5 Land converted to cropland 49 viii 4.6 Estimated large herbivore numbers and biomass in Mesolithic and modern Britain 4.7 Late Pleistocene extinct and living genera of large animals 4.8 Numbers of extinct animal species according to IUCN 4.9 \sland diversity at risk: birds 4.10 Threatened species 4.11 Number of threatened animal species in major biomes Boxes 4.1 Loss of diversity in agricultural genetic resources 4.2 ‘Lazarus species Cuapter 5: Terrestrial biodiversity Maps 5.1 Photosynthetic activity on land 5.2 Global land cover 5.3 Diversity of vascular plant species 5.4 Biodiversity at country level 5.5 Flowering plant family density 5.6 Terrestrial vertebrate family density 5.7 Current forest distribution 5.8 Non-forest terrestrial ecosystems FIGURE 5.1 A typical species-area plot TABLES 5.1 Global distribution of land area, by latitude bands 5.2 Different definitions of forest cover 5.3 Sample effects on forest area estimates of different forest definitions 5.4 Global area of five main forest types 5.5 Important families and genera, and numbers of species, in four areas of temperate broadleaf deciduous forest 5.6 Biomass and carbon storage in the world’s major forest types 5.7. Tree species richness in tropical moist forests 5.8 Estimated annual change in forest cover 1990-2000 100 106 108 77 71 80 81 81 83 84 87 OM 5.9 Boxes 5.1 oz. 5.3 5.4 Se) Global protection of forests within protected areas in IUCN categories I-VI Estimated plant species richness in the five regions of Mediterranean-type climate Defining ecosystems Species and energy Fire in temperate and boreal forest Temperate forest bird trends Grassland bird trends Cuapter 6: Marine biodiversity Maps 6.1 Coral reef hotspots 6.2. Shark family diversity 6.3. Marine turtle diversity 6.4 Mangrove diversity 6.5 Seagrass species diversity 6.6 Coral diversity 6.7. Marine fisheries catch and discards FIGURES 6.1 Species contributing most to global marine fisheries 6.2 Marine fisheries landings by major group 6.3. Global trends in the state of world stocks since 1974 6.4 Trends in global fisheries catch since 1970 6.5 Marine aquaculture production 6.6 Marine population trends TABLES 6.1. Area and maximum depth of the world’s oceans and seas 6.2 Relative areas of continental shelves and open ocean 6.3. Marine diversity by phylum 6.4 Diversity of craniates in the sea by class 6.5 Diversity of fishes in the seas by order 6.6 Marine tetrapod diversity 6.7 Regional distribution of breeding in seabirds 6.8 Diversity of mangroves 6.9 Current mangrove cover 97 105 74 78 82 85 104 117 126 128 30 34 136 40 148 45 146 147 147 150 158 117 118 122 123 124 125 129 132 133 6.10 Diversity of stony corals in 7.10 Thirty high-priority river basins 190 the order Scleractinia 38 Boxes 6.11 Coral reef area 39 7.1. Saline and soda lakes 164 6.12 Taxonomic distribution and 7.2 Wetland loss in Asian drylands 184 status of threatened marine animals 56 CuapTeR 8: Global biodiversity: Boxes responding to change 195 6.1 Life in sediments 19 Maps 6.2. Marine introductions 53 8.1 World protected areas 200 8.2 Centers of plant diversity 202 CHAPTER 7: Inland water biodiversity 163 8.3 Major areas of amphibian Maps diversity 204 7.1 Freshwater fish family 8.4 Endemic bird areas 206 diversity 170 8.5 Marine protected areas 208 7.2 Major areas of diversity of 8.6 International protected area inland water fish 76 agreements 210 7.3. Major areas of diversity of FIGURES inland water mollusks 80 8.1 Development of the global 7.4 Major areas of diversity of network of protected areas 198 selected inland water 8.2 Possible future scenarios crustacean groups 182 from GEO 3, evaluated with 7.5 Priority river basins 90 RIVM IMAGE 218 FIGURES 8.3 Possible future scenarios, 7.1 Reported global inland GLOBIO 220 fisheries production 79 TABLE 7.2 |nland water fish 8.1 Major global conventions introductions 86 relevant to biodiversity 7.3. Freshwater population trends 186 maintenance 214 7.4 River basin richness and Boxes vulnerability 189 8.1 Pioneering NGOs in 7.5. Changes in condition of a biodiversity conservation 197 sample of freshwater lakes 8.2 The precautionary principle 199 between 1950s and 1980s 91 8.3 Systematic conservation TABLES planning 203 7.1 Components of the hydrosphere 163 8.4 Negotiating a multilateral treaty 213 7.2 Physical and biodiversity features of major long-lived lakes 166 APPENDICES 7.3 Partial list of global hotspots 1 The phyla of living organisms 225 of freshwater biodiversity 168 7.4 \nsects of inland waters 172 2 Important food crops 244 7.5 Fish diversity in inland waters, by order 173 3 Domestic livestock 271 7.6 Tetrapod diversity in inland waters 75 4 Recent vertebrate extinctions 278 7.7 Major inland fishery countries 179 7.8 Numbers of threatened 5 Biodiversity at country level 295 freshwater fishes in selected countries 87 6 tmportant areas for 7.9. Taxonomic distribution and freshwater biodiversity 306 status of threatened inland water vertebrates 188 INDEX 329 7 * «z- M . art vo \) no =. Fs « ae oleae — LSS Pict a A 4 = = ; — rm — + = ~ E > 45 iF a ‘a = ‘i 1 = i ”} hes ; wor 7 i I. 5 ai = = ue Tone a ie ~ 1 ae a ee 5 se ( i baer.t ca Pie iin aoe - i =) > 4 ¢ >» @ lier ve ‘ os a E a eT etaciea ; fo (dhs ges Spe } po oa "= a Sr ee eee 3.0 , oa ee a =e a | we We Se Dy Flt ee 1 a ; k a a. p Bilis ie ull hf Rage a eawratyet te fiery Gat “i a ai ffl bach ener ep rhe Bay : vibes Foreword Klaus Topfer, Executive Director, United Nations Environment Programme It is a great pleasure for me to introduce this important new book from the UNEP World Conservation Monitoring Centre. Building on the analyses it carried out for the Earth Summit in Rio de Janeiro in 1992 and for the new millennium just two years ago, UNEP-WCMC has once again updated and revised its important overview of life on Earth in time for a major global event. In tune with the message of the Johannesburg World Summit on Sustainable Development, this new atlas places humankind firmly in the context of the species and ecosystems upon which we all depend for our livelihoods. We are a part of biodiversity and as such we should treasure it, use it wisely and share its benefits in our own enlightened self-interest. | commend this book to all who seek a greater appreciation of the inter-dependency between our own future and that of global biodiversity. In closing | should like on behalf of UNEP to thank most warmly the Aventis Foundation and the UK Department for Environment, Food and Rural Affairs for their support in the preparation of the World Atlas of Biodiversity. xi xii Preface Mark Collins, Director, UNEP World Conservation Monitoring Centre The diversity of life is the defining feature of planet Earth. It is unique - as far as we know - in the infinity of the universe. For 11000 years since agriculture began, humankind has increasingly appropriated the biological resources and natural productivity of lands and seas to support the expansion of civilizations and technologies. Everything that we have achieved has its origins in living animals, plants and the communities and ecosystems of which they are a part. But it is only in the past 30 years, since the United Nations Conference on the Human Environment in Stockholm in 1972, that we have begun to recognize the limits to nature's gifts. We now know that our own success is placing strain on nature’s ability to evolve, diversify, cleanse our air and water and provide us with the raw materials we need for food, fuel, fiber and health. Just ten years ago we began to take integrated and holistic action to ensure conservation and sustainable use of biological resources. The UN Conference on Environment and Development (UNCED) saw the signing of the Convention on Biological Diversity, the first global agreement on biodiversity that clearly positioned humankind as an integral part of the complex of life on Earth, rather than a special case somehow separate from nature and immune from its laws. The ‘ecosystem approach’ espoused by the Biodiversity Convention acknowledges that our relations with the rest of the living world are truly interactive, and that what we do to nature will in turn reflect on nature’s ability to respond to our own needs. The Convention foresaw a careful balance in the management of the Earth's living wealth through conservation, sustainable use and equitable sharing of costs and benefits. There could be no better time to launch a fresh assessment of the living world. This World Atlas of Biodiversity is published to coincide with the World Summit on Sustainable Development in Johannesburg, Republic of South Africa. The focus of the meeting is once again on sustainable development, but this time the emphasis is clearly on poverty alleviation. The message is clear: harmonized economic, social and environmental development will be but a dream while so many of the world’s people have no choices and no opportunities to take a planned approach to their lives. What is the relevance of this book in the context of the Johannesburg message? The reality is that this World Atlas of Biodiversity is of greater relevance now than at any time in the past. The world’s living wealth remains the cornerstone of sustainable livelihoods and quality lifestyles in both the industrialized and developing worlds. Recognition of this fact is spreading, and the value of biodiversity in people's lives, socially, economically and environmentally, has never been more apparent than it is today. This is not a textbook, it is a resource pack and a survival kit for the future. | hope that all who read it will find new insights into the significance of life on Earth to their own lives. And that they will take steps within their homes, communities and nations to utilize and enjoy living resources wisely, share’ the benefits and hold the capital in trust for future generations. Introduction OBJECTIVES The past ten years have seen a remarkable growth in concern for wildlife and the environment, with an increased appreciation of the links between the state of ecosystems and the state of humankind. Many analysts have concluded that achieving sustainable and equitable human development will require, among other measures, taking a more effective approach to managing human impacts on the biosphere. This was reinforced by the 1992 United Nations Conference on Environment and Development (the Earth Summit), at which the Convention on Biological Diversity (CBD) was opened for signature. Many conservation and management initiatives worldwide have arisen from efforts to meet the objectives framed by the CBD text. In principle, any kind of variation at any level of biological organization - encompassing genes, populations, species and communities - is biological diversity. The text of the CBD defines biological diversity as ‘the variability among living organisms from all sources including, inter alia, terrestrial, marine and other aquatic ecosystems and the ecological complexes of which they are part; this includes diversity within species, between species and of ecosystems’. In practice, the term is often contracted to ‘biodiversity’, and used to refer collectively to all such variation: in effect, as a convenient shorthand for the total complex of life in some given area, or on the Earth as a whole. In the present volume we aim to use the data now available to provide an overview of the current state of global biodiversity, using maps where helpful, and to ensure that this information is accessible to a wide readership. While biodiversity has many dimensions, attention is here focused on the diversity of living organisms and their populations, and on major aquatic and terrestrial ecosystem types. Far more space is given to the macro-scale organisms and landscape elements that may be subject to planning and management intervention than to microorganisms, despite the immense metabolic diversity of the latter, and their pivotal role in driving biosphere cycles. STRUCTURE OF THIS BOOK The eight chapters fall informally into four thematic sections. The first section opens with an outline of some fundamental aspects of material cycles and energy flow in the biosphere (Chapter 1). This is followed by a synopsis of the diversity of living organisms (Chapter 2) and of change in this diversity through geological time (Chapter 3). The second section (Chapter 4) is largely concerned with relationships between humankind and biodiversity, noting the increasing human impact on the environment from early modern humans onward, the use of biodiversity in human nutrition, and reviewing trends in recent time, focusing on depletion and extinction of species. The third section aims to characterize communities and biodiversity trends in the three basic biome types: terrestrial, marine and inland waters (Chapters 5, 6 and 7, respectively). Finally, Chapter 8 introduces some of the management and planning responses that have been implemented with a view to maintaining ecosystem health and putting human development on a sustainable foundation. i The biosphere Earth’s surface that contains all living organisms and the elements they exchange with the non-living environment. Water makes up about two thirds of an average living cell, and organic molecules based on hydrogen, carbon, nitrogen and oxygen make up the remaining one third. These and other elements of living cells cycle repeatedly between the soil, sediment, air and water of the environment and the transient substance of living organisms. The energy to maintain the structure of organisms enters the biosphere when sunlight is used by bacteria, algae and plants to produce organic molecules by photosynthesis, and all energy eventually leaves the biosphere again in the form of heat. Photosynthetic organisms themselves use a proportion of the organic material they synthesize; net primary production T HE BIOSPHERE IS THE THIN AND IRREGULAR ENVELOPE around and including the is the amount of energy-rich material left to sustain all other life on Earth. Humans now appropriate a large proportion of global net primary production, and have caused planetary-scale perturbations in cycling of carbon, nitrogen and other elements. THE LIVING PLANET The defining characteristic of the planet Earth is that it supports life, and has done so for at least 70 percent of its history (see Chapter 3). The position of the Earth relative to the sun, its size and composition appear to be the main factors that have allowed life to develop here. Most importantly, these factors have combined to ensure the permanent presence of a large amount of liquid water on the planet's surface, and this is the fundamental prerequisite of life as we know it. The space occupied by living organisms and the part of the planet that supports them is called the biosphere’. The non-living biosphere comprises the hydrosphere [the waters}, the soil and upper part of the lithosphere (the solid matter that forms the rocky crust of the Earth], and the lower part of the atmosphere (the thin layer of gas coating the planet's surface]. These domains interact in ways critical to the operation of the biosphere, and are linked in particular by the properties of water as a solvent and medium that fosters the chemical reactions basic to life. While providing the conditions necessary for life, the structure and composition of the non-living parts of the biosphere have them- selves been profoundly affected through time by living organisms. Most clearly, and from the human viewpoint most importantly, the presence of significant quantities of free oxygen in the atmosphere is entirely the product of oxygen-releasing photosynthesis by cyanobacteria starting more than 2000 million years ago. The idea that living organ- isms do not merely influence conditions in the biosphere but in some way regulate them to maintain the conditions conducive to life has received considerable attention in recent years, chiefly in terms of the Gaia hypothesis’ proposed in the 1970s. The extent of the biosphere At planetary scale, the biosphere can be pictured as a thin and irregular envelope around the Earth's surface, just a few kilo- meters deep on the globe's 6 371-kilometer radius. Because most living organisms depend directly or indirectly on sunlight, the regions reached by sunlight form the core of the The biosphere 3 4 WORLD ATL enermersertsere Map 1.1 Physical geography of the Earth The relative areas occupied by dry land and by water, and the general distribution of areas of extreme height or depth. AS OF BIODIVERSITY i biosphere: i.e. the land surface, the top few millimeters of the soil, and the upper waters of lakes and the ocean through which sunlight can penetrate. The biosphere is not homogenous, be- cause actively metabolizing living organisms are sparse or absent where liquid water is absent, such as in the permanent ice at the poles and on the very highest mountain peaks, but abundant where conditions are favorable. Nor are its boundaries sharply defined, because bacterial spores and other dormant forms of life passively disperse virtually everywhere, from polar icecaps to several tens of kilometers above the surface of the Earth (approaching the upper limit of the stratosphere], and living microorganisms occur’ within rocks more than 3 kilometers deep in the lithosphere. The whole of the sea is theoretically capable of supporting active life and con- stitutes therefore the vast majority of the volume of the biosphere (Figure 1.1). Depending on water clarity, the sunlit [photic] zone may reach just a few centimeters to a few hundred meters in depth, but the marine biosphere is extended into regions of total darkness, down to more than 10 000 meters in the ocean depths, by organisms that subsist on the rain of organic debris falling from the upper waters. In addition, there are animal communities on the sea floor based on microorganisms deriving their energy from hydrogen sulfide emitted from hydrothermal vents. Overall, however, the amount of living material in most of the sea - that part of the open ocean below the upper hundred or so meters - is relatively low. The atmosphere plays a vital role in the biosphere, not only in providing a source of essential gases, but also in buffering conditions at ground level, by regulating temperature and providing a shield against excessive ultraviolet radiation. Many organ- isms, from microscopic bacteria to bats and birds, spend part of their lives suspended in the atmosphere; however, no organism is known that passes its complete life cycle in the air, and living biomass per unit volume above the Earth’s solid or liquid surface is extremely low. Photosynthesis and the biosphere Life on Earth is based essentially on the chemistry of water and carbon. Indeed, in biochemical terms, living organisms are simply elaborate systems of organic macro- molecules dispersed in an aqueous medium. The average cell is about 70 percent water by weight; the remainder consists very largely of carbon-containing (organic) compounds com- posed mainly of the four elements hydrogen, carbon, nitrogen and oxygen. These com- 8 84 ~o Elevation (km] average elevation 840 m The biosphere SSA NE ET YS SS Figure 1.1 Hypsographic curve The horizontal baseline in this figure represents the Earth's total surface area of 510 million km’. The figure shows that 71% of this surface is covered by marine waters and 29% is dry land. It also shows the mean land elevation and mean ocean depth, and the amount of Earth's surface, in percentage terms, standing at any given elevation or depth. ei) & cs Gael 29% average depth 3 800 m o Depth (km) 5 6 WORLD ATLAS OF Energy from the sun drives photosynthesis, responsible for the vast majority of organic production. BIODIVERSITY pounds include four major types of large organic molecule - proteins, carbohydrates, lipids and nucleic acids - and about 100 different small organic molecules. A number of other elements are required in much smaller, though still vital, quantities. These include phosphorus, sulfur, iron and magnesium. All these elements cycle through the biosphere in a variety of forms, both organic and inorganic, following complex and interlinked pathways many of which are yet to be fully elucidated. Except for some micro- organisms that use energy derived from inorganic chemicals, the engine that drives the organic part of this turnover is photo- synthesis - the capture by living tissues of energy from the sun. Photosynthesis essentially involves the use of energy from sunlight to reduce carbon dioxide ([CO,) with a source of electrons {almost invariably hydrogen] to produce carbohydrates, water (HO) and, generally, a by-product from the hydrogen donor. In some bacteria the hydrogen donor is hydrogen gas, in others it is hydrogen sulfide; but, in cyanobacteria, algae and plants, water is the hydrogen donor and gaseous elemental oxygen (O.] is the by-product. This is over- whelmingly the predominant and most impor- tant form of photosynthesis on the planet, and is described by the following equation: 2nH,0 + nCO, + light > nH,0 + nCH20 + 2n0 The initial products of photosynthesis in plants are simple sugars such as glucose (C,H;20,). Larger carbohydrate molecules made from glucose include cellulose, the main component of plant cell walls and woody tissues, and starch, a key storage carbohydrate found in roots and tubers. Energy is needed to make the chemical bonds within these organic molecules, and energy is released when the bonds are broken down again. The controlled breakdown of these molecules within cells is the mechanism by which all cells obtain energy to do useful work. All organisms can break down sugars very directly without the need for oxygen. Many bacteria that live in aerobic conditions use only this method. Virtually all eukaryotes (see Chapter 2) have evolved a more complex additional pathway that requires oxygen but yields much more energy. This latter pathway - aerobic respiration - essentially reverses the basic photosynthetic reaction shown above. The major cycling process of the biosphere, therefore, consists of the photosynthetic fixing of carbon dioxide with water to produce organic compounds, in which energy is stored, and oxygen; this is followed by respiration of these compounds, in which the stored energy is released and carbon dioxide and water are produced. Photosynthesis therefore is not only responsible for the vast majority of organic production, but also for the maintenance of free oxygen in the atmosphere, without which aerobic organisms [the great majority of eukaryotic organisms, including humans) could not survive. Although photosynthesis is the primary engine of the biosphere, in the sense that it injects energy into the system and creates basic organic molecules, production of the full range of organic molecules on which life depends requires additional elements. Of the four key elements, nitrogen is often the one in limited supply, but it is an essential com- ponent of nucleic acids and proteins. Although the atmosphere consists of 79 percent nitrogen, this inert gaseous form of the element cannot be used by plants or most other organisms until combined (fixed) with other elements. In the biosphere, atmos- pheric nitrogen is fixed by a range of bacteria, including cyanobacteria, some free-living soil bacteria, and most importantly by specialized bacteria that live symbiotically in the root nodules of leguminous plants (peas, beans, etc.}] Some nitrogen is also fixed by lightning in electric storms and, in the modern world, industrially in the production of fertilizer. Fixed nitrogen is made available to plant roots through association with fungi (mychorrhizas) and as nitrogen-fixing organisms decay. From plant roots, it is transported to metabolizing plant cells. On death, these steps are reversed, and the fixed nitrogen may be immediately recycled or revert to elemental nitrogen. PRODUCTIVITY AND THE CARBON CYCLE About half of the solar energy reaching the upper atmosphere of the Earth is immediately reflected. Most of the remainder interacts with the atmosphere, ocean or land, where it evaporates water and heats air, so driving atmospheric and ocean circulation. Much less than 1 percent of the incoming energy is intercepted and absorbed by photosynthetic organisms. On land these photosynthesizers are overwhelmingly green plants, although cyanobacteria and algae are also present, the latter particularly in the symbiotic associations with fungi known as lichens. In aquatic habitats, particularly the sea, virtually all photosynthesis is carried out by cyanobacteria and algae, although green plants are also present in shallow coastal and inland waters. Photosynthesizers fix carbon and therefore accumulate organic mass or biomass (often measured in dry form - that is, the once-living tissues of an organism with the water extracted]. These organisms are the primary producers. The amount of carbon fixed is referred to as gross primary production and is typically measured in grams (g) of carbon (C) per unit of space (area or volume) per unit of time. The photosynthetic producers also respire to meet their own energetic needs. Under some circumstances, respiration of photo- synthesizers over a given period may balance their carbon fixation, so that there is no net accumulation of organic carbon. More nor- mally, however, there is a surplus of fixation over respiration, so that organic matter is accumulated over time. This accumulation is referred to as net primary production [{NPP]. The accumulated matter is available to the vast suite of organisms of all sizes, including humans, that cannot synthesize their own organic compounds from an inorganic base or harness energy from inorganic sources. Such organisms are referred to as heterotrophs, while photosynthesizers and the few kinds of microorganisms that use other energy sources to synthesize organic compounds are referred to as autotrophs. Food webs An organic product produced by a photo- synthesizer may pass through a number of heterotrophs before finally being broken down again to its inorganic constituents. Conventionally this can be viewed as a food chain. At macroscopic level, a green plant may be eaten by a herbivore - a grasshopper, say - which is eaten by a lizard, which is itself eaten by a hawk, which dies and is disassembled and partially consumed by animal scavengers, with the remainder decomposed by bacteria and fungi. In reality, this is an enormous over- simplification. The plant will almost certainly have a complex network of symbiotic fungi associated with its roots, which make use of some of the gross production of the plant but which also provide it with some essential nutrients. The plant itself may shed leaves which are directly broken down by other fungi, protoctists such as slime molds, The biosphere Organic products pass through the food chain, through processes such as predation 8 a ae WORLD ATLAS OF BIODIVERSITY Map 1.2 Primary production in the biosphere Global spatial variation in annual net primary production (NPP], in g C per m’ per year, calculated from an integrated model of production based on satellite indices of absorbed solar radiation. Source: Map created from data supplied by Chris Field and George Merchant, Department of Global Ecology, Carnegie Institution of Washington. See http://jasper.stanford.edu/chrisweb/flab/ flab.html, and Field et al”. g C per m’ per year 1 782 - 3 859 1107 - 1781 881-1106 671 - 880 487 - 670 341 - 486 230 - 340 143 - 229 61 - 142 0- 60 and many forms of bacteria. The grass- hopper is likely to be parasitized by a host of smaller organisms, some of which are themselves in turn parasitized. It will also support a host of benign microorganisms in its intestine that are themselves constantly growing and reproducing. The lizard may die and decompose and the hawk may eat the grasshopper directly. The overall pattern of feeding relationships thus forms a web of immense complexity in any but the simplest ecosystems. Each organism in the food web respires, releasing energy which is eventually dissi- pated in the form of heat, carbon dioxide and water. At each stage, therefore, some carbon is returned to the inorganic part of the carbon cycle. In addition, all living organisms produce waste products, some of which are incompletely metabolized organic compounds. Heterotrophic organisms are also not completely efficient in their appropriation of the organic material they consume, so that some proportion of this is excreted as waste product. These organic wastes are theor- etically available to other organisms in the food web. The assimilation efficiency of heterotrophic organisms may be anything from 20 percent [in the case of some terrestrial herbivores} to 90 percent (in the case of some carnivores}, with the remainder excreted. Of the amount assimilated, a high Proportion is expended as respiration, with the remainder available to add biomass, i.e. to enable the organism to grow and reproduce. The proportion available to add biomass is dependent on the organisms involved as well as a range of other factors. It can be as lowas 10 percent or less and as high as 50 percent or more. This proportion is a measure of the net growth efficiency of the organism. For purposes of ecological analysis, particularly involving productivity estimates, the gross growth efficiency is the most commonly used measure. This is simply the product of the assimilation efficiency and the net growth efficiency of a particular heterotroph and is a measure of the pro- portion of food consumed by that organism that, after excretion and respiration, is ultimately available for its growth. As a very coarse generalization, a value of 10 percent is widely used, although it is acknowledged that in terrestrial herbivores the figure is likely to be lower and in planktonic communities and_ terrestrial carnivores it is likely to be higher. Using the figure of 10 percent in the example above, for every kilo of plant matter eaten by the grasshopper, the latter would add 10 grams to its body weight. When the grasshopper was eaten by the lizard, this would add 1 gram to the lizard’s body weight, and when the lizard was eaten by the hawk, this would add 0.1 grams to the hawk’s weight. This explains why, at the species level, so-called higher predators are rarer than herbivores and in any given area have a lower biomass, while the biomass of primary producers exceeds that of all heterotrophs combined. Measures of local and global productivity Primary productivity varies enormously, both spatially and temporally, at all scales. Most ob- viously, under natural conditions productivity The biosphere 9 10 WORLD ATLAS OF BIODIVERSITY Table 1.1 Global annual net primary production Note: This represents one among several attempts to estimate global production; see text for further details Ocean data averaged 1978- 83, land 1982-90, units in petagrams (1 Pg = 10'°g) Source: Adapted from Field et al“ Biosphere units NPP (x 10g C) Ocean 48.5 Terrestrial 56.4 Tropical rainforest 17.8 Deciduous broadleaf forest 1.5 Broadleaf and needleleaf forest 3.1 Evergreen needleleaf forest 3.1 Deciduous needleleaf forest 14 Savannah 16.8 Perennial grassland 2.4 Broadleaf shrubs with bare soil 1.0 Tundra 0.8 Desert 0.5 Cultivation 8.0 effectively ceases every night. Seasonal variations in most parts of the world are also marked. Productivity is, however, difficult to measure, so that estimates at all scales are subject to considerable uncertainty. On land one major source of uncertainty is below ground productivity: in natural eco- systems, less than 20 percent of plant produc- tion is typically consumed by herbivores. The remainder enters the soil system, either through the plant roots or as leaf litter. Measuring this portion of terrestrial produc- tivity - probably over 80 percent of the total - is particularly problematic. In the past there has been a marked tendency to under- estimate it. Similarly, it had long been assumed that the nutrient-poor waters of the open ocean were extremely unproductive, but it is now known that large populations of extremely small photosynthesizing unicellular organisms - the so-called picoplankton - form the basis of a surprisingly productive eco- system in these regions. One approach to estimating global pro- ductivity is based on measurements in particular ecosystems and extrapolation from these using estimates of the global extent of those ecosystems. This suggests that net primary production on land is of the order of 45-65 x 10" g C per year (that is, 45-65 petagrams or thousand million metric tons}, and at sea is around 51 x 10" g C per year. This gives a global estimate of annual net primary production in the order of 100 x 10" g C. Gross primary production is estimated to be about twice this. However, global measures using a somewhat different technique, involving assessments of relative concentration of oxygen isotopes, have indicated that annual gross primary production on land may be greater than 180 x 10" g C, while that in the sea is around 140 x 10° g C. This would give a global figure of over 320 x 10" g C, implying global net primary production of more than 160 x 10° g C. This global figure is 60 percent higher than those global estimates based on summation of individual ecosystem measurements. Another approach‘ has used a com- prehensive set of satellite indices of photosynthetic activity in the ocean and on land, combined with a model of primary production, to generate a more integrated global estimate of NPP. There is considerable spatial and temporal variability, but on average annual NPP on land amounts to around 56 x 10" g C, while that in the sea is around 48 x 10" g C {see Table 1.1). The carbon cycle and global biomass estimates Carbon fixation by photosynthesis forms one crucial step in the carbon cycle®. Once fixed, the carbon will remain for a greater or lesser period within living tissues, that is, form part of the planet's biomass. As all cells and individual organisms have a limited lifespan, eventually the carbon will rejoin the non- living carbon pool {see Table 1.2]. It may, however, remain as organic carbon com- pounds for a far greater period than it remained part of the biomass; e.g. the woody tissues of Paleozoic forests were formed several hundred million years ago but remain, fossilized, as a source of coal and oil. Eventually all carbon will recycle through the inorganic pool, as carbon dioxide in the atmosphere, in the soil or dissolved in the sea, or as inorganic carbon compounds (car- bonates]) in rocks or dissolved in the sea. The great majority of carbon at any one time lies within the lithosphere, around 80 percent as carbonate and the remainder as organic carbon compounds. A large propor- tion of this carbon is effectively inaccessible to the biosphere in the short term but itself participates in the overall carbon cycle, mainly through tectonic activity. As seafloor crust is gradually consumed along subduction plate margins, carbon sediments are taken into the Earth's mantle and later released as carbon dioxide by volcanic and hydrothermal activity. Although the loss of carbon to the mantle is extremely slow, without volcanic activity tectonic processes would eventually exhaust the available carbon pool. Most of the carbon incorporated in living organisms is associated with green plants, and almost all of this is in the form of cellulose-rich woody tissues. The total terrestrial animal biomass appears to be insignificant in comparison, probably more than two orders of magnitude less. The cyanobacteria and algae that are the primary producers in the ocean are estimated‘ to amount to only 0.2 percent of the biomass of all primary producers globally, although they generate in the region of half the global NPP {they cycle organic material much more rapidly than land plants, which also sequester large amounts in woody tissues). The role of diversity in the biosphere The biota play the pivotal rile in the major biogeochemical cycles, with different groups of organisms [e.g. nitrogen-fixing bacteria and photosynthesizing plants) mediating different processes. Simplistically, therefore, at least some biological diversity is necessary to maintain the biosphere as it currently operates, and microbial diversity is partic- ularly fundamental. However, just how much diversity is needed, and how much redun- dancy, if any, is built into the system, remains unclear. Indeed, the relationship between biological diversity and a whole suite of ecological measures including stability, resilience and productivity remains incom- pletely understood’, although there is an increasing volume of theoretical and ex- perimental work”® indicating that diversity may play an important role in long-term ecosystem functioning. Human influence on the biosphere There are believed to be more than 6 billion humans on the planet at present. A significant proportion of global net primary production is diverted to support this population. Using a relatively conservative definition of appro- priation that takes account of the global agricultural and natural production used by humans’, that proportion has recently been estimated at around one third of the terrestrial global total’. This result is very similar to that obtained in an earlier study", but because of uncertainties in almost all the figures on which it and the earlier estimate were based, the margin for error is very high. Less conservative definitions also attempt to take into account the loss of overall net primary production that may result from human actions [e.g. through severe land degradation or accumulation of waste) but which is not directly used by humans. This is done by estimating the net primary pro- duction of what is thought would have been the prevailing biomes in the absence of humans. A detailed analysis in one European country using this approach” ’ suggested that around 50 percent of NPP there was appropriated by humans. Human efforts to appropriate the pro- ducts of photosynthesis and other actions Total carbon content on Earth Amount buried in sedimentary rocks: organic carbonate Active carbon pool near surface: of which Dissolved inorganic carbon in sea Atmospheric CO, Organic carbon in soil Biomass on land Biomass in the sea associated with the development of complex societies have had enormous impacts on natural biomes and biogeochemical cycles. Over large areas of the Earth's surface, humans have replaced complex and species- rich natural habitats with simplified modified habitats specialized for agricultural production. The biosphere 10%g 1.6x 107g 6.5 x 107g 40 000 x 10'°g 38 000 x 10"g 750 x 10° g 1500 x 10g 560 x 10°g 5-10 x 10"°g Table 1.2 Estimated global carbon budget and biomass totals Note: Biomass figure on land refers to plants Source: Adapted from Schlesinger 11 12 WORLD ATLAS OF BIODIVERSITY Clearance by fire, burning of fuelwood and charcoal, soil cultivation, and fossil fuel use all increase movement of organic carbon into the atmosphere. Global cycling of nitrogen, phosphorus and sulfur has also been per- turbed. Application of industrially produced fertilizer has doubled the rate at which nitro- gen in fixed form enters the terrestrial cycle, and industrial processes have doubled movement of sulfur from the lithosphere into the atmosphere. Increasing levels of nitrogen and phosphorus lead to shifts in nutrient availability which can cause radical change in natural communities, and sulfur is a major contributor to acidification phenomena. That human activities may have profound local impacts on natural biota is indisputable. What is now becoming clear is that these REFERENCES activities may also have planet-wide impacts, particularly on climate. Analysis of atmos- pheric samples trapped in polar ice cores indicates that present-day concentrations of atmospheric carbon dioxide and methane (CH,} are unprecedented in the past 420 000 years”. Although their absolute concentration in the atmosphere is low (CO, around 360 and CH, at 1.7 parts per million by volume) these two gases play an extremely important role in determining atmospheric temperature. It is indisputable that the rise in these gases is a result of human activities, so clearly these activities are having some impact on global climate. The extent of this impact, particularly when compared with natural climatic fluctuations, remains a subject of great controversy. 1 Hutchinson, G.E. 1970. The biosphere. Scientific American 233(3): 45-53. 2 Lenton, T.M. 1998. Gaia and natural selection. Nature 394: 439-447. 3 Parkes, R.J. 1999. Oiling the wheels of controversy. Nature 401: 644. 4 Field, C.B. et al. 1998. Primary production of the biosphere: Integrating terrestrial and oceanic components. Science 282: 237-240. 5 Schlesinger, W.H. 1997. Biogeochemistry: An analysis of global change. 2nd edition. Academic Press, San Diego and London. 6 Schulze, E.D. and Mooney, H.A. (eds) 1993. Biodiversity and ecosystem function. Springer, Berlin. 7 Grime, J.P. 1997. Biodiversity and ecosystem function: The debate deepens. Science 277: 1260-1261. 8 McGrady-Steed, J., Harris, P.M. and Morin, P.J. 1997. Biodiversity regulates ecosystem predictability. Nature 390: 162-165. 9 Wright, D.H. 1990. Human impacts on energy flow through natural ecosystems, and implications for species endangerment. Ambio 19(4): 189-194. 10 Rojstaczer, S., Sterling, S.M. and Moore, N.J. 2001. Human appropriation of photosynthetic products. Science 294: 2549-2552. 11 Vitousek, P.M. et al. 1986. Human appropriation of the products of photosynthesis. BioScience 36: 368-373. 12 Haberl, H. et al. 1999. Colonizing landscapes: Human appropriation of net primary production and its influence on standing crop and biomass turnover in Austria. /FF-Social Ecology Papers No. 57. Institute for Interdisciplinary Research of Austrian Universities, Vienna. Also see: http://www.cloc.org/conference/presentations/in4/npp-abstract.htm 13 Schulz, N. 1999. Effects of human land-use on the amount of biologically available biomass-energy throughout the landscape. An empirical case study about Austria. Conference paper available online in pdf at: http://www.univie.ac.at/iffsocec/conference?9/htmlfiles/blue.html {accessed January 2002). 14 Petit, J.R. et al. 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 399: 429-436. The diversity of organisms 2 The diversity of organisms congruent with the branching pattern of evolution. There is no single operational definition of what a species is, and taxonomy at all levels is subject to change as a result of new methods and data, but species diversity of better known organisms can often be assessed with useful accuracy. Globally, about 1.75 million species have been described and named, but the total including undescribed species might be up to ten times greater. All species known are assigned on the basis of shared patterns of form and function to one of about 100 major groups (phyla). There are marked differences between phyla in overall morphology, physiology and mode of life. These differences imply the existence of major genetic diversity, and contribute directly to structural, trophic and other dimensions of diversity within ecosystems. The phyla of living organisms fall into three primary lineages: the true bacteria, the archeans and other organisms. The first two are prokaryotes, the remainder (protoctists, S YSTEMATICS AIMS TO DEFINE SPECIES and sort them into a hierarchy of named groups animals, fungi and plants) are eukaryotes. EVOLUTION AND SYSTEMATICS A basic principle in evolution is that just as new individuals arise from ancestral individuals, so new populations arise from existing popu- lations, and ultimately new species arise from existing species. The chief mechanism by which this occurs is believed to be reproductive isolation. For example, physiographic or climatic change may divide an existing single population into two or more separate popu- lations, or individuals may colonize a new and geographically separate habitat. The genetic makeup of these isolated populations will diverge, mainly through natural selection acting on them, but probably also through other mechanisms. This genetic divergence may be manifested in various ways, physically, physiologically and behaviorally. If the period of isolation continues for long enough, the populations will diverge enough that they can be regarded as separate species. Each one of these species may itself in turn give rise to other species in due course, although some will die out without giving rise to any progeny. The surviving descendant species may themselves give rise to new species and so on through the long march of evolutionary time. The result of this process is a branching tree-like structure - a phylogenetic tree - rooted in the distant past. In archeans and bacteria, but probably rarely in other groups, elements of two separate branches may combine into one, giving a reticulate rather than exclusively tree-like pattern. If it is assumed that all life on Earth had a common origin in the distant past {see Chapter 3], then all existing organisms form the topmost extremities of a vast and unimaginably complex single phylogenetic tree. Systematics has two roles'®. The first is to name the immense variety of different sorts of organisms that exist. The second is to try to elucidate the relationships between all these different organisms, that is to develop hypotheses of where they are positioned in the phylogenetic tree. Systematics provides the basic framework for the whole of biology, and is a fundamental discipline for bio- diversity studies. Taxonomy is the subset of systematics that deals in particular with the 13 14 WORLD ATLAS OF BIODIVERSITY — a aa a I Relationships between organisms have to be inferred from their genetic, morphological, biochemical or behavioral characteristics. definition, naming and classification of species and, in some cases, subspecific populations. The traditional output consists of species descriptions or revisions, or lists of species in a given group (possibly with hypotheses of their evolutionary relation- ships), or checklists of all the species in some higher taxon in a site or region. Because the actual evolutionary events that generated the overall phylogenetic tree are lost in history, the relationships between organisms have to be inferred from the evidence to hand. The most important forms of evidence are the characters of organisms, both living and fossil. These characters may be genetic, morpho- logical, biochemical or behavioral. Methods to reconstruct phylogeny generally use two working assumptions: that species sharing a large number of characters are likely to be related, and that species sharing some uniquely complex and specialized feature are likely to be more closely related than species not possessing this feature. Groups and names In the current system for naming species (nomenclature), each has a two-part sci- entific name (binomial), based on Latin or latinized Greek, comprised of the genus name (e.g. Vipera) and specific epithet (e.g. berus). The author of the specific epithet may be given after the binomial. By convention, both parts of the binomial are italicized when printed, and the author name is shown in parentheses if the species was originally put into a different genus. Similar species (e.g. the European adder Vipera berus and asp viper Vipera aspis) are grouped together in the same genus [Vipera], similar genera in families (Viperidae}, families in orders (Serpentes), orders in classes (Reptilia) and classes in phyla (Craniata or Vertebrata) up to the highest level, the kingdom. An organism can only be assigned to a single species, genus, family, etc., and the taxonomic system forms a hierarchy with each lower taxonomic level being nested entirely within each increasingly inclusive higher level. Although the traditional Linnaean hierarchy includes only the seven obligatory categories above, intermediate categories are some- times used, and a further more inclusive category - the domain. Groups such as mammals or snakes, that because of shared unique characters are considered to contain all the living descen- dants of a common ancestor, are called monophyletic groups. Groups with no shared unique characters but only unspecialized or non-unique characters in common, are termed paraphyletic groups. These typically are groups of related species left over after one or more clearly monophyletic lineages that evolved within the group have been recognized and named. Examples are fishes (the craniate vertebrates without the unique features of tetrapods], reptiles (the amniote vertebrates without the unique features of birds or mammals], and the entire kingdom Protoctista. Groups defined on characters that appear to have evolved more than once are called polyphyletic groups. Although the goal of most systematists is to recognize only monophyletic groups in order to be able to retrieve evolutionary relationships from a classification, many paraphyletic groups, such as the three named above, continue to be very widely used in practice. Given a classification congruent with phylogeny, the taxonomic hierarchy becomes a device to store information on hypotheses about evolutionary history. Because hypoth- eses about relationships are always subject to revision as new information becomes available, or existing data are reinterpreted, the taxonomy of species is not fixed. The fact that species names are liable to change can cause confusion, for example when, as is commonly the case, conservation legislation uses a name no longer current. Species concepts and diversity assessment Despite the importance of ‘the species’, there is no unequivocal and operational definition of what a species is or how species can be recognized“°. There are several definitions, differing in mainly theoretical and often subtle ways, but much of the existing body of systematic knowledge has been built up around elements of ‘the biological species concept’. This defines a species as a population of organisms that actually or potentially interbreed in nature, and that are reproductively isolated by morphological, behavioral or genetic means from other such groups. It is, however, applicable only to organisms where sexual reproduction is the norm. In most real cases, especially where all the systematist has to hand is a collection of preserved specimens, whether criteria concerning reproductive isolation are met or not cannot in fact be tested, but an experienced worker will come to hold some particular level of morphological or other difference as deserving of species status. Where there is good evidence from fieldwork and geographic data attached to specimens that two somewhat similar populations occur in the same locality [i.e. are sympatric] but maintain their differences, they may be presumed not to interbreed, and will be treated and named as species. Different taxonomists will often use different criteria for the same group of organisms, so that one specialist may regard a group of fundamentally similar populations as a single species, whereas another will treat each smaller distinct population as a separate species. In the latter case, often associated with the ‘phylogenetic species concept’, the assumption is that each distinct lineage once established on its own evolutionary course is de facto a separate species. Many populations that were formerly described at subspecies level [i.e. somewhat distinct, but regarded as part of a single polytypic species complex) have subsequently been elevated to full species in this way, particularly if geo- graphically isolated. However, subspecies have often been named on the basis of just a few superficial features, not representative of the overall pattern of variation within species, and the formal subspecies category is now much less used than in the past (particularly among vertebrates). Different taxonomic criteria are used to classify species in different groups of organisms. Those used, for example, to define species of fungi are very different from those used to define species of bird, and their application characters and The diversity of organisms 15 demands taxonomists with relevant specia- lized knowledge. Some organisms are diffi- cult or logically impossible to accommodate in any species concept involving criteria that assume all individuals reproduce by out- breeding, that is by sexual reproduction with another individual. Many higher plants are self-fertile or make use of various forms of asexual reproduction; in the latter case all individuals in a lineage are identical clones {assuming a very low or non-existent mutation rate]. Among the prokaryotes, bacteria can readily receive extraneous genes through direct entry of genetic material from the fluid environment, or from viruses or other bacteria’, and this ‘horizontal’ transfer is independent of reproduction. Different kinds of bacteria have traditionally been defined by the cytological or biochemical properties of colonies in culture but, more Different taxonomic characters and criteria are used to classify species in different groups of organisms. 16 WORLD ATLAS OF BIODIVERSITY i SSIS recently, comparison of DNA and RNA sequences from sample collections has been | used to distinguish one lineage from another. In either case, the biological species concept, developed with reference to sexually reproducing, outbreeding animals and plants, is not appropriate. Such factors mean that the species’ cannot provide a standard unit in which to evaluate all biodiversity because it does not define a single level in the hierarchy and its significance is not equivalent across all groups of organisms; it can, however, serve this purpose for most plants and animals. The use of different species concepts by different systematists can make a very large difference to the number of species recognized in a group and to complications in nomenclature (these will both affect the outcome of biodiversity inventory, an The discovery of entirely new species of mammals and birds is rare, and often involves small and obscure forms. Remarkably, two large mammals previously unknown to science were discovered in one small area, the Vu Quang Nature Reserve in Truong Son, Viet Nam (along with many new species in other groups). The Vu Quang ox or soala Pseudoryx nghetinensis was described in 1993, followed a couple of years later by a giant muntjac deer Megamuntiacus vuquangensis from the same area. The world’s smallest muntjac deer, the Truong Son muntjac Muntiacus truongsonensis, was recently found in another part of the same region in Viet Nam. The soala is of particular interest because it does not appear to fit neatly in any of the main bovid groups as currently recognized. It is now known also to occur in adjacent parts of Laos. However, claims that another new bovid species existed in Southeast Asia, described as Pseudonovibos spiralis, were premature because the distinctive horns described as new were later shown to be domestic cattle horns, apparently reshaped artificially. important application of systematics). Usually, however, only a small or very small number of taxonomists is working on any one group of organisms at any time, so that while the species level taxonomy of organisms is in a continual state of flux, it is generally not subject to radical and wholesale change. Exceptions arise where new techniques, chiefly molecular infor- mation and DNA data in particular, are applied to previously neglected groups or ones that have not been revised for many years. Nevertheless, the key point appears to be that units corresponding more or less closely to the biologists’ model of ‘the species’ do indeed exist in nature and, in animals in particular, they define themselves to an extent through their reproductive behavior. It has thus been possible to reach some measure of consensus on species-level classification of well-studied groups of larger organisms such as terrestrial vertebrates, and to estimate and compare the number and kinds of species in different sites, areas or countries. NUMBERS OF LIVING SPECIES From a practical point of view it is more important to know how many species, and which ones, occur in some Spatially restricted area, such as a protected area or a country, than in the world overall. However, proper evaluation of each local situation requires some knowledge of the wider context and, where the goal is maintenance of global biodiversity in the face of increased risk, it is clearly important to have a sound appreci- ation of the full baseline range of diversity. This requires both an estimate of the number of known valid species, and an estimate of the number of unknown species, neither of which is readily available. The number of known species can be estimated by collating data from systematists and the taxonomic literature. Although many species names are synonyms (i.e. different names inadvertently applied to the same species), this can be done with reasonable precision for more familiar and well-reviewed groups of species. Recent calculations of this kind suggest that around 1.75 million of the probably far larger number that exist have been discovered, collected and later named by systematists”’. Any estimate of how many undiscovered and hence undescribed species are likely to exist in any given group, and in the biosphere overall, involves substantial uncertainty”®. In taxonomic groups where individuals are readily visible, popular or economically important, and subject to sustained system- atic attention, e.g. mammals and birds, the number of known species is certainly very close to the total number of species in the group that exist. On average around 25 new species of mammals and and five of birds have been described annually in recent years’. Changing systematic opinion on which populations should be regarded as separate species and which should not, rather than completely new discoveries, is the major source of change in the number of named species in such groups. The converse applies to groups whose individuals are small, difficult to collect, obscure and of no popular interest, e.g. many groups of invertebrate animals. Frequently there are so few systematists actively working on a group that the number of named species appears to be limited mainly by the rate at which collected specimens waiting on museum shelves can be studied and described, and changing opinion on which populations are separate species is insig- nificant. In some cases, where new sampling and collection methods have been used, unexpectedly large numbers of new species have been found (e.g. tropical forest canopy insects and marine sediment nematodes). If findings from such local work are extrapolated to global level the total number of species calculated to exist is many orders of mag- nitude greater than the number actually known. Some estimates suggest that most undescribed terrestrial forms are likely to be tropical forest beetles, but new molecular techniques are revealing unsuspected divers- ity among microorganisms’. Although the goal of systematics is to recognize and name species, and to maintain an ordered body of information on names and associated biological data, there is no master catalog of all known species. Developing such a resource has only become feasible with advances in information technology during the past ten years. However, while many systematic data, in the form of checklists and museum catalogs, are now available in digital form over the Internet, and more will become so, a harmonized catalog in this format of all known species remains a distant prospect. Recent estimates of the numbers of known and possibly existing species in the world biota are given in Table 2.1. These are mostly large numbers, and the fossil record suggests The diversity of organisms 17 \ I EL ST | LO Hy that overall diversity has been increasing for | some 600 million years up to the very recent | past, but the numbers themselves are of little significance except in a wider context. | Currently, much concern is focused on species numbers in relation to anthropogenic environmental change. The Convention on Biological Diversity decided in 1998 to establish a Global Taxonomic Initiative (GTI), in recognition of a major ‘taxonomic impediment’ to effective biodiversity management. The objective is to improve decision-making for conservation, sustainable use and benefit-sharing by increasing taxonomic knowledge and the number of trained taxonomists and curators. Similarly, the Global Biodiversity Information Facility (GBIF)" aims to develop an interoperable network of biodiversity databases and information technology tools that will enable users to access the world’s stores of biodiversity information. In the same field, Species 2000” {a global network based in the UK and Japan}, and the Integrated Taxonomic Information System (ITIS]" in North America have joined forces to create a unified Catalogue of Life, planned to cover all known species of living organisms. Basic reference data on 250 000 species had been collated by mid-2001, and the plan is to reach 500 000 by 2003. DIVERSITY AT HIGHER LEVELS Until van Leeuwenhoek observed micro- organisms through a primitive microscope in the late 17th century, humans had been aware only of organisms visible to the naked eye (macroscopic) and regarded all living things as either plants or animals. At the end of the 19th century, with improved cytological techniques and new views on evolution, a third kingdom of organisms (Protista) was recognized for bacteria and other unicellular organisms. Around this time it became widely accepted that the cell was the fundamental unit of organization of all living organisms. Subsequent work, in the mid-20th century, recognized a basic dis- tinction between two kinds of cellular organization - prokaryotic and eukaryotic. In prokaryotic organisms, the genetic material is free within the cell. In eukaryotes, the genetic material is linked to proteins and organized into chromosomes that are packed within a membrane-bounded cell nucleus. There are several other profound differences. In eukaryotes the enzymes needed to extract energy from organic molecules are organized 18 WORLD ATLAS OF BIODIVERSITY Table 2.1 Estimated numbers of described species, and possible global total into discrete membrane-bounded organelles (mitochondria) within the cell, and in the eukaryotes that photosynthesize (plants, some protoctists} the pigments and enzymes needed to fix solar energy are also in discrete organelles (chloroplasts) within the cell. On the basis of this, it became accepted that the major taxonomic divide in organisms was between the prokaryotes, containing only the bacteria, and eukaryotes, which included all other organisms, including many unicellular forms included in the original kingdom Protista. Systematics has traditionally generated information on similarity and hypotheses of relationship on the basis of characters restricted to particular sectors of the phylogenetic tree, but molecular sequencing has broadened the range of useful evidence, Archaea 175 7 Bacteria 10 000 ? Eukarya Animalia 1 320 000 Craniata (vertebrates), total 52 500 55 000 Mammals” 4 630 Birds* 9 750 Reptiles® 8 002 Amphibians* 4950 Fishes” 25 000 Mandibulata [insects and myriapods) 963 000 8 000 000 Chelicerata (arachnids, etc.) 75 000 750 000 Mollusca 70 000 200 000 Crustacea 40 000 150 000 Nematoda 25 000 400 000 Fungi 72 000 1 500 000 Plantae 270 000 320 000 Protoctista 80 000 600 000 I with the potential for radically dissimilar groups to be compared and their phylogeny estimated. Analysis of small subunit ribo- somal RNA (SSU rRNA] - a molecule that is universal, functionally constant (central to protein manufacture in all cells) and very highly conserved over time - has proved especially informative. This work has revealed that there are two kinds of prokaryote: the true bacteria and the archeans. The archeans were first known only by ‘extremophiles’, i.e. forms living under exceptional conditions of high temperature or salt concentration, but repre- sentatives are now known to be widespread alongside bacteria in less extreme habitats. At biochemical level, the bacteria and archeans are as different from each other as from eukaryotes, leading to the conclusion that all organisms can be assigned to three basic Notes: This table presents recent estimates of the number of species of living organisms in the high- level groups recognized, and in some selected groups within them. Vertebrate classes are distinguished because of the general interest in these groups. The described species column refers to species named by taxonomists. Most groups lack a formal list of species. All estimates are approximations. They are inevitably inaccurate because new species will have been described since publication of any checklist and more are continually being described, and other names turn out to be redundant synonyms. In general, the diversity of microorganisms, small-sized species, and those from habitats difficult to access, are likely to be seriously underestimated. Among Archaea and Bacteria the figures of 175 and 10 000 are very rough estimates of ‘species’ defined on features shown in culture’; there appears to be no sound estimate of the total amount of prokaryote diversity. The estimated total column includes provisional working estimates of the number of described species plus the number of unknown species; the total figure is highly imprecise. Only a small selection of animal phyla is shown, but the figure for Animalia applies to all. Figures in the total row are for all species in all domains. Source: Data mainly from United Nations Environment Programme’ and Hammond’; vertebrates from individual sources indicated. BACTERIA forms of life, or domains: Archaea, Bacteria and Eukarya, rather than two (prokaryotes and eukaryotes]. The distance between lineages, in terms of amount of change in rRNA sequence, and their branching sequence in evolution, have been represented as a ‘universal phylogenetic tree’. Where the root of the tree is located, and its basal branching pattern, i.e. which two of the three domains are more closely related than either is to the third, remains open to discussion’. It has been argued” that this deep branching took place in very early evolution, possibly before modern cell types had been established, when genes were widely spread by horizontal transfer {as persists to a lesser extent among extant prokaryotes) and in a sense, shared communally. Reticulate evolution would have been common, with distinct lineages of organisms emerging only when the evolving cell became sufficiently integrated that horizontal gene transfer was reduced and vertical transfer down closed ancestor-descendent series became the norm. It has also been per- suasively argued that the Eukarya arose through the permanent symbiotic fusion of a number of different prokaryotic organisms in one cell, with organelles such as mitoc- hondria and chloroplasts representing the vestiges of different lineages of formerly independent prokaryotes* ”. ARCHAEA The diversity of organisms 19 EUKARYA Animals Fungi Plants PROTOCTISTS Within the Eukarya, the fungi, animals and plants form insignificant clusters over- shadowed by the morphological and physio- logical diversity of the protoctists (protists). Protoctista is a name of convenience for the enormous and very diverse collection of all the small-to-microscopic eukaryotes that lack the distinguishing features of fungi, animals or plants. Further changes of pers- pective are expected as research on protoctist systematics continues. For example, a recent tendency” is to recognize the Chromista as another major group distinct from remaining eukaryotes. The chromists are aquatic species distinguished by structural and biochemical characters, and include organ- isms with the largest linear dimensions known (kelp), as well as microscopic but ecologically important organisms (e.g. diatoms, downy mildews). The only known organisms that are not cells, or assemblages of cells, are viruses. They exist on the very boundary of most definitions of life. Consisting only of nucleic acids and protein, they are much smaller than the smallest bacteria, they can only replicate inside other living cells, and they are totally inert outside other cells, when they can survive for years ina crystallized state. Each type of virus may be more closely related to the organism in which it grows than to other viruses’. They are not discussed elsewhere in this book. Some characters of the high level group- Figure 2.1 The phylogenetic tree Note: This diagram represents in highly simplified form the distance between the three domains of organisms and the general branching pattern within them; because of conflicting interpretations the root and branching sequence of the three domains are not represented. Source: Adapted from Woese”® 22 WORLD ATLAS OF BIODIVERSITY a RR Table 2.2 Key features of the major groups of living organisms Archaea Prokaryotic. Composition of the cell wall and of lipids in cell membranes differ from those in Bacteria. Distinctive SSU rRNA, more similar in some respects to Eukarya than to Bacteria. Reproduce asexually by cell splitting, or produce genetic recombinants without any fusion of cells by the Eukarya are commonly accepting genes from other bacteria, or from the fluid medium, or through mes CE Penae. of regarded as kingdoms, the cell reproduction. Flourish in habitats with radical extremes of temperature or salinity that are highest formal category of unavailable to other organisms (apart from bacteria in some cases}, but also occur in other the Linnaean system of environments. taxonomy. The Archaea and Bacteria have been treated as a single group, or as separate kingdoms, but a more recent tendency is to treat Archaea, Bacteria and Eukarya as Note: The four groups within Bacteria Prokaryotic. Reproduce asexually by cell splitting, or produce genetic recombinants without any fusion of cells by accepting genes from other bacteria, or from the fluid medium, or through viruses, independent of cell reproduction. Metabolically uniquely versatile; key mediators of major three separate domains in | biogeochemical cycles. Permeate the entire biosphere, including other organisms, although dominant recognition of the deep only in exceptional habitats. phyletic divergence between them, and fundamental Eukarya differences in RNA composition Rearetien Source Piocipally Matguleiand A Multicellular, mainly macroscopic, eukaryotes. Reproduce through fertilization of an egg by a Schwartz’, also University of California ee ‘ aap s fe and/University of Arizona’, sperm, the fertilized egg (now diploid, i.e. a duplicate set of chromosomes] is called a zygote and {except sponges] this forms a characteristic hollow multicelled blastula from which the embryo develops. All heterotrophic. Fungi Mainly multicellular, micro- to macroscopic eukaryotes. Fungi develop directly without an embryo stage from resistant non-motile haploid (one set of chromosomes] spores that can be produced by a single parent. Sexual reproduction also results in haploid spores. Most consist of a network of threadlike hyphae. Heterotrophs, vital to decomposition processes; form mycorrhizal symbioses with plants, facilitating exchange of soil nutrients. | Plantae Multicellular macroscopic eukaryotes. The fertilized egg develops into a multicelled embryo different from blastula of animals. Alternate spore-producing generations and egg or sperm- producing generations. Virtually all are terrestrial photosynthetic autotrophs. Protoctista | Mainly microorganisms. Possess the features of eukaryotes, but lack the characteristics of fungi, animals or plants. Extraordinary variation in life cycle and morphology. Early evolution probably based on symbiotic relationships between different kinds of bacteria forming lineages of composite organisms resulting in the protoctist grade of organization. Include photosynthetic algae (formerly classed as plants) and heterotrophs (formerly called ‘protozoa’). ings of organisms [domains and kingdoms] is presented in Appendix 1. The objective of | are outlined in Table 2.2. A synopsis of this material is to provide a convenient | information on each of the 96 phyla recognized overview of global organismal diversity, in in the most recent comprehensive synthesis’ terms of higher taxon diversity. The diversity of organisms a1 REFERENCES 1 Minelli, A. 1993. Biological systematics: The state of the art. Chapman and Hall, London. 2 Vane-Wright, R.|. 1992. Systematics and diversity. In: World Conservation Monitoring Centre. Groombridge, B. (ed.} Global biodiversity: Status of the Earth's living resources, pp. 7-12. Chapman and Hall, London. 3 Mayr, E. and Ashlock, P.D. 1991. Principles of systematic zoology. McGraw-Hill Inc., New York. 4 Gaston, K.J. fed.) 1996. Biodiversity: A biology of numbers and difference. Blackwell Science Ltd, Oxford. 5 Vane-Wright, R.|. 1992. Species concepts. In: World Conservation Monitoring Centre. Groombridge, B. (ed.] Global biodiversity: Status of the Earth's living resources, pp. 13-16. Chapman and Hall, London. | 6 Margulis, L. and Schwartz, K.V. 1998. Five kingdoms. An illustrated guide to the phyla of life on earth. 3rd edition. W.H. Freeman and Company, New York. 7 United Nations Environment Programme 1995. Heywood, V. ed.) Global biodiversity | assessment. Cambridge University Press, Cambridge. | 8 Hammond, P. 1992. Species inventory. In: World Conservation Monitoring Centre. Groombridge, B. {ed.) Global biodiversity: Status of the Earth's living resources, pp. 17-39. | Chapman and Hall, London. 9 Williams, D.M. and Embley, T.M. 1996. Microbial diversity: Domains and kingdoms. | Annu. Rev. Ecol. Syst. 27: 569-595. 10 GTI. http://www.biodiv.org/programmes/cross-cutting/taxonomy/ (accessed January 2002). 11 GBIF. http://www.gbif.org/ [accessed January 2002). 12 Species 2000. http://www.sp2000.org/ [accessed January 2002). 13 ITIS. http://www.itis.usda.gov/ (accessed January 2002). 14 Woese, C.R., Kandler, 0. and Wheelis, M.L. 1990. Towards a natural system of organisms: Proposal for the domains Archaea, Bacteria and Eukarya. Proceedings of the National Academy of Sciences 87: 4576-4579. 15 Woese, C.R. 2000. Interpreting the universal phylogenetic tree. Proceedings of the National Academy of Sciences 97: 8392-8396. 16 University of California, Berkeley, Museum of paleontology. http://www.ucmp.berkeley.edu/alllife/threedomains.html (accessed January 2002). 17 University of Arizona, Tree of Life project. http://phylogeny.arizona.edu/tree/phylogeny.html (accessed January 2002). 18 Margulis, L. 1998. The symbiotic planet: A new look at evolution. Weidenfeld and Nicolson, London. 19 Philippe, H. and Forterre, P. 1999. The rooting of the universal tree of life is not reliable. Journal of Molecular Evolution 49: 509-523. 20 Woese, C.R. 1998. The universal ancestor. Proceedings of the National Academy of Sciences 95: 6854-6859. 21 http://www.ucmp.berkeley.edu/chromista/chromista.html (accessed January 2002). 22 Woese, C.R. 1998. Default taxonomy. Proceedings of the National Academy of Sciences 95: 11043-11046. 23 Wilson, D.E. and Reeder, D.M. (eds) 1993. Mammal species of the world: A taxonomic and geographic reference. 2nd edition. Smithsonian Institution Press, Washington DC and London. Available (accessed January 2002) in searchable format at http://nmnhwww.si.edu/msw/ Covers existing or recently extinct species, as of book publication year. 24 Clements, J.F. 2000. Birds of the world: A checklist. Pica Press, Sussex. Other lists are available online, e.g. a list based on Sibley, C.G. and Monroe, Jr., B.L. 1990. Distribution and taxonomy of birds of the world. Yale University Press, New Haven and London is available (accessed January 2002) at http://www.itc.nl/~deby/SM/SMorg/sm.html 22 WORLD ATLAS OF BIODIVERSITY SS SSS SSS PSS TR 25 http://www.embl-heidelberg.de/~uetz/LivingReptiles.html (accessed January 2002). 26 Estimate is sum of species numbers for Anura [4 701), Caudata (505) and Gymnophiona (caecilians] (159) retrieved from online resource Amphibia Web at http://elib.cs.berkeley.edu/aw/lists/index.shtml (accessed January 2002). Also see Duellman, W.E. 1993. Amphibian species of the world: Additions and corrections. Special Publication No. 21, University of Kansas Museum of Natural History, Lawrence, Kansas, and database with later revisions available online in searchable format at http://research.amnh.org/herpetology/amphibia/ {accessed January 2002). 27 Eschmeyer, W.N. et al. 1998. A catalog of the species of fishes. Vols 1-3. California Academy of Sciences, San Francisco. Available online in searchable format at http://www.calacademy.org/research/ichthyology/catalog/fishcatsearch.html (accessed January 2002). Biodiversity through time 3 Biodiversity through time hand, new groups of organisms appear, diversify and generally persist for very long periods of time; on the other, most such groups and their included species eventually cease to exist. Most analyses of the fossil record show an erratic rise in overall biodiversity, increasing through the Mesozoic and Cenozoic and reaching a peak around the end of the Tertiary. However, diversity has been greatly reduced during several periods of radical environmental change during each of which more than half the multicellular species then living became extinct. Such mass extinction phases have provided important new opportunities for diversification in remaining lineages, and the spread of new communities. Evidently, the large number of species existing now on Earth is the result of a modest net T WO FUNDAMENTAL PATTERNS CAN BE DISTINGUISHED in the fossil record. On one excess of originations over extinctions during the 3 800 million years of evolution of life. THE FOSSIL RECORD Knowledge of the history of diversity through geological time is based on analysis of the fossil record. When fossils, as the inert mineralized parts or casts or imprints of dead organisms, are interpreted in a biological context they provide the only direct evidence of the history of life on the planet. The fossils discovered and described by paleontologists represent more than a quarter of a million species, virtually all of them now extinct’, but these are believed to make up only avery small fraction of all the species that have ever existed. For example, the fossil record of marine animals is far more comprehensive than that of terrestrial forms, but the marine sample is estimated to represent only about 2 percent of all the marine animals that have lived’. The fossil record overall may represent as little as 1 percent, or less, of all the species that have existed’. Clearly, statements about broad patterns in the evolution of life, and the ascendancy or extinction of groups of organ- isms, thus rest on a very narrow base of tangible evidence. Macroscopic animals with hard skeletons that lived in shallow marine environments, where their remains could be buried by sediment, petrified, and later exposed in uplifted rock strata, are by far the most likely to be both preserved and found. The fossil record from the past 600 million years is thus dominated by mollusks, brachiopods and corals’. For many other kinds of organism, exceptional circumstances are required if a dead individual is to become preserved and found. With some exceptions, microscopic or soft-bodied organisms rarely leave discern- able traces of their existence, and larger organisms are usually decomposed, dis- assembled and never discovered. Terrestrial vertebrate fossils are often of individuals that must have been preserved by the smallest of chances, perhaps sudden burial during a natural disaster, or as a result of the body falling into a rock crevice out of reach of scavengers. All else being equal, the probability of preservation will rise greatly the more widespread and abundant the species is, and the longer it persists through time. Conversely, there is a very low probability that any individual of a numerically rare or restricted-range species with a short persistence time will die in circumstances conducive to fossilization and subsequently be found. Factors such as these mean that even the relatively better known 23 2, WORLD ATLAS OF Figure 3.1 The four eons of the geological timescale Arrows indicate approximate age of oldest confirmed fossils of the groups named. Source: Adapted from Margulis and BIODIVERSITY TT SS groups are certain to be very incompletely known. The plant and animal species now living include a substantial number of rare or local species and it is difficult to imagine many of them being represented in the fossil record of our time as recovered in the future’. Because the fossil record gives an incomplete and biased view of the past history of life, the reconstruction of that history has been the subject of great debate. It has been generally accepted that the record can give a reasonable insight into past diversity in terms of taxonomic richness**, particularly at higher taxonomic levels. On the other hand, recent analysis of a new database of fossil occurrence has indicated that sampling effects (reflecting variation in the nature of the fossil record and the way it is reported) can make a very significant contribution to the shape of global diversity curves. Preliminary results of this analysis have suggested, for example, that the widely accepted post-Paleozoic increase in marine diversity {see below] may to some extent be an artifact of the analytic methods used’. PATTERNS OF DIVERSIFICATION The early history of life In its earliest history, there was no life on the Earth. Now, there are about 1.75 million different species of all kinds known, with perhaps many times that number still unknown. Self-evidently, life has both arisen and diversified. The planet Earth is between 4 500 million and 5 000 million years old. The oldest known rocks are about 4 000 million years (My) in age, and mineral grains (zircon) of apparently ‘chemofossils’ suggest that life had evolved by 3 800 My ago’ but periodic intense ocean heating caused by extraterrestrial impacts is suspected to have restricted early organisms to thermophilic non-photosynthetic forms, possibly associated with hydrothermal vents’®. Most of the principal biochemical pathways key to the modern biosphere were probably in place by 3 500 My ago’. The first tangible evidence of cellular organisms themselves in the fossil record consists of filaments and spheroids, believed to be the remains of prokaryote microorganisms, in rocks of about this age, including traces from an apparent Precambrian submarine thermal spring system of about 3 235 My in age’. Stromatolites (rock domes formed in shallow waters from multiple layers of sedi- ment and bacteria) also first appear at around 3500 My ago and are the most abundant fossils so far known during the 3 000 My up to the start of the Phanerozoic eon, marked by the start of the Cambrian period around 545 My ago. For much of this time, bacteria-like organisms were the only known life forms; gross morphological diversity was very low and many kinds apparently persisted for hundreds of millions of years, some outwardly indis- tinguishable from existing forms. Probable microbial mats, similar to stromatolites but developed on an exposed soil surface, have been interpreted as the first evidence of terrestrial ecosystems around 2 600 My ago”. The next major step in the evolution of life was the development of eukaryotic organisms. Biochemicals characteristic of eukaryotes have been found in shales 2 700 My old", far Schwartz’ greater age have been reported. Carbon earlier than the first fossils, reported from Eon HADEAN | | | Millions of 4500 4000 3900 3500 3 000 2600 25 years ago i origin oldest ‘chemofossils’ oldest evidence of of Earth Earth possible microfossils eukaryotes rocks evidence of bacteria of life shales of around 1 500 My age”. These earliest eukaryote microfossils include a number of ‘acritarchs’ (widely interpreted as the cysts or resting stages of marine algae) some of which show cytological features characteristic of protoctists (see previous chapter]. Diversity in acritarchs, and the rate at which different forms replaced one another in the record, were low until around 1000 My ago when both species number and species turnover increased markedly”. Radiations around the early Phanerozoic boundary For many years it was assumed that animals originated in the Cambrian period at the base of the Phanerozoic [the Phanerozoic is the eon of time characterized by presence of animal fossils; it includes the Paleozoic, Mesozoic and Cenozoic). This is now known not to be the case, aS a wide range of fossil animals, including recognizable arthropods and possibly echinoderms, is now known from about 100 My before the Cambrian. Most fossils from this time, however, appear completely unrelated to extant forms, and consist mainly of enigmatic frond- and disc-shaped soft-bodied animals: the so-called Ediacaran fauna. The lower Cambrian marks a dramatic change from this early fauna, with the sudden appearance in the fossil record [e.g. the Chengjiang and Burgess Shale faunas) of a wide range of animals, many with calcareous skeletons. It is generally accepted that this represents a genuine explosion of diversity that took place over only a few million years, and is not an artifact of the fossil record. The lower Cambrian thus represents the most important PROTEROZOIC 2 000 1500 oldest protoctist fossils Biodiversity through time period of high-level diversification in the his- tory of animal life on Earth. These archaic invertebrates had by the end of the Cambrian period, around 500 My ago, established all the basic body plans seen in extant animals, and many others besides. Each such basic lineage is recognized taxonomically at phylum level, and the range of morphological diversity was higher at 500 My ago than at any time before or since. As many as 100 different animal phyla may have existed during the Cambrian” including every well-skeletalized animal phy- lum living today (except perhaps the Bryozoal, whereas in the latest synthesis all extant animals are placed in 37 phyla”. Plants and animals began to extend into terrestrial habitats during the first half of the Paleozoic, with the first fossil material known from the late Silurian, around 400 My ago. At this point approximately 90 percent of the history of life to the present had already passed. Fossils suggest low diversity for the next 100 My until the later Devonian period. No new animal phyla appeared with the colonization of land, millions of years after the initial Cambrian radiation of animal phyla. Diversity of marine animals in the Phanerozoic The overall pattern of diversity [assessed as numbers of families) appears to show a possible early peak around the start of Phanerozoic time, followed by a plateau of somewhat higher diversity extending through most of the Paleozoic era, and then, after the end-Permian mass extinction (see below) a steady increase in diversity over remaining geological time‘. Some recent methods of analysis’ giving special attention to sampling 1 000 545 500 t oldest MH oldest oldest animal fungus plant fossils fossils fossils 26 WORED AMLAS OR BiVODIMER SIM SS LL TE SE Figure 3.2 Periods and eras of the Phanerozoic This is an expanded version of the most recent segment of the geological timescale shown in Figure 3.1. The Phanerozoic is the eon of time extending from the base of the Cambrian, some 545 million years ago, to the present, and to which the entire fossil record was formerly thought to be restricted. Source: Adapted from Margulis and Schwartz", Cambrian base date from International Subcommission on Cambrian Stratigraphy website” procedures suggest this broad view may be subject to significant revision. Although the number of phyla has decreased markedly since the Cambrian, diversity at all lower taxonomic levels appears to have either increased overall or, in a few cases, remained more or less level. The number of orders of marine animals present in the fossil record climbed steadily through the Cambrian and Ordovician, slowing towards the end of the Ordovician to a figure of between 125 and 140, which has been maintained throughout the Phanerozoic. The diversity of marine families repre- sented in the fossil record shows a similar pattern of increase through the Cambrian (possibly falling during the latter half of the period] and Ordovician, leveling off at around 500, a figure which was maintained until the late Permian mass extinction. This extinction event resulted in the loss of around 200 families, but diversity increased subsequently to the modern level of around 1 100 families, with a number of temporary reversals during minor extinction events. The trend in number of species in the fossil record is even more extreme. From the early Cambrian until the mid-Cretaceous, the number of marine species remained low; since then, that is during the past 100 million years, it appears to have increased dramatically, perhaps by a factor of 10. (See Figure 3.3.) Diversity of terrestrial organisms in the Phanerozoic Although low to moderate peaks and troughs are evident in the record, the overall pattern of family diversity in terrestrial organisms shows a continuing rise from the Silurian to the present, thus differing somewhat from Period Ordovician Era Millions of 545 years ago 490 the pattern shown by many analyses of marine animals. It is generally accepted that vascular terrestrial plants first arose in the Silurian, although some paleobotanists argue for a late Ordovician origin, a time when microfossils suggest that bryophyte-like (non-vascular]) plants already existed". Diversity increased during the Silurian, and then more rapidly during the Devonian, owing to the first appear- ance of seed-bearing plants, leading to a peak of more than 40 genera during the late Devonian. Diversity then declined slightly, but started to increase markedly during the Carboniferous, with 20 families and more than 250 species in the mid-Carboniferous record of the northern hemisphere. Following this, diversity increased only slowly until the end of the Permian. There was a marked decrease in diversity at the end of the Permian, coinciding with or preceding the mass extinction of animal species. Subsequent increase in diver- sity was slow, reaching around 400 species in the early Cretaceous, but apparently more rapid from the mid-Cretaceous. This overall pattern masks important changes with time in the composition of the flora, most notably in the relative importance of the three main groups of vascular plant: the pteridophytes, gymnosperms and _ angio- sperms (see Figure 3.4]. The Silurian and early Devonian are marked by a radiation of primitive pteridophytes. During the Carboniferous, more advanced pteridophytes and gymnosperms developed and underwent extensive diversifi- cation. Following the late Permian extinction event, pteridophytes were largely replaced by gymnosperms [although ferns remained abundant) and these became the dominant group until the mid-Cretaceous. The dramatic Devonian 438 408 360 800 -- —— Marine animals 700;- ——— Fishes —— Insects 600;- ——= Tetrapods Number of families pS {=J oO T Biodiversity through time 27 a A TP I DOS I I I CALLS | I GEST E E LLBE E IIT I I EP EOE EI IT hs ETB I a TEE RTT \ } 600 500 400 J Pon | 300 200 100 0 Geological time (10° years) increase in plant diversity since then is entirely due to the radiation of the angiosperms, fossils of which first appear in the lower Cretaceous, although an early to mid-Jurassic origin, around 170 My ago, has been argued”. Colonization of land by animals has occurred many times; although the oldest body fossils of terrestrial animals date from the early Devonian, it is generally accepted that the primary period of land invasion by animals was the Silurian. The overwhelming number of described extant species of terrestrial animals are insects. Their fossil record is more extensive than might be expected, but had been little studied until recently. Data on insect diversity at family level have been collated, based on nearly 1 300 families”. This analysis shows a very slow increase in families from the first appearance of insects in the Devonian, a rise Permian 286 248 213 Jurassic in numbers during the Carboniferous, and a steeply increasing rise throughout the Mesozoic to the Tertiary. The apparent explosion in insect diversity had previously been attributed to ecological opportunities provided by the expansion of flowering plants, but the insects are now known to have begun their diversification some 150 My before the flowering plants”. The fossil record of vertebrates includes around 1 400 families, with tetrapods (amph- ibians, reptiles, birds, mammals) somewhat outnumbering fishes. The bird record is much less substantial than that for other groups, probably because their light skeletons have been less frequently preserved. Terrestrial vertebrates first appear in the fossil record in the late Devonian. Diversity remained relatively low during the Paleozoic, with around 50 families, and may have MESOZOIC | 144 Figure 3.3 Animal family diversity through time The lines plotted represent the number of families in the fossil record. Notes: The blue line essentially represents marine invertebrate animals. Although a small number of vertebrate groups, notably fishes and a few tetrapod species, are included, these make up a tiny proportion of the total marine family diversity shown. The curve for fishes includes an increasing proportion of freshwater forms through the Cenozoic. Tetrapods are amphibians, reptiles, birds and mammals. The extent to which such results, based on sampling and interpretation of the fossil record, represent actual diversity has been subject to discussion. Source: Marine animals, adapted from Sepkoski’; fishes and tetrapods, adapted after Benton”: insects, adapted from Labandeira and Sepkoski'”. Tertiary Quaternary 28 WORLD ATLAS OF BIODIVERSITY Eas widespread, abundant and geologically longer- lived species, in effect, the extinction-resistant forms; and so do not represent the biota as a whole”. If most species survived for less than ' Pteridophytes 700 Gymnosperms MN) Angiosperms 500 |- ie) f=} i=) Number of land plant species ES oO oO N oO oO | 300 Figure 3.4 Plant diversity through time Notes: Pteridophytes are ferns Filicinophyta and allies, gymnosperms are conifers Coniferophyta and allies, angiosperms are flowering plants Anthophyta. Note changing numerical dominance of each group over time. This generalized diagram Is to illustrate changing abundance of major groups over time; it does not represent short-term extinction events Source: Adapted from Kemp™, after Niklas. l 200 100 0 Geological time (10° years) declined during the early Mesozoic. From the mid-Cretaceous the number of families started to increase rapidly, reaching a recent peak of around 340. Diversity of genera follows this overall pattern in a more exaggerated form. It appears that periodic increase in the number of tetrapod families is mainly a result of lineages becoming adapted to modes of life not already followed by other organisms, i.e. by adopting new diets or new habitats”. PATTERNS OF EXTINCTION If living species represent between 2 and 4 percent of all species that have ever lived”, almost all species that have lived are extinct, and extinction can be presumed to be the ultimate fate of all species. Numerous estimates have been made of the lifespan of species in the fossil record; these range from 0.5 My to 13 My for groups as varied as mammals and microscopic protoctists. Analysis of 17500 genera of extinct marine microorganisms, invertebrates and verte- brates, suggests an average lifespan of 4 My in these groups’. Given this average lifespan, at a very gross estimate, the mean extinction rate would be 2.5 species per year if there were around 10 million species in total. However, because of bias inherent in the fossil record, such lifespan estimates are likely to relate to 4 My, the overall extinction rate at any given time would have been correspondingly higher. Major extinctions in animals In general the Precambrian fossil record is too incomplete to allow detailed analysis of extinc- tion rates. However, there is good evidence of a major loss of diversity during the Vendian period in latest Precambrian times, around 550 My ago, when the entire Ediacaran fauna disappeared [along with many acritarchs). Another wave of extinction affected archaeo- cyathid sponges, mollusks and trilobites during the lower Cambrian, some 530 My ago. By far the most severe marine invertebrate mass extinction was in the late Permian (250 My ago). At that time, the number of families of marine animals recorded in the fossil record declined by 54 percent and the number of genera by 78-84 percent. Extrapolation from these figures indicates that species diversity may have dropped by as much as 95 percent. Other major extinctions in marine invertebrates occurred at the end of the Ordovician (440 My ago) {see Figure 3.3), when around 22 percent of families were Lost, and during the late Devonian and late Triassic (21 percent and 20 percent, respectively). Around 15 percent of marine families disappeared at the end of the Cretaceous. The vertebrate fossil record, especially for terrestrial tetrapods, is much less amenable to analysis of extinction rates than the inverte- brate record, chiefly because it is less complete and less diverse. However, studies indicate that fishes have been subject to at least eight important extinction events since their recorded origin in the Silurian, while tetrapods have experienced at least six such events since their appearance in the late Devonian. Some of these events coincide with each other and with those recorded for marine invertebrates; in particular, the five major mass extinction events outlined above are paralleled by losses in vertebrate diversity. The most significant is the late Permian event, which, in terms of percentage loss, is the largest recorded extinction both for fishes (44 percent of families disappearing from the fossil record) and tetrapods (58 percent of families disappearing]. The late Cretaceous event was more significant for tetrapods than for fishes, with at least 30 of the 80-90 families then in the fossil record disappearing at this time. These families were, however, virtually confined to three major groups which suffered complete extirpation — the dinosaurs, plesiosaurs and pterosaurs. Most other vertebrates were almost completely unaffected. Major extinctions in vascular plants Fewer major extinction events have to date been distinguished in the plant fossil record than in the animal record. Periods of elevated plant extinction appear in some cases more protracted than animal extinction events and not usually coincident with them”. However, there is good evidence for extensive reduction in woody vegetation at the end of the Permian, with widespread loss of peat forests in humid areas and of conifers in some semi-arid regions. A fourfold increase in atmospheric carbon dioxide around the Triassic-Jurassic boundary is correlated with a more than 95 percent turnover in the megaflora (i.e. leaf fossils, etc., as opposed to pollen or spores)”. The end-Cretaceous catastrophe appears to have had a major influence on the structure and composition of terrestrial vegetation and on the survival of species. Data from fossil leaves suggest that perhaps 75 percent of late 350 —— 300 |— ~ a =) | q oO I Number of familiy extinctions nN oO oO I =) =) | a i) I Biodiversity through time ag Frequency {=} S 0 20 40 % extinction Cretaceous species became extinct, although data from fossil pollens indicate a lower though still significant level of extinction. During the Tertiary there are two other periods of widespread heightened extinction rates, during the late Eocene and from the late Miocene to the Quaternary, although in the latter extinction of taxa at generic level and above appears to have been mainly regional rather than global. Mass extinctions The very many species extinctions represented in the fossil record are not distributed evenly through time, nor do they occur randomly. In paleontology much attention has been devoted to mass extinction periods, during which some 75-95 percent of species then living became Mesozoic 200 Geological time {10° years) Figure 3.5 Frequency of percent extinction per million year period See text for explanation Source: Adapted from Raup 80 100 Figure 3.6 Number of family extinctions per geological interval through the Phanerozoic Note: 76 geological intervals represented, average duration around 7 million years. 29 Source: Benton’. 30 WORLD ATLAS OF BIODIVERSITY SE NE RS LL Sg Table 3.1 The principal mass extinctions in the Phanerozoic fossil record Source: Summarized from Kemp™, Hallam and Wignall” extinct during geologically very short periods of time, In some cases possibly as little as a few hundred thousand years or even less. Five such phases [based chiefly on extinction of marine species) are recognized during Phanerozoic time, late in each of the Ordovician, Devonian, Permian, Triassic and Cretaceous periods (Table 3.1). Although each of the “Big Five’ mass extinctions had a very profound effect on then contemporary life, they are not isolated peaks standing out from a constant, very low, background rate. Rather, the extinction rate has varied continuously throughout the Phanerozoic with periods of more or less elevated rates of extinction, of which the most extreme can be characterized as mass extinc- tion events. A frequency plot (Figure 3.5) of the percentage of species becoming extinct in each |-My interval of the 600-My record of animal life provides an indication of this variation in extinction intensity’. The plot takes the form of Late Ordovician 440 85 Late Devonian 365 80 End Permian 250 95 End Triassic 205 80 End Cretaceous 66 75 The last but largest of several extinction events during the Ordovician. More than 25% of marine invertebrate families lost. Entire class Graptolithina reduced to a few species; acritarchs, brachiopods, conodonts, corals, echinoderms, trilobites, all much reduced. Mass extinctions came at the end of prolonged period of diversity reduction. Rugose corals lost >95% of shallow water species; stromatoporoid corals reduced by half and reefs disappeared; brachiopods lost 33 families; ammonoids and trilobites severely affected. Fishes suffer only major mass extinction; all early jawed fishes (placoderms] disappear and most agnatha. First major crisis in plants; diversity greatly reduced, but spread of first tree, the gymnosperm Archaeopteris. The most severe extinction crisis: metazoan life came within a few percent of total extinction. Tabulate and rugose corals terminated, complex reefs disappear (return after 8-My gap]; echinoderms almost wiped out; worst crisis in history of foraminifera; severe extinction in ammonites, brachiopods, bryozoa, mollusks. Some losses in early ray-fin fishes. Major loss of terrestrial vertebrates (75% of families} and insects (8 of 27 insect orders extinct). Mass extinction in plants: large plants including peat-forming trees lost [spread of small conifers, lycopods and quillworts); sudden unprecedented abundance of fungal spores at end of period. Mass extinction in marine invertebrates, especially brachiopods, cephalopods and mollusks; also mass disappearance of scleractinian corals and sponges. Several seed fern families lost; some land vertebrates lost, but evidence for mass extinction questionable. Radical change in planktonic foraminifera; 85% of calcareous nanoplankton lost, also all ammonites, belemnites and many bivalves; losses in echinoids and corals. Many marine reptiles extinct lichthyosaurs, plesiosaurs, mosasaurs): significant losses in freshwater and terrestrial vertebrates, including last dinosaurs (high turnover throughout dinosaur history - end Cretaceous unusual in that no replacements emerged]. Mass extinctions in plants: highest (possibly 60% species loss} in angiosperms, lowest in ferns. Cooling; Warming Marine regression Marine transgression and anoxia Marine transgression and anoxia Volcanism Warming Marine transgression Marine regression Impact of large meteor Volcanism Cooling Marine regression an even left-skewed curve, with no substantial discontinuity between rare periods of high extinction rate (more than 60 percent of species extinct per one million years) and the most frequent rate (10-15 percent]; the mean intensity is 25 percent of species extinct per 1-My interval. This suggests that mass extinc- tions may arise from causes not qualitatively different from those associated with extinctions at other times. It should also be remembered that because of the extremely long duration of the Phanerozoic during which species have been constantly becoming extinct, and the very short duration of the mass extinction events themselves, the latter account for only a small percentage [estimated at around 4 percent) of all extinctions. The precise causes and timespans of each of the mass extinctions have been the subject of much debate and study“. It is now widely accepted that the late Permian mass extinction was a long-term event, lasting for 5-8 million years. It appears to have been associated with geologically rapid global physical changes (including the formation of the supercontinent Pangaea], climate change, and extensive, tectonically-induced marine transgression and increased volcanic activity. The late Cretaceous extinction is probably the best known, but in terms of overall loss of diversity is also the least important. There is strong evidence that this extinction event was associated with climate change following an extra-terrestrial impact, although this remains somewhat controversial. The late Ordovician event appears to be correlated with the global Hirnantian glaciation, REFERENCES Biodiversity through time 31 eee aaaamaaamamaaaaaaamaaaamaaamaaaacaetaecmeaaanaaien with three separate episodes of extinction spread over only 500 000 years. In all cases, however, the ability to determine accurately the timing and periodicity of extinction is heavily dependent on the completeness of the fossil record and the reliability and precision of stratigraphic analysis. A mass extinction period is typically followed by a phase of 5-10 My of very low diversity, with a handful of species dominant in fossil faunas and floras. When diversity again increases the biota may be very different in composition from those preceding. In several instances, groups previously showing low diversity have radiated and spread widely following the demise of groups previously dominant; e.g. ray-finned fishes diversified following loss of placoderms; quillworts and seed ferns diversified for a time after loss of glossopterids; and the mammals radiated after loss of many terrestrial reptile groups. Reef organisms provide a classic example: reef communities have been lost several times during the Phanerozoic but with each reappearance the main reef-building forms have been different”. Important new evo- lutionary opportunities arise for the lineages that survive mass extinction events, which can radically redirect the course of evolution®”. The present-day diversity of living organisms is the result of a net excess of originations over extinctions through geological time. Figure 3.6 shows the number of family extinctions in each interval of geological time, and the cumulative diversity of families, both marine and terrestrial. 1 Raup, D.M. 1994. The role of extinction in evolution. Proceedings of the National Academy of Sciences 91:6758-6763. 2 Sepkoski, J.J. 1992. Phylogenetic and ecologic patterns in the Phanerozoic history of marine biodiversity. In: Eldredge, N. (ed.] Systematics, ecology and the biodiversity crisis, pp. 77-100. Columbia University Press, New York. 3 Jablonski, D. 1991. Extinctions: A paleontological perspective. Science 253: 754-757. 4 Benton, M.J. 1995. Diversification and extinction in the history of life. Science 268: 52-58. 5 Alroy, J. et al. 2001. Effects of sampling standardization on estimates of Phanerozoic marine diversification. Proceedings of the National Academy of Sciences 98: 6261-6266. 6 Mojzsis, S.J. et al. 1996. Evidence for life on Earth before 3,800 million years ago. Nature 384: 55-59. 7 Nisbet, E.G. and Sleep, N.H. 2001. The habitat and nature of early life. Nature 409: 1083-1091. 8 Farmer, J.D. 2000. Hydrothermal systems: Doorways to early biosphere evolution. GSA 32 WORLD ATLAS OF BIODIVERSITY NSO a Today 10. Available online [accessed January 2002) at http://www.geosociety.org 9 Rasmussen, B. 2000. Filamentous microfossils in a 3,235-million-year-old vulcanogenic massive sulphide deposit. Nature 405: 676-678. | 10 Watanabe, Y., Martin, J.E.J. and Ohmoto, H. 2000. Geochemical evidence for terrestrial | ecosystems 2.6 billion years ago. Nature 408: 574-578. 11 Brocks, J.J. et al. 1999. Archean molecular fossils and the early rise of eukaryotes. Science 285: 1033-1036. 12 Javaux, E.J., Knoll, A.H. and Walter, M.R. 2001. Morphological and ecological complexity in early eukaryotic ecosystems. Nature 412: 66-69. 13 Knoll, A.H. 1995. Proterozoic and early Cambrian protists: Evidence for accelerating | evolutionary tempo. In: Fitch, W.M. and Ayala, F.J. (eds). Tempo and mode in evolution: Genetics and paleontology 50 years after Simpson, pp. 63-68. National Academy Press, Washington DC. 14 McMenamin, M.A.S. 1989. The origins and radiation of the early metazoa. In: Allen, K.C. and Briggs, D.E.G. (eds). Evolution and the fossil record, pp. 73-98. Belhaven Press, a | division of Pinter Publishers Ltd, London. 15 Margulis, L. and Schwartz, K.V. 1998. Five kingdoms. An illustrated guide to the phyla of life on earth. 3rd edition. W.H. Freeman and Company, New York. 16 Wellman, C.H. and Gray, J. 2000. The microfossil record of early land plants. | Philosophical Transactions of the Royal Society of London. Series B. Biological Sciences 355: 717-731. 17 Wikstrom, N., Savolainen, V. and Chase, M.W. 2001. Evolution of the angiosperms: Calibrating the family tree. Proceedings of the Royal Society of London. Series B. Biological Sciences 268: 2211-2220. 18 Labandeira, C.C. and Sepkoski, J.J. 1993. Insect diversity in the fossil record. Science 261: 310-315. 1? Benton, M.J. 1999. The history of life: Large databases in palaeontology. In: Harper, D.A.T. {ed.) Numerical palaeontology: Computer-based modelling and analysis of fossils and their distributions. John Wiley and Sons, Chichester and New York. 20 May, R.M., Lawton, J.H. and Stork, N.E. 1995. Assessing extinction rates. In: Lawton, J.H. and May, R.M. {eds}. Extinction rates, pp. 1-24. Oxford University Press, Oxford. 21 Jablonski, D. 1995. Extinctions in the fossil record. In: Lawton, J.H. and May, R.M. leds]. Extinction rates, pp. 25-44. Oxford University Press, Oxford. 22 Crane, P.R. 1989. Patterns of evolution and extinction in vascular plants. In: Allen, K.C. and Briggs, D.E.G. (eds). Evolution and the fossil record, pp. 153-187. Belhaven Press, a division of Pinter Publishers Ltd, London. 23 McElwain, J.C., Beerling, D.J. and Woodward, F.l. 1999. Fossil plants and global warming at the Triassic-Jurassic boundary. Science 285: 1386-1390. 24 Kemp, T.S. 1999. Fossils and evolution. Oxford University Press, Oxford. 25 Hallam, A. and Wignall, P.B. 1997. Mass extinctions and their aftermath. Oxford University Press, Oxford. 26 Erwin, D.H. 2001. Lessons from the past: Biotic recoveries from mass extinctions. Proceedings of the National Academy of Sciences 98: 5399-5403. 27 International Subcommission on Cambrian Stratigraphy. http://www.uni- wuerzburg.de/palaeontologie/Stuff/casué.htm (accessed January 2002). 28 Benton, M.J. 1989. Patterns of evolution and extinction in vertebrates. In: Allen, K.C. and Briggs, D.E.G. {eds}. Evolution and the fossil record, pp. 218-241. Belhaven Press, a division of Pinter Publishers Ltd, London. 29 Benton, M.J. ed.) 1993. The fossil record 2. Chapman and Hall, London. Data available online in interactive format at http://ibs.uel.ac.uk/ibs/palaeo/benton/ (accessed January 2002); also see http://palaeo.gly.bris.ac.uk/frwhole/fr2.families Humans, food and biodiversity 33 A Humans, food and biodiversity apes, almost certainly in Africa around 5 million years ago. Fossils usually attributed to the genus Homo itself date from the late Pliocene, perhaps 2 million years ago, and anatomically modern humans appeared some 100 000 years or more before the present. Agriculture developed independently in several regions around 10500 years ago, in association with increased population density. The global human population continued to grow only slowly until the 19th century, when revolutionary developments in agriculture, industry and public health triggered an exponential rise that has continued to the present day. Agriculture is a means to channel the Earth’s resources into production of human bodies. Humans have converted large areas of terrestrial habitat and use more than one third of net primary production on land. They are strongly implicated in the extinction of many large terrestrial mammal and bird species in prehistory, and are responsible for habitat change and exploitation that have caused the decline and extinction of many species in recent times. T HE LINEAGE LEADING TO THE HUMAN SPECIES EMERGED from an ancestry among the HUMAN ORIGINS The human species evolved as a natural element of diversity in the living world, and it is a simple ecological imperative that humans depend on other species and communities to supply the basic requirements of existence and to maintain biosphere function. The creation of organic compounds by photosynthesizing organisms is the point at which the sun’s energy enters the biosphere (Chapter 1); humans and other animals are unable to capture energy in this way and must consume and digest either primary producers or other organisms that are themselves dependent on primary producers, in order to obtain these energy-rich organic compounds for their own activities. While humans are doing nothing funda- mentally different from other animals, with the benefits of society and technology, which serve to increase the rates of resource extraction, they are uniquely successful at it. Self-evidently, humans have not arrived at their extraordinarily dominant position on the planet overnight. The growth of their influence can be traced back several million years, to well before the Pleistocene when a stone-tool-wielding hominid first emerged somewhere in eastern Africa. Climate during the past 2 million years The Earth's climate appears always to have been in a state of flux. Generally the degree of accuracy in our understanding of global climate, and certainly the degree of resolution in the timescale of climate change, decreases the further back in time we go. It is therefore difficult to compare periods that in geological terms are recent with more distant times. However, it does seem that during the past 2 million years there have been numerous, intense climate changes that were at least as severe, or perhaps more severe, than any recorded earlier in the Earth’s history’. It is during this period that hominids very similar to modern humans first appear in the fossil record. By the early Holocene, some 10 000 years ago, technologically sophisticated humans had spread to all the major land masses except Antarctica, and had evidently started to exert a major, and ever-growing, impact on the biosphere. Indeed there is WORLD ATLAS OF BIODIVERSITY Map 4.1 Early human dispersal A highly generalized view of the colonization of the world by advanced humans from the early Pleistocene onwards. Coastline is shown 150 m lower than at present and, with northern hemisphere ice cover, represents an approximation of that at periods of glacial maximum. Dates represent earliest well-established time of arrival; arrows indicate general net direction of dispersal, not actual routes. Presence and dispersal of early humans are not represented: the lineage is believed to have arisen in Africa and later fossil material dated to between 1.8 and 1.4 million years before the present is known from several sites in Eurasia. Ice sweep. = Human dispersal Years before the present evidence that such impacts - for example in the extinction of large mammal and bird species - were already being felt con- siderably earlier than this. However, because the rise of humans coincides with a period of major climatic and ecological fluctuations, it is often difficult to disentangle the effects of the former from the latter, and so the precise nature of these impacts remains controversial. The Tertiary period, which began some 65 million years ago and ended with the start of the Quaternary (the Pleistocene and Holocene) 1.8 million years ago, is charac- terized overall by a gradual decrease in global temperature and increase in aridity. Superimposed on this general pattern were many oscillations, occurring on a timescale of thousands of years’. These oscillations are ee believed to be linked to cyclic variations in the Earth's position in its orbit around the sun, known as Milankovitch cycles. These cycles notwithstanding, the climate during virtually the whole of the Tertiary was notably warmer than at present. Around 1.8 million years ago, at the very start of the Pleistocene’, there was apparently rapid global cooling leading to the start of a period dominated by marked climate cycles of around 100000 years’ duration. For long periods of each cycle global temperatures were significantly lower than they are today, and extensive areas of the northern hemisphere land masses were covered in ice sheets. Detailed analysis of climate changes over the past 400 000 years or so, particularly through examination of Antarctic ice cores‘, indicate that each cycle over this period has been broadly similar, with a gradual decline in average temp- erature, although with many minor and some large oscillations (relatively colder periods being referred to as stadials and warmer ones as interstadials}, followed by a short period of intense warming in which temp- eratures rose from those of fully glacial conditions to those characterizing a warm interglacial state, perhaps sometimes over only a few decades. Mean global temperature during glacial periods was around 6°C cooler than during interglacials, with cooling more pronounced at the poles than the equator. The mid-latitudes and equatorial regions were probably some- what more arid than at present. The large amount of water locked up in the greatly expanded polar ice caps meant that mean sea Humans, food and biodiversity 35 Zealand) level during glacial periods was probably around 100-150 meters lower than at present. Until around 11000 years ago, each temperature peak was apparently followed by an almost immediate decline as the next glacial cycle began‘. Overall, it appears that during the past half million years the Earth’s climate has been as warm as, or warmer than, today’s for only around 2 percent of the time’. It seems likely that this also holds true for the early Pleistocene. The end of the last glacial cycle, around 11 000-12 000 years ago, marked the start of the Holocene. Temperatures similar to today’s have prevailed throughout the Holocene, making it by far the longest true interglacial period for at least the past half million years and probably for the last 1.8 million years. 36 WORLD ATLAS OF BIODIVERSITY eee en Human origins and dispersal The origins and early history of humans are among the most controversial subjects in paleontology. Remains are generally scarce and often open to varying interpretation. Nevertheless, consensus has emerged over the broad outlines. It is likely that the direct ancestors of humans - the hominid line - diverged from the apes in Africa during a cool, dry phase of the late Miocene, around 6 million to 5.5 million years ago. The primary evidence for this divergence is in the form of ‘molecular clocks’ calculated from comparison of human and ape genetic material’. Relevant fossil material from this time horizon is sparse and often of uncertain status, but material from Ethiopia dated to 5.8 million to 5.2 million years ago’ is believed to represent Ardipethecus, generally accepted as the earliest known member of the hominid branch. Other species of this genus are known from before 4 million years ago, with various species of Australopithecus and Paranthropus, all from eastern Africa, some- what younger in age. Sometime during the middle to late Pliocene, 2.5 million to 1.8 million years ago, the genus Homo is thought to have evolved from Australopithecus stock*’. Until com- paratively recently it had been assumed that early man then remained confined to Africa until less than 1 million years ago. However, two well-preserved skulls recently found in the Caucasus (southern Georgia] have been dated to about 1.8 million years’ (and variously attributed to Homo ergaster, otherwise known from Africa, or H. erectus). Elsewhere, the earliest stone tools from Turkey and southwest Asia have been dated to about 1.5 million years ago, and from northeast Asia to nearly 1.4 million years’, while tools and fossils from East and Southeast Asia might be as old as 1.9 million years. These finds suggest that populations of early forms of Homo spread through the southeastern fringes of Europe as well as into Asia within at most a few hundred thousand years of the genus originating in Africa’. The earliest hominid remains elsewhere in Europe date from 780 000 years ago in Spain", followed by somewhat different material known from several sites widely distributed in northern Europe of about 500 000 years in age or less''. Fossil material from these two periods may be attributed to H. antecessor and H. heidelbergensis, respectively. The diversity of hominid material and the range of dates suggest that the spread of early hominids was complex and multidirectional, with possibly three lineages moving into Eurasia from Africa. Modern humans, Homo sapiens, are believed to have appeared sometime after 200 000 years before the present. Anatomically modern human remains are known from around 100 000 in Africa and the Middle East, with more recent dates in East and Southeast Asia. The first appearance in Europe dates from about 40000 years ago", with expansion apparently from the east toward the west, where in places premodern (Neanderthal) humans remained until about 28 000 years ago. One interpretation of the several lines of genetic evidence suggests, on the basis of mathematical models, that the modern human population is derived from a small founding population of perhaps 10 000 breeding individuals that existed sometime between 130000 and 30000 years before the present’?"’. The earliest evidence of hominids outside Africa, Europe and Asia is much more recent. The age of the human skeleton found in 1974 at Lake Mungo in Australia, evidently the earliest human remains known on the continent, has been estimated at 62 000 years before the present", but a date close to 56 000 years IS now widely accepted for human arrival on the continent”. In the Americas, the oldest good evidence of human presence is that of a coastal settlement in Chile dated around 14000-15000 years ago”, although even this is far from universally accepted amongst archeologists. There are very controversial and now widely questioned claims for evidence of human settlement much earlier, most notably that dating from 32 000 years before the present from Pedra Furada in northeast Brazil". The earliest unequivocal evidence of widespread occupation in the Americas comes from the so-called Clovis hunting culture whose oldest remains are generally dated at around 12 000 years before the present”. Settlement of the Caribbean islands including Cuba and Hispaniola appears to have taken place considerably later, around 6 000 years ago”. Archeological evidence indicates that colonization of the Pacific Islands east of New Guinea began around 4 000 years ago, when much of Melanesia and Micronesia was settled. Fiji and Samoa were probably colon- ized around 3 500 years ago, and the outliers of Hawaii, Easter Island and New Zealand within the last 1500 years’. In the Indian Ocean, Madagascar was probably first settled around 1 500 years ago”. Technology The earliest evidence of tool manufacture by hominids is from the Gona River drainage in northern Ethiopia where stone tools have been dated to at least 2.5 million years ago. These tools are small (generally less than 10 centimeters long) and simple, but are already of relatively sophisticated manufacture, suggesting that even older artifacts will eventually be found™. They are of essentially the same design as those dated from 2.3 million to 1.5 million years ago from other sites in East Africa (e.g. Lokalalei in the Lake Turkana basin in northern Kenya and the Olduvai gorge in Tanzania] indicating effective technological stasis for at least 1 million years (from 2.5 million to roughly 1.5 million years ago}. Collectively these tools are referred to as products of the Oldowan stone tool industry”. Around 1.5 million years ago much more sophisticated and often larger tools including hand axes and cleavers suddenly appear in the archeological record in East Africa. These are referred to as Acheulian tools. The Acheulian tool industry spread into Europe, the Near East and India, and remained apparently relatively unchanged until around 200 000 years ago, showing similar temporal persistence to the Oldowan industry. Evidence of tools and artifacts made from organic materials from the Paleolithic is understandably extremely scarce. In 1995, however, three large wooden implements very similar in design to modern-day javelins Humans, food and biodiversity 37 gu SH and dated to around 400 000 years ago were discovered at Schdningen in Germany. These were found in association with a smaller wooden implement [probably a throwing stick], stone tools and the butchered remains of more than ten horses, and can be per- suasively interpreted as throwing spears used in systematic, organized hunting”. There is also intriguing although indirect evidence from Southeast Asia dating to around 900 000 years ago that hominids at this time, at least in this region, were capable of repeated water crossings using watercraft. The evidence is in the form of stone tools dated to this age from the island of Flores in eastern Indonesia. Even at the time of the last glacial maximum (when global sea levels would have been at their lowest), reaching Flores from the Asian mainland would have required crossing three deep-water straits with a total distance of at least 19 kilometers. The impoverished - and typically island - nature of the Paleolithic fauna of Flores would appear to preclude the existence of any now submerged land bridge®. The next oldest, again indirect, evidence for the use of watercraft is the colonization of Sahul, probably sometime around 40 000 to 60 000 years ago. Even at times of lowest sea level, this would have necessitated crossing some 100 kilometers of open sea. Fire At some point in their evolutionary history, hominids clearly learnt to control, manipu- late and, presumably later, to start fires. Determining even very approximately when this may have happened is difficult, and as Watercraft may have played an important role in human dispersal for millennia. 38 WORLD ATLAS OF BIODIVERSITY i ee Large-scale burning of vegetation is one of the major human impacts on the biosphere. with all else to do with human evolution, controversial. This is chiefly because the existence of natural fires caused by lightning strikes and volcanic activity greatly compli- cates the interpretation of the archeological and geological record - an association between artifacts and evidence of burning does not necessarily indicate a direct link between the two”. The earliest dated associations between artifacts and burning that could be construed as deliberate use of fire are from Africa. Stone tools and splintered bones, which can be interpreted as evidence of butchery, have been found associated with clay baked at several hundred degrees for several hours at Chesowanja in Kenya, in deposits around 1.4 million years old”:””. The characteristics of the ‘clay are consistent with formation beneath a campfire, but are also consistent with formation around a slow-burning tree stump that could be associated with a natural bushfire”. Charred animal bones and other evidence of human occupation from just over 1 million years ago have been found at Swartkrans cave in South Africa, although similar problems of interpretation apply. Numerous sites in Europe and Asia provide evidence of human occupation and associated fire from the mid-Pleistocene, some 400 000 years ago. The best known of these is at Zhoukoudien in China, although even here the evidence for deliberate use of fire is widely considered equivocal”. Others are at Torralba-Ambrona (Spain), Terra Amata {France}, Westbury-sub-Mendip [England] and Vertesszollos (Hungary]”. Large-scale burning of terrestrial vege- tation is undoubtedly one of the major present-day impacts by humans on the biosphere, and is believed to constitute around one third of current annual anthropo- genic carbon dioxide emissions. Evidence of earlier impact is invariably circumstantial. It is difficult to demonstrate widespread biomass burning in the fossil or subfossil record and even harder to demonstrate a link with human activities - natural fires may be caused by storms or volcanic activity and may be expected to vary in extent, frequency and intensity according to prevailing climatic and ecological conditions. However, the abundance of elemental carbon in marine sediments off Sierra Leone in Africa can be persuasively interpreted as a measure of intensity of biomass burning in sub-Saharan Africa. Analysis of a core from these sediments covering the past million years or so indicates that inferred fire incidence in the region was low until around 400 000 years ago. Since then, five episodes of intense burning of vegetation can be inferred, all except the most recent coinciding with periods when the global climate was changing from interglacial to glacial. The current peak is unique in that it is occurring during an interglacial period. The change from inter- glacial to glacial climate is generally associated with increased aridity, so that vegetation may be expected to be more vulnerable to fire. It would also be expected that a substantial fuel base in the form of woody biomass would have accumulated during the warmer, wetter, interglacial. It may well be merely coincidental, but it is intriguing that the period of increased fire incidence in the Sierra Leone core - around 400 000 years ago - coincides with the timing of the first widespread evidence for hominid use of fire. Food resources The early australopithecine members of the hominid lineage appear, on craniodental evidence, to have been well adapted to consume hard but brittle items, such as nuts and seeds, and soft items, such as fruits”. Use and development of stone tools by later hominids is usually thought of as associated with meat consumption, and cut and hammer marks on large mammal bones from hominid sites are evidence of this, but microscopic wear patterns found on some tools indicate they were used to scrape wood and cut coarse plants such as reeds or grass”. Oldowan and Acheulian stone tools are rarely found more than 10-20 kilometers from their rock sources, suggesting that their owners ranged over a relatively small area’, so if hominid populations at the time were small in size, the environmental impacts of targeted resource extraction would have been minimal. The ratio of different forms of carbon and nitrogen incorporated in human bone colla- gen can provide information on the relative importance of food from terrestrial, coastal or freshwater habitats. Such ‘stable isotope analysis’ of collagen extracted from archeo- logical remains, coupled with other material evidence, suggests that pre-modern humans (Neanderthals) in Europe, although omni- vorous and opportunistic, behaved as preda- tory carnivores™. Similar evidence for slightly later human remains suggests a significant broadening of the resource base, owing particularly to increased use of freshwater resources’. Remains found at Mediterranean sites suggest further diversification of human diet and the means of food gathering”, Humans, food and biodiversity 39 associated with an apparent rise in human numbers during the later Paleolithic and approaching the start of the Neolithic. Here, there was increasing use of agile small game, such as partridges and hares, and apparently decreasing reliance on slow-maturing, easily collected forms such as tortoises and shell- fish, the average size of which decreased over time at a rate consistent with the effects of excess harvesting, possibly indicative of increasing human density”. Origins of agriculture Until the end of the Pleistocene, humans evidently depended on hunting and gathering of wild resources for their sustenance. Around the end of this period a radical change was initiated - the emergence of crop-based agriculture and domestication of livestock - phenomena which appear to have arisen independently in Africa, Eurasia and the Americas. The earliest direct evidence for animal domestication, of the dog Canis, dates from around 14000 years ago in Oberkassel in Germany”. However, analysis of ‘molecular clocks’ appears to show that the dog diverged from its wild ancestor, the wolf Canis lupus, far earlier than this - perhaps more than 100 000 years ago”. It is quite possible that pS SSS SSS hy The dog diverged from its wild ancestor, the wolf, more than 100 000 years ago 40 WORLD ATLAS OF BIODIVERSITY ec ceceecnereneeeremme ener nccmcccmcememnc domestication of the dog as a guard animal and aid in hunting predates domestication of other animals and plants as food sources by many tens of thousands of years, but in the absence of archeological evidence this remains speculative. The study of plant and animal domestication for agricultural purposes is a rapidly expand- With the rise of industrial-scale agriculture and commercial breeding, many local agricultural genetic resources, both crops and livestock, have been lost and replaced by modern, genetically uniform types specialized for superior production in higher input systems. There is much evidence that many local varieties possessed features of adaptive value in a particular environment and cultural context, and the precautionary principle argues that the diversity of domesticated forms and their wild relatives should be conserved where possible in order to maintain options for future breeding improvements. A complete picture of the global reduction in local genetic resources is not available, but there is abundant evidence at national level of the enormous scale of genetic erosion in crop plants. China Wheat varieties Korea Garden landraces (Rep.] Mexico Maize varieties USA Varieties of apple, cabbage, field maize, pea, tomato About 1 000 {10 percent) of 10 000 varieties used in 1949 remained in 1970s About 25 percent of landraces of 14 crops grown in home gardens in 1985 remained in 1993 Only 20 percent of maize varieties planted in 1930s remain Only 15-20 percent of varieties grown in 1804- 1904 available at present A growing number of countries have documented and evaluated livestock resources, see Appendix 3 for summary of mammal breed status. ing field, based on study of fossil pollen records, archeological remains, the genetics of present-day crops and their close relatives, and remaining indigenous agricultural systems”. The development of agriculture began independently in different continents and proceeded at different rates, while early cultivators undoubtedly continued to rely heavily on hunting and gathering from the wild. The first development of agriculture as an integrated system for food production was based in the Fertile Crescent, composed of the uplands of Anatolia and western Iran and the arid lowlands to the south, with the first tangible evidence approaching 11 000 years in age. Wheat Triticum, barley Hordeum, rye Secale, pea Pisum and lentil Lens, cattle Bos, sheep Ovis, goat Capra and pig Sus were all domesticated in this region, and formed the basis of the Neolithic peasant economy that spread steadily through much of western Eurasia into surrounding areas after about 10 000 years before the present. The system integrated the use of food plants, cereals especially, and domesticated animals for fertilizer and power as well as food. Although early plant domesticates are known elsewhere in the world, integrated agricultural systems appear to have taken longer to develop, perhaps in part because of the absence of domesticated animals. For example, in Middle America seeds of domes- ticated squash Cucurbita are known from about 10 000 years ago, maize from 6 300 years and beans from 2300, with village-based farming economies evidently taking several thousand years to develop”. Farming systems based on rice cultivation in Asia appear to date from 7 500 years ago. Analysis of mitochondrial DNA has indicated that a multiple maternal origin is general among domestic livestock species, i.e. female animals from more than one wild stock have contributed, at different times and places, to the present genetic diversity among each breed‘. For each of the four major breeds (cattle, sheep, goats, pigs), this evidence is consistent with archeological findings in suggesting a primary center of origin around the Fertile Crescent region, but also suggests additional centers in Asia. Most of the major crops on which humans presently depend have been grown con- tinuously since the early or middle Holocene. They have been constantly selected over this period and have developed large amounts of useful genetic variation. Indeed the success of individual crop species over wide geo- graphical areas is partly determined by their flexibility in evolving and sustaining genotypes suitable for local environments. Conventional breeding involves the selection and crossing of desirable phenotypes within a crop in order to create more productive genotypes. The process of harvest, storage and sowing alone may have assisted in the selection of traits such as non-shattering seed heads, uniform ripening of seeds, uniform germination, large fruits and seeds, and easy storage. Breeding methods have increased the rate of introgression of desired genetic traits into new cultivars; genetic modification is now possible at the level of incorporating individual genes directly into genomes. Domesticated crops and livestock have been transported around the world probably since full-scale agriculture began, e.g. wheats are recorded in areas outside their presumed center of origin at least 8 000 years before the present. Some crops became increasingly widely distributed after the 1500s when European colonists moved out of their home continent. CURRENT FOOD AND NUTRITION Perhaps as many as 7000 of the 270000 described plant species have been collected or cultivated for consumption”, but very few, some 200 or so, have been domesticated, and only a handful are crops of major economic importance at global level“. The variety of species used is limited more by production and cultural factors, such as tradition and palatability, than by nutritional value. Twelve crop plants together provide about 75 percent of the world’s calorie intake. These comprise (in alphabetical order): bananas/plantains, beans, cassava, maize, millet, potatoes, rice, sorghum, soybean, sugar cane, sweet potatoes and wheat“. More than 40 mammal and bird species have been domesticated, around 12 of which are important to global agricultural production (see Appendix 3]. Some, such as cattle, pigs, goats, sheep and chickens, are fundamental to many non-industrial agri- cultural systems, and provide a wide range of products in addition to foodstuffs. Although the global consumption of wild terrestrial species cannot be assessed acc- urately, the amount is likely to be insig- nificant in comparison to products from just three domestic forms: pigs, cattle and chickens. On the other hand, fishes from wild sources are an important nutrient source, and the amount harvested and consumed is known to be greatly under-reported in official statistics. Humans, food and biodiversity SSS SES At world level, cereals are the most important single class of food commodities, providing around 50 percent of daily calorie intake and 45 percent of protein. In contrast, meat provides around 15 percent of the protein, and fishery products only some 6 percent, or 15 percent of all animal protein. In each case, when comparing sources of calories, protein and fat in the global human diet, just two or three commodities stand out from a large number of commodities of lesser importance. Rice and wheat together provide around 40 percent of the world supply of both calories and protein, while milk and pigmeat are key sources of fat, calories and protein {see Table 4.1). At country level, a much wider variety of plant species are important in that they make a significant contribution to human nutrition”. Around 22 species and groups of species provide more than 5 percent of the per capita supply of either calories, protein or fat in at least ten countries of the world. Notes on the origin, uses and genetic resources in these 22 crops are tabulated in Appendix 2, together with briefer information on the remaining 50 or so crops that are also nutritionally important but to a smaller number of countries. Although the vast bulk of human food supply on a global level is derived from Table 4.1 Top ten food commodities, ranked by percentage contribution to global food supply Note: This information is partly determined by the way food commodities are aggregated for reporting purposes; for example, fishery products collectively provide about 6% of the hypothetical global protein supply, but no individual fishery commodity is important enough to appear in the table. Source: FAO food balance sheets“*, 1997 data. f Map 4.2 Livestock breeds: numbers and status Color represents the number of mammal breeds in each country. Recording of breeds is incomplete globally. Pie charts represent the proportion of all mammal breeds associated with each FAO region assessed as threatened (gray] or extinct {black}. Source: Charts calculated from FAO World Watch List for Domestic Animal Diversity (3rd edition, 2000); country data derived from FAO Domestic Animal Diversity Information System (DAD-I|S} database, available online at http://www. fao.org/dad-is/ [accessed February 2002) 42 WORLD ATLAS OF BIODIVERSITY Number of breeds 110-285 WM 4c-10° a 17-45 [__] Insufficient data Status 2 % threatened — % extinct farming, a large number of people worldwide, including many who are principally agricul- turalists or pastoralists, make extensive use of wild resources. In parts of Africa, bushmeat (meat from wild animals) supplies most of the protein intake; similarly some 80 percent of people in sub-Saharan Africa are believed to rely largely or wholly on traditional medicines derived almost exclusively from wild sources. Indeed, traditional medicine continues to be the source of health care for the majority of people living in developing countries, and is widely incorporated in primary health care systems. Wild plants in many areas are extremely important as famine foods when crops fail, or may provide important dietary supplements, and use of fuelwood and charcoal from wild sources is almost universal in the developing world. A small number of people continue to derive most of their requirements from wild sources by hunting, fishing and gathering. Ina large sample (220) of such societies, nearly 40 percent have a high dependence on fishing, around one third are highly dependent on gathering and 28 percent on hunting of terrestrial resources’. Fishing and gathering tend to be alternative activities, both of them complementary to terrestrial hunting. In contemporary conditions, fishing tends to be more important where temperature is lower (especially in high northern latitudes} while the converse applies to gathering. People in such communities also derive shelter, medi- cine, fuelwood and esthetic or spiritual fulfill- ment from wild species, and the immense variety of plant and animal species used has been well documented. Global food supply National data on reported food commodity supplies are collated by the Food and Agriculture Organization of the United Nations (FAO). Given information on the food value of these commodities, and the size of the human population, it is possible to estimate the average national food supply per person. Dietary food value can be broadly assessed in terms of energy or materials. The former is conventionally measured in calories (calories/person/day), the latter in terms of weight of protein or fat {grams/person/day). These standard measures take no account of the vitamins and minerals (micronutrients) that are required for maintenance of full health. The nutritional value of the human diet varies geographically and over time, and to a great extent according to the state of Humans, food and biodiversity 43 development and purchasing power of the people concerned. The human population of the world doubled between 1950 and 1990, reached around 6 billion in the late 1990s, and will continue to grow for decades, albeit more slowly because of decreased reproduction rates. At global level, there has been enough food available in recent years to supply the entire human population with a very basic diet, largely vegetarian, providing 2 350 kcal per day. In 1992 the average food supply was estimated at 2718 kcal daily (after losses during storage and cooking), made up of 2 290 kcal from plants and 428 kcal from livestock products’. Thus the global food supply was nominally sufficient for the calorific needs of a population 15 percent larger than the estimated population; a similar small annual hi Dietclass Class 1 Class 2 Rice Maize Human population {thousands} 2920923 514911 % of world population 94 9 Increase in supply needed by 2050 x 2.37 x 1.96 Table 4.2 World diet classes Notes: Population calculation based on 1997 data, from FAO website. Excludes Japan and Malaysia, in anomalous position using six-part classification. Total population in 117 countries in sample: 5 462 493 000. The penultimate row in fact shows the percent each class forms of this total population in the sample countries; this - is assumed to be an acceptable surrogate for the total world population. Source: FAO’”. 44 WORLD ATLAS OF BIODIVERSITY ne ee nee eraser rere esc ee nce c acne cane Class 3 Class4 Class5 Class 6 Wheat Milk,meat, Millet, Cassava, plantain Wheat sorghum taro, yams 664507 942924 61867 357 361 12 17 1 tl xX 2.84 x 1.13 x 4.82 xX7.17 surplus has existed since the 1970s“. Taking these aggregated global data at face value, it appears that a more than sufficient amount of food has been produced annually during the past two decades to maintain the world’s human population. Global food supply has doubled since the 1940s; the increases in supply over the last two decades are attri- buted mainly to increased productivity (69 percent) and secondarily to an increase in production area (31 percent). Each year during the past two decades, between 850 million and 900 million people have been undernourished”. Given that there has been sufficient food available in the world overall, undernourishment must be an effect of unequal access to the appropriate amount and type of food. Unequal distribution is evident at macro and micro scales: some countries are more productive and richer in resources than others and, whether at nation- al or village level, high-status social groups secure better diets than others, compounded in some cases by gender differences. Poverty is a key cause of undernutrition, and often also an effect of it, forming a self-reinforcing cycle from which it is difficult to escape without appropriate outside intervention. Put in different terms food insecurity is a problem of lack of access resulting from either inadequate purchasing power or an inade- quate endowment with the productive resources that are needed for subsistence”. Although the 1996 World Food Summit in Rome called for a 50 percent reduction in the global number of chronically malnourished people by 2015, the absolute number of people affected remains high, and if trends continue this target will not be met”. Regional variation The ‘global average diet’ is a simplifying abstraction that ignores an enormous amount of regional, national and local variation in food sources and in supply. Nor does it take account of micronutrient availability (vitamins, miner- als, trace elements], i.e. substances that do not directly contribute to energy or protein intake but which are nevertheless essential elements of a healthy diet. The human diet can be assessed in several ways. The FAO devised a simple classification to serve as the basis for an analysis of food requirements in relation to population growth*’. In this scheme, diet is assessed in terms of calorie sources as recorded at national level in the FAO food balance database, and countries with similar diet structures are clustered together in one of six diet classes (Table 4.2, Map 4.3). Each class is named after the food product that best characterizes the diet (although this is not necessarily the principal calorie source). The countries in each food class tend to share broad demographic features. Rice countries (class 1) have high population densities, higher than average mortality rates and little diet diversification. Maize countries (class 2] generally have population densities near the world average and low mortality, especially infant mortality. Wheat countries on average have low population density, but this masks serious land and water shortages in many. Class 4 countries, with a diet described as mixed ‘milk-meat-wheat’, include the world’s most highly developed nations, with fertility, mortality and popu- lation growth rates well below the global average. Millet countries (class 5} in general tend to have high population growth, high fertility and low life expectancy; the diet provides only a marginal surplus of energy i i a supplies over requirements. Diet class 6 countries contain the human populations most at risk from food insecurity. The diet is characterized by roots and tubers, and on average does not provide basic energy requirements. These populations show high fertility, high mortality and a rapid growth in numbers. Poverty is widespread and the infrastructure weak, but there are significant reserves of under-exploited arable land. These major diet patterns have been used to help characterize the improvements to production and food supply that may be needed in order to meet the per capita energy needs of the population projected to exist in 2050. In terms of diet class, availability in class 4 could remain little changed, supply in classes 1, 2 and 3 would need to approxi- mately double, and classes 5 and 6 would need at least a fourfold increase. In some cases, food security may be better ensured by diversification rather than increasing yields. The limitations of agri- culture based on uniform varieties and a high input of fertilizers and pesticides become more acute for farmers who rely on poor resources or marginal environments. By growing diverse and locally adapted crops farmers can bring about greater security in food production and more efficient use of limiting resources. Many traditional agri- cultural systems manage to varying degrees a high diversity of both cultivated and wild food species (Table 4.3), and in a sense reduce the distinction between wild plants and domesticated crops. Strong selection pres- sures exerted by natural processes and humans over several millennia and wide geo- graphical areas have resulted in thousands of varieties within most crop species. High agricultural diversity can not only provide insurance against crop failure in difficult agricultural environments, but tends to have nutritional benefits. Transition to more uniform diets, with high intake of fats and sugar, has resulted in declining nutri- tional status among numerous indigenous groups”. Low dietary diversity is associated with micronutrient deficiency, a problem far more common than general protein-energy malnutrition, and particularly prevalent in Humans, food and biodiversity 45 SSS a children, pregnant women and breastfeeding mothers. Diverse cropping systems in back- yard and home gardens, whether rural or urban, can lead to direct improvement in family nutrition and in some cases provide cash income. Even a small mixed vegetable garden Is capable” of providing 10-20 percent of the recommended daily allowance of Wa Table 4.3 Examples of diversity in agricultural systems 74-86 Source: Multiple sources Brazil Kayapo Over 45 tree species planted for food or to attract game; 86 varieties of food plants grown Ecuador Siona-Secoya Major staples: 15 varieties of manioc, 15 of plantain, 9 of maize; pre-1978 traditional gardens yield 12 300 kilos of food or 8.8 million calories; 72% calories and 14.8% protein, 22.2% fats, 90.9% carbohydrates; post-1978 horticulture provides 67.8% calories, 10.2%, purchased Indonesia Java 500 species in home gardens in a single village; ability to support 1 000 people per km* Indonesia WestSumatra 6 main tree crops, many vegetables and fruit; 53 species of wild plants also protected and harvested Kenya Bungoma 100 species of fruit and vegetables; 47% households collect wild plants, 49% maintain wild plants in gardens Kenya Chagga Over 100 species produced in gardens Mexico Huastec Over 300 useful species found in managed forest plots called te'lom, 81 food species Mexico Migrant rural 338 species of plants and animals in home gardens, community including 62 species of wild plants in southeast Papua New Gidra Approximately 54% calories and 82% of protein come Guinea from non-purchased sources (wild, sago, coconut gardens) Peru Bora 22 varieties of manioc and 37 tree species are planted; 118 useful species found in fallow fields Peru Santa Rosa 168 species identified in 21 home gardens Philippines Hanunoo System of intercropping with 40 crops ina single field; over 1 500 plants considered useful, of which 430 may be grown in swiddens Sierra Gola forest Of food items 14% are hunted, 25% are from fallow land, Leone 8% are from plantations, 19% are from farm, swamp or garden, 21% are from streams and rivers, 13% is bought or given Thailand Lua 110 varieties of food plants and 27 wild food plants are found in swidden fallows 44 WORLD ATLAS OF BIODIVERSITY a Bacteria” 400 000 7.84 x 10" 4x10” 1.39 x 10° 7.06 x 10" - 1.18 x 10” -6x10° -214x10' - 1.09 x 10° Collembola® 6 500 2.4 x 10’ 2x 10" 64 5x 10? Termites” 2760 2.3 x 10° 2.4 x 10” 1 400 1.44 x 10" Antarctic krill 1 1.43 x 10’ 5x 10" 4.29 x 10° 1.5 x 10" Birds” 9946 1600-3 200 2x 10" -4x 10" Elephants” 2 0.07 - 0.1 4.26 x 10° 85 - 126 5.34 x 10° - 6.31 x 10° - 7.29 x 10° Great whales” 10 <0.01 2.83 x 10° 52 - 65 1.89 x 10” - 3.6.x 10° - 2.33 x 10” Domestic livestock excluding pets!“ ca 15 7.3 x 10" Humans 1 6 x 10° 3.9 x 10" Humans plus livestock 11.2 x 10" Table 4.4 protein, 20 percent of iron, 20 percent of Number of individuals and biomass, selected organisms Notes: Calculations based on . data in sources cited after group names. Biomass is estimated dry mass standardized on 30% wet mass. Number of birds per km? estimated by dividing global bird population (2-4 x 10") by land area minus area of extreme desert, rock, sand and ice {approx 125 million km’). Estimates of abundance for Collembola represent minimum global totals. Whale estimates presume a Sex ratio of 1:1 Mean Asian elephant mass estimated at 3 500 kilos; range estimated at 500 000 km African elephant mass estimated at 4 250 kilos calcium, 80 percent of vitamin A and 100 percent of vitamin C. HUMAN NUMBERS AND IMPACTS Human population size Information on early human population numbers is based heavily on inference from circumstantial evidence, and remains on an uncertain footing even when written historical material becomes available in some abun- dance for the past few hundred years. Highly speculative estimates based on extrapolations of population densities of great apes” and on studies of contemporary human hunter-gatherers” indicate that the global late Pleistocene human population may have been between 5 million and 10 million. It seems likely that any increase in human population up to then had been a result of increasing the total area occupied by the species, rather than by any major increase in population density in already occupied areas. Information on human population size in historic times is fragmentary, and populations lacking written records are likely to be inadequately represented. One analysis” distinguishes three main phases of population change. First is a cycle of primary increase in Europe, Asia and the Mediterranean brought about by the spread and further development of Neolithic agriculture, which appears to have allowed a great increase in population density. At the start of the Iron Age in Europe and the Near East, some 3000 years ago, the world population may have been doubling every 500 years, and the total probably reached 100 million around this time or soon after. Growth appears to have slowed to reach near zero by around year 400, possibly because the limits of then current technology had been reached. After the Dark Ages in Europe, a second growth cycle began around the 10th century in Europe and Asia, with numbers rising to a peak of around 360 million during the 13th century, followed by a slight fall. The global population then increased slowly until the 19th century, when an increasingly rapid rise began as a result of revolutionary developments in agriculture, industry and public health (Figure 4.1). Crucially, the rate of global population growth peaked during the late 1960s: it was then at just over 2 percent per annum, but is now about 1.7 percent. The absolute increase per annum has also peaked; it was around 85 million more people per annum in the late 1980s and is now about 80 million. T T 12 000 11 000 10 000 9 000 5 000 Such trends suggest that the present global total of some 6 billion may not itself double, as all previous totals have done. The medium variant of the current UN long-range forecast suggests the total in 2050 may be 9.3 billion®. Although several countries in Africa have yet to shift to lower reproduction rates, thus making a further doubling quite possible, it may be that ‘children born today may be thinking about their retirement at a time when the global population count will have stabilized - or even begun to decline”. The exponential rise in abundance of a single species, to a position of global eco- logical dominance, in the sense of using a disproportionate share of natural resources, is without known precedent in the history of the biosphere. This was not achieved without significant, often adverse, effects on the environment, many of which stem from the agricultural activities required to maintain human numbers. There is no single species in which so many individuals, of such large body size, are distributed so widely over the planet. There appear to be few macroscopic species, the Antarctic krill Euphausia suberba being one of these few, in which the number of individuals approaches the size of the present human population. No animal of comparable size has a population remotely similar in number to that of humans. Biomass provides a measure of the way in which global net primary production or NPP (see Chapter 1) is partitioned. If the standing crop of domestic livestock is added to the human biomass, amounting to around 11.2 x 10''kilos in total, then only the global biomass of bacteria as a whole is higher (Table 4.4). The current human population is unevenly spread across the land surface of the Earth. While some areas, such as most of Antarctica, the interior of Greenland, and hyperarid hot deserts, have no permanent human presence, in others human densities may locally reach 4000 3 000 Humans, food and biodiversity 47 Oa ee ’ an extreme of 1 000 inhabitants per hectare (e.g. in Calcutta and Shanghai). Map 4.4 shows the current density of human popu- lations over the Earth, based on census counts within administrative units of varying size. Among the most striking features of this map are the large areas of very high population density in parts of China and the Ganges-Brahmaputra lowlands, also on Java, and the large area of high population density extending over most of Europe. Human activities have now made them- selves felt throughout the biosphere, but it might be expected that the degree of trans- formation of terrestrial landscapes would be related to ease of access and proximity to population centers. Settlements, ranging in size from villages to cities of many million inhabitants, are connected by networks of paths, railways and waterways that allow the influence of humans to diffuse far beyond these settlements. Map 4.5 shows the results of a GIS-based analysis of the relative distance of points on the Earth’s surface from all such human constructions. Those points most remote can be given a high ‘naturalness’ value [i.e. ignoring other possible impacts, they are likely to be least disturbed) and, conversely, points surrounded by a high density of human structures can be given a low value. Although human population density was not part of the analysis, there are, unsurprisingly, strong similarities between Maps 4.4 and 4.5. 6 5 ae 2 a o 5 ze = a [°} oa 2 000 1000 0 Figure 4.1 Human population This graph shows the long period of many thousands of years during which the world human population remained small, followed by an exceptionally rapid rise to a total of 6 billion at present. Source: Data from McEvedy and Jones™; FAOSTAT database” iy 48 WORLD ATLAS OF BIODIVERSITY Map 4.3 FAO world diet classes A classification of country dietary patterns, based mainly on calorie sources. Each class is named after the foodstuff that best characterizes the diet. The classification of a few countries (e.g. Japan, Malaysia] appears anomalous, and some countries are not included in the classification. Source: Analysis by FAO’, Diet class Class 1 Rice Class 2 Maize Class 3 Wheat Class 4 Milk, meat, wheat Class 5 Millet, sorghum Class 6 Cassava, plantain, taro, yams Not assigned to a class The human share of global resources Agriculture is a set of activities designed to secure a greater and more reliable share of the energy and materials in the biosphere for the benefit of the human species, i.e. to divert an increased amount of available energy toward production of human bodies. Naturally occurring plants and animals are replaced by specially cultivated or bred varieties that can produce nutrients efficiently from available resources, and in a form that humans can conveniently use. The growth and persistence of these selected species is subsidized by humans. In less-developed agricultural sys- tems, this subsidy may be very small, perhaps just the removal of competition for light or grazing, |.e. plots are cleared or weeded and wild herbivores discouraged. Efficiency can be high but output is usually low. In western industrialized agriculture the subsidy is enormous. Competitors, pests and predators are removed from vast areas (through the use of herbicides and pesticides); fossil fuel is consumed to process, transport and apply any nutrients that limit production (nitrogenous fertilizer] and to store produce. Efficiency is high in some respects, e.g. use of space and labor, but much lower if all hidden costs of fossil fuel use and waste impact are considered. Output can be very high. The area of land devoted to agricultural production is now a significant proportion of the global land surface. Five thousand years ago, the amount of agricultural land in the world was negligible. There is no direct evidence from the greater part of this period on the rate of expansion of agricultural land. Useful historical data relate to the past few hundred years, and this evidence suggests that about 250 to 300 million hectares (ha) of land globally were devoted to crops in 1700. At present, arable and permanent cropland covers approximately 1 500 million ha of land, with some 3 400 million ha of additional land classed as permanent pasture (this figure includes rangeland and wooded land used for grazing). This represents a nearly sixfold increase in cropped land over the past three centuries (Table 4.5). The current extent of cropland is represented in Map 5.2. All the cropland is used to produce domestic plant material, and much of the land classed as pasture, together with large parts of the grassland and open shrubland landcover classes in Map 5.2 lof which ‘permanent pasture’ is a subset) is used to produce domestic herbivore biomass. Humans, food and biodiversity 49 Most domestic herbivores are destined to become human biomass or to meet other human requirements, so can be counted as surrogate humans. The rise in livestock numbers has been accompanied by a decline in wild herbivores. For example, the extinct wild ox Bos primigenius of Eurasia and North Africa has been replaced in its former range by domestic cattle, of which approximately 1 360 million head exist in the world“. Similarly, the American bison Bison bison was reduced from perhaps 50 million before European arrival on the continent to Table 4.5 Land converted to cropland Source: 1700-1950 estimates from Richards’; 1980 and 1999 data from FAOSTAT database {complete time series from 1961 available from this source]. 50 WORLD ATLAS OF BIODIVERSITY rrr Table 4.6 Estimated large herbivore numbers and biomass in Mesolithic and modern Britain Source: After Yalden”. Wild boar Sus scrofa 1 357 740 Wild ox Bos primigenius 99 250 Red deer Cervus elephus 1472870 Roe deer Capreolus capreolus 1 083 810 Moose Alces alces 67 490 108 619200 Pigs 39 700 000 “147 287 000 21 676 200 13 498 000 under 1 000 at the end of the 19th century; although there has been significant recovery under management, it has been replaced over much of its former range by domestic cattle, of which 100 million head are present in the United States. Data on the fauna of the Bialowieza forest in Poland (a forest remnant with populations of large wild herbivores and predators] have been used” to estimate the possible number and biomass of wild herbivores in heavily forested Mesolithic Britain. These estimates were compared with current herbivore populations in Britain. Indications are that there are now 40 times more domestic cattle than there were wild ox and 20 times more domestic sheep than there were wild deer; the overall large herbivore biomass has increased by a factor of ten {see Table 4.6). Similar values are likely to apply to other heavily populated industria- lized countries. Not only are domesticated mammals far more abundant than their wild Sheep 20364 600 916 407 000 853.000 127 950 000 Cattle 3908900 2149 895 000 Red deer 360 000 54 000 000 Roe deer 500 000 10 500 000 Introduced deer 111 500 5 017 500 Feral sheep and goats 5 700 256 500 Domestic herbivore total 25126500 3194 252 000 Wild herbivore total 977200 69774000 relatives ever were, the latter are in many cases extinct or near extinction. As mentioned in Chapter 1, calculations at global level and in a study of one country (Austria) estimate the proportion of terrestrial global production appropriated or diverted for human use at 30-50 percent (depending on how much change beyond direct harvesting is incorporated). One approach to assessing the effects of energy appropriation has used the empirical relationship between energy and species number”. There is much evidence suggesting that at a range of spatial scales and within different taxonomic groups, the diversity of species present in an ecosystem tends to be positively correlated with the amount of energy available [see Chapter 5]. Accepting the empirical evidence for this relationship, it has been argued that the number of species present will decline if the amount of energy available for use declines. At global level, a conservative estimate using this relationship predicted that 3-9 percent of terrestrial species would be extinct or endangered by the year 2000”. The evidence from the Austrian study” is consistent with the species-energy relationship: the curve developed by Wright” predicts that with 41 percent of potential NPP at country level now being appropriated by humans, 5-13 percent of species should have been extirpated from the country; in fact, 8 percent of birds and 7- 14 percent of reptiles have been lost. The significance human resource appro- priation has for other species, and for biosphere function, in part depends on the extent to which the resources are limited in availability. Space clearly is limited, and it may be that the main effects of appropriation will be exerted through direct changes to land- cover, and diminished availability of material resources for wild species, rather than through a diversion in energy flow. SPECIES EXTINCTION AND HUMANS Tracking extinction Change in biological diversity has principally been assessed in terms of declining popu- lations, species, communities and habitats. There has always been special concern about extinction, because this is a threshold from which there is no turning back. As discussed in Chapter 3, the extinction of species is natural and expected, and self-evidently there have always been species at risk of extinction. It seems very likely, however, that recent and current extinction rates are considerably higher than would be expected without the influence of humans. For several reasons it is difficult to record contemporary extinction events with precision. The species involved may well be unknown. Even if they have been discovered and named, they may be too small to be noticed without special sampling procedures. The entire process of decline and eventual extinction may take place over many years or even centuries in the case of particularly long-lived organisms such as many trees. The near- terminal stages in the process of species extinction are unlikely to be observed. Where such observations have been possible, it is because the species has been destroyed in part by unusually extreme hunting pressure (e.g. the passenger pigeon Ectopistes migratorius) or extreme ecological events (e.g. extinction of many native land snails in French Polynesia and Hawaii following introduction of the predatory snail Euglandina rosea), and has been the subject of sufficient interest to be closely monitored. In other cases, positive evidence of extinction is lacking. Typically, many years elapse before sightings of a species become sparse enough to generate concern, and many more years are likely to pass before negative evidence li.e. failure to find the species despite repeated searches} accumu- lates to the point where extinction is the most probable explanation. In other words, unless circumstances are exceptional, monitoring of recent extinction events has a resolution limit measured in years or decades. This is why it is not possible to state with precision how many species became extinct in a given month, year or even decade, nor to predict exactly how many species, let alone which ones, are going to become extinct this year or decade or century. The search effort and chance both play a part in determining whether a species not seen for decades is rediscovered, as shown by the occasional new encounter with Humans, food and biodiversity 51 | Neen eee eee en enn y// Standing out as some good news against a background of widespread species depletion, a small but significant number of species have been considered extinct but rediscovered after a gap of several decades. In some cases, prolonged directed searches have been carried out, but others have emerged by chance. Sometimes a ‘rediscovery’ is in part a consequence of new systematic work confirming the species status of long-neglected populations, e.g. differentiation of the pygmy mouse lemur Microcebus myoxinus from other mouse lemurs on Madagascar. The coelacanth Latimeria is an extreme ‘Lazarus taxon’, being the living representative of an entire order (Coelacanthiformes} of early fishes thought to have become extinct some 80 million years ago. There are at least two population groups, one off southeast Africa (L. chalumnae) first discovered in 1938, and one in Indonesia (named L. menadoensis) discovered in 1998. * Bavarian pine vole Microtus bavaricus: an alpine species believed lost when a hotel was built near its single known locality, but recently rediscovered nearby in the Austrian Alps. ¢ Fiji petrel Pterodroma macgillivrayi: known by one specimen collected in 1855, but regarded as extinct until a bird flew into a researcher's headlight one night in 1984. ¢ Jerdon’s courser Rhinoptilus bitorquatus: known by just two museum skins and last recorded in 1900, until rediscovered in 1986 in a patch of scrub forest in southern India. * Jamaican iguana Cyclura collei: believed to have gone extinct during the 1940s but rediscovered in 1990 in the remote Hellshire Hills. « The Cranbrook pea Gastrolobium lehmannii: endemic to Western Australia where last collected from the wild in 1918 and since listed as extinct; rediscovered in 1995. species once feared extinct [see Box 4.2). Hidden survivors are even more likely in plants, which may have propagules that can remain viable but unseen for extremely long periods. Extinctions in the recent past are likely to be recorded with significant accuracy either where circumstances favor preservation of hard remains in good number (e.g. in caves, potholes or kitchen middens) or where early naturalists recorded the fauna or flora with sufficient care that they set a firm baseline against which the composition of the modern biota may be assessed. The detailed record of bird extinc- tion in Hawaii is a result of the former circumstance, and the record of mollusk extinction on many islands a result of the latter. Table 4.7 Late Pleistocene extinct and living genera of large animals Source: Adapted from Martin”. Africa 7 42 49 14 No peak North America 33 12 45 73 11 000-13 000 years ago South America 46 12 58 79 11 000-13 000 years ago Australia 19 3 22 86 ca 50 000 years ago Map 4.4 Human population density The relative density of human population based on census data relating to administrative units of various sizes. Source: CIESIN, gridded population of the world, version 2, data available online at http://sedac.ciesin.columbia.edu/ plue/gpw/ (accessed April 2002) 52 WORLD ATLAS OF BIODIVERSITY Population density High Early human impacts on biodiversity A wide range of factors affects the frequency of occurrence of particular species in the fossil record, of which the abundance of that species in life is only one. There is not necessarily therefore any direct relationship between the former and the latter, so that deducing past changes in abundance of species from the fossil record is a problematic exercise. Cataloging, though not dating, extinctions is rather less contentious, al- though even this may be problematic as evinced by the existence of so-called ‘Lazarus’ taxa [those presumed extinct that are rediscovered alive. One of the unusual features of the Quaternary period (the Pleistocene and Holocene} has been the disproportionately high extinction rates in the largest ter- restrial species, particularly mammals and birds (Table 4.7). These species are gen- erally referred to as the ‘megafauna’, often defined as those with an adult mass of 44 kilos or more, although the term has not been used consistently. The extinct American fauna include such well-known genera as the sabretooth cats Smilodon, giant ground sloths Eremotherium, glyptodonts Glyptotherium and mammoths Mammuthus, as well as a number of scavenging and raptorial birds, including the giant Teratornis and Cathartornis. Those in Australia include the marsupial equivalents of rhinoceroses [family Diprotodontidae] and lions (Thylacoleo}, giant wombats (Phascolonus, Ramsayia and Phascolomis), the large emu-like Genyornis and the giant monitor lizard Megalania. |In all, some 40 species of the larger Australian land mam- mals, reptiles and birds became extinct across the entire continent about 46 000 years ago", approximately 10 000 years after the first known human colonists. These extinctions have been followed by a similar series on islands during the Holocene. On New Zealand, the moas (giant flightless ratite birds in the family Anomalopterygidae) became extinct after the first humans came to the islands. On Madagascar, two endemic hippopotamus species Hippopotamus lemerle/ and H. madagascariensis, the elephantbird Aepyornis maximus, and a number of large to very large lemur species all appear to have died out 500 to 900 years ago. Similarly on the Caribbean islands, a number of large mammals, including several ground sloths (order Xenarthra, family Megalonychidae]) Humans, food and biodiversity 53 appeared to have survived until human occupation but to have died out at some point since then. The precise causes of all these extinctions have been the subject of endless debate, which centers chiefly on the role of humans. At one extreme lies the ‘blitzkrieg’ hypo- thesis, applied particularly to the apparently sudden collapse (i.e. over a few hundred years] of the North American megafauna at the hand of humans. At the other are those who maintain that in most, if not all, cases the impact of early humans was negligible and climate change, particularly increasing aridity, was the cause. Several features of the phenomenon seem to point persuasively to humans having played a pivotal role in most, if not all, of these extinctions. The most compelling is their Wy Map 4.5 Terrestrial wilderness The wilderness value of any given point is essentially a measure of remoteness from human influence, assessed on the basis of distance from settlement, access routes and permanent manmade structures. Source: GIS analysis by R. Lesslie [ANU], method developed for the Australian Heritage Commission 54 WORLD ATLAS OF BIODIVERSITY Wilderness level High Low timing. In each case the arrival of humans seems to have preceded the major spate of extinctions {in as much as these can be dated], with no or very few such extinctions having been recorded prior to human arrival (compare Map 4.1 and Table 4.7). The cumulative weight of this coincidence is difficult to counter, and is supported by population modeling. For example, new archeological evidence suggests that the first Polynesian settlements in New Zealand date from the late 13th century, and moas were becoming scarce by the end of the 14th. This evidence, coupled with a mathematical model of population and predation, indicates that all 11 moa species were driven to extinction within 100 years of human arrival’. Similarly, a computer simulation’ of population dynamics of humans and large herbivores in North America accurately models mega- faunal extinction in accord with archeological evidence for significant human arrival around 13 400 years ago and the first wave of extinction starting some 1 000 years later. The situation is somewhat different in Africa and Eurasia, where megafaunal extinctions were relatively few, and were spread out over the whole of the Pleistocene. In Africa the peak was in the lower Pleistocene (21 genera extirpated from the region between 1.8 million and 700 000 years ago compared with nine between 700 000 and 130 000 years ago, and seven later than 130000 years ago). It is noteworthy that this more gradual, earlier and less extreme pattern of extinctions has typified the region where humans evolved. It is difficult to formulate an entirely climate-based model of extinction that can account for this asynchronicity - outside Africa and Eurasia, these species survived a series of climatic changes at least as extreme as those they faced at the start of the Holocene. The recent studies of Genyornis newtoni and other megafauna in Australia’ * seem to indicate a widespread and largely synchronous dis- appearance from a wide range of habitat types during a period of relative climatic stability, strongly implying that some other agent was responsible. It is also noteworthy that the fossil and archeological evidence indicates that, on continents at least, these extinctions were not matched by parallel extinctions of smaller species (for example, as far as is known no insect species at all became extinct in Europe in the entire Pleistocene’). If climate change were the cause, it would be expected that these species would be at least as affected Humans, food and biodiversity 55 as larger ones as in general their opportunities for long-range dispersal and migration are much more limited. Even if humans are accepted as the major agents in these extinctions, evidence for the mechanisms involved in most cases remains elusive. It seems likely that extensive use of fire, and direct hunting of a fauna that had evolved in the absence of humans and was therefore unlikely to recognize them as potential predators, may have been sufficient cause to exterminate the large herbivores. The large carnivores and scavengers may then have suffered population collapses owing to the disappearance of their prey base. Extinction in modern times From the relatively sparse evidence that is available, it appears that amongst animals Wy Map 4.6 Vertebrate extinctions since AD1500 An indication of the number and former occurrence of recently extinct mammal, bird and freshwater fish species. Each of these three vertebrate classes is represented by a differently colored symbol, sized according to number of extinct species. Only extinctions that are fully resolved according to firm Committee on Recently Extinct Organisms (CREO) criteria are covered. The exception is Lake Victoria, indicated by a blue circle: some reports have suggested that up to a third of the 500 or so endemic fishes are extinct, but others note that available evidence is inconclusive. In many cases, including most islands and lakes, the position of the symbol indicates last record or core of former range. Where several species ranged more widely over a country, the symbol is positioned at the centre of that country. Source: Based on several sources; see Appendix 4 56 WORLD ATLAS OF BIODIVERSITY con ST ee eee ee ee ee ® Vertebrates e Birds ® Fishes ® Mammals e O Possibly-extinct Lake Victoria fishes Number of species 10-25 6-9 3-5 =2 more than 300 vertebrates, including at least 60 and possibly more than 80 mammals, more than 120 birds and around 375 invertebrates, have become extinct during the past 400 years {see numerical summary in Table 4.8 and list of extinct vertebrates in Appendix 4). Data for plants are much more equivocal, owing in part to the uncertain taxonomic status of many extinct plant populations. One source” lists some 380 extinct plant taxa and a further 370 or so classified as extinct or endangered (these include a number of infraspecific taxa and a number that although believed extinct in the wild are extant, and sometimes abundant, in cultivation). Because mammals and birds tend to be relatively well recorded, and because they leave recognizable macroscopic skeletal remains, it iS principally among these groups that known oe e@ o@ ; e te © e es e . e @ extinctions may be reasonably representative of actual extinctions. In these groups the known extinction rate over the past 400 years, based on data in Appendix 4, averages out at around 20-25 species per 100 years. A crucial question then, is how this observed extinction rate compares with some hypothetical or expected background extinc- tion rate. It is of course impossible to derive such a rate from observation of the modern world, as this has already been highly modified by human activity. The only reason- able comparison is thus with historical records, for which we must turn to the fossil record, discussed in Chapter 3. Although extinction rates have evidently been highly variable during the history of life on Earth, it seems that the average persistence time of species in the fossil record is around 4 million years. If 10 million species exist at any one time, on this basis the extinction rate would amount to around 2.5 species annually. However, it is unclear how representative the fossil record is of species as a whole. It is likely that many rare species or those with very restricted distribution never appear in it at all. These rare species may almost by definition be expected to be inherently more prone to extinction than the species that are recorded and may therefore be expected to have a lower persistence time. This would mean that average species duration was less - perhaps much less - than 4 million years and the actual extinction rate in geological time considerably higher than the rate observed in the fossil record. Applying a mean persistence time of 4 million years to birds and mammals [and Humans, food and biodiversity 57 TE Wy assuming some 10 000 species of the former and around 5000 of the latter), the back- ground extinction rate would be around one species every 500 years and 1000 years, respectively, so that current rates would be some 100 or 200 higher than background. Even if background rates in these groups were Table 4.8 Numbers of extinct animal species according to IUCN Note: Alternative criteria used by the Committee on Recently Extinct Organisms (CREO) result in a lower number of mammals and fishes being regarded as certainly extinct than is given in this table [see full list in Appendix 4). Source: Hilton-Taylor’”. Map 4.7 Threatened mammal species Color represents number of globally threatened mammal species in each country in 2000. Pie charts represent the proportion of the mammal fauna assessed as threatened at the national level in a small sample of countries. This is a highly generalized comparison because of differences in status assessment methods. Note: To reduce ambiguity Alaska (United States} has for the purposes of this map been assigned to the same class as adjacent Canada rather than the conterminous United States Source: Global data from Hilton- Taylor’”, country data from selection of national Red Data books. 58 WORLD ATLAS OF BIODIVERSITY Numbers threatened W140 a 22-3 a 3-27 0-12 National status — % threatened ten times this, the currently observed extinction rates would still be 10-20 times those expected. It seems therefore that, even if a high background extinction rate is postulated, recent extinctions are still much higher than might be expected (the alternative explanation is that the background rate was higher still). This elevated rate is particularly noteworthy as the Holocene appears to have been a period of relative climatic and geological stability, in which extinction rates might have been expected to be low. Most known extinctions have occurred on islands, including 42 of the 61 resolved mammal extinctions (68 percent), and 105 of the 128 bird extinctions (82 percent). Reasons for the former are probably twofold. First, island species do appear to be particularly extinction-prone, by virtue of their limited Mexico ranges and usually small population sizes, and also because they have often evolved in the absence of certain pressures (e.g. ter- restrial predators, grazing ungulates); if faced with these pressures (usually through human intervention] their populations may collapse completely. Second, it is much easier to arrive at some certainty that a given species is no longer present on an island of limited extent than that it has disappeared completely from a continental range, where the limits of its range were probably uncertain to start with. Most known or probable continental extinctions have been among freshwater organisms, particularly fishes and mollusks. Many freshwater biota appear to have the characteristics of island organisms, in that they have limited and highly circumscribed ranges and they are similarly often sensitive to Humans, food and biodiversity 59 hy) a Sp aye Phitippines Nifesg, jo Tomé 1 Principe South Africa e Y gf external pressures [e.g. introduction of preda- tory fish species, complete habitat destruction through drainage or dam construction). Figure 4.2 represents the number of accepted extinction events among mammals, 50 - birds and fishes in each third of the five centuries since AD1500 [see also Map 4.6). The information in Appendix 4 allows some elements of this broad global picture to be disentangled: GM Fishes | |) Birds (8) Mammals ~ (=) Ww i=) T ¢ early extinctions among the remaining archaic fauna of large islands, notably mam- mals on Hispaniola, Cuba and Madagascar; ¢ somewhat later extinctions, especially among ground birds, on many isolated small islands, such as St Helena, the ip} =} = Number of species S T Figure 4.2 Vertebrate extinctions by period since AD1500 This graphic is derived from information in the ‘Period’ column of the list of extinct vertebrates given in Appendix 4. ‘Early’ refers to the first four decades of a century, mid’ to the next three decades, ‘late’ to the final three decades. Only the extinction events regarded by Committee on Recently Extinct Organisms (CREO) criteria as ‘resolved’ are shown, and only the three groups recently assessed are represented. Source: Derived from Appendix 4. Mascarenes and others, during European colonial consolidation; early mid late early mid late early _ late early mid 16th century | 17th century | 18th century | 19th century | 20th century late early mid late Continental countries Brazil China India Colombia Peru Small islands French Polynesia Solomon Islands Mauritius (and Rodrigues) Sao Tome and Principe Saint Helena and dependencies Table 4.9 Island diversity at risk: birds Note: The first five rows are the continental countries (Indonesia and Philippines excluded} with most threatened bird species; the other five rows are the small island groups with most threatened bird species. 7 Source: BirdLife International” 60 WORLD ATLAS OF BIODIVERSITY Fr 1492 113 8 1100 23 i 923 68 7 1695 Th ig 5 1538 73 5 60 23 38 163 23 14 27 y 33 63 9 14 53 13 25 e increasing evidence of 20th century extinctions among fishes in many countries, and among marsupial mammals on the Australian mainland, with continuing losses of island species everywhere. The apparent reduction in extinction rate from the late 19th century onwards may be in part attributed to management action designed to maintain highly threatened species, and there is indeed good evidence that populations of a small number of target species have recovered significantly, but it is sure also to reflect the difficulties of observing and documenting extinction events. As mentioned above, many years may pass before a possibly extinct species can be treated with certainty as extinct, and the declining extinction rate toward the end of the 20th century is most probably in part an artifact of this monitoring process. Clearly, in view of our very incomplete knowledge of the world’s species, and the fact that only a minute proportion of living species are being actively monitored at any one time, it is extremely likely that more extinctions are occurring than are currently known. Indeed, most predictions of present and near-future extinctions suggest extremely high rates. Most are based on combining estimates of species richness in tropical forest with estimates of rate of loss of these forests, and predict species extinction on the basis of the general species-area relationship (which predicts a decline in species richness as area declines, see figure 5.1). It is widely believed that the great majority of all terrestrial species occur in tropical forests, and most of these species will be undescribed arthropods (notably beetles). At present rates of forest loss, it has been predicted that between 2 and 8 percent of forest species will become extinct, or committed to extinction, between 1990 and 2015°. Depending on whether higher or lower estimates of tropical species richness are used, extinction at this rate could entail loss of up to 100 000 species annually. As a cautionary note, it should be observed that few extinctions to date have actually been recorded in continental tropical moist forests, although monitoring species in these habitats presents great difficulty. Threatened species assessment Except for species lost as a result of random environmental factors, extinct species must largely be drawn from a pool of species that could be assessed as in decline or at risk, all of which face eventual extinction if negative trends or threats to their populations are not reversed. Various national and other programs have developed methods to assess the relative severity of risks faced by species, and to label species with an indicative category name. Conservation activities can then be prioritized on the basis of relative risk, taking account of other relevant factors, such as feasibility, cost and benefits, as appropriate. The system‘ developed by the Species Survival Commission of IUCN-the World Conservation Union (IUCN/SSC) and collabo- rators In conjunction with its Red Data Book” and Red List Programme” has been designed to provide an explicit and objective frame- work for assessment of extinction risk, and to be applicable to any taxonomic unit at or below the species level, and within any specified geographical or political area. To be categorized as threatened, any species has to meet one of five sets of criteria formulated to permit evaluation of all kinds of species, with a wide range of biological characteristics. The criteria are defined on population reduction, population size, geographic area and pattern of occurrence, and quantitative population analysis. The size and connectedness of different populations of a species influence the like- lihood of its survival. In general, small isolated populations will be more sensitive than larger connected ones to demographic factors (e.g. random events affecting the survival and reproduction of individuals] or environmental factors (e.g. hurricanes, spread of disease, changes in the availability of food). Islands tend to have a much higher proportion of their biota at risk than continental countries because they start with many fewer species, all or many of which face the risks associated with a small range size [see Table 4.9 for an example using bird data]. Biogeographic theory based around the species-area relationship, supported by much empirical evidence, predicts that each ‘habitat island’ created by fragmentation of a continuous habitat area will come at equilibrium to contain fewer species than previously. Human activities everywhere tend to promote fragmentation of natural and often species-rich habitats (e.g. primary tropical forest or temperate meadow grassland) and the spread of highly managed species-poor habitats [e.g. eucalypt plantations or cereal croplands). As a result, many species occur in just the kind of fragmented pattern that increases the risk of extinction. Recent declines Reduction in population numbers, or com- plete loss of a species from a site or an individual country (often termed extirpation) are far easier to observe than global species extinction, and appear liable to occur wherever humankind has modified the en- vironment for its own ends. The conservation status of most species is not known in detail, and this certainly applies to the many million as yet undescribed species, but two large animal groups - the mammals and birds - have been comprehensively assessed and may be representative of the status of biodiversity in general. Approximately 24 percent (1 130) of the world’s mammals and 12 percent (1 183) of the world’s bird species are regarded on the basis of IUCN/SSC Humans, food and biodiversity 6 TT bd criteria as threatened [see Table 4.10). Proportions are a great deal lower in other vertebrates, but none of these has been assessed fully. Empirical observations such as these give sufficient grounds for serious concern for biodiversity maintenance, regard- less of any hypotheses that have been pro- posed regarding the future rate of extinction. Interestingly, the ratio of threatened mam- mals to threatened birds is near 1:1, as is the ratio of recorded, recently extinct mam- mals to birds, giving some indication that threatened species categories may be a reasonably reliable indicator of proneness to extinction, and also that mammals may as a group be somewhat more susceptible than birds to extinction. Countless other species, although not yet globally threatened, now exist in reduced num- bers and fragmented populations, and many of these are threatened with extinction at national level. The significance of loss of diversity at gene level implied by loss of local populations of species is not clear, although it has been argued that loss of resilience in response to environmental change is inevitable. Table 4.10 Threatened species Notes: Includes species assessed as globally threatened and assigned to categories critically endangered’, endangered’ or vulnerable’ under IUCN criteria. Mosses not included here; most plant taxa listed were assessed for the World list of threatened trees” in 1998, Source: Hilton-Taylor’’; BirdLife International’. 62 WORLD ATLAS OF BIODIVERSITY re Map 4.8 Critically endangered mammals and birds The general distribution of almost all the 362 mammal and bird species categorized as critically endangered’, the highest risk category, In 2000. Each circle represents one distribution record; some species are known from a single point locality, others are represented by a cluster of localities. On this map, a high density of symbols can represent many records of a single species, or single records of many separate species. Red and blue symbols represent mammals and birds, respectively. 9 Source: Mammals: species selection based on Hilton-Taylor”, distribution research and mapping by UNEP-WCMC; birds: spatial data provided by BirdLife International (February 2002), further information in BirdLife International”. ° Critically endangered Mammals Birds At global level, most of the species assessed as threatened are terrestrial forms (Table 4.11]. The preponderance of terrestrial species is because the great majority of mammals and birds are terrestrial, and little or nothing is known of the population status of most aquatic species in most groups. Where significant numbers of aquatic species have been assessed, e.g. among crustaceans and mollusks, the proportion of threatened aquatic species rises markedly. Among fishes, the high number of freshwater species doubtless in part reflects the general lack of data on marine species, but to some extent indicates relative risk - many fresh- water species being restricted to small and isolated habitat patches. Evidence of the vulnerable nature of freshwater habitats and the risk faced by 1) conn Pa ica) i gro ie} 3, oo (oo) (a) cs) many aquatic groups is accumulating. For example, in the United States, freshwater groups are considerably more threatened than terrestrial groups (specifically, nearly 70 percent of the mussels, 50 percent of the crayfish and 37 percent of the fishes)”. Forest is an important habitat for a high pro- portion of the threatened terrestrial verte- brates. Among birds, for example, about 70 percent of the 1186 species assessed as threatened in the year 2000 occur in forest, and 25 percent in grassland, savannah and scrub habitats”. Of the threatened forest birds, 41 percent occur in lowland moist forest and about 35 percent in montane moist forest”. Among the 515 mammals regarded as threatened in 2000 that were assigned to a habitat category, 33 percent occur in lowland moist forest and 22 percent in montane formations”. , food and biodiversity 63 I Bey ) rs} bate) o Ay roy co) ; o J y oO. c) ° fe) 9 5) PR Saee z = fo} ae Eig. no ° e 7AM) b fe) aX ° : 2 eae of g 2 rac} ? 2. ° & % cS} } i) oo > N/ ‘= fa) i ) 9)! if ae 2 : : a 6° ( = ° fs} =. S io)e) fo} Q o # ® po ~ 2 ~~ ° ] yess o@ se f 13) =a o & 23 Oo, ° ° ° : 2 re g 8 ea "Oe C) Sin % 0 f » 6 3 L cay 9 3 1 oO Le, ° ome ° Numbers of globally threatened species can be mapped at country level [e.g. Map 4.7, mammals]. Maps of this kind provide at best a broad overview of the occurrence of such species, shown as direct numbers or as a proportion of the total number in that group in the country. As with other country-level biodiversity analyses, this information could be used to help focus efforts to slow or reverse biodiversity decline. Data with improved spatial resolution (although plotted in a highly simplified way) are shown in Map 4.8, representing the individual ranges of all the mammals and birds assessed as ‘critically endangered’. BirdLife International has gen- erated individual distribution maps of all threatened birds. Among other applications, these are being used to develop the first-ever world map of all the threatened species in an entire major group of organisms. A preliminary version of this, as a global density surface, is shown in Map 4.9. Being geo- referenced rather than country based, and at relatively fine spatial resolution, this has considerable potential to focus bird con- servation efforts effectively. Table 4.11 Number of threatened animal species in major biomes Notes: Counts include globally threatened species tabulated in each biome (IUCN categories ‘critically endangered’, endangered’ or ‘vulnerable’). Only mammals and birds have been comprehensively assessed. Some species, e.g. amphibians and migratory fishes, are counted in more than one biome row. Plants not tabulated: almost all assessed as globally threatened are terrestrial. Inland water includes saline wetlands, cave waters, etc., as well as freshwaters. Source: Hilton-Taylor’”. Marine 315 25 105 9 0 Inland water 1932 31 1B Nid 131 Terrestrial S627) lil till il 4a mene COC umes) Map 4.9 Threatened bird species density BirdLife International has developed digital range maps of all threatened bird species. This map is based on distribution data for the 1 186 species assessed as threatened in 2000. The data are plotted as a global density surface, representing the number of species potentially present in each location. The map is the first such treatment of any large group of threatened species. Source: Density surface provided by BirdLife International, further information on threatened birds in BirdLife International” 66 WORLD ATLAS OF BIODIVERSITY FO re ee Sere eS ere ea area rere Threatened bird species > density High a i | Low Proximate causes of recent declines The continuing conversion of natural habitats to cropland” and other uses typically entails the replacement of systems rich in bio- diversity with monocultures or systems poor in biodiversity. Habitat modification, from agricultural conversion and a variety of other causes, is in general the most important factor acting to increase species’ risk of extinction. Among species assessed as globally threatened in 2000°, habitat modification is the principal threat affecting more than 80 percent of the mammals, birds and plants; it is similarly predominant in several other major groups, notably in 95 percent of threatened bivalve mollusks, and is the main cause of loss in 75 percent of extinct freshwater fishes”. A second major source of biodiversity loss is the widespread introduction of species outside their natural range where they typically induce change at the community and ecosystem level. The effects of alien species are especially pronounced in closed systems such as lakes and islands. Introduced species, such as rats and cats, are cited as a cause of extinction in nearly 40 percent of the approximately 200 species where cause could be attributed; the majority were island forms“. Seven endemic snails in French Polynesia have been extirpated following the late 1970s introduction of a carnivorous snail species (Euglandina rosea) intended to control another introduced species (the giant African snail Achatina fulica), itself an agricultural pest. Accidental introduction of the brown tree snake Boiga irregularis to Guam in 1968 led to decline of the entire avifauna, of which one species is now thought extinct and one is extinct in the wild”. Introduction of the Nile perch (Lates niloticus) to Lake Victoria has contributed to the decline or extinction of nearly 200 native and endemic cichlid fishes. Overall, intro- duction events probably number in the thousands at the global level” and, although cases involving animals have been cited to illustrate the scope of the problem, intro- duced plants are in many places as pervasive and damaging. High trade demand for certain species and products, whether for international markets consuming hardwoods, sea fish, live animals, and plants and derivatives, or local markets consuming commodities such as bushmeat or turtle eggs, can readily push exploitation beyond the production capacity of the Humans, food and biodiversity 65 resource. The direct impact of hunting, collecting and trade is the second most important threat category among globally threatened mammals and birds, with around 35 percent in each group affected’. Rapid environmental change, such as that associated with El Nino Southern Oscillation (ENSO} events, can have significant impacts on natural habitats. For example, in 1997-98 climate fluctuation associated with El Nino was implicated in the persistence and spread of fires in Brazil, Indonesia and elsewhere: an estimated 1 million hectares of savannah woodland burned in Brazil and a similar area of forests in Indonesia were affected by fire. The effect of events of this type will be multiplied many times wherever habitats are already fragmented and species are depleted. | Wy 6 WORLD ATLAS OF BIODIVERSITY Be a ET REFERENCES 1 Coope, G.R. 1995. Insect faunas in ice age environments: Why so little extinction? In: Lawton, J.H. and May, R.M. (eds). Extinction rates. Oxford University Press, Oxford. | 2 Adams, J. 1999. Sudden climate transitions during the Quarternary. Progress in Physical Geography 23(1): 1-36. 3 Van Couvering, J.A. {ed.) 1997. The Pleistocene boundary and the beginning of the Quarternary, Cambridge University Press, Cambridge. 4 Petit et al. 1999. Climate and atmospheric history of the past 420,000 years from the Vostok ice core, Antarctica. Nature 399: 429-436. 5 Larick, R. and Ciochon, R.L. 1996. The African emergence and early Asian dispersal of the genus Homo. American Scientist November-December. | 6 Haile-Selassie, Y. 2001. Late Miocene hominids from the Middle Awash, Ethiopia. Nature | 412: 178-181. 7 Wood, B. and Brooks, A. 1999. We are what we ate. Nature 400: 219-220. Gabunia, L. et al. 2000. Earliest Pleistocene hominid cranial remains from Dmanisi, Republic of Georgia: Taxonomy, geological setting, and age. Science 288: 1019-1025. § Zhu, R.X. et al. 2001. Earliest presence of humans in northeast Asia. Nature 413: 413-417. 10 Bermudez de Castro, J.M. et al. 1997. A hominid from the Lower Pleistocene of Atapuerca, Spain: Possible ancestor to Neandertals and modern humans. Science 276: 1392-1395. 11 Dennell, R. 1997. The world’s oldest spears. Nature 385: 767-768. 12 Cann, R.L. 2001. Genetic clues to dispersal in human populations: Retracing the past from the present. Science 291: 1742-1748. 13 Rogers, A.R. 2001. Order emerging from chaos in human evolutionary genetics. Proceedings of the National Academy of Sciences 98: 779-780. 14 Barbujani, G. and Bertorelle, G. 2001. Genetics and the population history of Europe. Proceedings of the National Academy of Sciences 98: 22-25. Thorn, A. et al. 1999. Australia’s oldest human remains: Age of the Lake Mungo skeleton. Journal of Human Evolution 36: 591-612. 16 Roberts, R.G. et al. 2001. New ages for the last Australian megafauna: Continent-wide extinction about 46,000 years ago. Science 292: 1888-1892. Dillehay, T.D. 1997. Monte Verde: A Late Pleistocene settlement in Chile, Volume 1: The archaeological context and interpretation. Smithsonian Institution Press, Washington DC. 18 Guidon, N. and Delibrias, G. 1986. Carbon-14 dates point to man in the Americas 32,000 years ago. Nature 321: 769-771. 19 Haynes, C.V. 1984. Stratigraphy and Late Pleistocene extinction in the United States. In: Martin, P.S. and Klein, R.G. {eds). Quaternary extinctions, pp. 345-353. University of Arizona Press, Tucson. 20 Harcourt, C.S. and Sayer, J.A. 1995. The conservation atlas of tropical forests: The Americas. IUCN-the World Conservation Union, Cambridge and Gland. | 21 Pimm,L.S., Moulton, M.P and Justice, L.J. 1995. Bird extinctions in the central Pacific. In: Lawton, J.H. and May, R.M. (eds). Extinction rates, pp. 75-87. Oxford University Press, Oxford. 22 Dewar, R.E. 1984. Mammalian extinctions and stone age people in Africa. In: Partin, P.S. and Klein, R.G. (eds). Quaternary extinctions, pp. 553-573. University of Arizona Press, Tucson. 23 Semaw, S. et al. 1997. 2.5-million-year-old stone tools from Gona, Ethiopia. Nature 385: 333-336. 24 Thieme, H. 1997. Lower Palaeolithic hunting spears from Germany. Nature 385: 807-810. 25 Morwood, M.J. et al. 1998. Fission-track ages of stone tools and fossils on the east Indonesian island of Flores. Nature 392: 173-176. ol “I 26 27 28 29 30 3 = 32 33 34 35 36 37 38 39 40 4 a 42 43 44 45 46 47 48 49 Humans, food and biodiversity 67 ua a I EE ba Bird, M.I. and Cali, J.A. 1998. A million-year record of fire in sub-Saharan Africa. Nature 394: 767-769. Gowlett, J.A.J. et al. 1981. Early archaeological sites, hominid remains and traces of fire from Chesowanja, Kenya. Nature 294: 125-129. Isaac, G. 1981. Early hominids and fire at Chesowanja, Kenya. Nature 296: 870. de Heinzelin, J. et al. 1999. Environment and behavior of 2.5 million year old Bouri hominids. Science 284: 625-629. James, S.R. 1989. Hominid use of fire in the Lower and Middle Pleistocene: A review of the evidence. Current Anthropology 30: 1-26. Hoopes, J.W. 1996. In search of nature. Imagining the precolombian landscapes of ancient Central America. Working paper for the Nature and Culture Colloquium, University of Kansas, Lawrence. Teaford, M.F. and Ungar, P.S. 2000. Diet and the evolution of the earliest human ancestors. Proceedings of the National Academy of Sciences 97: 13506-13511. Ambrose, S.H. 2001. Paleolithic technology and human evolution. Science 291: 1748-1753. Richards, M.P. et al. 2000. Neanderthal diet at Vindija and Neanderthal predation: The evidence from stable isotopes. Proceedings of the National Academy of Sciences 97: 7663-7666. Richards, M.P. et al. 2001. Stable isotope evidence for increasing dietary breadth in the European mid-Upper Paleolithic. Proceedings of the National Academy of Sciences 98: 6528-6532. Stiner, M.C. 2001. Thirty years on the Broad Spectrum Revolution’ and paleolithic demography. Proceedings of the National Academy of Sciences 98: 6993-6996. Clutton-Brock, J. 1995. Origin of the dog: Domestication and early history. In: Serpell, J. ed.) The domestic dog: Its evolution, behaviour and interactions with people, pp. 8-20. Cambridge University Press, Cambridge. Vila, C. et al. 1997. Multiple and ancient origins of the domestic dog. Science 276: 1687-1689. Smith, B.D. 1998. The emergence of agriculture. Scientific American Library. Smith, B.D. 2001. Documenting plant domestication: The consilience of biological and archaeological approaches. Proceedings of the National Academy of Sciences 98: 1324- 1326. Luikart, G. et al. 2001. Multiple maternal origins and weak phylogeographic structure in domestic goats. Proceedings of the National Academy of Sciences 98: 5927-5932. FAO 1998. The state of the world’s plant genetic resources for food and agriculture. Food and Agriculture Organization of the United Nations, Rome. World Conservation Monitoring Centre 1992. Global biodiversity: Status of the Earth's living resources. Groombridge, B. {ed.] Chapman and Hall, London. Prescott-Allen, R. and Prescott-Allen, C. 1990. How many plants feed the world? Conservation Biology 4(4): 365-374. Palsson, G. 1988. Hunters and gatherers of the sea. In: Ingold, T., Riches, D., Woodburn, J. Hunters and gatherers 1: History, evolution and social change, pp. 189-204. BERG, Oxford. FAOSTAT database. Food and Agriculture Organization of the United Nations. Available online at http://apps.fao.org/ (accessed February 2002). FAO 1996. Food requirements and population growth. World Food Summit technical background document TBD VI E4. Food and Agriculture Organization of the United Nations. Available online at http://www.fao.org/wfs/index_en.htm {accessed February 2002). Uvin, P. The state of world hunger. Hunger report. Gordon and Breach. Available online at http://www.brown.edu/Departments/World_Hunger_Program/hungerweb/intro/ StateofWorldHunger.pdf (accessed April 2002). Committee on World Food Security 2001. Mobilizing resources to fight hunger. 2002 World Food Summit background document CFS:2001/Inf.7. 68 WORLD ATLAS OF BIODIVERSITY eee; ;;;;;;;; xx 50 de Garine, |. and Harrison, G.A. (eds) 1988. Coping with uncertainty in food supply. Clarendon Press, Oxford. Fernandes, E.C.M., Okingati, A. and Maghembe, J. 1984. The Chagga homegardens: A multi-storied agroforestry cropping system on Mount Kilimanjaro (northern Tanzania). Agroforestry Systems 2: 73-86 52 Juma, C. 1989. The gene hunters: Biotechnology and the scramble for seeds. Zed Books, London. 53 Marsh, R.R. and Talukeder, A. 1994. Effects of the integration of home gardening on the production and consumption of target interaction and control groups: A case study from Bangladesh. Unpublished paper. 54 McEvedy, C. and Jones, R. 1978. Atlas of world population history. Penguin Books, Harmondsworth. 55 Roberts, N. 1998. The Holocene. An environmental history. 2nd edition. Blackwell Publishers, Oxford. 56 UN Population Division 2001. World population prospects. The 2000 revision. Draft. ESA/P/WP.165. Available online at http://www.un.org/esa/population (accessed April 2002). 57 Smil, V. 1999. How many billions to go? Nature 401: 429. 58 Yalden, D.W. 1996. Historical dichotomies in the exploitation of mammals. In: Taylor, V.J. and Dunstone, N. The exploitation of mammal populations, pp. 16-27. Chapman and Hall, London. 59 Wright, D.H. 1990. Human impacts on energy flow through natural ecosystems, and implications for species endangerment. Ambio 19(4]: 189-194. 60 Haberl, H. 1997. Human appropriation of net primary production as an environmental indicator: Implications for sustainable development. Ambio 26(3): 143-146. Holdaway, R.N. and Jacomb, C. 2000. Rapid extinction of the moas (Aves: Dinornithiformes]: Model, test, and implications. Science 287: 2250-2254. Alroy, J. 2001. A multispecies overkill simulation of the end-Pleistocene megafaunal mass extinction. Science 292: 1893-1896. 63 Miller, G.H. et al. 1999. Pleistocene extinction of Genyornis newtoni: Human impact on Australian megafauna. Science 283: 205-208. 64 UNEP-WCMC species conservation database (plants). Available online at http://www.unep-wemc.org/species.plants/plants-by-taxon.htm#search 65 Reid, W.V. 1992. How many species will there be? In: Whitmore, T.C. and Sayer, J.A. {eds}. Tropical deforestation and species extinction, pp. 55-73. Chapman and Hall, London. 66 IUCN 1994. IUCN Red List categories. IUCN-the World Conservation Union, Gland. 67 Hilton-Taylor, C. (compiler) 2000. 2000 IUCN Red List of threatened species. |IUCN-the World Conservation Union, Gland and Cambridge. Also available online at http://www.redlist.org/ [accessed April 2002). 68 The Species Survival Commission of IUCN-the World Conservation Union. Red List material. Available online at http://iucn.org/themes/ssc/red-lists.htm (accessed February 2002). 69 Master, L.L., Flack, S.R. and Stein, B.A. (eds) 1998. Rivers of life: Critical watersheds for protecting freshwater biodiversity. Nature Conservancy, Arlington. BirdLife International 2000. Threatened birds of the world. Lynx Edicions and BirdLife International, Barcelona and Cambridge. Richards, J.F. 1990. Land transformation. In:Turner Il, B.L. et al. (eds). The Earth as transformed by human action, pp. 163-178. Cambridge University Press, with Clark University. 72 United Nations Environment Programme 1995. Heywood, V. led.} Global biodiversity assessment. Cambridge University Press, Cambridge. 5 “ 6 6 N ~s 7 oO 7 Humans, food and biodiversity 69 73 74 75 76 77 78 79) 80 8 = 82 83 84 85 86 87 88 89 90 9 = 92 Martin, P.S. 1984. Prehistoric overkill: The global model. In: Martin, P. and Klein, R.G. (eds). Quaternary extinctions, pp. 354-403. University of Arizona Press, Tucson. Juma, C. 1989. Biological diversity and innovation: Conserving and utilizing genetic resources in Kenya. African Centre for Technology Studies, Nairobi. Alvarez-Buylla Roces, M.A., Lazos Chavero, E. and Garcia-Burrios, J.R. 1989. Homegardens of a humid tropical region in South East Mexico: An example of an agroforestry cropping system in a recently established community. Agroforestry Systems 8: 133-156. Alcorn, J.B. 1984. Development policy, forests and peasant farms: Reflections on Haustec- managed forest contribution to commercial and resource conservation. Economic Botany 38(4]: 389-406. Davies, A.G. and Richards, P. 1991. Rain forest in Mende life: Resources and subsistence strategies in rural communities around the Gola North Forest Reserve [Sierra Leone]. A report to the Economic and Social Committee on Overseas Research (ESCOR}, UK Overseas Development Administration, London. Michon, G., Mary, F. and Bompard, J. 1986. Multistoried agroforestry garden system in West Sumatra, Indonesia. Agroforestry Systems 4(4): 315-338. Kunstadter, P., Chapman, E.C. and Sabhasri, S. 1978. Farmers in the forest: Economic development and marginal agriculture in Northern Thailand. An East-West Center Book, University Press of Hawaii, Honolulu. Padoch, C. and de Jong, W. 1991. The house gardens of Santa Rosa: Diversity and variability in an Amazonian agricultural system. Economic Botany 45(2): 166-175. Denevan, W.M. and Treacy, J.M. 1987. Young managed fallows at Brillo Neuvo. In: Denevan, W.M. and Padoch, C. (eds). Swidden-fallow agroforestry in the Peruvian Amazon. Advances in Economic Botany 5, pp. 8-46. New York Botanical Garden, New York. Ohtsuka, R. 1993. Changing food and nutrition of the Gidra in lowland Papua New Guinea. In: Hladik, C.M. et al. {eds). Tropical forests, people and food. Biocultural interactions and applications to development. Man and the Biosphere Series. Volume 13. UNESCO and Parthenon Publishing, Paris and Carnforth. Conklin, H.C. 1954. An ethnoecological approach to shifting agriculture. Transactions of the New York Academy of Sciences 17: 133-142. Posey, D.A. 1984. A preliminary report on diversified management of tropical forest by the Kayapo Indians of the Brazilian Amazon. Advances in Economic Botany 1: 112-126. Vickers, W.T. 1993. Changing tropical forest resource management strategies among the Siona and Secoya Indians. In: Hladik, C.M. et al. (eds). Tropical forests, people and food. Biocultural interactions and applications to development. Man and the Biosphere Series. Volume 13. UNESCO and Parthenon Publishing, Paris and Carnforth. Michon, G. 1983. Village forest gardens in West Java. In: Huxley, P. {ed.] Plant research and agroforestry, pp. 13-24. International Centre for Research in Agroforestry, Nairobi. Whitman, W.B., Coleman, D.C. and Wiebe, W.J. 1998. Prokaryotes: The unseen majority. Proceedings of the National Academy of Sciences 95: 6578-6583. Rusek, J. 1998. Biodiversity of Collembola and their functional role in the ecosystem. Biodiversity and Conservation 7: 1207-1219. Zimmerman, P.R. et al. 1982. Termites: A potentially large source of atmospheric methane, carbon dioxide, and molecular hydrogen. Science 218: 563-565. Nicol, S. Time to krill? Antarctica Online: The website of the Australian Antarctic Program: http://www.antdiv.gov.au/science/bio/issues_krill/krill_features.html Gaston, K.J. and Blackburn, T.M. 1997. How many birds are there? Biodiversity and Conservation 6: 615-625. Said, M.Y. et al. 1995. African elephant database 1995. IUCN-the World Conservation Union, Gland. Wu) 70 WORLD ATLAS OF BIODIVERSITY nner nnn n nnn rrr nrnnnnnnrcnnne aE 93 Santiapillai, C. 1997. The Asian elephant conservation: A global strategy. Gajah. Journal of the Asian Elephant Specialist Group 18: 21-39. 94 Nowak, R.M. 1991. Walker’s mammals of the world. 5th edition. Volume II. Johns Hopkins University Press, Baltimore. 95 Jefferson, T.A., Leatherwood, S. and Webber, M.A. 1993. Marine mammals of the world. FAO Species Identification Guide. United Nations Environment Programme, Rome. 96 Kemf, E. and Philips, C. 1995. Whales in the wild. WWF Species Status Report, WWF - Worldwide Fund for Nature, Gland. 97 Oldfield, S., Lusty, C. and MacKiven, A. 1998. The world list of threatened trees. World Conservation Press, Cambridge. 98 Harrison, |.J. and Stiassny, M.L.J. 1999. The quiet crisis: A preliminary listing of the freshwater fishes of the world that are extinct or ‘missing in action’. In: MacPhee, R.D.E. (ed.] Extinctions in near time: Causes, contexts, and consequences, pp.271-332. Kluwer Academic/Plenum Publishers, New York. Terrestrial biodiversity m EI TOG EIT EEE SEL LEC TE IAL TELIA PEN FLT PEE EI LH Terrestrial biodiversity Earth’s surface but they are more accessible than aquatic habitats and so are far better known. The land supports fewer phyla than the oceans but most of the global diversity of species, and is characterized above all by an extensive cover of vascular plants, with associated animals and other groups of organisms. Forest and woodland ecosystems form the predominant natural landcover over most of the Earth’s surface. These systems generate around half the terrestrial net primary production, and forests in the tropics are believed to hold most of the world’s species. Approximately half the area of forest developed in post-glacial times has since been cleared or degraded by humans, and the amount of old-growth forest continues to decline. Grassland, shrubland and deserts collectively cover most of the unwooded land surface, with tundra on frozen subsoil at high northern latitudes. These areas tend to have lower species diversity than most forests, with the notable exception of Mediterranean-type shrublands, which support some of the most diverse floras on Earth. Humans have extensively altered most grassland and shrubland areas, usually through conversion to agriculture, burning and introduction of domestic livestock. They have had less immediate impact on tundra and true desert regions, although these remain vulnerable to global climate change. T ERRESTRIAL ECOSYSTEMS EXTEND OVER LITTLE MORE than one quarter of the THE TERRESTRIAL BIOSPHERE scapes tending to replace trees above Land and water on the Earth's surface Land extends over nearly 150 million square kilometers (km*] or about 29 percent of the total surface of the planet, the remainder being covered by oceans. With the continents in their present position, more than two thirds of the land surface is in the northern hemisphere, and the area of land situated in the northern hemisphere above the Tropic of Cancer slightly exceeds that in the rest of the world put together (Table 5.1). A small proportion of the total land area is occupied by inland water ecosystems: lakes and rivers cover around 2 percent and swamps and marshland a similar amount. About half of the land surface, approximately 52 percent, is below 500 meters {m) in elevation, and the mean elevation is 840 m. A minor but significant proportion of the land surface is mountainous in nature, with alpine tand- 1 000 m at higher latitudes and above 3 500 m in the tropics. The land as an environment for living organisms The environmental conditions prevailing in any given place determine what kinds of organism can live there. Relevant environ- mental conditions include various aspects of the physical environment, and the other Northern hemisphere North of Tropic of Cancer Equator north to Tropic of Cancer Equator south to Tropic of Capricorn South of Tropic of Capricorn (including Antarctica) Southern hemisphere Table 5.1 Global distribution of land area, by latitude bands SS] bid! 72 WORLD ATLAS OF BIODIVERSITY Most terrestrial organisms occupy an environment where water loss is a constant threat. species that any given species directly or indirectly interacts with. These interactions may occur through a range of mechanisms, including competition, predation, symbiosis, mutualism and parasitism. For many living species, interactions with humans are now a significant feature of their environment. The most fundamental distinction, at least for macroscopic organisms, is between ter- restrial and aquatic environments. All living organisms require water because the basic processes of life take place in the aqueous medium found in cells. However, most terrestrial organisms, in contrast to aquatic ones, occupy an environment where water loss is a constant threat. The anatomical and physiological solutions to this problem are many and varied, including epidermal and/or cuticular layers with reduced permeability to water, water storage tissues, and metabolic or behavioral processes that conserve water. However, the greater the risk of dehydration, the more carbon and energy are needed for water conservation mechanisms, leaving less available for other adaptations. Thus, life tends to occur at low density and at low diversity in hyperarid environments, whether exceptionally hot or exceptionally cold. There are a number of other extremely important differences between air and water as a medium for living things. It takes far less energy (about 500 times less} to raise the temperature of a given mass of air by 1 degree Celsius than to heat the same mass of water by the same amount, and water conducts heat far more rapidly than air. As a result, while aquatic organisms are buffered against rapid fluc- tuation in their surroundings, terrestrial organisms can be subjected to wide extremes in temperature, corresponding to daily and seasonal variation in insolation. The air surrounding terrestrial organisms is much less dense {about 800 times less) than water, and so land organisms must support themselves against the full effects of gravity, but are not subject to the large forces exerted on aquatic organisms by moving water. Oxygen is freely and uniformly available in the atmosphere, but in water is far less concentrated and much more variable in time and space. On the other hand, mineral nutrients dissolve readily in water and it is possible for aquatic organisms to extract nutrients directly from their immediate surroundings (although in many aquatic habitats such nutrients are only present in low concentrations}. On land, in contrast, mineral nutrients occur in soil where their distribution and concentration are spatially variable, and they are directly accessible primarily to microorganisms, plants, fungi and algae. Global variations in terrestrial habitats The wide variations in water availability and temperature regimes prevailing on the land surface interact in often complex ways with a number of other factors, including geology, soil, terrain, wind, fire regimes and human activities, to generate the immense range of environments apparent on the Earth’s land surface. Classification of these into a manage- able system is a major problem in biology, one not merely of theoretical interest but of considerable importance in the management and conservation of the biosphere. The problem arises largely from a need to divide the natural environment into a series of discrete bounded units for the purposes of mapping, measuring and monitoring habitats, whereas the world in reality appears to forma highly variable continuum. Where gradients exist between different physical regimes (particularly of water and temperature], habitat types tend to intergrade imperceptibly, and it is impossible without arbitrary definitions to distinguish, for example, grassland with a few trees from open woodland with grass ground cover. Even broad categor- ies, such as ‘forest’ or ‘wetland’, inevitably require arbitrary limits to be set, e.g. for the density of tree cover necessary before an area can be called a forest, or for the duration of flooding necessary before an area can be classified as a wetland rather than a terrestrial system. In such circumstances, it is important to keep in mind the inherent variability of ecosystems, rather than attach undue signifi- cance to their labels. Vegetation and chlorophyll The global distribution of actively growing vegetation can be visualized without classi- fication or criteria relating to structure, to physiognomy or to species composition. Advanced very high resolution radiometer (AVHRR) satellite sensors measure the reflectance of vegetation, primarily of the green photosynthetic pigment chlorophyll, in the visible and the near infrared part of the spectrum. On land this can be interpreted as broadly equivalent to the density and vigor of green plant growth, represented as the normalized difference vegetation index (NDVI). This is plotted, aggregated over one year, in Map 5.1. This reveals a clear distinction between areas rich in standing growth of plants, whether cropland or natural vegetation, and areas where standing plant growth is sparse or absent - these are essentially the drylands and rangelands of conventional landuse classifications. Spatial variation in chlorophyll density is only indirectly related to variation in primary production levels; net primary production depends further on soil and climate conditions and community dynamics within ecosystems (see Chapter 1]. So, while Map 5.1 shows high chlorophyll in both tropical and high latitudes, net primary production is far higher in the former than the latter, where it is restricted mainly by seasonally unfavorable climatic conditions (compare with Map 1.2). Landcover and ecosystems Landcover classification is more directly con- cerned with differences in the physical aspects Terrestrial biodiversity of ground cover, mainly for landuse planning and management, than with biodiversity or the community aspects of vegetation cover. Many current landcover maps (Map 5.2] are based on interpretation of remote-sensing data that have the virtue of being quantitative in nature and available in time series suitable for monitoring applications. Because the source data are typically interpreted in terms of a classification system developed to take account of conditions on the ground, landcover maps are subject to some of the definition problems mentioned above, no matter how generalized the categories are at the highest levels (e.g. ‘forest’, ‘cropland’, urban’). One of the most useful ecological distinctions to be made is between areas with extensive or significant tree cover, and areas with few or no trees. Terrestrial plant growth is favored by high soil water availability and relatively elevated temperature during all or a major part of the year. Trees tend to be the main plant growth form in such conditions, and forest or woodland the main vegetation. Conversely, primary production is strongly limited by a shortage of soil water. Grasses and low shrubs tend to be the main plant growth forms in such dryland regions and, where vegetation exists, it consists mainly of grassland, savannah or shrubland. Tree growth remains insignificant in those parts of the world where water is present in some form, but temperatures are too low for growth during all or part of the year. Grasses and low shrubs are the main plant growth forms in dryland regions. 73 Ni) 7% WORLD ATLAS OF BIODIVERSITY e = > Se eee < 2 = 22 Sen Map 5.1 Photosynthetic activity on land A map illustrating the normalized difference vegetation index (NDVI) calculated from AVHRR satellite data and aggregated over the year > 1998. NDVI values vary with Photosynthetic activity absorption of red light by plant chlorophyll and the High reflection of infrared radiation by water-filled leaf cells, and provide an indication of photosynthetic activity. The map reveals a clear distinction between areas rich in standing growth of plants, whether cropland or natural vegetation, and areas where standing plant growth is sparse or absent. Source: Adapted from image provided by the SeaWiFS Project, NASA/Goddard Space Flight Center and ORBIMAGE Low For example, in polar regions there is an so unavailable to living organisms. In tundra abundance of water, but because of permanent areas at high latitudes the subsurface is low temperature it is mainly in solid form and permanently frozen and plant growth is restricted to the few summer months when thawing of the superficial layers makes liquid water available. In upper alpine regions, liquid water is often present in seasonal abundance, Box 5.1 Defining ecosystems a The word ‘ecosystem’ was introduced by the plant ecologist Tansley in 1936, to refer to the but temperatures are too low over the year as communities of plants, animals, other organisms and the physical environment of any given a whole to allow tree growth. place. The concept reflected a new and visionary approach to biological research which Many areas theoretically have sufficient focused on the system-level flow of materials and energy between components at any given rainfall and a temperature regime suitable study site. The spatial boundaries of the system were originally of little or no significance. for supporting significant tree cover but do The term is now widely used, particularly in the context of environmental planning, to refer not do so. The most important reason for to broad biological communities of similar appearance, usually defined by physical, climatic, this is undoubtedly human intervention but structural or phenological features. Ecosystem diversity is generally understood to refer to in some areas this may also be a result of the range of different kinds of ecosystem, in this sense, within some defined area. In this natural causes. Nutrient availability may be usage, the ecosystem’ is treated as a map unit and the spatial boundaries of the system limiting. Slopes may be too steep to allow assume major significance. formation and retention of soil, so that trees cannot anchor themselves. Minerals may be 5s a 7 meaner od — present in toxic concentrations, or the fre- quency and intensity of natural fires or floods may be too high. Large herbivores may also prevent establishment of significant tree cover, although as noted in Chapter 4 many of those large herbivores in the Americas and Australasia that might have exerted such an influence in the past are now extinct. Landcover classifications can be used as the basis for habitat or ecosystem classi- fications (see Box 5.1}, by incorporating elements of species composition and com- munity structure. Ecosystem maps should have more direct application to biodiversity conservation and management than maps of landcover; however the development of such maps is constrained by the same problems of defining boundaries and developing a consis- tent approach to classification. Map 5.7 is a Bl ee representation of global forest cover, derived by applying a simple classification of five forest and woodland habitat types to the landcover data plotted in Map 5.2. Forest classification is discussed further below. GLOBAL VARIATIONS IN TERRESTRIAL SPECIES DIVERSITY Just as habitat types show great variation across the world’s land surface, so does the number of species that may be found in any given place. The spatial heterogeneity, wide range of present physical conditions and complex history of the world’s land surface all contribute to extreme variation in terrestrial biological diversity. Variation in species number is not strictly related to variation in habitat type because two areas that are structurally similar may have very different Terrestrial biodiversity 75 Syeprpemereore 8 76 WORLD ATLAS O Map 5.2 Global land cover Map adapted from the global landcover classification developed by the University of Maryland. The Maryland classification includes 13 classes and was based on AVHRR remote-sensing data with a spatial resolution of 1 km. For presentation purposes the data have here been generalized to a 4-km grid. Also for clarity, the two needleleaf forest classes in the original classification have been combined, as well as the two broadleaf forest classes, and the urban and built-up category has been omitted. Source: Data from University of Maryland Global Land Cover Facility. For 98 full description see Hansen . F BIODIVERSITY SE Cover class Broadleaf forest Needleleaf forest Mixed forest Woodland Wooded grassland Closed shrubland Open shrubland Grassland Cropland Bare ground Water bodies numbers of species present. The most impor- tant components of this variation can be expressed in three different, though not com- pletely independent, ways: e the kinds of organisms, particularly pri- mary producers, that live in any one place — this influences the kinds of habitats found there and is a reflection of ecosystem diversity; ¢ the numbers of different kinds of organisms that live in any one place - this is usually assessed by measures of species diversity, although other classification systems (e.g. guilds, functional groups, higher taxa] can also be used; the individual taxonomic identity of the organisms that live in any one place - this is determined by biogeography and influences, among other things, how a given area contributes to global biodiversity. Measuring species diversity Comparing the diversity of different parts of the world is complex because of the way diversity changes with scale. A wide range of observations has demonstrated that, as a general rule, the number of species recorded in an area increases with the size of the area, and that this increase tends to follow a predictable pattern known as the Arrhenius relationship, whereby: logS = c + zlogA where S = number of species, A= area and c and zare constants [see Figure 5.1). The slope of the relationship (z in the equation) varies considerably between surveys, although it is generally between 0.15 and 0.40, but some survey data do not fit the relationship at all. A —— common generalization from this finding is that a tenfold reduction in an area [i.e. loss of 90 percent of habitat) will result in the loss of between 30 percent (with z = 0.15) and 60 percent (with z = 0.40) of the species originally present, or approximately half the species. The relative diversity of different sites will often partly depend on the scale at which diversity is measured. Thus 1 m* of semi- natural European chalk grassland may contain many more plant species than 1 m’ of lowland Amazonian rainforest, whereas for any area larger than a few square meters this could be reversed. In other words, when an area is sampled the number of species recorded increases with the size of the area, but this rate of increase varies from area to area, i.e. the slope of the Arrhenius relationship is not constant everywhere. The reason for this increase may be quite straightforward. When small areas are sampled they are likely to be relatively homogeneous in terms of habitat type. Log number of species 0 T T 1 2 Log area wo pe Terrestrial biodiversity 77 Figure 5.1 A typical species-area plot Note: The data, consisting of species counts in a Series of areas of different size, are plotted on logarithmic axes resulting in a straight-line graph, the slope of which (z} indicates the rate at which species number changes with changing area. Rests 78 WORLD ATLAS OF atic a ee ree a a er a TS TE EE EC TT BIODIVERSITY At a small scale, as sample area increases, so an increasing proportion of the species present in that habitat is likely to be included in the sample. Beyond a certain point, however, as larger areas are sampled so an increasing number of different habitats will be included in the sample area, each with new species that are likely to be included in the sample. The species/area relationship therefore increasingly reflects habitat hetero- geneity at larger scales. Ecologists attempt to take account of this by recognizing different kinds of biological diversity. The diversity within a site or habitat is often referred to as alpha (a) diversity while the differences between habitats are referred to as beta (§) diversity. Thus an area with a wide range of dissimilar habitats will have a high fB- diversity, even if each of its constituent habitats may have low a-diversity. Differences in site diversity over large areas, such as continents, are sometimes referred to as gamma [y) diversity. Measures of diversity can refer simply to species richness but can also be more sophisticated statistical measures that take into account the relative abundance of different species in a given place. A variety of different measures of this kind has been developed (of Latitudinal variation in diversity on land is strongly correlated with, and may be largely explained by, variation in incident energy over the Earth's surface. The relationship between diversity and productivity, and related measures, has been the subject of long-standing debate in ecology, but recent studies have shown that at global or continental scale, organismal diversity, particularly as measured at higher taxonomic levels, is strongly correlated with available energy’. This kind of relationship has been demonstrated, for example, for flowering plants, for trees, Lepidoptera, land birds and land mammals in a range of countries and continents*, and for fishes in river basins at a global level’. One simplistic explanation for this may be that higher energy availability leads to increased net primary production (NPP], and this broader resource base allows more species to coexist. While the general relationship appears robust, the details are complex. Energy availability can be measured in several ways: as heat eneray, as potential (PET) or actual evapotranspiration (AET], or as NPP, and which is the best predictor of diversity has yet to be determined. Some measure of the simultaneous availability of water and radiant energy may provide the best general predictor of potential macro-scale species richness'’. More complete explanation for richness variation would need also to consider the roles of topography, history and edaphic factors. which H’, the Shannon-Wiener function, is a commonly used example). With many of these measures, an area in which all species are of similar abundance would generally be given a higher diversity measure than an area with the same number of species, a few of which were very abundant and the remainder rare. Deriving these statistical measures requires intensive sampling; for this reason, simpler measures of species richness tend to be more useful at larger scales. Major patterns of variation in global species diversity Despite the difficulties of establishing strictly comparable quantifiable measures, a wealth of empirical observations indicates that species richness in eukaryotes tends to vary geo- graphically according to a series of fairly well- defined rules. In terrestrial environments: ¢ warmer areas hold more species than colder ones; * wetter areas hold more species than drier ones; areas with varied topography and climate hold more species than uniform ones; less seasonal areas hold more species than highly seasonal ones; e areas at lower elevation hold more species than areas at high elevation. The single most obvious pattern in the global distribution of species is that overall species richness increases as latitude decreases toward the equator [see Box 5.2]. At its simplest this means that there are more species in total and per unit area in temperate regions than in polar regions, and far more again in the tropics than in temperate regions. This applies as an overall general rule, and within most individual higher taxa (at order level or higher), and within most equivalent habitats. The pattern can be seen in Maps 9.3, illustrating vascular plant species diversity, and 5.4, which represents country-level diversity in terrestrial vertebrates and vascular plants. There is good evidence that moist tropical forests are the most species-rich environ- ments on Earth. If current estimates of the number of as yet unknown species (see Chapter 2) in the tropical forest microfauna are accepted, these regions, extending over perhaps 7 percent of the world’s surface, may conceivably hold more than 90 percent of the world’s species. If tropical forest small insects are discounted, then coral reefs and, for flowering plants at least, areas of Mediterranean climate in South Africa and Western Australia may be similarly rich in species. Topographic heterogeneity may be expec- ted to play a significant part in determining species number for two reasons. First, such heterogeneity will increase habitat variability, thereby increasing the range of niches that can be occupied by different organisms. Second, depending on the size and mobility of organisms, the chances of geographic isolation and speciation increase in topographically diverse landscapes. The role of topography has been demonstrated statistically at continent scale for North American mammals’, and at both landscape and patch scale for vascular plants”. The available information on distribution of species is geographically very incomplete, and relates to only a small fraction of the 1.7 million known species. Geographically, western Europe has been more thoroughly sampled than elsewhere, while large areas in the tropics, particularly of South America and central Africa are very poorly known. Taxonomically, the larger mammals, birds, vascular plants and a few invertebrate groups, such as Odonata, are better known than other groups of species. Because of the uneven availability of information there has been considerable interesi in identifying groups of individual species that may serve as surrogates for biodiversity more gener- ally, or higher taxa, such as families, that might predict patterns of richness in their included species". Maps 5.5 and 5.6 represent, respectively, the distributions of flowering plant families (phylum Anthophyta) and of terrestrial (non-aquatic) vertebrate families (phylum Chordata), plotted as a global family diversity surface for each group. The resulting maps share key features, notably a marked latitud- inal gradient in family richness, but also have striking differences, particularly in areas of high diversity. For example, Africa appears to be very rich in vertebrate families, most notably in moist forest areas around the Gulf of Guinea and in the east, including less humid woodland and savannah habitats, but is relatively poor in plant families compared with other continents in the tropics. For flowering plants, this mirrors the sequence evident at species level, where there may be around 90000 species in the neotropics, 40 000 in tropical Asia and 35 000 in tropical Africa’. Biogeography and endemism While ecological factors influence which kinds of species, and how many of them, can persist in a given area, history has already deter- mined which actual lineages are present. A complete explanation for global variation in biodiversity must therefore involve both historical events and current ecological pro- cesses. The former are implicit in any explan- ation of the origin of diversity, the latter in explanations of its maintenance; these being two separate, although intimately linked, problems. On land, continental drift resulting from plate tectonics, climate change, moun- tain building or sea-level change, and prob- ably the evolutionary lability of different lineages, are among the important historical factors. Geographic features commonly re- strict or prevent the further dispersal of species: for example, a large river can pre- sent a barrier to a terrestrial species, the sea is a barrier to non-flying island forms and land is a barrier to freshwater species. Barriers to dispersal explain why the fauna and flora of ecologically similar areas in geographically separated parts of the world tend to be composed largely of different individual species. They also underlie the phenomenon of endemism. An endemic species is one restricted to some given area, which may be a continent or country, or more significantly a relatively small area, such as a mountain block, island or lake. Discrete areas of complex topography, particularly in the tropics, often have high endemism in a range of taxonomic groups, possibly because Terrestrial biodiversity 79 mia so WORLD UNESCO” US classification standards” FAO" Table 5.2 Different definitions of forest cover ATLAS Closed forest Woodland Closed tree canopy Open tree canopy Forest Other wooded land OF BIODIVERSITY climate change has encouraged speciation by isolating different lineages at different times. Communities and ecoregions The most comprehensive attempts to describe and classify habitats try to combine elements of all three sources of variation outlined above. At fine scales these are generally based on community ecology. On land this is exemplified by the phyto- sociological approach developed principally in continental Europe during the 20th century. Trees 2 5 m tall with crowns interlocking Trees 2 5 m tall with crowns not usually touching but with canopy cover 2 40% Trees with crowns interlocking, with crowns forming 60-100% cover Trees with crowns not usually touching forming 10-60% or 25-60% cover Land with tree canopy cover > 10% and > 0.5 ha in area; trees should be able to reach a minimum height of 5 m Land with either a crown cover of 5-10% of trees able to reach a height of 5 m at maturity; or crown cover of more than 10% of trees not able to reach a height of 5 m at maturity; or with shrub or bush cover > 10% This was intended to describe and classify plant communities on the basis of dominant and other associated species, by inspection or quadrat analysis of vegetation patches, taking into account species identity, growth form and abundance. One major problem with this is that the more precisely a community is defined, the more site-specific it becomes, and hence the more limited its use in higher- level analysis and planning. A more recent related approach” attempts to delimit ‘eco- regions, these being defined as relatively large units of land each of which contains a distinct assemblage of natural communities and species, with boundaries similar to the original extent prior to anthropogenic change. These are nested within a hierarchy having a NR traditional biogeographic realms and biome systems [ecosystem types) as the first and second levels. FORESTS Forests and woodlands probably once covered about half of global land area and now cover about one quarter. They provide habitat for half or more of the world’s species. They are responsible for just under half of the global terrestrial annual net primary production (Table 1.1], and they and their soils house about 50 percent of the world’s terrestrial carbon stocks. In addition to carbon storage, forests perform many other important ecosystem services, such as regulating local hydrological and nutrient cycles, and stabiliz- ing soils and watersheds. Forests also provide a wide variety of products, including food and fuel, medicines, construction materials and paper, which are important both for human subsistence and for economic activity. Wood products are one of the most economically important natural resources. In the region of 3.3 billion cubic meters (m*] of wood is extracted from forests and other habitats annually, the equivalent of several hundred million trees. Just over half of this volume is used as fuelwood and charcoal, of which developing countries consume 90 percent’. The remainder is industrial roundwood, which is processed into various wood products. Forests are frequently important culturally and play a significant role in the spiritual life of communities worldwide. What is a forest? Despite their importance in a number of different human contexts and the large amount of research that has focused on forest ecosystems, a precise definition of ‘forest’ remains elusive. Although it is generally accepted that the term indicates an eco- system in which trees are the predominant life form, the problem arises because of the broad range of systems in which trees occur and the difficulty, even, in deciding what constitutes a tree. For example, tree species may dominate at high altitude, but be barely recognizable as trees because of their spreading prostrate forms. Savannahs may NEI DIE IT II EL EE possess large numbers of trees, but as they can occur at low density in association with other life forms it may be difficult to define precisely in which areas they are dominant. A variety of different definitions of forest have been proposed by organizations that evaluate and monitor natural resources (Table 5.2). Estimates of forest area may vary widely depending on the definition adopted. Of sing- ular importance is the degree of canopy cover used as the threshold for dividing forests from non-forests (Table 5.2]. As a consequence, the precise definitions employed should be borne in mind when comparing forest cover data provided by different institutions. Forest types There is great variation in the forms and types of forest distributed throughout the world. Information about this variation and the distribution of forest vegetation types is crucial to understanding the different roles of forests in supporting biodiversity, in carbon and hydrological cycles and other ecosystem processes, and in supplying wood and non- wood forest products. However, if deriving a satisfactory definition of forest is problematic, arriving at consensus on how to classify forests is even more difficult. A number of global classification systems have been proposed, but as yet none has gained universal acceptance. The UNESCO (United Nations Educational, Scientific and Cultural Organization) system proposed by Ellenberg and Mueller-Dombois” is one such system. It includes nearly 100 forest and woodland ‘subformations’ and allows for yet finer subdivisions, but many of the charac- teristics that separate categories can only be determined in the field. Other classifications, such as the EROS Data Center seasonal landcover regions, with nearly a thousand classes, reflect more strongly the nature of landcover data obtained from Earth-orbiting satellites and the methods used in analyzing and classifying them’. This complex system has been translated into a much less complex one in the International Geosphere-Biosphere Programme ([IGBP) classification, which includes seven forest and woodland types that reflect phenology and canopy closure world- Australia Tree canopy cover > 20% Tree canopy cover > 70% Senegal Tree canopy cover 2 10% includes dry woodland] ‘Closed’ forest (canopy cover > 40%} wide, but provides little other information on forest physiognomy, composition or environ- ment within the class names'. Map 5.7 represents the global distribution of a range of forest types, based on forest physiognomy and phenology, here aggregated into five broad categories, discussed further below. Temperate and boreal needleleaf forests Distribution, types and characteristic taxa Temperate and boreal needleleaf forests cover a larger area of the world than other forest types. They mostly occupy the higher latitude regions of the northern hemisphere, as well as high-altitude zones and some warm temper- ate areas, especially on nutrient-poor or otherwise unfavorable soils. These forests are composed entirely, or nearly so, of coniferous species (Coniferophyta]. In the northern hemisphere, pines Pinus, spruces Picea, larches Larix, silver firs Abies, Douglas firs Pseudotsuga and hemlocks Tsuga dominate the canopy, but other taxa are also important. In the southern hemisphere coniferous trees, including members of the Araucariaceae, Cupressaceae and Podocarpaceae, often occur in mixtures with broadleaf species in systems that are classed as broadleaf and mixed forests. Temperate and boreal needleleaf 13.1 Temperate broadleaf and mixed 1S Tropical moist 11.7 Tropical dry 2.5 Sparse trees and parkland 6.9 Total 41.7 Terrestrial biodiversity 8 TE pare » - Ali 78 689 3 934 Table 5.3 Sample effects on forest area estimates of different forest definitions Table 5.4 Global area of five main forest types Note: Based on forest cover as shown in Map 5.7. Estimates of this kind vary significantly with different source data and classifications | eos 82 WORLD ATLAS OF BIODIVERSITY SSS a ET SE Structure and ecology The structure of temperate and boreal needleleaf forests is often comparatively simple, as conifer canopies are efficient light absorbers, reducing the possibilities for development of lower strata in the canopy. The tallest of these forests, the giant redwood forests of the west coast of the United States, may reach 100 m in height, but most are much shorter, and indeed some pine forests at high altitude or in arid environments are quite stunted. The distribution of temperate and boreal needleleaf forest is limited at high altitudes and latitudes by lack of enough days with temperatures suitable for growth, and at lower altitudes and latitudes by competition with broadleaf species. In about a quarter of the area of these forests, deciduous conifers of the genus Larix replace the evergreen species. This is especially true in far northern continental areas with extremely low winter temperatures. In many areas, wildfire is an important factor affecting the dynamics and main- tenance of the forest ecosystem. Many coni- ferous species produce resins that increase flammability, and many are characterized by thick bark that increases the resistance of adult trees to fire-induced mortality. A number of tree species, such as the jack pine In general, as the frequency of fire increases, the intensity of individual fires tends to decrease, because of reductions in the standing amount of fuel. Increased fire frequency also tends to increase the diversity of the herb layer by severely affecting the shrub layer. Fires are caused by natural events such as lightning strikes and by human activities. Changing forest management has significantly altered fire regimes in coniferous forests. The 1988 fires that affected over 5 000 km’ in and around Yellowstone National Park, United States, were attributed to the accumulation of fuel in the forests resulting from a long-term policy of fire suppression. Climatic variation plays an important part in determining fire occurrence and severity. The Yellowstone fires, Canadian fires in 1989, and the 1987 Black Dragon fire in the boreal forest region of China and Siberia in 1987, were all ascribed to unusual drought conditions. The Black Dragon fire burned over 70 000 km’, and qualified as the largest forest fire in recorded history’. There is concern that global climate change may increase the frequency and impact of fires in boreal coniferous forests. Pinus banksiana, have serotinous cones, which depend on the high temperatures of forest fires to open and release their seeds. Non-coniferous species are generally less resistant to fire than conifers, so periodic fires are an important factor in maintaining the composition and extent of these forests. Biodiversity Although tree species richness is low in most temperate and boreal needleleaf forests, many conifer species are of great con- servation concern. A well-known example occurs in the giant redwood forests of northern California, where the redwood Sequoiadendron giganteum is considered vulnerable to extinction’. About 22 percent (140) of the world’s 630 conifer species have been assessed as globally threatened". Most of the threatened taxa are characteristic of mixed forests, particularly in the southern hemisphere. Old growth conifer stands, which may be many centuries in age, represent an irreplaceable gene pool and are an important habitat for many other organisms. Species richness in these forests is commonly increased by a relatively high diversity of mosses and lichens, which grow both on the ground and on tree trunks and branches. For example, there are at least 100 species of moss growing in the coniferous forests between 1 300 and 2 000 m altitude on Baektu Mountain, on the Chinese-Korean border’. Mosses and lichens are important sources of food for many animals of coni- ferous forest. Vertebrate richness is generally lower in boreal needleleaf forests than in broadleaf temperate and tropical forests. Many species are wide-ranging generalists, often with a holarctic distribution, e.g. wolf Canis lupus, brown bear Ursus arctos. There are a number of animals of con- servation concern that are dependent on temperate needleleaf forests. The northern spotted owl Strix occidentalis caurina requires large expanses of old-growth coniferous forest in the northwest United States to provide nesting habitat and adequate food resources; Kirtland’s warbler Dendroica kirtlandii needs young regrowing jack pine as a nesting habitat. Terrestrial biodiversity 83 a \ Fire suppression programs have reduced the Temperate broadleaf and mixed forests | Table 5.5 available habitat for this species to critical Distribution, types and characteristic taxa | Important families and levels. While there is relatively little infor- Temperate broadleaf and mixed forests cover | genera, and numbers of mation available on the conservation status of some 7.5 million km’ of the Earth’s surface species, in four areas of invertebrates, many common old-growth [Table 5.4]. They include such forest types as | temperate broadleaf species are known to become much rarer in the mixed deciduous forests of the United | deciduous forest modern managed forests, often through the States and their counterparts in China and | loss of essential microhabitats’. Source: After Rohrig’. ' Family Genus Common Northeast Europe EastAsia South Role in carbon cycle name America America Temperate and boreal needleleaf forests make Payarene quia a “5 i F re a significant contribution to the global carbon Lithocarpus 1 47 balance, accounting for more than a third of Castanopsis 45 the carbon stored in forest ecosystems (Table Cyclobalanopsis Asian oak 30 Castanea Chestnut 4 1 7 5.6) and about 8 percent of global annual net Fagus Beech 1 2 7 primary production (Table 1.1}. Furthermore, Nothofagus Southern beech ° 10 the soils under these forests store large Aceraceae = Acer Maple 10 9 66 amounts of carbon (up to 250 metric tons per Dipteronia ; 1 hectare), some of which may be liberated by Betualceae —Betulus Birch 6 4 36 increasing decomposition rates related to finds alley 2 te ; 2 EF 3 : : Salicaceae Salix Willow 13 35 97 climate change. Of particular note in this : : i Juglandaceae Carya Hickory 1 4 context are the giant conifer forests of the Juglans Walnut 5 1 4 Pacific northwest of the United States. These Platycarya 2 forests may store more than twice as much Leguminosae Cercis Redbud 1 1 2 carbon per hectare as tropical rainforests. Gleditsia Honey locust 2 7 1 Gymnocladus 3 Maackia 3 Use by humans Robinia Locust 1 Global industrial roundwood production is Hie Acacia i l dominated by coniferous species. Pines Pinus, : ny, : : ae ; : Magnoliaceae Liriodendron _ Tulip tree 1 spruces Picea, larches Larix, silver firs Abies, Magnolia Magnolia 8 50 Douglas firs Pseudotsuga and hemlocks Tsuga Oleaceae Fraxinus Ash 4 20 from the needleleaf forests of the northern Osmanthus 2 10 hemisphere are the major sources of softwood. Rosaceae Malus Apple 1 8 Some conifer species from the southern eras oy) : i Z hemisphere and the tropics also provide eee Mountain ash 3 5 18 excellent timbers. Large-scale exploitation of Kageneckia 2 natural coniferous and mixed forests is taking 1 EUnet 1 place around the Pacific Rim, notably in North Tilliaceae Tita Basswood, : ! : : lime, linden 4 3 20 America, Russia and Chile. Temperate and age Lauraceae Phoebe 16 boreal needleleaf forests are also a principal Gassafras Gaesaiiaes 1 D source of pulpwood for paper production. Beilschmidia 1 Cryptocarya ] : Persea { Other ecosystem services é st ; Myrtaceae © Amomyrtus 1 These forests, like others, stabilize soils on Myrceungenella y) sloping topography, especially in mountainous Myrceugenia 8 Nothomyrcia 1 regions. Recognition of this function in early 20th century Switzerland was the basis for a Ulmaceae es ey f ! i new program of forest planting to control Zelkova 3 avalanches in the Alps. Coniferous forests Carpinaceae Carpinus Hornbeam 2 2 5 have high recreational and cultural values, Ostrya 1 1 3 especially in regions such as northern Europe. Table 5.6 Biomass and carbon storage in the world's major forest types Note: The carbon storage figures are over- rather than underestimates as they incorporate no weighting for anthropogenic disturbance or for the variation in biomass and area among different forest classes within the broad types. Source: After Adams*” and Huston™~ 86 WORLD ATLAS OF BIODIVERSITY er ne rn RN Japan, the broadleaf evergreen rainforests of Japan, Chile, New Zealand and Tasmania, and the sclerophyllous forests of Australia, the Mediterranean and California. These last are characterized by a predominance of often evergreen trees with small, hard, leathery leaves. Trees belonging to the Anthophyta and the Coniferophyta grow in mixtures in many of these forests, especially in the southern hemisphere. For example, the Valdivian and Magellanic rainforests of Chile include mixtures of Nothofagus {an anthophyte) with Podocarpus and members of the Cupressaceae [both conifers)’. Much of the forest of New Zealand was originally a mixture of the conifers Podocarpus, Phyllocladus, Dacryocarpus and Dacrydium, with such Anthophyta as Metrosideros, Elaeocarpus and Weinmannia. |In North American mixed forests, pines Pinus, hem- locks Tsuga and cypress Taxodium, among other conifers, are mixed in various propor- tions with oaks Quercus, maples Acer, ashes Fraxinus, hickories Carya, beeches Fagus and other hardwoods. The beech family (Fagaceae) is generally important in temperate broadleaf forests (Table 5.5), with such genera as Castanopsis and Cyclobalanopsis playing an important rolein Japan and China, and many different Quercus species being important elements of hardwood forests in North America, Asia and Europe and of sclerophyllous forests in California and the Mediterranean. Sclerophyllous forests in Australia are largely made up of Eucalyptus species, as are the wet forests of Tasmania, which may also include Nothofagus’. Structure and ecology Depending on the precise forest type, these forests tend to be structurally more complex than pure coniferous forests, having more layers in the canopy. It is not uncommon for an upper canopy layer to have as many as six distinct subcanopy and understorey layers below it. The tallest of these forests, the mixed forests of southern Chile and some Eucalyptus forests of Australia, can reach over 50 m in height. On the other hand, some sclero- phyllous forests barely reach 5 m. Deciduous forests may support rich herb layers, which depend on the increased penetration of sunlight early in the growing season. As in needleleaf forests, fire plays an important role in many types of temperate broadleaf and mixed forest. Both natural and anthropogenic fires are important ecological factors affecting the maintenance of forest structure and composition, especially in the Eucalyptus forests of Australia and the sclerophyllous forests of the Mediterranean and California. Spatial variation in forest structure and composition may be influenced by fire and other kinds of natural and anthropogenic disturbance. When canopy trees die, the resulting gaps in the canopy increase light availability locally, and such areas may be colonized by a different subset of the forest flora. These gap dynamic pro- cesses are important in maintaining stand diversity. Relatively few old-growth broadleaf and mixed forests remain in the temperate zones because of the historical exploitation of these forests by human populations. Biodiversity As might be expected from their structural diversity, temperate broadleaf and mixed forests tend to be richer in species than coniferous forests. Southern mixed hardwood forests in the United States are commonly composed of as many as 20 canopy and subcanopy tree species and may include as Temperate needleleaf 200-1 500 300 (700 giant conifer} 394, many as 30 overstorey species”. In comparison, Temperate broadleaf European forests tend to be less species rich, and mixed forest 150-300 350 262 while the deciduous forests of East Asia may be Tropical moist forest 195-500 300 350 the richest of all” {Table 5.5). Tropical dry forest 98-320 250 62 In the late 1990s, some 370 temperate dicotyledonous tree taxa worldwide were considered to be of conservation concern”. Japan alone has 43 threatened endemic tree species, which are mostly characteristic of its temperate broadleaf forests”. Fitzroya cupressoides, an endangered large conifer once important for the local and international timber trade, is characteristic of mixed forests in southern Chile. It has been over- exploited and proves to be highly dependent for its regeneration on large-scale natural disturbances, such as landslides and light- ning strikes. F. cupressoides is now listed in CITES (Convention on International Trade in Endangered Species of Wild Flora and Fauna] Appendix |, and commercial international trade is accordingly prohibited. While temperate broadleaf forests gen- erally support moderate animal diversity, species richness is often lower than in comparable tropical habitats. However, there is considerable geographical variation in richness in the northern hemisphere; the forests of East Asia are generally the most species rich’. As in the boreal needleleaf forests, many of the mammal and bird families of northern broadleaf forests are holarctic in distribution and can be found in other habitat types. In contrast to most northern hemisphere forests, the temperate forests of southern South America, Australia and New Zealand contain several restricted-range mammals and birds. Analysis” of habitat requirements of Australian mammals indicates that the forests of southeast Australia and Tasmania are of particular importance for wildlife conservation. Examples of temperate broadleaf forest species of special conservation concern include the huemul deer Hippocamelus bisulcus of the southern Andes, threatened by both habitat loss and hunting; a number of New Zealand forest birds including the kakapo Strigops habroptilus and some kiwi species Apteryx spp., which are threatened mainly by introduced predators; Leadbeater’s possum Gymnobelideus leadbeateri, which is threatened by the loss of specific habitat within the montane ash forests of Victoria {Australia]; the Amami rabbit Pentalagus furnessi of Amami Island (Japan), threatened Above-ground biomass of temperate decid- the former Soviet Union ranges from 140 to 500 age and altitude, among other factors”. Soil and to the inherent greater decomposability of where between 135 and 160 metric tons per the forest types included in this broad cate- Terrestrial biodiversity 85 pee a tit by habitat loss and introduced predators; and_ | the European bison Bison bonasus of central and eastern Europe, with low genetic diversity and at risk from disease. The figure shows the living planet index for a sample of 170 forest-occurring birds in North America and Europe. The prevailing trend during 1970-99 was for a small net increase over the period. This could be interpreted as reflecting a phase of relative stability in these forests during recent decades, following centuries of decline in area, but may also be correlated with local increase in forest area through the spread of plantations, and management in some forests may have exerted a positive effect. The separate North America and Europe samples (123 and 47 species} show extremely similar trends. 140 120 100 40 T T T T 1 1970 1975 1980 1985 1990 1995 Source: UNEP-WCMC, from data collated for Loh”. Role in carbon cycle uous forests in Europe, the United States and metric tons per hectare depending on stand carbon storage in these systems is lower than in the needleleaf forests, owing both to climat- ically favorable conditions for decomposition leaf litter from broadleaf trees. One survey” suggests that average soil carbon storage in these forests is some- hectare. Thus, global carbon storage across gory may total as much as 231 petagrams (Table 5.6). 8 WORLD ATLAS OF BIODIVERSITY Tropical moist forests are often particulary rich in palms. Use by humans Temperate broadleaf and mixed forests have provided large amounts of timber over the centuries, but they have largely been replaced by tropical forests as the primary source of hardwood timber in global trade. Hardwood production continues in the temperate zones, for products such as wood chips, furniture and finishing wood. Some of the non-timber products of temperate mixed forests include camphor from Cinnamomum camphorum in Japan (though this has now largely been replaced by synthetics], and sweet chestnuts Castanea sativa from southern Europe. Mushroom production is also a major income source in some parts of Europe, North America and Asia. Other ecosystem services Services provided by temperate broadleaf and mixed forests include soil and watershed protection. This is especially important in the southern Andes” and other areas where steep topography is responsible for a high incidence of landslides, but it is also a benefit in Mediterranean regions where soils are prone to degradation. Natural temperate forests are important reservoirs of genetic material of trees such as eucalypts that are now commonly grown as plantation species. These forests, perhaps more than any other type, experience significant recreational use in many areas. ee ST Tropical moist forests Distribution, types and characteristic taxa Tropical moist forests cover more than 11.5 million km? of the humid tropics (Table 5.4 and Map 5.7] and include many different forest types. The best known and most extensive are the lowland evergreen broadleaf rainforests, which make up over half the total area. These include, for example: the season- ally inundated varzea and igapo forests and the terra firme forests of the Amazon basin; the peat forests and moist dipterocarp forests of Southeast Asia; and the high forests of the Congo basin. Mountain forests are generally divided into upper and lower montane formations on the basis of physiognomy. These include cloud forest - the middle- to high-altitude forests that derive a significant part of their water supply from cloud, and support a rich abundance of epiphytes. Mangrove forests (see Chapter 6] also fall within this broad category. The high diversity of many tropical forests (see below) makes it difficult to characterize them taxonomically. However, some plant families are more prevalent than others. In neotropical moist forests, the legumes Leguminosae are particularly abundant and are often the most species-rich family. Other families that are generally among the richest in tree species in lowland neotropical moist forests are Moraceae, Lauraceae, Annonaceae, Sapotaceae, Myristicaceae, Meliaceae, Euphorbiaceae and Palmae®. In Southeast Asian lowland moist forests the dominant family is the Dipterocarpaceae; the Myrtaceae is also very speciose®. In African rainforests legumes are again important, and the ten richest families usually include the Olacaceae, Sterculiaceae, Dichapetalaceae, Apocynaceae, Sapindaceae and Ebenaceae. The other most abundant families in both Africa and Asia are often the same as those in the Americas”. Structure and ecology Many tropical moist forests have canopies 40 to 50 m tall, and some have emergent trees that rise above the main canopy to heights of 60 m or more. Such large-stature forests are characteristic of lowland forests and some lower montane forests on relatively nutrient- rich soils. Another characteristic of these forests is a relatively high frequency of woody lianas and, especially in the neotropics, palms”. Moist tropical forests are also known for a high abundance and diversity of vascular epiphytes, which take advantage of the higher light availability found in the canopy and can survive because of abundant rainfall and high atmospheric moisture. On more nutrient- poor soils and at higher altitudes forest stature decreases substantially; communities such as those on white sands (bana and campina) and in upper montane environments (elfin forests] may be no more than a few meters tall. With increasing altitude, decreas- ing forest stature is accompanied by a reduc- tion in the frequency of lianas and palms, and an increase in tree ferns (Cyatheaceae, Blechnaceae) and non-vascular as well as vascular epiphytes”. Unlike the other forest types discussed here, tropical moist forests have relatively little seasonal limitation to their growth, though seasonal drought may be a limiting factor, particularly in the semi-evergreen formations. However, the tropical moist forest environment is an intensely competitive one. Though solar energy inputs are high, canopy closure and complexity are also substantial, resulting in efficient capture of incident radiation and understorey light availability frequently much less than 2 percent of that above the canopy. This in turn limits the growth of understorey species and regener- ating trees. Some species can tolerate low light availability, while others grow or regenerate only in gaps in the canopy. Such gaps are formed by the death of one or more canopy trees and represent a significant contribution to overall environmental hetero- geneity, which is an important contributor to high diversity within tropical forests”. Infrequently, catastrophic disturbances such as blowdowns caused by hurricanes or convective storms may create large areas of regenerating forest” and perhaps alter the long-term forest composition, as can logging and other forms of forest disturbance caused by human activities (see below). Terrestrial biodiversity 87 any Soil nutrients are another limiting re- source in many forests. Soils in the humid regions of the tropics are notoriously poor in nutrients owing, among other factors, to loss through leaching by the high annual rainfall and to retention in the high-standing biomass. The formation of gaps in the canopy allowing regeneration of tree species is important for increasing the heterogeneity of available nutrients as well as the availability of solar energy. Forests such as the Amazonian varzea, which are seasonally inundated by sediment-bearing rivers, are an exception to this nutrient limitation. Biodiversity In numerical terms, global terrestrial species diversity is concentrated in tropical rain- forests. Many theories have been proposed to explain this phenomenon”. Generally speak- ing, the wet tropical forests of Africa have a lower tree species richness than those of Asia and America (Table 5.7). However, there is great local variation in species richness. Within the Amazon basin, tree species rich- ness ranges from 87 species per hectare in the east”! to 285 species in central Amazonia” and nearly 300 species in the west”. The high diversity of tree species in lowland evergreen rainforests is mirrored in the diversity of epiphytes and lianas, which is also much higher in neotropical forests than in other regions™. Fifty-three families in the Anthophyta and at least nine pteridophyte (Filicinophyta and allies) families include epiphytes. Of nearly 25000 species of vascular epiphytes, around 15000 belong to the Orchidaceae. Nearly a thousand others are members of the pineapple family Bromeliaceae, which is primarily neotropical. Other groups having a high diversity of Africa Southeast Asia Americas Table 5.7 Tree species richness in tropical moist forests Source: After Phillips et al.” 56-92 108-240 56-285 88 WORLD ATLAS OF BIODIVERSITY V8 rn eh areas Map 5.3 Diversity of vascular plant species This map shows the species richness of vascular plants, plotted as a world density surface. It is based on some 1 400 literature records from different geographic units, with richness values as mapped calculated ona standard area of 10 000 km* using a single species-area curve. Value categories range between extremes of more than 5 000 species and fewer than 100 species per 10 000 km’. Source: Data and analysis © Wilhelm Barthlott (Botanic Institute and Botanic Gardens, University of Bonn). Reproduced by permission, with modification to colors. For further details see Barthlott”’ and website http://www.botanik.uni-bonn.de/system/ biomaps.htm#worldmap [accessed March 2002). Diversity High Low epiphytic species are the cactus family Cactaceae, the aroids Araceae, the pepper family Piperaceae and the African violet family Gesneriaceae. Not all tropical moist forests have high species richness. Mangrove ecosystems have a low diversity of tree species despite their sometimes high productivity and high animal diversity (see Chapter 6]. Extremely nutrient- poor soils, such as white sands, lead to the development of low-diversity forests including bana and campina®. As climate becomes more seasonal, tree species richness tends to decline [see dry forests, below). Increasing altitude also tends to reduce species rich- ness, with montane forests typically having fewer tree species than lowland ones”. Regionally, the forests of Asia and South America are rich in animal species, generally more so than those of Africa’. Many forest animals are largely confined to moist forests, e.g. the okapi Okapia johnstoni, but some are widespread outside, such as leopard Panthera pardus®. In Africa, the Guineo- Congolean forest block contains more than 80 percent of African primate species, and nearly 70 percent of African passerine birds and butterflies*. About half of the 1100 South American reptile species are found in moist forests, with around 300 of these endemic to the habitat’. Amphibian species are partic- ularly diverse in tropical moist forests®; 90 percent of 225 species identified in the Amazon basin forests are endemic. The importance and diversity of the fish communities in forest streams and rivers is often overlooked; the Amazon basin has the richest freshwater fish fauna known, with at Terrestrial biodiversity 89 least 2 500 species, many important as major seed predators and dispersal agents”. Local species richness of insect and other arthro- pod groups in tropical forest canopies is much higher than in temperate forests“. Around one third of the animal biomass of the Amazon terra firme rainforest consists of ants and termites, and each hectare of soil is estimated to contain more than 8 million ants and 1 million termites*’. Numerous tropical moist forest species are of conservation concern. Notable animals include the Sumatran rhino Dicerorhinus sumatrensis of Southeast Asia, endangered by habitat fragmentation and hunting; the bonobo Pan paniscus of the Democratic Republic of the Congo, threatened by habitat destruction and hunting for food; the Philippine eagle Pithecophaga jefferyi of the Philippines, reduced to small fragmented populations through habitat loss and hunting; and the indri Indri indri of eastern moist forest on Madagascar, threatened by habitat destruction as is the recently rediscovered Edward's pheasant Lophura edwardsi of Viet Nam. Role in carbon cycle Lowland evergreen broadleaf rainforests can have high above-ground biomass (Table 5.6), though not as high as some giant conifer forests. Soil carbon, however, is relatively low in most tropical moist forests, with the exception of the peat forests of Southeast Asia and some swamp forests. On this basis it has been estimated that the remaining tropical moist forests store over 300 petagrams of organic carbon, or about one fifth of global terrestrial organic carbon. They account for 90 WORLD ATLAS OF Map 5.4 Biodiversity at country level Country-level biodiversity, represented by an index based on species diversity in the four terrestrial vertebrate classes and vascular plants, adjusted according to country area. Countries at the high end of the scale have a higher value of the index than would be expected on area alone. The index is unreliable for smaller countries [such as Togo and Luxembourg in this plot). Note: To reduce ambiguity Alaska (United States] has for the purposes of this map been assigned to the same class as adjacent Canada rather than the conterminous United States. Source: Based on national biodiversity indices developed by UNEP-WCMC; see Appendix 5. Diversity High Low nearly a third of global terrestrial annual net primary production (Chapter 1], and are therefore key to the global carbon cycle and in regulating global climate. Use by humans Tropical hardwood species contribute almost one fifth of world industrial roundwood production’. Of the several thousands of species [in more than 200 families in the phylum Anthophyta) that show commercial potential, a few hundred may be found in international trade. Important families include Dipterocarpaceae, with species of meranti and balau Shorea, and keruing Dipterocarpus; Meliaceae, with mahogany Swietenia and Khaya, and cedar Cedrela and Toona; and Leguminosae with rosewood Dalbergia and Pterocarpus. The exploitation of tropical hardwood from moist forests has been the subject of much publicity in the past two decades, coinciding as it has done with increased rates of deforestation and forest degradation. Timber supplies from some countries are now widely exhausted, generating openings for other countries to take over as suppliers. A few major producers, however, continue to dominate supply. Indonesia, Malaysia, Brazil and India accounted for 80 percent of tropical log production in International Tropical Timber Organization (ITTO) countries in 1999- 2000°. Important non-timber products from tropi- cal moist forests include rattans, which are the second most important source of export earn- ings from tropical forests. A few other craft products and some medicinal products, such as Terrestrial biodiversity m1 = AS AN SE SE SE I TT DPS LEY Ay the bark of Prunus africana, are significant in international trade. Brazil nuts Bertholettia excelsa and native rubber Hevea brasiliensis are other extractive products from natural tropical moist forests that are important in international markets, and many tropical moist forests provide fruit, bushmeat and other products for local markets. Other ecosystem services Like other forest types, tropical moist forests often play an important role in soil and water- shed protection. The high rainfall regimes of the humid tropics mean that exposed soil is particularly liable to leaching of mineral nutrients and to erosion. Montane forests, and especially cloud forests, serve to intercept and store water, thus regulating local and regional hydrological cycles“. Lowland forests are also important in hydrological cycles; it has been estimated that about half the rainfall in the Amazon Basin is derived from water recycled by forest transpiration”. Regenerating forest, or forest fallow, is an important part of the cycle of shifting culti- vation, which is vital for restoring fertility to areas that have been previously cultivated. This regeneration can only take place if nearby forest cover is adequate to provide a source of propagules. Tropical dry forests Distribution, types and characteristic taxa Tropical dry forests are characteristic of areas in the tropics affected by seasonal drought. Such seasonal climates characterize much of the tropics, but less than 4 million km? of tropical dry forests remain. The seasonality of 92 WORLD ATLAS OF BIODIVERSITY rainfall is usually reflected in the deciduous habit of the canopy trees, with most being leafless for several months of the year. However, under some conditions, e.g. less fertile soils or less predictable drought regimes, the proportion of evergreen species with leaves highly resistant to water loss increases (‘sclerophyllous’ forest]. Thorn forest, a dense forest of low stature with a high frequency of thorny or spiny species, is found where drought is prolonged, and especially where grazing animals are plenti- ful. On very poor soils, and especially where fire is a recurrent phenomenon, woody savannahs develop” (see ‘sparse trees and parkland’ below). Perhaps the best-known tropical dry forest tree species is teak Tectona grandis (Verbenaceae], a deciduous hardwood charac- teristic of the seasonal forests of south and Southeast Asia, widely exploited for furniture and other uses, and now an important plantation species. In Southeast Asia the dipterocarps Shorea and Dipterocarpus, and Lagerostroemia (Lythraceae), and a number of species of legumes [Leguminosae] are also important components of seasonally dry forests’. In Africa, dry forests occur both north and south of the equatorial rainforests. In the north they are characterized by Afraegele (Rutaceae], Diospyros (Ebenaceae], Kigelia (Bignoniaceae) and Monodora (Annonaceae), among other taxa, while in the south the characteristic genera are Entandophragma (Meliaceae], Brachystegia {Leguminosae}, Diospyros, Parinari (Chrysobalanaceae), Syzigium (Myrtaceae) and Cryptosepalum {Leguminosae}. In the neotropics, tropical dry forests occur in Yucatan and Pacific slopes of Central America, on the leeward sides of Caribbean islands, in Venezuela, Caribbean Colombia, in northeast Brazil and in the Chaco region of Bolivia and Paraguay (see Map 5.7). The neo- tropical dry forests are quite rich in species and include Leguminosae, Bignoniaceae, Rubiaceae, Sapindaceae, Euphorbiaceae, Falcourtiaceae, and Capparidaceae as the families with the largest numbers of species.Important genera include Tabebuia (Bignoniaceae], Trichilia (Meliaceae], Eryth- FS ZF SS a roxylum (Erythroxylaceae], Randia (Rubi- aceae], Capparis (Capparidaceae), Bursera (Burseraceae], Acacia (Leguminosae) and Coccoloba (Polygonaceae}””. Structure and ecology Tropical dry forests are generally of lower stature than moist forests, with canopy heights ranging from only a few meters to 30 m or occasionally 40 m*. The taller forests have multilayered canopies. Dry forests tend to have more small trees than moist forests and a lower above-ground biomass. The trees have a greater proportion of their total biomass below ground, as more extensive root systems help the trees to obtain water from the soil and avoid drought. Dry forests have a much lower incidence of epiphytes than wet forest, and tend to have both higher frequencies and higher diversity of vines and lianas’. Plants with specialized mechanisms for avoiding drought or conserving water are an important feature of these forests. The proportion of deciduous tree species is thought to increase steadily with decreasing annual rainfall, but factors such as the sub- strate and the between-year variation of seasonal rainfall patterns are also important. Many species have water storage tissues such as succulent stems or tubers, and specialized photosynthetic mechanisms that conserve water are especially common among the epiphytes. Most tropical dry forest trees tend to flower and sometimes to re-leaf before the end of the dry season, and stored water within the plant is essential to this pattern. As in tropical moist forests, environ- mental heterogeneity is linked with in- creased species diversity. Gallery areas along water courses are one source of such variation, and they serve as a refuge for animals during the dry season“. Termite mounds provide an important source of environmental variation in African dry forests, adding to local topography and supplying high-nutrient microenvironments to the system, increasing tree species richness by 40-100 percent‘. Fire is also a major factor determining the dynamics and extent of tropical dry forests. Biodiversity Though of lower species richness than tropical moist forests, dry forests still have appreciably more tree species than most temperate forests. The richest neotropical dry forests, which are not the wettest ones but those in western Mexico and in the Chaco of southeast Bolivia, have around 90 woody species per 0.1 hectare sample’”. Although a comprehensive assessment has yet to be made, dry forests are thought to have high rates of plant species endemism relative to wet forests in the tropics”. Sixteen percent of the plant species of the Chamela dry forest in western Mexico are local endemics, and 20 percent of the flora of Capeira, Ecuador, are endemic to western Ecuador”. Many of the dipterocarps in Thailand’s seasonal forests are national endemics and distinct from the species in the country’s moist forests’. Vertebrate species diversity is lower in dry forests than in moist forests, but many dry forests have high rates of endemism among mammals, especially among groups such as insectivores and rodents, characterized by low body weights, low mobility and short gener- ation times”. Among neotropical dry forests, those of Mexico and the Chaco have the highest numbers of mammal endemics (26 and 22 respectively”). Remaining areas of dry forest are often important refuges for once widespread species. The Gir forest of Gujarat (India) contains the only population of Asiatic lion Panthera leo persica, once found throughout much of southwest Asia; the dry forests of western Madagascar are inhabited by around 40 percent of the island-endemic lemurs; some, such as red-tailed sportive lemur Lepilemur ruficaudatus, are almost entirely confined to this habitat. Invertebrate species richness tends to be poorly known, but in groups such as lepidoptera and hymenop- tera richness in some dry forest areas may be comparable to adjacent wet forest”. Because of their high degree of endem- ism and because degradation and conversion have progressed further than in wet forests, the biota of tropical dry forests are often highly threatened. Hunting, especially for the wildlife trade, and habitat conversion are Terrestrial biodiversity 93 important pressures on dry forest animal species. Threatened dry forest species include Spix’s macaw Cyanospitta spixii, which is nearly extinct globally as a result of trapping and habitat loss in Brazil's northeastern caatinga region; the Chacoan peccary Catagonus wagneri, which was rediscovered in the Gran Chaco of central South America during the 1970s and is threatened by overhunting, habitat loss and disease; Verreaux's sifaka Propithecus verreauxi of western Madagascar, which is at risk from loss of spiny and gallery forest habitat; and the Madagascar flat-tailed tortoise Pyxis planicauda, which is restricted to the western Andranomena forest of Madagascar and is believed to be declining as a result of habitat destruction. Many of the endemic lemurs of Madagascar are found in dry forest 9% WORLD ATLAS OF BIODIVERSITY Map 5.5 Flowering plant family density Global diversity of flowering plants represented as a density surface derived from distribution maps for 284 non-aquatic plant families. Scale indicates number of families present, up to a maximum of 182, divided into 15 classes. Source: Compiled primarily from Heywood'”’ by UNEP-WCMC. Sa TE Density High Low Role in carbon cycle Their lower biomass means that tropical dry forests represent a smaller reservoir of stored carbon per unit area than the other forest types discussed so far. With a total biomass ranging from 98 to 320 metric tons per hectare” and soil carbon storage in the region of 100 metric tons per hectare”, it is unlikely that relatively undis- turbed tropical dry forests store more than 250 metric tons of carbon per hectare (Table 5.6). This, combined with the fact that little intact tropical dry forest remains worldwide, suggests that the total contribution of seasonally dry tropical forests to global carbon storage is far less than that of other forest types. Use by humans Notable among the economically important species of seasonally dry tropical forests is teak Tectona grandis, which accounts for about 1 percent of reported global tropical timber exports. More than ten species of Thailand's seasonal forests are significant for the timber trade’, with timber species such as mahogany Swietenia and several species of Tabebuia (Bignoniaceae) characteristic of neotropical dry forests. The southern dry forests of Africa also contain useful timber species such as Entandrophragma spp. Tropical dry forests also provide large quantities of fuelwood for local populations. A number of food plants are native to tropical dry forests and medicinal uses have been reported for many dry forest plant species’. Craft products are also important. Other ecosystem services Protection of relatively fragile soils is a vital ecosystem service provided by tropical dry forests. Rains may be intense during the wet season, and erosion can be a severe problem in tropical dry forest areas, where soils are often thin and soil formation processes slow™. Tropical dry forests may also be important resources for native pollinators as well as nectar sources for domestic bees. Many dry forest trees produce conspicuous flowers with specialist pollination mechanisms. Their mass flowering provides a major nectar resource for pollinating insects at the end of the dry season when other such resources may well be limited’” *°. Honey production is one of the liveli- hoods being promoted for local communities in dry forest areas in Mexico and elsewhere. Sparse trees and parkland Sparse trees and parkland are forests with open canopies of more than 10 percent crown cover. They occur principally in areas of transition from forested to non-forested landscapes. The two major zones in which these ecosystems occur are the boreal region and the seasonally dry tropics. At high latitudes, north of the main zone of boreal forest or taiga, growing conditions are not adequate to maintain a continuous closed forest cover, so tree cover is both sparse and discontinuous. This vegetation is variously called open taiga, open lichen woodland or forest tundra®. It is species poor, has high bryophyte cover, and is frequently affected by fire. It is important for the livelihoods of a number of groups of indigenous people, inclu- ding the Saami and some groups of Inuit. In the seasonally dry tropics, decreasing soil fertility and increasing fire frequency are related to the transition from closed dry forest Terrestrial biodiversity 95 Hy), 98 WORLD ATLAS OF BIODIVERSITY The large savannah vertebrates of Africa are largely absent from other continents through open woodland to savannah. The open woodland ecosystems include the more open Brachystegia and /soberlinia woodlands of dry tropical Africa and parts of both the caatinga and cerrado vegetations of Brazil”. Open woodlands in Africa are more species rich than either closed dry forest or savannah. The cerrado supports a high diversity of woody plants, though many of them are of shrubby habit. Animal diversity is generally low in forest tundra; few species are restricted to this habitat, many also occurring in boreal forest or tundra proper. The sparsely wooded tropi- cal savannahs are generally more species rich than temperate forests or grasslands”. Wooded savannahs vary greatly. Those of America are relatively species poor, while African savannah sometimes attains a rich- ness not far below rainforest in the same continent. Sparsely wooded areas in Australia are amongst the richest wildlife habitats on the continent, sometimes more so than adjacent wet forests”. The large savannah vertebrates present in such high diversity in Africa are largely absent from other continents” {see Chapter 4). The density and biomass of tropical savannah soil inver- tebrates [mostly earthworms, ants and termites] is generally lower than temperate grasslands, but greater [at least in biomass] than tropical rainforests”. Species of conservation concern include the black rhinoceros Diceros bicornis, threat- ened primarily by hunting for their horn, and RA BS I I the golden-shouldered parrot Psephotus chrysopterygius of northern Queensland (Australia), threatened by the burning of seeding grasses during the breeding season and predation by feral cats. Forest plantations Forest plantations, generally intended for the production of timber and pulpwood, increase the total area of forest worldwide. FAO! estimates that forest plantations covered 187 million hectares in 2000, of which Asia accounted for 62 percent. This represents a significant increase from the 1995 estimate of 124 million hectares. Commonly mono- specific and/or composed of introduced tree species, plantation forests tend to be less valuable as a habitat for native biodiversity than are natural forests. However, they can be managed in ways that enhance their habitat value. Plantations are also important pro- viders of ecosystem services, such as main- taining nutrient capital and protecting water- sheds and soil structure as well as storing carbon. They may also alleviate pressure on natural forests for timber and fuelwood production. In some countries, wood from plantation sources makes up a significant portion of the industrial wood supply. For example, New Zealand is more than self-sufficient in wood production based on plantations’'. New forest plantation areas are increasing globally at a rate of 4.5 million hectares per year, with particularly high rates of increase in Asia and South America®’. In future, forest plantations are likely to play an increasingly important role in mitigation of greenhouse gases, as encouraged by the Kyoto Protocol. Changes in forest cover About half of the forest that was present under modern (i.e. post-Pleistocene} climatic conditions, and before the spread of human influence, has disappeared, largely through the impact of human activities. The spread of agriculture and animal husbandry, the har- vesting of forests for timber and fuel, and the expansion of populated areas have all re- duced forest cover. The causes and timing of forest loss differ among regions and forest types, as do the current trends in change in forest cover. The temperate forests of western Europe have diminished by far more than the 50 percent estimated for forests globally. Much of this deforestation occurred between 7000 and 5000 years ago as Neolithic agriculture expanded’. The expansion of human populations and increasing demand for fuel during classical times and the Middle Ages put further pressure on European forests. Between Neolithic times and the late 11th century, forest cover in what is now the United Kingdom decreased by 80 percent. As European forests dwindled they became an increasingly valuable resource that was more carefully managed. Forest cover stabilized during the 19th century in much of western Europe in response to both improved man- agement and reduced demand for forest products (owing to the increasing use of fossil fuels and changes in construction materials). Since the early 20th century, forest cover in Europe has expanded, often through the establishment of conifer plantations. In eastern and central Europe and in Russia, forest clearance accelerated during the 16th and 17th centuries as sedentary agriculture expanded. One estimate” sug- gests that around 1 million km’ of forest had been cleared in the former Soviet Union up to 1980. Timber exploitation continues to drive forest clearance in the coniferous forests of Siberia and parts of eastern Europe. In North America, indigenous groups had impacts on the forests from at least 12 000 years ago, but most forest clearance took place after European settlement. Forest cover in eastern North America reached its minimum around 1860, but then increased with the westward movement of the agri- cultural frontier and subsequent urbanization and industrialization. Forests west of the Appalachians suffered the most severe impacts in the late 19th and early 20th centuries, but are still under pressure from demand for timber and pulp. In Oceania, just as in North America, indigenous groups had significant impacts on the forests before the arrival of Europeans. This was especially true of the aboriginal Terrestrial biodiversity 97 Nee ee a a NS ES LR SS Africa 650 Asia 548 Europe 1039 North and Central America 549 Oceania 198 South America 886 World total 3 869 use of fire in Australia, but European colon- ization greatly increased the rate of forest conversion. Over 230 000 km’ of forest and 120 000 km* of woodland in Oceania are estimated to have been converted to cropland between 1860 and 1980”. In tropical Asia, Africa and Latin America, large-scale deforestation was precipitated by European colonial activities, including agriculture and timber exploitation. It is estimated that more than 1 million km? of forest and a similar amount of woodland in tropical developing countries were converted to cropland between 1860 and 1980*. The bulk of this conversion was in south and Southeast Asia, where forest area declined by 39 percent from 1880 to 1980. Globally, tropical dry forest has lost the greatest proportion of its original area of the four major types of closed forest, nearly 70 percent. About 60 percent of the original area of temperate broadleaf and mixed forests has disappeared, and tropical moist and temperate needleleaf forests have lost about 45 percent and 30 percent of their original area, respectively. Current trends in change in forest cover, which are shown in Table 5.8, reveal that the rates of deforestation continue to be high in Temperate and boreal needleleaf 675 470 Temperate broadleaf and mixed 457 535 Tropical moist 1 382 004 Tropical dry 413 524 Sparse trees and parkland 274 401 Total 3 202 934 Table 5.8 Estimated annual change in forest cover 1990-2000 Note: Figures refer to natural forests and plantation forests combined. Source: FAO‘'. Table 5.9 Global protection of forests within protected areas in IUCN categories I-VI Note: Forest areas and protection as assessed in 1999. Source: UNEP-WCMC”’. 5.4 7.0 12.2 11.2 5.8 8.3 Wy 98 WORLD ATLAS OF BIODIVERSITY ES SS ET ESS EEE ERE Fuelwood and charcoal consumption more than doubled between 1961 and 1991. the developing countries of the tropics, in both absolute and proportional terms. In contrast, temperate countries are losing forests at lower rates, or indeed showing an increase in forest area”. Pressures on forest biodiversity The principal pressures on forests and their biodiversity are conversion to other landuses, principally forms of agriculture, and logging. Conversion of forest to agriculture is the main cause of tropical moist forest loss. This is largely due to expanding populations and the use of shifting cultivation at an intensity that does not permit adequate fallow periods. Government resettlement programs that have moved large numbers of poor farmers have increased the rate of land colonization and clearance in parts of Southeast Asia and Latin America. In some areas, land has been converted to ranching principally as a means of gaining title in order to permit speculation in land values. Thus, population growth, poverty and inequitable land tenure are among the causes underlying deforestation by conversion to agriculture. Timber extraction puts great pressure on biodiversity in both tropical and temperate forests. Global consumption of industrial roundwood was more than 1 521 million m* in 1998" and, on current trends, is projected to continue to rise. Although some timber species are naturally abundant, a factor that can help ensure their survival from com- mercial exploitation, many have suffered extensive and irreversible population and genetic losses. Furthermore, rare species that are indistinguishable in the field from their commercially important relatives are in danger of extinction through overexploitation. This particular problem exists, for example, among the dipterocarp groups meranti, balau and keruing. A few hundred species may be traded under these names, and a significant proportion of these are geographically and ecologically restricted and so at high risk of extinction. Furthermore, logging operations create access to forest areas that may otherwise have remained isolated. This improved access facilitates hunting and other activities that exert pressure on forest biodiversity, and may ultimately lead to colonization and conversion of the land to agricultural use. There is also strong evidence that logging can increase the probability of wildfire in temperate forests and even in tropical moist forests not usually subject to burning”. Particularly in the drier areas of the tropics, fuelwood extraction can have serious impacts on forests and open woodlands. Fuelwood and charcoal consumption more than doubled between 1961 and 1991, and is projected to rise by another 30 percent to 2 395 million m* by 2010. About 90 percent of the consumption is in developing countries, but wood fuel may play an increasing role in some developed countries, increasing demand for wood still further’. In addition to loss of area, forest conver- sion and logging lead to changes in the condition or quality of the remaining forest. These can include fragmentation of large areas of continuous forest. Tropical forest fragments are distinct from continuous for- ests in both ecology and composition”. There are physical and biotic gradients associated with fragment edges, and forest structure undergoes radical change near the edges as a result of the impacts of wind and increased tree mortality. Some animal species are “edge-avoiders’ and decline in abundance in forest fragments, while others become more abundant. Some non-forest and even non- native species of plants and animals success- fully invade forest fragments but not con- Tr tinuous forest. In addition to directly affecting canopy composition, removal of large timber trees may also affect the availability of seed for regeneration and may affect animal species that depend on the timber species. Other factors that affect forests and their biodiversity include acid rain and global climate change. So far, most of the effects of acid precipitation, which is caused by indus- trial air pollutants, have been documented in temperate needleleaf forests and associated waterways of Europe and North America. The likely impacts of global climate change on forests are still being debated, but there seems to be general consensus that the boreal coniferous forests are particularly vulnerable to both range restrictions and increasing fire frequency”. Another forest type that has been shown to be vulnerable to climate change is tropical montane cloud forest, which depends upon clouds to supply it with atmospheric moisture. Research has shown that the mean cloud base is moving upwards on tropical moun- tains as a result of climatic shifts. The forest Species are not able to migrate at a comparable rate and, in any case, range shifts will be limited by the land area existing at higher elevations. Local extinctions in cloud forest amphibians, including the golden toad Bufo periglenes assessed as critically endangered, have been attributed to climatic fluctuations that may be linked to long-term global climate change*. NON-FOREST ECOSYSTEMS The parts of the Earth that are too cold, too dry or too severely affected by fire and/or grazing do not support forest or woodland ecosystems. However, as can be seen from Map 5.1, many of them do support active plant growth. Natural non-forest ecosystems in- clude tundra (both arctic and montane}, grasslands and savannahs, and shrublands. Less productive, but with unique elements of biodiversity, are the deserts and semi-deserts (Map 5.4). Tundra Tundra is the vegetation found at high latitudes beyond the limits of forest growth; a the same term is sometimes used for similar vegetation at high elevation at lower latitudes, but these may be distinguished as ‘polar tundra’ and ‘alpine tundra’, respectively. In the Arctic, polar tundra occurs north of the northern tree line, which is determined by a number of climatic factors including the summer position of arctic air masses” and the depth of permafrost (permanently frozen subsurface soil). Similarly, alpine tundra occurs above the climatic tree line on mountains, and its elevation varies in a complex fashion with latitude, continental or oceanic climate, and the maximum elevation and overall size of the massif”. The characteristics of polar and alpine tundra environments differ in many respects. In high-latitude polar tundra systems, temperatures are low for much of the year, while permafrost limits both drainage and root extension, and the growing season may last for as little as six to ten weeks. Rainfall is low, usually less than 200 mm per year, and at extreme latitudes may be so low that the environment is described as polar desert. At high elevation in temperate regions temperatures may be similarly low, although permafrost is rare. However, at high altitudes in the tropics, although low temperatures occur every night, high insolation causes warming during the day so that adequate temperatures for active plant growth occur In polar tundra systems, temperatures are low for much of the year. — wv 100 WORLD ATLAS OF BIO Map 5.6 Terrestrial vertebrate family density Global diversity of terrestrial vertebrates represented as a density surface derived from distribution maps for all 350 non-aquatic families of mammals, birds, reptiles and amphibians. Scale indicates number of families present, up to a maximum of 126, divided into 15 classes. Source: Compiled from multiple sources by UNEP-WCMC. DIVERS Iii EERE ET ETT Density High Low throughout the year and the diurnal temper- ature range is large. Despite these environmental differences, polar and alpine tundra vegetation have some features in common. Both lack trees, but contain woody species growing in dwarf or prostrate forms, especially in locations with less extreme climates. As latitude or altitude increases, grasses, sedges, bryophytes and lichens increase in importance while shrubs decrease. Many plants have tussock or cush- ion growth forms. At extremes of latitude or elevation a high proportion of bare ground is characteristic. In the arctic tundra, plants cover 80-100 percent of the ground’, and cover decreases along the climatic gradient to polar desert. The important woody plants are birches Betula, willows Salix and alders Alnus. These are all genera that occur as trees in temperate regions but as spreading, prostrate or dwarf forms, sometimes less than 20 cm tall, in tundra. Other shrub species are also important, including Dryas, Vaccinium and Empetrum. Interspersed with the shrubs and of increasing importance in more extreme sites are the sedges Carex and the cotton grasses Eriophorum, among other gramin- oids, and a high diversity of mosses and lichens. Biomass is often much lower above than below ground, and annual production is low. This combination means that the poten- tial of these ecosystems for recovery following disturbance is limited”. Temperate alpine systems are charac- terized by many of the same taxa as the arctic tundra, but at lower latitudes other groups become important. At high altitude in the Terrestrial biodiversity 101 ‘= : : a aa ea I tropics, giant rosette plants are a notable feature of alpine communities. These dis- tinctive plants, which have a number of morphological and physiological adaptations to the high insolation, large temperature fluctuations and desiccating conditions on tropical mountains, include Espeletia and Puya in the paramos of the Andes, and Senecio and Lobelia in the high mountains of Africa”. In comparison with forested ecosystems, both polar and alpine tundra are relatively species poor. It is estimated that species richness declines by a factor of between three and four between the boreal and arctic zones’, and species richness in the polar desert is one fifth that of the tundra. The entire North American Arctic has a vascular flora of about 600 species’. Bryophytes and lichens may add more than 300 additional species to the circumpolar flora. The tropical alpine systems are richer; Venezuelan paramos include more than 400 angiosperm species”, but this is still much lower than in surrounding forests. Because most arctic plant species have wide geographic distributions, few are of significant conservation concern, but many alpine areas are isolated by lowlands with contrasting climates so increasing endemism in many alpine floras. The floras of isolated mountains in North America have significant rates of endemism”; about 15 percent of the flora above the timber line in the Alps are locally endemic species” and about 80 percent of the flora at high altitude in East Africa and Ethiopia are endemic”. Restricted distrib- utions tend to make species more vulnerable 1022 WORLD ATLAS OF BIOD 2S RR I SB IVERSITY to extinction, while the harsh environment of alpine regions adds to the risk. Animal species richness tends to be low; groups represented by several species in boreal forest are often reduced in diversity by up to one third in tundra habitat”’. In contrast, a few groups, particularly water birds and waders, are able to exploit the large numbers of invertebrates found in tundra soil and can be both diverse and abundant”. Although the species richness of the most common in- vertebrate groups (Collembola and oribatid mites) decreases with increasing latitude, their total abundance may increase. There are relatively few globally threaten- ed species that are completely dependent on tundra. An exception is one of the world’s most severely endangered species, the once abundant Eskimo curlew Numenius borealis of the Americas, which nests almost ex- clusively within this habitat. Two globally threatened bird species, Steller’s (Polysticta stelleri) and spectacled eider (Somateria fischeri] remain within the Arctic throughout the year. Although low in number of species, the Arctic is home to most of the world’s geese and calidrid sandpipers”. Although the tundra accounts for less than 2 percent of global annual net primary production (Table 1.1], the high below-ground biomass and soil carbon in arctic tundra means that it makes an important contribution to global carbon stocks. Total biomass in tundra communities of the Russian Arctic falls in the range 7-30 metric tons per hectare, of which some 60-70 percent is below ground”. Sedge-moss communities in the North American Arctic may have 15-30 times as much biomass below ground as above ground”, and three to eight times the amount of dead as live material may accumulate. Tundra soils may store around 200 metric tons of carbon per hectare”. Because of its inhospitable climate, tundra is not subject to severe pressure for conversion for other landuses. However, its lack of ecological resilience means that disturbances, e.g. those associated with settlements or long-distance pipelines, tend to have long-lasting effects. It is anticipated that the effects of global warming on the arctic tundra will be significant, as relatively large temperature increases are predicted for this zone”. These will cause changes in the permafrost regime and decomposition of accumulated soil organic matter, which in turn will release additional carbon dioxide into the atmosphere. Evidence suggests that the period of active plant growth has recently lengthened in parts of North America. There is also evidence that species are already migrating northwards in response to climate change, so that arctic tundra is likely to be compressed into a much smaller area of remaining appropriate climate. Grasslands and savannahs Grassland ecosystems may be loosely defined as areas dominated by grasses (members of the family Gramineae excluding bamboos) or other herbaceous plants, with few woody plants“. Grassland ecosystems are typically maintained by drought, fire, grazing and/or freezing temperatures”. In addition, they are often associated with soils of low fertility”. Savannahs are tropical ecosystems charac- terized by dominance at the ground layer of grasses and grass-like plants. They form a continuum from treeless plains through open woodlands to closed-canopy woodland with a grassy understorey. Some savannah areas therefore meet general definitions of grass- land (fewer than 10-15 woody plants per hectare”) while others meet the definition of woodland. Some polar and alpine tundra communities also meet the definition of grassland. Around 20 percent of the Earth’s land surface (excluding Antarctica) supports grassland, with these regions differing greatly in naturalness”. Temperate grasslands make up approximately one quarter of this area, and savannahs the remainder (Map 5.8). The most extensive areas of temperate grasslands are the prairies of North America, the pampas and campos of southern South America and the steppes of central Europe, southwest and central Asia and Russia. Temperate grass- lands are sometimes divided into formations of tall grass, mixed grass and short grass, which differ both floristically and ecologically; short-grass communities are usually associ- ated with drier climatic regimes”. Tropical grasslands and savannahs include the llanos of the Orinoco basin in Venezuela and Colombia, the cerrado of central Brazil, and the savannahs of tropical and subtropical Africa. In addition to many species of grasses, the sedges (Cyperaceae] and many different groups of dicotyledonous herbs are also important. Key ecological factors in grasslands are grazing pressure and the effects of fire. Most natural grasslands had, at one time, large populations of native grazing mammals. These have been replaced to a great extent by domesticated ungulates, which also exert a significant degree of grazing pressure [the magnitude depending on stocking densities). Grazing tends to increase abundance of less palatable species and to increase species richness in productive areas, or decrease it in less productive areas. At intermediate fre- quencies, fire tends to increase diversity and suppress invasion by woody species. Frequent fires favor grasses, which usually recover easily, whereas low fire frequency may allow the density of woody species to increase. These factors have important conse- quences for vegetation structure in grassland and savannah systems. A high proportion of plant biomass, in the form of roots and rhizomes, is located underground; there is a high turnover of those parts of the plant above ground”. One important consequence of this is that grassland soils, especially in more humid Terrestrial biodiversity 103 environments, are often rich in organic matter and are therefore particularly vulnerable to conversion to cropland, with replacement of native grasses by their domestic derivatives {cereals} and other plants”. At very fine spatial scales, natural grass- lands can be extremely species rich. For example, a square meter of ‘meadow steppe’ in the former Soviet Union may have 40-50 plant species”, a tall grass prairie remnant of less than 2 hectares may contain 100 species, and 250 hectares may contain 250-300 plant species”. However, grassland communities tend to be similar over large areas, and structurally simple, so that at the landscape scale diversity is relatively low compared with tropical moist forest or Mediterranean-type ecosystems. Grasslands tend to have low rates of endemism; however, the climatic and soil gradients within them have led to substantial ecotypic variation and high genetic diversity”. The world’s grasslands and savannahs support distinctive plant and animal com- munities. Although species diversity tends to increase towards the tropics, it tends to be moderate or low at the landscape scale and above. Little more than 5 percent of the world’s mammal and bird species are pri- marily dependent on grasslands habitats”. Alt these ecosystems hold, or formerly held, an array of native herbivores, and these in turn support a number of high-profile mammalian and avian predators. The world’s grasslands and Savannahs support distinctive plant and animal communities. 10% WORLD ATLAS OF BIO — seeeeenoeed —— DIVERSITY The figure illustrates the living planet index in a sample of 29 grassland birds from North America (25 species] and Europe (4). The clear trend over four decades is downward, and as with forest birds the separate continent samples show a similar overall pattern. 140 - 120 - 100 80 60 - AD jy 1970 Source: UNEP-WCMC T 1975 T T 1 1980 1985 1990 , from data collated for Loh?”. The savannah communities of East Africa are typified by large herds of ungulate herbivores, including a remarkable diversity - more than 70 species — of antelope and other medium- to large-sized bovids. The biomass of ungulate herbivores here may rise to 30 metric tons per km’, which is the highest recorded in any terrestrial environment”. Most grassland invertebrate biomass is found within the soil {most commonly nematodes, enchytraeid worms and mites] and may be in the order of 100 to 1000 times as great as vertebrate biomass; soil invertebrate biomass above the soil is often dominated by Orthoptera’’. Many grassland birds require large areas of habitat to take full advantage of sparsely distributed food resources, and fragmentation of natural grasslands has made conservation of these wide-ranging species difficult’. An analysis of the location of bird species in the neotropics indicated that nearly 12 percent of threatened birds (38 species) are confined to grasslands. In this region, the grasslands of southern Brazil and northern Argentina are particularly threatened as a result of agricultural improvement®™. In North America, grassland species have experienced the most consistent declines of any group of birds monitored in a national survey of breeding birds; available data suggest that there has been a constant decline in these species over the past 30 years™. Among notable threatened mammal species in grasslands are the greater one- horned rhinoceros Rhinoceros unicornis, associated with tall riverine grassland in northern India and southern Nepal, where it is threatened by poaching and further loss of its restricted habitat. Other large mammals are the saiga antelope Saiga tatarica of central Asian grasslands, and the vicuna Vicugna vicugna in arid grasslands and plains in the Andes; both of these are threatened in parts of their range. Among birds, the short grassland habitat of the plains wanderer Pedionomus torquatus of southern Australia continues to be lost to cultivation; Rudd’s lark Heteromirafra ruddi inhabits the montane grassland plateaus of eastern South Africa and is threatened by habitat degradation; while the lesser florican Sypheotides indica, of western India and Nepal, is critically endangered through loss of suitable grass- land habitat. Grassland biomass ranges from around 2 to over 80 metric tons per hectare’. As a rule, more than 50 percent of this is below the soil surface, and the ratio of root to shoot biomass ranges from below five in warm humid grasslands to as much as 30 in the desert grassland of Mongolia’. Usually more than half of the below-ground biomass is in the upper 10 cm of the soil, and soil carbon stocks may be as high as 250 metric tons per hectare®. Annual production reaches 30-50 metric tons per hectare per year in some warm humid grasslands. It is estimated that grasslands and savannahs together account for about a third of global terrestrial net primary production (Table 1.1), so despite their relatively low biomass grasslands play an important role in the global carbon balance. Because the high below-ground biomass tends to increase the fertility of their soils, grasslands have been subject to high rates of conversion to agriculture. Less than half of North American grasslands remain in natural or semi-natural states“, the steppes of the former Soviet Union have been extensively irrigated and converted to agriculture and much of the pampas has been converted to agriculture or grazing land. Some anthropo- genic grassland consists of short-term monospecific sown pasture, while some areas support species-rich semi-natural grassland created over centuries by pastoral- ists in conjunction with livestock grazing. Domestic livestock grazing is the most extensive human use of unconverted or anthropogenic grassland ecosystems, and of most arid or semi-arid ecosystems. Livestock have an impact on ecosystems through trampling, removal of plant biomass, alteration of plant species composition through selective grazing, and competition with native species. The impact of this on the biological diversity of these ecosystems has been variable. In some areas where the native vegetation is well adapted, the impact on plant species diversity has been relatively small. In other areas, where vegetation has not evolved in the presence of hooved herbivores, the changes have been great. Sometimes, particularly in tropical and semi- tropical grasslands, the dominant com- ponent of vegetation has shifted from grass to woody plants. Overgrazing can lead to reduction in plant cover, loss and degradation of soil, and invasion by non-native plant species. In almost all cases, wild animal diversity has been greatly affected (mostly through compe- tition and hunting, but also through spread of pathogens}, so that the biomass of domestic livestock greatly exceeds that of native wild herbivores. In some areas, feral species [e.g. rabbits, camels, donkeys, horses, goats] may also have a marked impact on natural or semi-natural ecosystems. Shrublands Shrub communities, where woody plants, usually adapted to fire, form a continuous cover, occur in all parts of the world with 200- 1000 mm of rainfall”. In more arid areas including some semi-desert systems, shrubs are the dominant life form, but cover is discontinuous. The most distinctive and best- known shrublands are those of Mediterranean climate regions. Mediterranean climates are typified by cool, wet winters and warm, or hot, dry sum- mers. However, no single climatic or bio- climatic definition of a Mediterranean eco- system has yet been established, so that these areas remain rather loosely defined. Mediterranean ecosystems encompass a wide range of habitat types including forest, woodland and grassland, but are typified by a low, woody, fire-adapted sclerophyllous shrubland (maquis, chaparral, fynbos, mallee] on relatively nutrient-poor soils. These sys- tems occur in five distinct parts of the world: the Mediterranean basin; California (United States]; central Chile; Cape Province (South Africa]; and southwestern and south Australia. Each of these regions occurs on the west side of a continent and to the east of a cold ocean current that generates winter rainfall. They cover around 2.5 million km’ in total, or between 1 percent and 2 percent of the Earth’s surface (according to definition). More than two thirds of the total Mediterranean- type ecosystem area is found within the Mediterranean basin. Cape Province, South Africa 0.09 Southwestern Australia 0.31 California 0.32 Chile 0.14 Mediterranean basin 1.87 Differences in vegetation structure between regions are in part a consequence of differ- ences in the annual distribution of rainfall. In South Africa the sclerophyllous fynbos community contains an abundance of erica- ceous species as an understorey to low broader-leaved shrubs including members of the Proteaceae and Myrtaceae™. The Australian heaths are structurally similar, with Epacridaceae replacing Ericaceae. Californian shrublands, known as chaparral, are characterized by Adenostoma [Rosaceae] and a high richness of Arctostaphylos species and other members of the Ericaceae. The shrublands, or matorral, of Chile include many Terrestrial biodiversity 105 ayy 8 550 ca 8 000 5 050 ca 2 100 25 000 Table 5.10 Estimated plant species richness in the five regions of Mediterranean- type climate Source: UNEP”, MH, meen 10 WORLD ATLAS OF BIODIVERSITY aa Map 5.7 Current forest distribution Map adapted from the global landcover classification developed by the University of Maryland. The Maryland classification includes 13 classes and was based on AVHRR remote-sensing data with a spatial resolution of 1 km The map shown was derived by reclassifying the Maryland landcover data to accord with an ecologically based classification of five major forest types. For presentation purposes the data have here been generalized to a 4-km grid. Source: Data from University of Maryland Global Land Cover Facility. For full description see Hansen Forest type (eee Tropical moist Tropical dry Temperate broadleaf and mixed Temperate and boreal needleleaf Sparse trees and parkland of the same genera as those of California, while in the Mediterranean basin itself Ericaceae, Cistaceae, Leguminosae and Oleaceae are all important. Species richness in Mediterranean-type ecosystems, particularly among plants, is generally high - approaching values for moist tropical forest areas - and levels of endemism are also very high. Among the five Mediterranean-type ecosystems, species richness appears highest on the poorer soils of South Africa and southwest Australia (Table 5.10}, and lower on the richer soils of California, Chile and the Mediterranean basin”. Countries around the Mediterranean Sea hold some 25000 vascular species {about 10 percent of all vascular plants) of which around 60 percent are endemic to the Mediterranean region. The remaining four Mediterranean-type ecosystem regions are all considered to hold a disproportionately high floristic diversity in relation to their area’’. At fine scale, mean plant richness in the fynbos of South Africa is moderate, i.e. around 16 species per m’, but many species have small ranges, and there is a uniquely high turnover in the species composition of plant communities along ecological and geogra- phical gradients. At landscape scale, richness accordingly rises to very high values, for 2 256 Species occur in 471 km* on the Cape peninsula and the entire Cape floristic region (including some non-fynbos vegetation) holds some 8 550 species, about 70 percent of which are endemic. The Mediterranean-type ecosystems in general have a relatively high proportion of Terrestrial biodiversity 107 = ES SIE EEE LEE TE La LOOP OE IT LEIA PERE IEEE EO ; — their species categorized as threatened. The Cape flora, largely within a Mediterranean- type ecosystem, occupies only 4 percent of the land area of southern Africa, but accounts for nearly 70 percent of the region’s threatened plant species. About one third of the natural vegetation has been transformed by human activity; the remaining natural vegetation Is at risk from a number of invasive introduced woody plants, and the effects of an introduced ant (that suppresses native seed-storing ants and thus renders seed liable to destruction by rodents or fire]. Around 10 percent of the Californian flora is considered threatened (equivalent to approximately one quarter of all threatened plants in the United States). In Australia, heath habitats, primarily in the southwest Mediterranean-type ecosystem region, rank third after ‘woodland’ and ‘scrub’ in numbers of ‘endangered’ category plants. Given their much smaller extent, this indi- cates that a far higher proportion of their flora is threatened than in either woodland or scrub habitats. Vertebrate diversity tends to be lower than in plants. To take an example, in the Cape Mediterranean-type ecosystem, reptile diver- sity is only moderate while bird and mammal diversity is relatively low. The absence of large mammals in California and the Mediterranean basin may be linked to overhunting by humans during the late Pleistocene™. Several threatened animal species rely on shrubby or scrub habitat. The Iberian lynx Lynx pardinus, found in the light woodland and maquis of Spain and Portugal, where habitat loss and hunting have led to decline, is possibly the most threatened cat species. Map 5.8 Non-forest terrestrial ecosystems Map adapted from the global landcover classification developed by the University of Maryland. The Maryland classification includes 13 classes and was based on AVHRR remote-sensing data with a Spatial resolution of 1 km. The map shown was derived by reclassifying the Maryland landcover data to accord with a highly generalized classification of non-forest terrestrial ecosystem types. For presentation purposes the data have here been generalized to a 4-km grid. Source; Data from University of Maryland Global Land Cover Facility. For 9 full description see Hansen 108 WORLD ATLAS OF BIODIVERSITY SSRIS ET Ecosystem type Tundra Closed shrubland Open shrubland Grassland and savannah Cropland Desert Water body The riverine rabbit Bunolagus monticularis of South Africa is restricted to a small area of riverine bush in the central Karoo, where it is threatened by further loss of this habitat to agriculture. Among birds, the island cisticola Cisticola haesitatus is endemic to the island of Socotra in the western Indian Ocean, where it is threatened by loss of light scrub and grassland habitat, possibly through over- grazing by goats. Mediterranean-type shrublands are not notably high in either biomass or net primary production. Biomass at mature fynbos sites is typically 15-16 metric tons per hectare, and in chaparral may be twice that®’. Combined with their relatively low rates of primary production, related to both climate and soil fertility factors, the incidence of fire tends to reduce the accumulation of carbon in these ecosystems and their soils. Total carbon storage is probably between 100 and 150 metric tons per hectare in Mediterranean-type shrublands”. The Mediterranean basin itself has for many centuries been subject to intense human activities, including forest clearance and grazing, such that little genuinely natural vegetation remains. It has been suggested that the plant diversity is locally high because of the number of species that have evolved as components of successional vegetation in response to frequent disturbance. In other Mediterranean-type shrublands, expanding human populations and conversion of land to agricultural or residential use are important pressures. These changes are often accom- panied by changing fire and grazing regimes, and both these changes tend to facilitate invasion by non-native plant and animal species, which threaten native species pop- ulations, especially in California and South Africa”. Deserts and semi-deserts Nearly 10 million km? of the Earth's land area is hyperarid, or true desert, where rainfall is extremely low and unpredictable in space and time, so that in some years none falls at all. These areas have a ratio of rainfall to potential evapotranspiration (P/PET) of less than 0.05”. The Sahara desert alone makes up nearly 70 percent of the world hyperarid zone. Other extensive areas are found in the Arabian Peninsula and central Asia, with smaller areas in southwest Africa, the Horn of Africa, western South America and western North America. In semi-deserts, areas with less arid climates, the vegetation is usually Terrestrial biodiversity 109 more substantial than in deserts, but covers no more than 80 percent of the ground. Temperate deserts and semi-deserts cover nearly 6 million km’ in Eurasia and North and South America”. Polar regions and some high mountain areas with a permanently cold, dry climate also meet the definition of desert, but have completely different ecological characteristics from true drylands and are not usually considered with them. Plant cover in desert ecosystems ranges from areas without vegetation to areas with low densities of small shrubs and perennial grasses, often with populations of annuals that vary in density depending on seasonal precipitation. Deserts are often character- ized by short periods of relatively high pro- ductivity during rainy periods, interspersed with long periods of very low productivity. As uh 110 WORLD ATLAS OF BIODIVERSITY Desert species show a wide range of adaptations to extreme environments. Dating back some 15 million years, the Namib appears to be Earth's oldest desert. a result, most herbivorous insects develop only one generation per season, leading to a time lag in the interaction between herbi- vorous insects and their food resources”. Consequently, most of the primary produc- tivity in deserts ends up as dry plant material or seeds, which accumulates because micro- bial decomposition is limited by the low moisture availability. These resources form the basis for populations of detritivores such as termites, darkling beetles and isopods, and seed predators such as ants. These groups can support a rich fauna of predatory arthropods and reptiles as well as omni- vorous birds and mammals. It is often assumed that all deserts have low species diversity because of the harsh environmental conditions, but among animals almost all terrestrial higher taxa are represented, and their species diversity may in some situ- ations be comparable to that of more mesic habitats”. True desert species show a wide range of adaptations to the conditions of an extreme environment. Characteristic plants include the Cactaceae in the Americas and the succulent Euphorbiaceae in Africa. Semi- desert species include salt bush Atriplex, and creosote bush Larrea. Amongst animals, groups that are intrinsically adapted to very low moisture environments include reptiles and many arthropods, although species in a wide range of other groups have also evolved to cope with these conditions. Strategies for survival amongst both plants and animals often include long periods of dormancy [as seeds, in the case of many plants) between rainfall events. In the particular conditions prevailing in the so-called fog deserts (notably the Atacama desert and the Skeleton Coast desert], different strategies have evolved. Here plants and animals make use of the regular moisture-laden fogs, which roll in from the cold offshore currents, to obtain a low but predictable supply of water. The often overlooked inland water habitats of deserts may contain a particularly high proportion of locally endemic species; the pools of Cuatro Ciénegas, Mexico, for example, contain numerous mollusk and fish species found nowhere else. Many of the larger desert and sub-desert vertebrates are threatened; the openness of these arid areas means that species such as antelopes and other bovids are more con- spicuous than forest species and thus more vulnerable to overhunting. Threatened verte- brates include the wild bactrian camel Camelus bactrianus with a few remnant populations in the Gobi desert of Mongolia and China, and Przewalski’s gazelle Procapra przewalskii of China’s subdesert steppes, now restricted by overhunting and habitat loss to a few small areas surrounding Lake Quinghai. The Mexican prairie dog Cynomys mexicanus is confined to prairies and intermontane basins with herbs and grasses where it is threatened by persecution and continuing SES EEE a habitat loss. The Addax antelope Addax nasomaculatus was originally widespread from the western Sahara to Egypt and Sudan, but as a result of uncontrolled hunting it persists only in small and scattered local populations and is extinct through much of its former range. Principal threats to desert environments include the activities of domestic livestock such as cattle, which cause soil compaction through trampling, and can damage vege- tation and waterholes. Introduced species, such as rabbits in Australia, have been highly damaging to some desert environments. Other human activities that have affected desert habitats include use of off-road vehicles, irrigation and afforestation schemes, and housing projects. In drylands, most REFERENCES Terrestrial biodiversity 11 OEE adverse impacts that lead to some form of land degradation can be categorized as ‘desertification’. Under the UN Convention to Combat Desertification, the latter term is defined explicitly as ‘land degradation in arid, semi-arid and dry sub-humid areas resulting from various factors, including climatic variations and human activities’. According to the above definition, hyperarid lands (true deserts) are not susceptible to desertification, because their productivity is already so low that it cannot be seriously decreased by human action. The effects of desertification on arid and semi-arid areas promote poverty among rural people, and by placing greater pressure on natural resources, poverty tends to reinforce any existing trend toward desertification. 1 Loveland, T.R. et al. 1999. An analysis of the IGBP global land-cover characterization process. Photogrammetric Engineering and Remote Sensing 65: 1021-1032. 2 ITTO 2001. Annual review and assessment of the world timber situation 2000. International Tropical Timber Organization. 3 Salisbury, H.E. 1989. The great black dragon fire: A Chinese inferno. Little, Brown and Co., Boston. 4 Farjon, A. and Page, C.N. 1999. Status survey and conservation action plan: Conifers. IUCN-the World Conservation Union, Gland. 5 Hoang Ho-dzung 1987. The moss flora of the Baektu mountain area. In: Yang, H. et al. (eds). The temperate forest ecosystem, pp. 29-31. Institute of Terrestrial Ecology Symposium 20, National Environment Research Council, UK. 6 Vaisanen, R., Bistrom, O. and Heliovaara, K. 1993. Sub-cortical Coleoptera in dead pines and spruces: Is primeval species composition maintained in managed forests? Biodiversity and Conservation 2(2): 95-113. 7 Rundel, P.W. and Boonpragob, K. 1995. Dry forest ecosystems of Thailand. In: Bullock, S.H., Mooney, H.A. and Medina, E. (eds). Seasonally dry tropical forests, pp. 93-123. Cambridge University Press, Cambridge. 8 Rohrig, E. 1991. Floral composition and its evolutionary development. In: Rohrig, E. and Ulrich, B. (eds). Ecosystems of the world 7: Temperate deciduous forests, pp. 17-24. Elsevier, Amsterdam. 9 Ovington, J.D. and Pryor, L.D. 1983. Temperate broad-leaved evergreen forests of Australia. In: Ovington, J.D. [ed.) Ecosystems of the world 10: Temperate broad-leaved evergreen forests, pp. 72-102. Elsevier, Amsterdam. 10 Barnes, B.V. 1991. Deciduous forests of North America. In: Rohrig, E. and Ulrich, B. (eds). Ecosystems of the world 7: Temperate deciduous forests, pp. 219-344. Elsevier, Amsterdam. 11 Williams, P.H., Gaston, K.J. and Humphries, C.J. 1997. Mapping biodiversity value worldwide: Combining higher-taxon richness for different groups. Proceedings of the Royal Society, Biological Sciences 264: 141-148. 12 Thomas, W.W. 1999. Conservation and monographic research on the flora of tropical America. Biodiversity and Conservation 8: 1007-1015. mA ww 12 WORLD ATLAS OF BIODIVERSITY = EE | 13 Olson, D.M. et al. 2001. Terrestrial ecoregions of the world: A new map of life on Earth. BioScience 51: 933-938. Also see website http://www.panda.org/resources/programmes/ global200/pages/list.htm {accessed March 2002). 14 FAO 2001. State of the world’s forests 2001. Food and Agriculture Organization of the United Nations, Rome. 15 UNESCO 1973. International classification and mapping of vegetation. United Nations Educational, Scientific and Cultural Organization, Paris. 16 Hilton-Taylor, C. [compiler] 2000. 2000 IUCN Red List of threatened species. |\UCN-the World Conservation Union, Gland and Cambridge. Available online at http://www.redlist.org/ {accessed March 2002). | 17 Veblen, T.T., Schlegel, F.M. and Oltremari, J.V. 1983. Temperate broad-leaved evergreen forests of South America. In: Ovington, J.D. (ed.} Ecosystems of the world 10: Temperate broad-leaved evergreen forests, pp. 5-32. Elsevier, Amsterdam. 18 Ching, K.K. 1991. Temperate deciduous forests in East Asia. In: Rohrig, E. and Ulrich, B. {eds}. Ecosystems of the world 7: Temperate deciduous forests, pp. 539-556. Elsevier, Amsterdam. 19 Lear, M. and Hunt, D. 1996. Updating the threatened temperate tree List. In: Hunt, D. {ed.) Temperate trees under threat, pp. 161-171. International Dendrology Society, UK. 20 Ohba, H. 1996. A brief overview of the woody vegetation of Japan and its conservation status. In: Hunt, D. (ed.) Temperate trees under threat, pp. 81-88. International Dendrology Society, UK. 21 Schaefer, M. 1991. The animal community: Diversity and resources. In: Rohrig, E. and Ulrich, B. [eds]. Ecosystems of the world 7: Temperate deciduous forests, pp. 51-120. Elsevier, Amsterdam. 22 Ovington, J.D. and Pryor, L.D. 1981. Temperate broad-leaved evergreen forests of Australia. In: Ovington, J.D. {ed.} Ecosystems of the world 10: Temperate broad-leaved evergreen forests, pp. 73-99. Elsevier, Amsterdam. 23 Rohrig, E. 1991. Biomass and productivity. In: Rohrig, E. and Ulrich, B. (eds). Ecosystems of the world 7: Temperate deciduous forests, pp. 165-174. Elsevier, Amsterdam. 24 Zinke, P.J. et al. 1984. Worldwide organic soil carbon and nitrogen data. Environmental Sciences Division, Publication No. 2212. Oak Ridge National Laboratory, US Department of Energy. 25 Gentry, A.H. 1988. Changes in plant community diversity and floristic composition on environmental and geographical gradients. Annals of the Missouri Botanical Garden 75: 1-34. 26 Whitmore, T.C. 1984. Tropical rain forests of the Far East. Clarendon Press, Oxford. 27 Grubb, P.J. 1977. Control of forest growth on wet tropical mountains. Annual Review of Ecology and Systematics 8: 83-107. 28 Huston, M.A. 1994. Biological diversity: The coexistence of species on changing landscapes. Cambridge University Press, Cambridge. 29 Nelson, B. et al. 1994. Forest disturbance by large blowdowns in the Brazilian Amazon. Ecology 75: 853-858. 30 Tanner, E.V.J., Kapos, V. and Healey, J.R. 1991. Hurricane effects on forest ecosystems in the Caribbean. Biotropica 23: 513-521. 31 Pires, J.M. 1957. Nocoes sobre ecologia e fitogeografia da Amazonia. Norte Agronémico 3: 37-53. 32 de Oliveira, A.A. and Mori, S.A. 1999. A central Amazonian terra firme forest. |. High tree species richness on poor soils. Biodiversity and Conservation 8: 1219-1244. 33 Gentry, A.H. 1988. Tree species richness of upper Amazonian forests. Proceedings of the National Academy of Sciences 85: 156-159. 34 Benzing, D.H. 1989. Vascular epiphytism in America. In: Lieth, H. and Werger, M.J.A. (eds). Ecosystems of the world 14B: Tropical rain forest ecosystems: Biogeographical and ecological studies, pp. 133-154. Elsevier, Amsterdam. 35 Prance, G.T. 1989. American tropical forests. In: Lieth, H. and Werger, M.J.A. (eds). Ecosystems of the world 14B: Tropical rain forest ecosystems: Biogeographical and ecological studies, pp. 99-132. Elsevier, Amsterdam. 36 Jenkins, M. 1992. Biological diversity. In: Sayer, J.A., Harcourt, C.S. and Collins, N.M.H. The conservation atlas of tropical forests: Africa, pp. 26-32. Macmillan Publishers, London. 37 Harcourt, C.S. and Sayer, J.A. (eds) 1996. The conservation atlas of tropical forests: The Americas. Simon and Schuster, New York. 38 Lynch, J.D. 1979. The amphibians of the lowland tropical forests. In: Duellman, W.E. [ed.} The South American herpetofauna: Its origin, evolution, and dispersal, pp. 189-215. Monograph No. 7 of the Museum of Natural History Kansas, Lawrence. 39 Goulding, M., Leal Carvalho, M. and Ferreira, E.G. 1988. Rio Negro: Rich life in poor waters: Amazonian diversity and foodchain ecology as seen through fish communities. SPB Academic Publishing, The Hague. 40 World Conservation Monitoring Centre. 1992. Groombridge, B. (ed.) Global biodiversity: Status of the Earth's living resources. Chapman and Hall, London. 41 Holldobler, B. and Wilson, E.0. 1990. The ants. Harvard University Press, Cambridge MA. 42 Hamilton, L.S., Juvik, J.O. and Scatena, F.N. 1993. Tropical montane cloud forests. East- West Center, Honolulu. 43 Salati, E. and Vose, P.B. 1984. Amazon basin: A system in equilibrium. Science 225: 129-138. 44 Mooney, H.A., Bullock, S.H. and Medina, E. 1995. Introduction. In: Bullock, S.H., Mooney, H.A. and Medina, E. (eds). Seasonally dry tropical forests, pp. 1-8. Cambridge University Press, Cambridge. 45 Sarmiento, G. 1992. A conceptual model relating environmental factors and vegetation formations in the lowlands of tropical South America. In: Furley, P.A., Proctor, J. and Ratter, J.A. (eds). Nature and dynamics of forest-savanna boundaries, pp. 583-601. Chapman and Hall, London. 46 Menaut, J.C., Lepage, M. and Abbadie, L. 1995. Savannas, woodlands and dry forests in Africa. In: Bullock, S.H., Mooney, H.A. and Medina, E. (eds). Seasonally dry tropical forests, pp. 64-92. Cambridge University Press, Cambridge. 47 Gentry, A.H. 1995. Diversity and floristic composition of neotropical dry forests. In: Bullock, S.H., Mooney, H.A. and Medina, E. {eds}. Seasonally dry tropical forests, pp. 146-194. Cambridge University Press, Cambridge. 48 Murphy, P.G. and Lugo, A.E. 1995. Dry forests of Central America and the Caribbean. In: Bullock, S.H., Mooney, H.A. and Medina, E. [eds]. Seasonally dry tropical forests, pp. 9-28. Cambridge University Press, Cambridge. 49 Ceballos, G. 1995. Vertebrate diversity, ecology and conservation in neotropical dry forests. In: Bullock, S.H., Mooney, H.A. and Medina, E. [eds]. Seasonally dry tropical forests, pp. 195-220. Cambridge University Press, Cambridge. 50 Dodson, C.H. and Gentry, A.H. 1991. Biological extinction in western Ecuador. Annals of the Missouri Botanical Garden 78: 273-295. 51 Janzen, D.H. 1988. Tropical dry forests, the most endangered tropical ecosystem. In: Wilson, E.0. 1988 (ed.} Biodiversity, pp. 13-137. National Academy Press, Washington DC. 52 Murphy, P.G. and Lugo, A.E. 1986. Ecology of tropical dry forest. Annual Review of Ecology and Systematics 17: 67-88. 53 Bye, R. 1995. Ethnobotany of the Mexican tropical dry forests. In: Bullock, S.H., Mooney, H.A. and Medina, E. (eds). Seasonally dry tropical forests, pp. 423-438. Cambridge University Press, Cambridge. Terrestrial biodiversity 113 - “ . hy) Wi 114 WORLD ATLAS OF BIODIVERSITY = TSS eT AY A QE RR PL 54 Maass, J.M. 1995. Conversion of tropical dry forest to pasture and agriculture. In: Bullock, S.H., Mooney, H.A. and Medina, E. (eds). Seasonally dry tropical forests, pp. 399-422. Cambridge University Press, Cambridge. 55 Bullock, S.H. 1995. Plant reproduction in neotropical dry forests. In: Bullock, S.H., Mooney, H.A. and Medina, E. (eds). Seasonally dry tropical forests, pp. 277-304. Cambridge University Press, Cambridge. 56 Tukhanen, S. 1999. The northern timberline in relation to climate. In: Kankaapaa, Tasanen, S.T. and Sutinen, M.-L. (eds). Sustainable development in northern timberline forests, pp. 29-62. Finnish Forest Research Institute Research Papers 734. 57 Solbrig, 0.T., Medina, E. and Silva, J.F. 1996. Determinants of tropical savannas. In: Solbrig, 0.T., Medina, E. and Silva, J.F. (eds). Biodiversity and savanna ecosystem processes: A global perspective, pp. 31-41. Springer, Berlin. 58 Newsome, A.E. 1983. The grazing Australian marsupials. In: Bourliére, F. [ed.) Ecosystems of the world 13: Tropical savannas, pp 441-459. Elsevier, Amsterdam. 59 Solbrig, O.T. 1996. The diversity of the savanna ecosystem. In: Solbrig, 0.T., Medina, E. and Silva, J.F. (eds). Biodiversity and savanna ecosystem processes: A global perspective, pp. 1-27. Springer, Berlin. 60 Lavelle, P. 1983. The soil fauna of tropical savannas. |. The community structure. In: Bourliere, F. (ed.) Ecosystems of the world 13: Tropical savannas, pp. 477-484. Elsevier, Amsterdam. 61 FAO 2001. Global forest resources assessment 2000. Food and Agriculture Organization of the United Nations, Rome. 62 Williams, M. 1989. Deforestation: Past and present. Progress in Human Geography 13: 176-208. 63 Williams, M. 1991. Forests. In: Turner II, B.L. et al. (eds). The Earth as transformed by human action, pp. 179-201. Cambridge University Press with Clark University, Cambridge. 64 FAO 1999. State of the world’s forests 1999. Food and Agriculture Organization of the United Nations, Rome. 65 Holdsworth, A.R. and Uhl, C. 1997. Fire in Amazonian selectively logged rain forest and the potential for fire reduction. Ecological Applications 7: 713-725. 66 Laurance, W.F. and Bierregaard, R.O. [eds] 1997. Tropical forest remnants: Ecology, management and conservation of fragmented communities. University of Chicago Press, Chicago. 67 Smith, T.M., Leemans, R. and Shugart, H.H. 1992. Sensitivity of terrestrial carbon storage to CO,-induced climate change: Comparison of four scenarios based on general circulation models. Climatic Change 21: 367-384. 68 Pounds, J.A., Fogden, M.P.L. and Campbell, J.H. 1999. Biological response to climate change on a tropical mountain. Nature 398: 611-615. 69 Bliss, L.C. 1997. Arctic ecosystems of North America. In: Wielgolaski, F.E. {ed.) Ecosystems of the world 3: Polar and alpine tundra, pp. 551-683. Elsevier, Amsterdam. 70 Wielgolaski, F.E. 1997. Introduction. In: Wielgolaski, F.E. (ed.) Ecosystems of the world 3: Polar and alpine tundra, pp. 1-6. Elsevier, Amsterdam. 71 Chernov, Yu |. and Matveyeva, N.V. 1997. Arctic ecosystems in Russia. In: Wielgolaski, F.E. {ed.] Ecosystems of the world 3: Polar and alpine tundra, pp. 361-507. Elsevier, Amsterdam. 72 Diaz, A., Péfaur, J.E. and Durant, P. 1997. Ecology of South American paramos with emphasis on the fauna of the Venezuelan paramos. In: Wielgolaski, F.E. (ed.) Ecosystems of the world 3: Polar and alpine tundra, pp. 263-310. Elsevier, Amsterdam. 73 Campbell, J.S. 1997. North American alpine ecosystems. In: Wielgolaski, F.E. (ed. Ecosystems of the world 3: Polar and alpine tundra, pp. 211-261. Elsevier, Amsterdam. Terrestrial biodiversity 15 ee a eee eed FEED Wh 74 Grabherr, G. 1997. The high-mountain ecosystems of the Alps. In: Wielgolaski, F.E. (ed.] Ecosystems of the world 3: Polar and alpine tundra, pp. 97-121. Elsevier, Amsterdam. 75 Hedberg, 0. 1997. High-mountain areas of tropical Africa. In: Wielgolaski, F.E. [ed.] Ecosystems of the world 3: Polar and alpine tundra, pp. 185-197. Elsevier, Amsterdam. 76 Zockler, C. 1998. Patterns of biodiversity in arctic birds. WCMC Biodiversity Bulletin 3: 1-15. 77 Bazilevich, N.I. and Tishkov, A.A. 1997. Live and dead reserves and primary production in polar desert, tundra and forest tundra of the former Soviet Union. In: Wielgolaski, F.E. (ed.) Ecosystems of the world 3: Polar and alpine tundra, pp. 509-539. Elsevier, Amsterdam. 78 United Nations Environment Programme 1995. Heywood, V. {ed.) Global biodiversity assessment. Cambridge University Press, Cambridge. - 79 Risser, P.G. 1988. Diversity in and among grasslands. In: Wilson, E.0. and Peter, F.M. (eds). Biodiversity, pp. 176-180. National Academy Press, Washington DC. 80 Coupland, R.T. 1992. Overview of the grasslands of North America. In: Coupland, R.T. [ed.] Ecosystems of the world 8A: Natural grasslands: Introduction and western hemisphere, pp. 147-150. Elsevier, Amsterdam. 81 Coupland, R.T. 1993. Review. In: Coupland, R.T. (ed.) Ecosystems of the world 8B: Natural grasslands: Eastern hemisphere and resumé, pp. 471-482. Elsevier, Amsterdam. 82 Lavrenko, E.M. and Karamysheva, Z.V. 1993. Steppes of the former Soviet Union and Mongolia. In: Coupland, R.T. ed.) Ecosystems of the world 8B: Natural grasslands: Eastern hemisphere and resumé, pp. 3-60. Elsevier, Amsterdam. 83 Wege, D.C. and Long, A.J. 1995. Key areas for threatened birds in the neotropics. Birdlife International, Cambridge. 84 Sauer, J.R. et al. 1997. The North American breeding bird survey results and analysis Version 96.3. Patuxent Wildlife Research Center, Laurel. Available online at: www.mbrpwrc.usgs.gov/bbs/bbstext.html [accessed April 2002). 85 Adams, J. 1997. Estimates of preanthropogenic carbon storage in global ecosystem types. Available online at http://www.esd.ornl.gov/projects/qen/carbon3.html [accessed April 2002). 86 Cowling, R.M. and Holmes, P.M. 1992. Flora and vegetation. In: Cowling, R.M. {ed.} The ecology of fynbos: Nutrients, fire and diversity, pp. 23-61. Oxford University Press, Cape Town. 87 Myers, N. 1990. The biodiversity challenge: Expanded hot-spots analysis. Environmentalist 10: 243-256. 88 Davis, G.W., Richardson, D.M., Keeley, J.E. and Hobbs, R.J. 1996. Mediterranean-type ecosystems: The influence of biodiversity on their functioning. In: Mooney, H.A. et al. (eds). Functional roles of biodiversity: A global perspective, pp. 151-183. John Wiley and Sons, Chichester. 89 Keeley, J.F. 1992. A Californian’s view of fynbos. In: Cowling, R.M. {ed.) The ecology of fynbos: Nutrients, fire and diversity, pp. 372-388. Oxford University Press, Cape Town. 90 Richardson, D.M., Macdonald, |.A.W., Holmes, P.M. and Cowling, R.M. 1992. Plant and animal invasions. In: Cowling, R.M. (ed.) The ecology of fynbos: Nutrients, fire and diversity, pp. 271-308. Oxford University Press, Cape Town. 91 Middleton, N. and Thomas, D. (eds) 1997. World atlas of desertification. Second edition. United Nations Environment Programme. Arnold, London. 92 West, N.E. 1983. Approach. In: West, N.E. (ed.] Ecosystems of the world 5: Temperate deserts and semi-deserts, pp. 1-2. Elsevier, Amsterdam. 93 Polis, G.A. 1991. Desert communities: An overview of patterns and processes. In: Polis, G.A. {ed.] The ecology of desert communities, pp. 1-26. University of Arizona Press, Tucson. 94 Phillips, O.L. et al. 1994. Dynamics and species richness of tropical rain forests. Proceedings of the National Academy of Sciences 91: 2805-2809. 1146 WORLD ATLAS OF BIODIVERSITY EE 95 FGDC 1995. FGDC Vegetation Classification Standards. Federal Geographic Data Committee, Reston. Unpublished. 96 UNEP-WCMC 1999. Contribution of protected areas to global forest conservation. In: International forest conservation: Protected areas and beyond. A discussion paper for the intergovernmental forum on forests. Commonwealth of Australia, Canberra. 97. Loh, J. ed.) 2000. Living planet report 2000. WWF - World Wide Fund for Nature, Gland. 98 Hansen, M. et al. 2000. Global land cover classification at 1 km resolution using a decision tree classifier. International Journal of Remote Sensing 21: 1331-1365. 99 Barthlott, W. et al. 1999. Terminological and methodological aspects of the mapping and analysis of global biodiversity. Acta Botanica Fennica 162: 103-110. 100 Heywood, V.H. led.) 1993. Flowering plants of the world. B.T. Batsford Ltd, London. 6 Marine biodiversity times the average elevation of the land, making the open sea by far the largest ecosystem on Earth. Despite this volume, marine net primary production remains similar to or less than that on land because photosynthesis in the sea is carried out by micro- scopic bacteria and algae restricted to the sunlit surface layers (plants are virtually absent). The diversity of major lineages (phyla and classes) is much greater in the sea than on land or in freshwaters, and many phyla of invertebrate animals occur only in marine waters. Species diversity appears to be far lower, perhaps because marine waters are physically much less variable in space and time than the terrestrial environment. Marine fisheries are the largest source of wild protein, derived from fishes, mollusks and crustaceans. The world catch from capture fisheries has grown fivefold over the past five decades, but appears to have declined during the 1990s despite increased fishing effort. More than half of the world’s major fishery resources are now in need of remedial management, mainly because of excess exploitation. M OST OF THE PLANET IS COVERED BY OCEAN waters whose average depth is four THE SEAS Oceans cover 71 percent of the world’s surface. They are on average around 3.8 kilo- meters [km] deep and have an overall volume of some 1 370 million km’. The whole of the world ocean [all contiguous seas) is theor- etically capable of supporting life, so that the remains far less well known and understood than the terrestrial part of the globe. The main reason for this is, quite simply, that marine part of the biosphere is far larger peclleteral Lee ey 7 F 3 Atlantic Ocean 82.217 000 than the terrestrial part. However, as on land, s life in the oceans is very unevenly distributed Li Eee beet Be tonichingle productive iC aageaal 14.056 000 - some parts are astonishi ucti ‘ ; P ‘ eS as Mediterranean Sea 2.505 000 and diverse while others are virtually barren. ‘ , : ; South China Sea 2 318 000 With the present configuration of land ; Bering Sea 2 269 000 masses, a major part (37 percent) of the world ; : ate : Caribbean Sea 1 943 000 ocean is within the tropics, and about 75 : : ‘ Gulf of Mexico 1 544 000 percent lies between the 45° latitudes. The d sie ie Sea of Okhotsk 1 528 000 largest continental shelf areas are in high : : East China Sea | 248 000 northern latitudes (Table 6.2), but about 30 Bg ; Yellow Sea 1 243 000 percent of the total shelf area is in the tropics. ae : ; ; Hudson Bay 1 233 000 Within the tropics, the shelf is most extensive Seachem 1 008 000 in the western Pacific (China Seas south to Agere ia) North Sea 575 000 north Australia). pte: Black Sea 461 000 Although knowledge of the functioning Red Sea 438 000 of the marine biosphere has increased enor- : y Baltic Sea 422 000 mously in the past few decades, overall it Marine biodiversity 17 Table 6.1 Area and maximum depth of the world’s oceans and seas 65 Source: Couper WJ 118 WORLD ATLAS OF BIODIVERSITY Total area (million km‘) Latitude bands (% of total] Polar and boreal (45-90°) Temperate (20-45°) Tropical {0-20°) Table 6.2 Relative areas of continental shelves and open ocean Source: Adapted from Longhurst and Pauly', after Moiseev. 360.3 26.7 26.6 40.9 36.8 28.8 36.6 30.3 much of it is inaccessible to humans. Study of any part below the top few meters requires specialized equipment and is expensive and time consuming. Knowledge of most of the sea is thus based largely on a range of remote-sensing and sampling techniques and often remains sketchy. As these techniques become more sophisticated, so our under- standing of marine ecosystems, particularly those away from the coastal zone, is under- going constant revision. Sea water and ocean currents Sea water is a complex but relatively uniform mixture of chemicals. Most of the 92 naturally occurring elements can be detected dissolved in it, but most only in trace concentrations; the most abundant are sodium (as Na+] and chlorine (as Cl-], which occur at a concen- tration some ten times higher than the next most abundant element, magnesium. The term used to quantify the total amount of dissolved salts in sea water is salinity, a dimensionless ratio, which generally ranges between 33 and 37 and averages 35. Most of the substances dissolved in sea water are unreactive and remain at relatively stable concentrations; however those that play a part in biological systems can be highly variable in time and space. The world’s sea waters are constantly in motion, at all scales from the molecular to the oceanic. Large-scale ocean circulation plays a vital role in mediating global climate as well as influencing the functioning of marine ecosystems. It is driven by complex interactions between a number of physical variables, notably latitudinal variations in solar radiation [and consequent heating and cooling), precipitation and evaporation, trans- fer of frictional energy across the ocean surface from winds, and forces resulting from the rotation of the planet. Surface currents are largely driven by pre- vailing winds. The most important features are vast, anticyclonic gyres in the subtropical regions of the world’s oceans. These are pri- marily driven by the westerly trade winds in the Roaring Forties and circulate clockwise in the northern hemisphere and counter- clockwise in the southern hemisphere. Also important are the eastward-flowing Antarctic circum-polar current and equatorial current. At smaller scales, eddies and rings are ubiqui- tous and are analogous to weather systems in the atmosphere; typical oceanic eddies may be 100 km across and persist for a year or more. The density of sea water increases with increasing salinity and with decreasing temp- erature until it freezes at around -1.9°C. Sea water of different density does not mix readily, and so the oceans tend to be well stratified vertically, with bodies of less dense sea water {warmer and less saline) sitting on top of cooler, more saline and denser bodies. The stratification is rarely stable over the long term, however, as the influence of climate changes the properties of surface waters and causes various forms of vertical mixing. The single most important factor controlling large- scale deep-water circulation appears to be the generation of cold, high-salinity water near the surface in the Weddell Sea and off Greenland in the North Atlantic. Here, during winter, the sea freezes and the floating ice sheet is virtually free of salt, leaving the underlying water more saline, cold and dense. This sinks to the bottom and moves south along the Atlantic floor, some passing south of the equator and circulating throughout the world ocean, producing what is known as the Great Conveyor - the interlocking system of major circulation currents in the deep sea. Because it originates near the surface it is well oxygenated as well as cold, and is the major reason why aerobic organisms can thrive in the deep sea. Major zones of upwelling occur along the western boundaries of continents where trade winds blow towards the equator, causing surface waters to be pushed offshore to be replaced by cooler deep waters. There are five major upwelling areas: the Humboldt current region off the Chilean and southern Peruvian coast of South America; the California current region off western North America; the Canary current off the coast of Mauritania in north- west Africa; the Benguela current region off southern Africa; and the northwest Arabian Sea. Marine production is much enhanced in these regions which typically support major pelagic fisheries. Outside these upwelling zones, stratification tends to persist through- out the year in the tropics and subtropics. MAJOR MARINE ZONES The continental shelf Marine waters around major land masses are typically shallow, lying over a continental shelf which may be anything from a few kilometers to several hundred kilometers wide. The most landward part is the littoral or intertidal zone where the bottom is subject to periodic exposure to the air. Water depth here varies from zero to several meters. Seaward of this the shelf slopes gently from shore to depths of one to several hundred meters, forming the sublittoral or shelf zone. Waters below low-tide mark in the continental shelf region are referred to as neritic. The extent, gradient and superficial geo- logy of continental shelf areas are determined by many factors, including levels of tectonic activity in the Earth’s crust. More than 80 percent of the global volume of river-borne sediment is deposited in the tropics {and an estimated 40 percent of it by just two river systems: the Huang He or Yellow River and the Ganges-Brahmaputra]’, and this is ref- lected in the extent of shelf areas in parts of the tropics, and in the high turbidity of coastal waters in monsoon regions. Most shelf areas in the tropics are overlain by sands or muds composed of sediment of terrestrial origin (terrigenous deposits]. Shelf regions support the marine com- munities most familiar to humans, and many of the marine resources of particular value to them. Although mangrove and coral reefs are two of the best known tropical coastal ecosystems, they dominate only a minor part of the world coastline: the former mainly in deltaic or other low-lying coastal plains, and the latter only in shallow waters where Marine biodiversity 119 ccm I terrestrial sedimentation is very low. Soft- bottom habitats with sparse vegetation are probably the most widespread coastal marine ecosystem type, and virtually the entire seabed away from the coastline is covered in marine sediments (see Box 6.1). The deep sea At the outer edge of the shelf there is an abrupt steepening of the sea bottom, forming the continental slope which descends to depths of 3-5 km. The sea bottom along the slope is referred to as the bathyal zone. At the base of the continental slope are huge abyssal plains which form the floor of much of the world ocean. Open waters above these plains make up the oceanic pelagic zone. Given that the oceans cover some 71 percent of the globe, and that the shelf area is relatively narrow, the oceanic pelagic zone is by far the most extensive ecosystem on Earth. The plains are punctuated by numerous sub- marine ridges and sea mounts which may break the surface to form islands. There are also a number of narrow trenches which have depths of from 7000 to 11000 meters [m). These constitute the hadal zone. Ocean Spaces between the particles of marine sediments contain water, air, detritus and organisms, the last divided by size into two main categories. The microfauna includes archeans, bacteria and protoctists. These include a number of primary producers at the base of the food web in shallow waters, and have a role in interstitial sulfur chemistry and oxygenation of sediment. The meiofauna includes sediment-living forms between 0.1 and 1 mm in size, and is a major component of seabed ecosystems, particularly in the deep sea. Four phyla contain only marine meiofauna: the Gastrotricha, Gnathostomulida, Kinorhyncha and Loricifera. Nematodes are typically the most numerous component, with harpacticoid copepods and foraminiferans also important. High concentrations occur around the burrows of deposit-feeding mollusks and polychaetes, with high bacterial numbers and elevated nutrient flux. They play a key role in the flow of nutrients from the microfauna to larger organisms, of which several distinctive species exist in and on the seabed. Sea cucumbers, crinoids, polychaete worms, sea spiders, isopods and amphipods are most abundant. Some taxa are only found here. Pogonophorans occur mostly below 3 000 m and many of the known species are restricted to the sediments of deep ocean trenches (more than 6 000 m). Small-scale variation in food supply, such as the fortuitous appearance of a large animal carcass, may enhance spatial structure and provide opportunity for other species to colonize. A micro-landscape of hills and valleys is created by the burrowing and fecal mounds of echiurid and polychaete worms. Wh 120 WORLD ATLAS OF BIODIVERSITY Coral reefs develop within the photic zone because of the symbiotic relationship between some coral species and zooxanthellae. trenches are formed as a consequence of plate tectonic processes where sectors of ex- panding ocean floor are compressed against an unyielding continental mass or island arc, resulting in the crust buckling downwards (subducting) and being destroyed within the hot interior of the Earth. CLASSIFYING THE MARINE BIOSPHERE Much less effort has been devoted to catego- rizing and mapping units within the marine biosphere than to terrestrial environments. As on land, basic marine habitat types are commonly defined by some combination of structural, climatic and community features (e.g. ‘tropical coral reef’) but, probably in part reflecting issues of scale and resolution, few global maps of marine ecosystem types are available. The principal global scheme intended to classify the entire world ocean within an objective oceanographic system is based on long-term data on sea surface color {obtained by the CZCS radiometer carried by the Nimbus orbiting satellite during 1978- 86}’*. These data reflect chlorophyll concen- trations and provide a basis for estimating primary production rates, and changes over time, on a 1-degree grid. These values, together with numerous other data sets, have been used as the basis of a classification of the world ocean into four ecological domains and 56 biogeochemical provinces. Taking a less quantitative approach, some 64 large marine ecosystem (LME] units have been defined*°, with a view to improving planning and management. These are large regions of marine space, some 200 000 km’ or larger, that extend from river basins and estuaries out to the seaward boundary of the conti- nental shelf, with distinct combinations of bathymetry, hydrography and production. With similar aims, the conservation organi- zation WWF has distinguished 43 large marine ecoregions, based broadly on oceano- graphic and community features, and also intended to be globally representative’. THE BASIS OF LIFE IN THE SEAS In the sea, as on the land, photosynthesis is the driving force behind maintenance of life. Because photosynthesis in nature depends on sunlight, with few exceptions primary pro- ductivity is confined to those parts of the ocean that are sunlit. Water absorbs sunlight strongly so that light intensity decreases rapidly with increasing depth. Red wave- lengths are most rapidly absorbed, except where turbidity is high, while blue-green wavelengths penetrate the deepest. Even in the clearest waters the latter are completely absorbed by around 1 km depth, this marking the extreme limit of the so-called photic zone. Photosynthesis is thus limited to the contin- ental shelf area and the first few hundred meters of surface waters {and often much less] of the open ocean which together make up avery small proportion of the total volume of the oceans. The portion of the photic zone where sunlight is strong enough to support appreciable amounts of photosynthesis is the euphotic zone. Virtually all other marine organisms, including those of the unlit middle depths and the deep sea, are dependent ultimately on growth of primary producers in areas that may be widely distant from them in time and space. The most important exceptions to this are the archeans and bacteria living around hydrothermal vents associated with rift zones in the ocean floor. The water here can be 10°C warmer than in adjacent areas and these microorganisms are able to grow using hydro- gen sulfide gas emitted at the vents as an energy source, and they in turn are used by other organisms. With some exceptions primary production per unit area tends to be lower in marine environments than in terrestrial ones, espec- ially if highly managed terrestrial agricultural systems are considered. This is because over the vast majority of the ocean the euphotic zone is far distant from the lithosphere. The latter provides essential nutrients for life, and these therefore have to be transported to the euphotic zone to allow life processes to continue. There is also usually a steady loss of nutrients and organic compounds from the euphotic zone, owing to the sinking of particles and bodies into the dark regions of the sea where no photosynthesis can take place. Continued productivity in the open sea is contingent on the replacement of the lost nutrients. The latter may originate either on land, in the form of river outflow, or from mar- ine sediments. Their replacement in the pelagic euphotic zone is dependent on the mixing or vertical movement of the water column. At latitudes higher than 40°, winter mixing allows replenishment of the euphotic zone, partic- ularly in continental shelf areas. However, in permanently stratified subtropical and tropical oceanic waters there is little vertical mixing, and therefore little influx of new nutrients. Productivity in these areas is correspondingly low. In zones of upwelling, surface waters are regularly replaced by nutrient-rich bottom waters and very high levels of productivity can be achieved, at least seasonally. Until the 1980s it had been believed that photosynthesis in the pelagic ocean was car- ried out only by single-celled phytoplankton, between 1 and 100 microns in diameter (1 micron = 0.001 mm), and also that vast expanses of open ocean where phytoplankton could not be detected were, in terms of productivity, the marine equivalent of deserts. New observational techniques have since revealed the presence in great abundance of exceptionally small and previously unknown organisms, collectively termed picoplankton. These appear to be predominantly photo- synthesizing unicellular cyanobacteria 0.6-1 micron in size, such as Prochlorococcus. Because of their extraordinary abundance {some 100 million cells may be present in 1 liter], and despite their minute size, these organisms play a crucial role in the produc- tivity of open ocean waters’ and have led to marked upward revisions in estimates of overall marine productivity. The major role played by microscopic organisms in marine productivity has a number of important implications for marine ecology. Although much remains unknown about the structure and dynamics of pelagic food webs, it seems that a high proportion of marine primary production is used directly by microscopic organisms (both autotrophic and heterotrophic) and cycled back into non- living forms (dissolved carbon dioxide and organic carbon) rather than supporting populations of larger organisms. Oceanic primary producers divert a high proportion of their energy into reproduction rather than accumulating biomass, in con- trast to terrestrial primary producers (plants). As a result of this, average standing biomass per unit area in the oceans has been esti- mated at around one-thousandth that on land. Population turnover of oceanic primary pro- ducers is also several orders of magnitude higher than turnover of major terrestrial pri- mary producers [plants]. The small size and rapid turnover of oceanic primary producers mean that there are no organisms directly analogous to the woody plants that so enrich terrestrial environments by providing structurally com- plex habitats for other organisms. The nearest Marine biodiversity 121 SS ES Ve ui 122 WORLD ATLAS OF BIODIVERSITY I a a Table 6.3 Marine diversity by phylum Notes: Strictly marine groups shown in bold. Estimates are from a variety of sources and in some cases, e.g. Mollusca, differ in detail from those in Table 2.1. equivalents are the large brown algae known as kelp [phylum Phaeophyta), whose structure is less complex and which are much more narrowly distributed. Structurally complex habitats may in contrast be created by animals, particularly corals (phylum Cnidaria} and, to a lesser extent, sponges [phylum Porifera], mollusks and serpulid worms (phylum Annelida]. BIOLOGICAL DIVERSITY IN THE SEAS It is well known that diversity at higher taxonomic levels (phyla and classes) is much greater in the sea than on land or in freshwater. Of the 82 or so eukaryote phyla currently recognized (see Chapter 2], around 60 have marine representatives compared with around 40 found in freshwater and 40 on land. Amongst animals the preponderance is even higher, with 36 out of 37 phyla having marine representatives (Table 6.3). Some 23 eukaryote phyla, of which 18 are animal phyla, are confined to marine environments. Most of these are relatively obscure and comprise few species. The major exceptions are the Echinodermata, of which some 7000 species are known, and the Foraminifera, with around 4000 known, ex- tant species. A number of other important phyla including the coelenterates (Cnidaria), sponges (Porifera) and brown and red algae (Phaeophyta and Rhodophyta, respectively) Archaea 2 phyla ? Ectoprocta Ectoprocts 5 000 Phoronida Phoronid worms 16 Bacteria 12 phyla 2 Brachiopoda Lamp shells 350 Mollusca Mollusks 775 000 Eukaryota Protoctista 27 phyla, including: Priapulida Priapulids 8 Chlorophyta Green algae 7000 Sipuncula Sipunculans 150 Phaeophyta Brown algae 1 500 Echiura Echiurids 140 Rhodophyta Red algae 4000 Annelida Annelid worms 12 000 others incl. 23 000 Tardigrada Water bears few Foraminifera Chelicerata Chelicerates 1000 Mandibulata Mandibulate Animalia arthropods few Placozoa 1 Crustacea Crustaceans 38 000 Porifera Sponges 10 000 Pogonophora Beard worms 120 Cnidaria Coelenterates 10000 Bryozoa Bryozoans 4000 Ctenophora Comb jellies 90 Echinodermata Echinoderms 7000 Platyhelminthes Flatworms 15 000 Chaetognatha = Arrow worms 70 Nemertina Nemertines 750 Hemichordata Hemichordates 100 Gnathostomulida Gnathostomulids 80 Urochordata Tunicates and Rhombozoa Rhombozoans 65 ascidians 2 000 Orthonectida Orthonectids 20 Cephalochordata Lancelets 23 Gastrotricha Gastrotrichs 400 Craniata Craniates 15 000 Rotifera Rotifers 50 Kinorhyncha Kinorhynchs 100 Fungi 3 phyla 500 Loricifera Loriciferans 10 Acanthocephala Spiny-headed worms 600 Plantae Anthophyta 50 Entoprocta Entoprocts 170 Nematoda Nematodes 12 000 Nematomorpha Horsehair worms <240 Total ca 250 000 are very largely marine, each with only a small number of non-marine (usually fresh- water] species. The reason for this predominance of marine higher taxa (particularly amongst animals) is believed to be because most of the fundamental patterns of organization and body plan, i.e. the different basic kinds of organism that are distinguished as phyla, originated in the sea and remain there, but only a subset of them has spread to the land and into freshwaters. It is noteworthy that only a third or so of marine phyla are found in the pelagic realm, the remainder being confined to sea bottom (benthic) areas - the habitat where eukaryotic organisms are believed to have evolved. In contrast, known species diversity in the sea is much lower than on land - some 250 000 species of marine organisms are currently known, compared with more than 1.5 million terrestrial ones. Much of this difference is because of the large number of described terrestrial arthropods, for which there is no marine equivalent. Amongst fishes, almost as many freshwater species as marine are known, despite the fact that freshwater habitats account for only around one ten-thousandth of the volume of marine ones. Similarly, the most diverse known marine habitats (coral reefs] are far less diverse in terms of species number than the moist tropical forests that are often taken as their terrestrial counterparts. The apparent lower total species diversity of the marine biosphere is likely in part at least to be a result of the physical characteristics of water, particularly its high heat capacity and its ability to mix. Because of these, marine environments (particularly deep-water ones) tend to show much less variation in time and space in their physical characteristics than terrestrial ones. This lack of physical variation seems to result in a similar lack of ecological variation over wide areas. In contrast to terrestrial faunas, where one phylum - the Mandibulata [insects and relatives) - vastly outnumbers all others in terms of known species, marine species are much more evenly distributed across higher taxa. The largest marine phyla - Mollusca and a TI TIE GS ET Marine biodiversity 123 Meh Myxini Hagfishes Cephalaspidomorphi Lampreys Elasmobranchii Sharks, skates and rays Holocephali Chimaeras Actinopterygii Ray-fin fishes Sarcopterygii Lobe-fin fishes Reptilia Reptiles Aves Birds Mammalia Mammals Crustacea - each comprise far fewer than 100 000 known marine species, in contrast with the Mandibulata, of which around 1 million terrestrial species have been identified to date. The only major eukaryote phyla [i.e. those with 10000 or more de- scribed species) that are believed to have comparable levels of diversity both on land and in the sea are the Platyhelminthes, Nematoda, Mollusca and Craniata. As on land, vertebrates are by far the best-known group of marine organisms. Of the 50 000 or so described extant species, more than 15 000 may be considered marine [Table 6.4), the overwhelming majority of which are fishes (Table 6.5) and a few tetrapods [Table 6.6). Marked latitudinal gradients in species richness have been described in a number of groups, for example in benthic isopods and mollusks’, and it is generally true that coastal waters within the tropics tend, in parallel with terrestrial environments, to be richer in number of species than those at higher latitudes (although many exceptions to this are known, such as penguins, pinnipeds and auks). Although marine biogeography had advanced relatively little since the early 20th century, improved distribution data and more robust methods have recently been developed. These have been applied to definition of coastal biogeographic regions’, and important areas for marine biodiversity analogous to those delimited on land. For example, 18 centers of endemism within the world tropical reef zone have been identified using data on reef fish, coral, snails and lobsters (Map 6.1); ten centers were regarded as ‘hotspots’ because of their higher threat score”. Table 6.4 Diversity of craniates in the sea by class Note: Because of changing taxonomy, incomplete information and occupation of multiple habitat types, these estimates are indicative only. 124 WORLD ATLAS OF BIODIVERSITY See Sa a Myxiniformes Petromyzontiformes Chimaeriformes Heterodontiformes Orectolobiformes Carcharhiniformes Lamniformes Hexanchiformes Squaliformes Squatiniformes Pristiophoriformes Rajiformes Coelacanthiformes Acipenseriformes Salmoniformes Stomiiformes Ateleopodiformes Aulopiformes Myctophiformes Lampridiformes Polymixiiformes Ophidiiformes Gadiformes Batrachoidiformes Lophiiformes Mugiliformes Table 6.5 Hagfishes 1 6 43 Atheriniformes Silversides 8 47 139 Lampreys ] 6 9 Beloniformes Needlefishes, sauries, Chimaeras 3 6 31 flyingfishes, halfbeaks 3) 38 140* Bullhead sharks and Cyprinodontiformes Rivulines, killifishes, horn sharks 1 1 8 pupfishes, four-eyed fishes, Carpet sharks 7 14 31 poeciliids, goodeids 8 88 13 Ground sharks 7 47 207 Stephanoberyciformes Gibberfishes, pricklefishes, Mackerel sharks 7 10 16 whalefishes, hairyfish, Cow sharks 2 4 5 tapetails 9 28 86 Dogfishes and Beryciformes Fangtooths, spinyfins, sleeper sharks 4 23 74, lanterneyefishes, roughies, Angel sharks i 1 12 pinecone fishes, squirrelfishes 7 28 123 Saw sharks 1 2 5 Zeiformes Dories, boarfishes, Rays 12 62 432* oreos, parazen 6 20 39 Coelacanths 1 1 2 Gasterosteiformes Pipefishes, seahorses, Sturgeons 2 6 12 sticklebacks, sandeels, Salmonids 1 11 21 seamoths, snipefishes, Lightfishes, hatchetfishes, shrimpfishes, trumpetfishes 1 71 238* barbeled dragonfishes 4 51 321 Synbranchiformes Swamp-eels 3 12 3 Jellynose fishes 1 4 12 Scorpaeniformes Gurnards, scorpionfishes, Greeneyes, pearleyes, velvetfishes, flatheads, waryfishes, sablefishes, greenlings, lizardfishes, sculpins, oilfishes, poachers, barracudinas, lancetfishes 13 42 219 snailfishes, lumpfishes 25 266 1 219* Lanternfishes 2 35 241 Perciformes Perches, basses, sunfishes, Oarfishes, ribbonfishes, whitings, remoras, jacks, crestfishes, opahs I 12 19 dolphinfishes, snappers, Beardfishes i | 5 grunts, damselfishes, Pearlfishes, cusk-eels, dragonfishes, wrasses, brotulas ~ 5 92 350* butterflyfishes, etc. 148 1496 7371* Cods, hakes, rattails 12 85 481 Pleuronectiformes Plaice, flounders, soles 1 123 564* Toadfishes 1 19 64* Tetraodontiformes Triggerfishes, Anglerfishes, goosefishes, puffers, boxfishes, frogfishes, batfishes, filefishes, molas 9 100 327* seadevils 16 65 297 Mullets 1 17 65* Total (conservative working estimate] ca 14 000 Algae Diversity of fishes in the seas by order Note: Strictly marine orders in bold; other orders that are mainly marine (more than 50% of species) marked with an asterisk*. Source: After Nelson”” The macro-algae are superficially plant-like protoctists that lack the vascular tissue used by higher plants to transport water and nutrients. They are almost exclusively aquatic; three of the four principal groups consisting of large-sized species are mainly marine in occurrence. These three, the green, brown and red algae (‘seaweeds’), are all cosmo- politan in distribution and occur in a range of environments, although some constituent families have somewhat restricted ranges. There are more marine species of red algae (Rhodophyta) - around 4000 - than the greens (Chlorophyta, ca 1000) and browns (Phaeophyta, ca 1 500) combined. As with pinnipeds and seabirds, the cold and cool temperate regions of the world appear to be surprisingly rich in species. On present incomplete information, the region ———— around Japan (northwest Pacific], the North Atlantic, and the tropical and subtropical western Atlantic hold the most species of marine algae. Southern Australia is not so species rich but appears to have the highest proportion of endemics. There are few species of larger algae in regions of cold- water upwelling; small isolated islands and polar regions also have few species. In contrast, coral reefs support a unique and generally diverse algal flora that includes many crustose coralline algae (more species of which are likely to be discovered). Mangrove areas also support a well-defined algal vegetation. Sandy coastlines hold few species of large algae and often form barriers to seaweed dispersal. Marine fishes Fishes are considered a paraphyletic group (see Chapter 2): that is, all living species are thought to share a common ancestor but to share this ancestor with another group [in this case, the tetrapods) not categorized as fishes. Apart from some 50 or so species of generally parasitic lampreys and hagfishes in the superclass Agnatha, fishes are divided into two unequal-sized groups, the cartilaginous fishes (Chondrichthyes) including chimaeras (class Holocephali) and sharks and rays (class Elasmobranchii), and the bony fishes (Osteichthyes}, including the ‘typical’ ray- finned fishes (class Actinopterygii) and the lobe-finned coelacanths and lungfishes (class Sarcopterygii). Some 60 percent of all known living fish species [i.e. around 14 000 species) occur in marine habitats. They range in size from an 8 mm-long goby Trimmatom nanus in the Indian Ocean to the 15-m whale shark Rhincodon typus, respectively the smallest and largest of all fish species, and occur in virtually all habitats, from shallow inshore waters to the abyssal depths. The elasmobranchs (sharks, skates and rays) are an overwhelmingly marine group with around 850 living species in ten orders. Although far less diverse than the bony fishes, the cartilaginous fishes include many of the largest fish species, anumber of which are top predators in marine ecosystems. There tend to be more shark species at lower latitudes, but at family level richness tends to be higher on the edge of the tropics (Map 6.2]. The bony fishes are a remarkably diverse group, with an enormous range of morphological, physio- logical and behavioral adaptations. Of the 26 orders of bony fishes with marine represen- tatives (Table 6.5), by far the largest and most diversified is the Perciformes. This is the largest of all vertebrate orders and dominates vertebrate life in the ocean, as well as being the dominant fish group in many tropical and subtropical freshwaters. As with other groups of organisms, the majority of fishes in the sea are strictly marine, occurring only in salt water. A Reptilia Chelonia Dermochelyidae Cheloniidae Elapidae Acrochordidae Iguanidae Anatidae Scolopacidae Laridae Squamata Aves Anseriformes Ciconiiformes Phaethontidae Sulidae Phalacrocoracidae Pelecanidae Fregatidae Spheniscidae Procellariidae Balaenidae Balaenopteridae Eschrichtiidae Neobalaenidae Delphinidae Monodontidae Phocoenidae Phystereidae Platanistidae Ziphiidae Trichechidae Dugongidae Mustelidae Odobenidae Otariidae Phocidae Mammalia Cetacea Sirenia Carnivora Marine biodiversity 125 acer Table 6.6 Marine tetrapod diversity Notes: Birds follow the list of seabirds recognized in Croxall et al.” with the additional inclusion of four eider ducks and three steamer ducks in the family Anatidae and the red phalarope Phalaropus fulicaria {[Scolopacidae] Monroe™. Figures in inland. Strictly marine families shown in bold. Leathery turtle 1 Sea turtles 6 Sea snakes and sea kraits 59 File snakes 1 Iguanas i 7 1 Gulls, terns, skuas, auks, skimmers 120 (13) Tropicbirds 3 Gannets and boobies 9 Cormorants andshags _—-36 (2] Pelicans 2 Frigatebirds 5 Penguins 17 Petrels, albatrosses, shearwaters 115 Right whales 3 Rorquals 6 Gray whale i Pygmy right whale 1 Dolphins 32 Beluga and narwhal 2 Porpoises 6 Sperm whales 2 River dolphins 1 Beaked whales 19 Manatees 1 Dugong 1 Otters and weasels 2 Walrus i Eared seals 14 Earless seals 17 Taxonomy follows Sibley and parentheses indicate species that breed largely or entirely Niyl 12 WORLD ATLAS OF BIODIVERSITY ne ee Map 6.1 Coral reef hotspots The location of 18 areas defined by high endemism in reef fishes, corals, snails and lobsters, including the ten areas identified as ‘hotspots’ on the basis of high threat score Source; Adapted from Roberts” Endemic-rich areas cane Endemic-rich area at higher risk Endemic-rich area proportion, however, may also occur in inland waters, often passing a particular part of their life cycle there. Species that spend most of their life in marine waters but ascend rivers to breed, such as many salmonids {family Salmonidae, order Salmoniformes) and sturgeons (family Acipenseridae, order Acipenseriformes], are referred to as anad- romous. Those that breed at sea but spend their lives otherwise in freshwater, such as most eels in the family Anguillidae (order Anguiliformes), are referred to as catadro- mous. Species with a wide salinity tolerance that may occur in marine, brackish and fresh waters (e.g. some sawfishes, family Pristidae, order Rajiformes) are referred to as euryhaline while those with narrow tolerances, be they to marine, brackish or fresh water, are referred to as stenohaline. Reptiles Present-day diversity of reptiles in the seas is low. One important reason for this appears to be that modern reptilian kidneys cannot tolerate high salinities and thus the only reptiles that have adapted to marine environ- ments are those which have developed specialized salt-excreting glands. The most thoroughly marine reptiles are undoubtedly the sea snakes in the subfamily Hydrophiinae (family Elapidae]. These spend their entire lives in the sea, giving birth to live young there. Although largely air-breathing like other reptiles, they can also absorb some oxygen directly from sea water and are thus able to remain submerged for long periods. Around 50 species are known, widely distributed in tropical parts of the Indo-Pacific region. In addition the little file snake Acrochordus granulatus [family Acrochordidae), from northern Australia and Southeast Asia is also entirely aquatic, but occurs in brackish estu- aries as well as sea water. Five species of sea krait in the subfamily Laticaudinae are also largely marine, feeding mainly on eels. However they return to land to breed, generally on small tropical islands. They, too, are confined to the Indo-Pacific region. One species of lizard, the Galapagos marine iguana Amblyrhynchus cristatus (family Iguanidae], feeds underwater on mar- ine algae but spends a considerable proportion of time on land. Several other reptile species regularly enter sea water, most notably a number of homalopsine mangrove snakes (family Colubridae) from the Indo-Pacific and the estuarine crocodile Crocodylus porosus (family Crocodylidae) from the same region. Undoubtedly the most prominent group of marine reptiles is the sea turtles, comprising the leathery turtle Dermochelys coriacea in the family Dermochelyidae and six members of the family Cheloniidae. All species are large {ranging from 70-centimeter (cm) adult carapace length in Lepidochelys kempii to, exceptionally, 250 cm in Dermochelys coriacea) and most are widely distributed in tropical and subtropical waters (Map 6.3). Sea turtles are almost completely marine; only the females emerge to nest on land, mostly within the tropics. One species, the loggerhead Caretta caretta, nests largely in temperate areas of the northern hemisphere. Sea turtles typically have a long period to maturity [often up to 25 years in the case of the green turtle Chelonia mydas) and a long lifespan. Females often nest only every two or three years. They Marine biodiversity 127 12 WORLD ATLAS OF BIODIVERSITY ee ne Map 6.2 Shark family diversity | Diversity in sharks, based on the distribution of all 30 families plotted as a density surface. Most sharks are coastal in occurrence and for illustration purposes the family density is shown within a band extending out 400 km from the coastline (slightly further than the 200 nautical-mile EEZ limit). Level of diversity High Source: Prepared by UNEP-WCMC; family distributions aggregated from the species fenge maps published by Compagno’ Low habitually return to the same nesting beaches, sometimes undergoing protracted migrations from feeding grounds. They may lay two or three clutches in a season, sometimes consis- ting of more than 100 eggs each, depending on the species. Nest, hatchling and juvenile mortality are often high. Birds Defining marine birds, or seabirds, is somewhat more problematic than defining marine species in other groups. All birds breed in terrestrial habitats, but a large number [almost all of them non-passerines) obtain all or much of their food from aquatic or littoral habitats. Some of these, including all frigatebirds [(Fregatidae), tropicbirds (Phaethontidae}, gannets and boobies (Sulidae}, penguins (Spheniscidae] and petrels, alba- trosses and shearwaters (Procellariidae] are indisputably marine, in that they obtain all their food from marine habitats, almost invariably breeding along coastlines and spending most or all of their time when not breeding out at sea. Many others, however, have less clear-cut habits. Some, such as a number of cormorants and shags (Phalacrocoracidae) have both resident in- land and marine populations. Others, such as a number of gulls and terns (Laridae) and some ducks and geese [(Anatidae], may breed inland but spend the rest of the year living in coastal areas or out at sea. Yet others, such as sandpipers (Scolopacidae) and other waders, typically feed in littoral or intertidal habitats rather than in the sea itself; many of these species also occur inland. Adopting a somewhat arbitrary division, and excluding all wading birds with the exception of the red phalarope Phalaropus fulicaria (a truly pelagic species outside the breeding season), over 300 species of birds can be considered wholly or largely marine. In common with pinnipeds, seabirds show a latitudinal distribution in which diversity is much higher at higher latitudes [temperate and polar regions) than it is in the tropics. Two thirds of all seabirds are confined as breeding species to these latitudes, compared with only 7 percent that are exclusively tropical. This stands in sharp contrast to the pattern found in most major terrestrial groups [see Chapter 5) and many marine groups such as sea turtles, mangroves and reef-building corals (see, respectively, Maps 6.3, 6.4, 6.6] in which species diversity increases dramatically with decreasing latitude. Diversity is also markedly higher in the southern than in the northern hemisphere, with over half of all seabird species breeding in southern temperate and polar latitudes. Dominance of this region is even more marked in the Procellariidae, the family with the greatest number of truly marine species, in which over 60 percent of species breed at these latitudes and half are confined to it (Table 6.7). Marine biodiversity 129 aaenee ee reer reer ee ee TE TE ET Ty My Table 6.7 Regional distribution of breeding in seabirds Notes: Several species breed in more than one latitudinal band so that overall totals exceed actual number of seabirds. Numbers in parentheses indicate approximate number confined to the region. Northern temperate and polar 79.5 24 (17) Northern tropical 75.2 15 (5) Southern tropical 78.1 25 (8) Southern temperate and polar 130.1 70 (58) 50 (35) 50 (36) 27 (0) 19 (0) 61 (5) 30 (4) 26(5) 81 (17) 124 (88) 34 (15) 56 (41} 160 (114) 1309 WORLD ATLAS OF BIODIVERSITY PR a Map 6.3 Marine turtle diversity An overview of marine turtle diversity, represented by the number of turtle species nesting in any one area, Each symbol shows the location of a nesting site or area, colored according to the number of species present, up toa maximum of five species in some parts of the tropics. Source: Prepared using spatial data from a GIS database of marine turtle nesting beaches maintained at UNEP-WCMC Number of species nesting in area SS Mammals Wholly aquatic mammals (those that never normally emerge on to land) are confined to two orders, the Cetacea and the Sirenia. The Cetacea comprises some 78 species, all except five marine, distributed throughout the world’s seas. They include the largest living animals - the rorquals in the family Balaenopteridae. All cetaceans are carniv- orous; the baleen or whalebone whales (families Balaenidae, Balaenopteridae, Esch- richtiidae and Neobalaenidae) are filter feeders, feeding on organisms several orders of magnitude smaller than they are. Of four living members of the order Sirenia, only one - the dugong Dugong dugon - is exclusively marine, occurring widely in coastal waters of the Indo-Pacific. One other, the Caribbean manatee Trichechus manatus, is found in both marine and inland waters while the other two (the Amazonian manatee Trichechus inunguis and West African manatee Trichechus senegalensis] enter coastal waters marginally if at all. One other species, the very large Steller’s sea cow Hydrodamalis gigas, survived in waters around Bering and Copper Islands in the North Pacific until the early 18th century. All sirenians are herbivores; marine populations feed mainly on seagrasses. The remaining marine mammals are all included in the order Carnivora. Two New World otters in the family Mustelidae, the sea otter Enhydra lutris from the north temper- ate Pacific coast and the marine otter Lutra felina from the south temperate Pacific coast, feed very largely or exclusively in marine waters; other otter species may frequent coastal areas but are predominantly inland ES SS a TR a SS Sa Hy Mis - water animals. Members of the three pinni- ped families Odobenidae (the walrus}, Otariidae {eared seals) and Phocidae [earless seals) are all largely aquatic, emerging on land to breed and rest, particularly when molting; all are marine with the exception of one or two species of Phocidae [the Baikal seal Phoca sibirica and, if the Caspian is regarded as a lake rather than a sea, the Caspian seal Phoca caspica). One member of the family Phocidae, the Caribbean monk seal Monachus tropicalis, has become extinct this century. All species are carnivorous. In contrast to most terrestrial mammal families, pinnipeds are considerably more diverse and more abundant at higher rather than lower latitudes. Of the 32 extant or recently extant species, only five occur within the tropics (two marginally). Part of the explanation for this undoubtedly lies in the greater availability of suitable habitat at higher latitudes: as noted above, 70 percent of continental shelf waters and just over 60 percent of the world’s marine area are found outside the tropics. However, this in itself is unlikely to account for the entire difference. It is probable that the greater productivity of shelf waters at high latitudes, discussed above, and of upwelling areas at mid- latitudes (e.g. the Benguela current off the western coast of South Africa and the Humboldt current off Chile and Peru) plays a major part. The isolated character of many island breeding sites, such as the Galapagos group, in temperate and sub-polar parts of the southern hemisphere may also have en- couraged speciation of pinnipeds here. Marine biodiversity 131 MN, 1322 WORLD ATLAS OF BIODIVERSITY a a a Table 6.8 Diversity of mangroves Note: Where two figures are given, second figure indicates number of hybrids. Families composed solely of mangrove species are shown in bold Source: Adapted from Duke’ and Spalding et al ug COASTAL AND SHALLOW WATER COMMUNITIES Mangroves | Mangrove woodland is indeed a truly hybrid terrestrial/marine ecosystem, unique in that | terrestrial organisms can occur in the canopy and marine species at the base''’. Mangroves, | or mangals, are a diverse collection of shrubs and trees [including ferns and palms] which live in or adjacent to the intertidal zone and are thus unusual amongst vascular plants in that they are adapted to having their roots at least periodically submerged in sea water. A wide variety of organisms is associated with mangroves including a number of epiphytic, parasitic and climbing plants, and large numbers of crustaceans, mollusks, fishes and birds”. Mangrove species are generally divided into those found only in mangrove habitats and those that may also be found elsewhere but which are nevertheless an important component of mangrove habitats. Both groups come from a wide range of families. The former includes around 62 species and seven hybrids in some 22 genera (Table 6.8). The appearance of mangroves is far from uniform: they vary from closed forests Filicopsida Plumbaginales Theales Malvales Ebenales Primulales Fabales Myrtales Rhizophorales Euphorbiales Sapindales Lamiales Scrophuliariales Rubiales Arecales Adiantaceae 3 Plumbaginaceae 2 Pelliciceraceae i Bombacaceae 2 Sterculiaceae 3 Ebenaceae 1 Myrsinaceae 2 Leguminosae 2 Combretaceae 4+] Lythraceae ] Myrtaceae 1 Sonneratiaceae 6+3 Rhizophoraceae 17+2 Euphorbiaceae 2 Meliaceae 2+1 Avicenniaceae 8 Acanthaceae 2 Bignoniaceae 1 Rubiaceae 1 Palmae i 40-50 m high to widely separated clumps of stunted shrubs less than 1 m high’. They are only able to grow on shores that are shelter- ed from wave action, and are particularly well developed in estuarine and deltaic areas; they may also extend some distance upstream along the banks of rivers, e.g. some 300 km up the Fly River in Papua New Guinea. Mangrove communities are largely re- stricted to the tropics between 30°N and 30°S, with extensions beyond this to the north in Bermuda and Japan, and to the south in Australia and New Zealand“ (Map 6.4). They occur over a larger geographical area than coral reefs {see below] and, unlike reefs, are well developed along the western coasts of the Americas and Africa. They have a more restricted distribution than coral reefs in the South Pacific. There are two main centers of diversity. The eastern group occurs in the Indo-Pacific (the Indian Ocean and western part of the Pacific Ocean) and is the most species rich’ "*. The western group is centered around the Caribbean and includes mangrove com- munities along the west coast of the Americas and Africa. Global mangrove area is believed to slightly exceed 180000 km’, divided regionally as shown in Table 6.9. Mangroves occur in over 100 countries [including dependent terri- tories) but exist in very small areas in many of these. Four countries (Indonesia, Brazil, Australia and Nigeria] between them account for over 40 percent of the world’s mangrove area, and Indonesia alone possesses nearly one quarter of the global mangrove area”. Although it is known that mangrove eco- systems in most parts of the world have been extensively degraded and cleared, it is difficult to obtain reliable data on the global extent of mangrove loss over time. Mangroves by their very nature occupy highly dynamic and unstable environments so that even without human action the location and extent of mangrove cover would be constantly chang- ing. One assessment” suggested that more than 50 percent of the world’s mangrove forest cover had been destroyed. Mangroves stabilize shorelines and de- crease coastal erosion by reducing the energy of waves and currents and by holding the bottom sediment in place with their roots. They also act as windbreaks and provide protection from coastal storms. They are generally highly productive ecosystems and are important habitats for crustaceans, shellfish and finfishes. Most of the larger commercial penaeid shrimps are mangrove dependent; these and other species are harvested both on a subsistence basis and commercially, and may provide a major source of income in some countries. As well as providing habitat for adults of many species of finfish and invertebrates, Mangroves serve as spawning and nursery areas for many others, often of major economic importance. The wood provides building material, used locally in houses, as fence poles and to build fish traps, and is also harvested on a large scale for production of pulp and particle board. In many areas mangroves are also an important source of fuel, as firewood and charcoal. Mangrove foliage may provide an important source of fodder for domestic livestock in some coun- tries, particularly during dry seasons when other sources of greenery are in short supply. Salt marshes Salt marshes are coastal communities of rooted salt-tolerant (halophytic) plants of terrestrial origin, dominated by grasses, herbs and dwarf shrubs. They share many characteristics with mangroves but replace them geographically in higher latitudes, except for some overlap at the extremes of mangrove distribution (in the Gulf of Mexico, Japan, southern Australia and northern New Zealand). Globally, salt marshes are estimated to cover around 350000 km? in total'®. They tend to develop in sheltered areas of mud and silty sand that are flat and slow draining; as plant growth and sedimentation elevates the marshland, so the period of tidal submergence decreases and a network of creeks develops. The two genera which are most prominent as pioneer saltmarsh plants are Salicornia, or samphire, and Spartina, or cord grass. Species of Puccinella, Scirpus and Juncus are also common. Salt marshes are highly productive, around 2 500 g C per m’ per year. Occurring in highly seasonal latitudes, salt marshes rapidly take up and accumulate nutrients during the growing season. In the autumn, when plants die or become dormant, the uptake of nutri- ents is greatly diminished and dead organic matter may be transported out of the marsh. The physical features of salt marshes mean that pollutants are not easily or rapidly flushed out. Other threats include infilling, especially around the North Sea”, and in- creased erosion through channel dredging. Rocky shores Rocky shorelines are generally exposed to oceanic swells and extreme wave action (except for more sheltered fjordlands) but are topographically variable, occurring as wide platforms, steep cliffs or other formations, depending on local geology”. They provide a unique habitat for plant growth, with a stable substrate for attachment, and shallow well- lit water that tends to be turbulent and rich in nutrients. Consequently macro-algae communities, dominated by kelps (Laminaria, Ecklonia, Macrocystis) and fuccoids (Fucus, Ascophyllum) flourish, mainly in temperate regions but also in areas of the tropics where seasonal upwellings of cold water occur, such as Chile and the southern Arabian coast’. The net primary productivity of kelp forests is comparable to tropical rainforests, and Macrocystis pyrifera, the giant kelp, can attain growth of up to 45 cma day. Two physical factors - water movement and desiccation - have influenced the diversity of intertidal rocky shore species. Strong wave action has favored use of crevices, dense aggregations of individuals, and the evolution of strong attachment devices [algal holdfasts, South and Southeast Asia 75 000 Americas 49 000 West Africa 28 000 Australasia 19 000 East Africa and Middle East 10 000 Total 181 000 Marine biodiversity 133 RS I SS, Table 6.9 Current mangrove cover Source: Spalding et at."6, 42 Hil Map 6.4 Mangrove diversity The location of current mangrove forest, together with contours representing gradients of mangrove species richness. Note that graphic presentation at this scale enormously exaggerates actual forest area. Source: Reproduced with modification, 73 from Spalding 134 WORLD ATLAS OF BIODIVERSITY Mangrove locality and species richness e Mangrove forest High diversity Low diversity cementation of barnacles, the byssus threads of mussels and the adhesive feet of gastro- pods and echinoderms]. The need to avoid being washed off the rocks means that most organisms on rocky shores are sessile or have limited motile ability. Competition for space is therefore intense and organisms inhabiting rocky shores occupy well-defined zones. It also means that they can be particularly vulnerable to disturbances such as oil spills, especially if these occur during calm periods which extend the residence time of pollutants. However, because they are generally exposed and high-energy environments, rocky shores tend to be less vulnerable than most to pollution. These features also limit some destructive human activities, such as construction, responsible for degrading other marine habitats. Seagrasses Seagrasses are a mixed group of flowering plants (not true grasses) that are adapted to live submerged in shallow marine and estuar- ine environments at a wide range of latitudes. About 58 species are recognized by many authorities, in four families, all within the monocotyledons. They occur from the littoral region to depths of 50 or 60 m but are most abundant in the immediate sublittoral area. There are more species in the tropics than in the temperate zones, and of the 12 seagrass genera seven are confined to tropical seas and five to temperate seas”. Most seagrass species are similar in external morphology, with long thin leaves and an extensive rhizome root system which enables them to fasten to the substrate. A variety of substrates are occupied from sand and mud to granite rock, but the most extensive beds occur on soft substrates”. While species diversity [see Map 6.5) is highest in Southeast Asia (as with mangroves and corals) there are two important centers in temperate regions: in mainland Japan and in southwest Australia. Secondary centers of diversity include East Africa, the Red Sea and the Mediterranean. Although seagrasses themselves are not a diverse group, they support considerable diversity in some assoc- iated species, including an estimated 450 species of epiphytic algae”. In many areas seagrasses form extensive but simple communities, referred to as sea- grass beds or seagrass meadows. Seagrass beds have high productivity and contribute significantly to the total primary production of inshore waters. Seagrasses are particularly important in nutrient-poor (oligotrophic) waters because they can extract some nutri- ents from sediments, unlike other marine primary producers. Many commercially impor- tant species are dependent on seagrass beds, often as nursery habitat, providing shelter from predators and adverse sea conditions. These include mollusks [such as the queen conch Strombus gigas), shrimp, lobster, holo- thurians and many finfish {such as grunts, Haemulidae; rabbitfish, Siganidae; emperors, Lethrinidae; and snappers, Lutjanidae). A small but important number of threatened species depend on seagrasses, including sirenians, the green turtle Chelonia mydas and many species of seahorse (Syngnathidae). In addition to such direct values, seagrass beds also play an important role in binding sediments, providing some protection from Marine biodiversity 135 Map 6.5 Seagrass species diversity This first global map of seagrass diversity indicates the extent of seagrass habitat inventory sites) and shows contours of species richness. Diversity contours not shown for parts of West Africa because species inventory incomplete. Source: Preliminary plot, compiled using multiple sources, from a seagrass atlas in preparation at UNEP-WCMC. 136 WORLD ATLAS OF BIODIVERSITY Seagrass locality and species richness « % ® Seagrass High diversity Low diversity coastal erosion, and may help remove excess nutrients and toxins from coastal waters. A preliminary estimate” suggested that seagrass beds extend over some 600 000 km* globally, but a precise figure is not yet available. Although seagrasses can be highly dynamic ecosystems, with individual seagrass beds undergoing significant shifts in distri- bution over relatively short periods, there is considerable evidence that there have been major net losses over the past century. Extensive losses along the Atlantic coasts of Europe and North America in the 1930s followed disease caused by a marine slime mold; some evidence suggests that other environmental impacts (possibly increased turbidity in coastal waters) may have increas- ed susceptibility to disease. More recently, nutrient loading has continued to increase in coastal waters worldwide, often leading to enhanced growth of phytoplankton, epiphytes and macroalgae, all of which can out- compete seagrasses for available sunlight. Sedimentation also has a major impact, reducing the passage of light through the water column and physically smothering seagrass plants. To a lesser degree, toxic pollutants and physical disturbance from activities such as trawling and dredging have also played a role in seagrass losses”. Shallow tropical coral reefs The term ‘coral reef’ applies to a variety of calcium carbonate structures developed by stony corals. They are tropical shallow-water ecosystems, typically with high biodiversity and largely restricted to coastal seas between the latitudes of 30°N and 30°S*. They are most abundant in shallow, well-flushed marine environments characterized by clear, warm, low-nutrient waters that are of oceanic salinity”. There are two basic categories: shelf reefs, which form on the continental shelves of large land masses, and oceanic reefs, which are surrounded by deeper waters and are often associated with oceanic islands. Within these two categories there are a num- ber of reef types: fringing reefs, which grow close to shore; barrier reefs, which develop along the edge of a continental shelf or through land subsidence in deeper waters, and are separated from the mainland or island by a relatively deep, wide lagoon; and atolls, which are roughly circular reefs around a central lagoon and are typically found in oceanic waters, probably originating from the fringing reefs of long-submerged islands. Two other less clearly defined categories are patch reefs, which form on irregularities on shallow parts of the seabed, and bank reefs, which occur in deeper waters, both on continental shelf and in oceanic waters”. The global extent of coral reefs is not known with certainty. A recent estimate, derived by measuring the total reef extent in a comprehensive set of national maps, sug- gests a world total of 284 300 km’. This is equivalent to a little more than 1 percent of the total world ocean shelf area. New information and mapping techniques may increase the accuracy of such estimates, but the total is unlikely to exceed 300 000 km’. A regional breakdown is provided in Table 6.11. Although coral reefs occur in around 110 countries and territories, just five countries - Indonesia, Australia, Philippines, Marine biodiversity 137 138 WORLD ATLAS OF BIODIVERSITY er eT ee ee ee Table 6.10 Diversity of stony corals in the order Scleractinia Note: Includes only the reef- forming scleractinians with zooxanthellae 27 Source: Veron France {overseas departments and territories in the Indian Ocean and Pacific) and Papua New Guinea - account for over half of the global total. Five different orders within the phylum Cnidaria (coelenterates) include species with calcified skeletons, or ‘stony corals’. Many of these are reef building (hermatypic] and some are solitary (ahermatypic]. The great majority of hermatypic corals belong to the order Scleractinia, the true stony corals, although not all scleractinians are reef-builders. Together with sea anemones and sea fans, the Scleractinia make up the class Anthozoa. Most scleractinian coral polyps have symbiotic algae (zooxanthellae) within their tissues; these use the nitrates, phosphates and carbon dioxide produced by the coral, and through photosynthesis generate oxygen and organic compounds that provide much of the polyps’ nutrition. The zooxanthellae give corals their color and, because they photosynthesize, restrict the corals that contain them to the photic zone*. Corals without zooxanthellae typically do not form reefs and can exist in deeper colder waters {see deep-water reefs, below). Because of difficulties with synonymy and in defining species, it is difficult to estimate precisely how many extant species Acroporidae 4 262 Agaricidae 6 43 Astrocoenidae 4 13 Caryophylliidae 1 i Dendrophyllidae 4 14 Euphyllidae 5 14 Faviidae 24 125 Fungiidae 13 56 Meandriniidae 7 8 Merulinidae 5 12 Mussidae 13 50 Oculinidae 4 15 Pectiniidae 5 28 Pocilloporidae 3 30 Poritidae 5 92 Rhizangiidae 1 1 Siderastreidae 6 28 Trachyphyllidae | 1 of reef-building coral there are. A recent synthesis” deals with around 800 zooxan- thellate scleractinians (see Table 6.10); another” lists more than 1 300 scleractinians in all, and 260 calcified hydrozoans. Not all reefs are constructed primarily by corals. Within the red algae (phylum Rhodophyta) and the green algae [phylum Chlorophyta) in particular several genera grow heavily calcified encrustations which bind the reef framework and in places are the main contributors to shallow reef growth. Coral reefs are among the most productive and diverse of all natural ecosystems. The main center of diversity for reef-building corals is Southeast Asia, with an estimated minimum of 450 species found associated with reefs around the Philippines, Borneo, Sulawesi and associated islands. This area is part of a single, vast, Indo-West Pacific biogeographic province that extends from the Red Sea in the west to the Pitcairn Islands in the east. Many coral genera and a significant number of species are found throughout the region, although overall diversity in the pro- vince decreases on leaving this center. In the east of this region, the central and eastern Pacific forms a series of somewhat distinct subregions, characterized by a number of genera and species (particularly in Hawaii) not found further west. This area also shares many species with the Indo-West Pacific province but overall has much lower diversity than most of the latter. The Atlantic, including the Caribbean and the Gulf of Mexico, forms a distinct province with few species in common with the Indo-West Pacific. It is also very depauperate compared with most of the latter (Map 6.6). Deep-water reefs Beyond the coral reefs of shallow tropical waters, a few species of coral that lack zooxanthellae form deep-water reefs on hard substrates in high-current areas associated with topographic rises, such as ridges and pinnacles. The best known is Lophelia pertusa, a colonial coral that forms structures ranging from patches a few meters in width to reefs many hundreds of meters in size, at depths of 100-3 000 m and temperatures of 4-8°C. Although best known through trawling and oil exploration activities in the northeast Atlantic, these deep-water reefs are global in occurrence. The complex matrix of living and dead branches of Lophelia increases spatial heterogeneity above that of the surrounding seabed and provides a habitat for many species. Boring sponges, anemones, bryo- zoans, gorgonians, polychaetes, barnacles and bivalves occur in large numbers” and their diversity is comparable to some shallow- water tropical systems”. Although large aggregations of fish are associated with Lophelia reefs, and reef areas support higher catches than adjacent seabed”, the species diversity of fish and coral is much lower than in tropical coral reefs (some 23 species of fish have been recorded on Lophelia reefs in the northeast Atlantic). The reefs are delicate structures easily destroyed by demersal fishing gear which routinely operate to depths of 2000 m. The total destruction of some Norwegian reefs has already been documented and an esti- mated 30-50 percent of others have been damaged. Slow growth, in the region of 4-25 mm per year, severely limits the ability of Lophelia to recover. The trawl fishery for orange roughy Hoplostethus atlanticus and oreos Allocyttus niger on sea mounts south of Tasmania has been responsible for sub- stantial destruction of Solenosmilia variabilis reefs, with some reduced to more than 90 percent bare rocks”. If recovery ever occurs it will take hundreds of years. OCEANIC PELAGIC COMMUNITIES There is a fundamental distinction between the processes and patterns observed in open oceans, dominated by global winds and large- scale vertical and horizontal movement of water masses, and those observed nearer to coasts, where shelf bathymetry, coastal winds and local input of nutrients, pollutants and sediments generate a diversity of smaller- scale phenomena. The oceanic pelagic zone is dominated by the activity of plankton in the euphotic surface waters. Plankton are by definition drifting or weakly swimming organisms, comprising a Marine biodiversity 139 a a a ean ea eC TM i Indo-Pacific 261 200 Red Sea, Gulf of Aden 17 400 Arabian Sea, Persian Gulf 4 200 Indian Ocean 32 000 Southeast Asia 91 700 Pacific 115 900 Wider Atlantic 21 600 Caribbean 20 000 Atlantic (excl. Caribbean) 1 600 Eastern Pacific 1 600 Total 284 300 wide range of small to microscopic animals, protoctists and bacteria. Free-swimming pelagic organisms, predominantly fishes but also cetaceans and cephalopod mollusks (squid), are collectively termed nekton. These organisms, when adult, are predators of plankton or smaller nekton. They in turn - as vertically migrating fishes or larvae, and as dead organic material - provide food for deep- sea and benthic organisms. With few excep- tions, the only other food source for creatures in the aphotic zone is the ‘rain’ of organic matter, such as feces, molted crustacean exoskeletons, and a variety of other organic material derived from plankton in the surface waters of the ocean. Plankton and larger free- swimming organisms tend strongly to con- centrate along major circulation currents (gyres], contact zones and upwellings, and this can give rise to significant local variation in diversity. The marked vertical gradients within the pelagic zone - of light, temperature, pressure, nutrient availability and salinity - lead to vertical structuring of pelagic species assem- blages. Several zones based on changes in species composition with depth have been recognized, including epipelagic [usually taken as from the surface to a depth of 200- 250 m and including the euphotic zone]; mesopelagic, which underlies the epipelagic zone to a depth of 1000 m or so; and below this the bathypelagic which changes in a somewhat less well-defined fashion to abyssopelagic at around 2 500-2 700 m depth. Table 6.11 Coral reef area Source: Spalding et al”? Map 6.6 Coral diversity The location of coral reefs, together with contours representing gradients of species richness among reef-building scleractinian coral species. Note that graphic presentation at this scale enormously exaggerates actual reef area. Source: Reproduced by permission, with modification, from Veron at revised by Veron, pers. comm. November 2001 1440 WORLD ATLAS OF BIODIVERSITY Oe ee ee Coral locality and species richness Coral High diversity Low diversity These zones, however, tend to fluctuate in time and space. As well as seasonal changes in water characteristics, many components of the epipelagic and mesopelagic nekton under- go marked diel migrations [i.e. on a 24-hour cycle), ascending to surface waters at night to feed and descending, sometimes over 1 km, during the day. Many species of nekton, particularly cetaceans and larger fishes, are also highly migratory, ranging over enormous expanses of ocean in more or less regular and predictable patterns. It has generally been assumed that bio- mass in the pelagic zone everywhere below the euphotic zone is low. However, recent studies have indicated that biomass of tropical mesopelagic animals may be surprisingly high. Study of the mesopelagic fauna has been limited to date, as it requires the use of expensive high-seas research vessels; know- ledge of taxonomy, distribution and biology of most of the species concerned remains very incomplete’. One study* recognized around 160 fish genera in 30 families as important components of the fauna. Most species are small (less than 10 cm in length) and often bizarrely shaped. Estimates based on a variety of surveys carried out indicate that global biomass of this stock may be large: a figure of 650 million tons [some six to seven times total current marine fisheries landings) has been suggested, although this should be regarded with extreme circumspection'. From available data, it would appear that the mesopelagic biomass is greatest in the northern Indian Ocean, and particularly in the northern Arabian Sea, one of the five major upwelling zones. Surveys here indicated extremely high biomass (25-250 g per m’) in the Gulf of Aden and Gulf of Oman as well as off the western coastline of Pakistan. These figures are around an order of magnitude higher than those recorded elsewhere in the tropics, indicating either a great overestimate for the northern Indian Ocean or an under- estimate elsewhere. Alternatively this region genuinely is ten times as productive as the rest of the tropical ocean system. Although this appears as yet unresolved, it is nevertheless apparent that there is substantial global mesopelagic fish biomass anid that the Arabian Sea is particularly rich in these species. DEEP-SEA COMMUNITIES Approximately 51 percent of the Earth's surface is covered by ocean over 3 000 m in depth, so deep-sea communities are preva- “ ¢ ma “g | bY lent over a significant proportion of the planet. All deep-sea habitat is in the aphotic zone, well below the distance sunlight can pene- trate. As deeper and deeper levels are reached biomass falls exponentially”. Despite their enormous volume, the deep oceans appeared to be relatively simple ecosystems and to make little contribution to global species diversity, but discoveries during the past decade or so have shown that in some regions species diversity in the benthic community increases with increasing depth. This was revealed by novel sampling techniques, principally the epibenthic sled®. The rate of discovery of new species and the proportion of species currently known from only one sample both suggest that a great number remain to be discovered” ”. As with arthropods in tropical moist forests, Marine biodiversity 11 Cee EE ES , 1442 WORLD ATLAS OF BIODIVERSITY I estimates for the number of unknown species vary widely, with some suggestions that there may be as many as 10 million undescribed species in the deep sea”. Others consider that the true figure is more likely to be around 500 000”. Ocean trenches Ocean trenches are typically close to land masses and tend to have high rates of sedimentation, a significant amount of which is of organic origin and an important available food source for trench communities. Several trenches also underlie highly productive cold- water upwelling zones, the organic fallout from which contributes greatly to productivity there. The water within trenches generally Originates from the surrounding bottom water, which is derived from cold surface water at high polar latitudes and is relatively well oxygenated”. Trenches tend to be isolated linear systems that because of their seismic activity form a habitat that is unstable and unpredictable compared with the relative environmental stability of the adjacent abyssal plains”. Trench faunas are not rich in species but are often high in numbers of endemic species. There are some 25 genera restricted to the ultra-abyssal (hadal) zone, representing some 10-25 percent of the total number of genera present, and two known endemic hadal families: the Galatheanthemidae (Cnidaria) and Gigantapseudidae (Crustacea). The latter family contains a single species: Giganta- pseudes adactylus. The greatest number of endemic species known from a single trench is a Sample of 200 from the Kurile-Kamchatka trench; this may be compared with 10 endemic species known from the Ryukyu and Marianas trenches. Hydrothermal vents Hydrothermal vent communities were first discovered in 1977, at a depth of 2 500 m on the Galapagos Rift, but contribute to what might be one of Earth’s most archaic eco- systems. They are now known to be assoc- iated with almost all known areas of tectonic activity at various depths. These tectonic regions include ocean-floor spreading centers, subduction and fracture zones and back-arc basins’. Cold bottom water permeates through fissures in the ocean floor close to ocean floor spreading centers, becomes heated at great depths in the Earth’s crust and finds its way back to the surface through hydrothermal vents. The temperature of vent water varies greatly, from around 23°C in the Galapagos vents to around 350°C in the vents of the East Pacific Rise, and they may be rich in metalliferous brines and sulfide ions“. Most species live out of the main flow at temperatures of around 2°C, the ambient temperature of deep-sea water. The biomass of vent communities is usually high compared with other areas of similar depth, and dense colonies of tube worms, clams, mussels and limpets typically constitute the major components. Hydrothermal vent communities are of particular interest because they flourish in the dark at high pressures and low temperatures”, and unique because they are supported by chemolithoautotrophic archeans and bacteria, notably Thiomicrospira species (phylum Proteobacteria], which form dense microbial carpets in the rich hydrothermal fluid and derive their energy chiefly from oxidizing hydrogen sulfide‘’’. Many of the eukaryote vent species filter-feed on these microorganisms, whilst others rely on sym- biotic sulfur bacteria for energy”. The overall species diversity at vents is low compared with other deep-sea soft sediment areas”, but endemism is high. More than 20 new families or subfamilies, 50 new genera and nearly 160 new species have been recorded from vent environments, including brine and cold seep communities‘’“. Vent communities are separated by gaps of between 1 and 100 km, and although they may persist only for several years or decades, sites of vent activity move relatively slowly allowing dispersal of vent organisms”. Two further seep patterns are known. Cold sulfide and methane-enriched groundwater seeps occur near the base of the porous limestone of the Florida escarpment, as well as in the Gulf of Mexico and elsewhere. These support a community similar in taxonomic composition to the hydrothermal vents of the east Pacific. Tectonic subduction zone seeps are more diffuse and lower in temperature than hydrothermal vent seeps and are rich in dissolved methane. They are known to occur off Oregon and in the Guaymas basin in the Gulf of California, and cold seeps occur at a depth of 1000 m in Sagami Bay near Tokyo and in the subduction zones of the trenches off the east coast of Japan. HUMAN USE OF AND IMPACT ON THE OCEANS The seas provide many biological resources used by humans. A wide range of animal species, notably fishes, mollusks and crus- taceans, contribute to marine fisheries and these provide by far the most important source of wild protein, of particular impor- tance to many subsistence communities around the world. Marine algae are also an increasingly important foodstuff, notably in the Far East, with current annual world production of around 2 million metric tons. Marine organisms are also proving extremely fruitful sources of pharmaceuticals and other materials used in medicines. Relatively minor although locally important uses include exploitation of coastal resources for building materials (e.g. coral limestone, mangrove poles} and other industrial products (e.g. tannins from mangroves). Access to marine resources Is not equitably distributed amongst the world’s nations. Most obviously, some 39 states are landlocked, i.e. have no seaboard [five of these border the inland Caspian Sea}. Those that do have sea- boards show great variation in length of coastline, and area of territorial waters and exclusive economic zones (EEZs}, both abso- lutely and relative to their land areas. They also show great variation in their capacities to exploit marine resources, both on the high seas and within their territorial waters and EEZs. Human activities, directly and indirectly, are now the primary cause of changes to marine biodiversity. Approximately one third of the world’s human population lives in the coastal zone [within 60 km of the sea) and indications are that this proportion will rise during the 21st century. Pressures exerted by the human population on the marine biosphere are substantial and increasing. Most identified threats relate to coastal and inshore (continental shelf] areas. How- ever, threats to the oceanic realm are undoubtedly increasing: fisheries and their attendant physical effects, such as habitat alteration owing to dredging and trawling, have entered deeper continental slope waters having previously been largely confined to the epipelagic zone, and deep-water oil and gas mining is planned. Even abyssal and hadal areas are susceptible to human impact. A small, steady increase in abyssal temperature of 0.32°C in 35 years has been attributed to global climate change brought about by human activities. Ocean waste dumping and the potential for deep-water mining and mineral extraction are also causes for Marine biodiversity 143 Many species of fishes, mollusks and crustaceans provide humans with their largest source of wild protein. 144 WORLD ATLAS OF BIODIVERSITY eee Just five species of finfish, among them the Atlantic herring, account for a quarter of the global catch. concern, as are the changes in biomass and species composition in the waters above these regions”. The following five activities have been identified as the most important agents of present and potential change to marine bio- diversity at genetic, species and ecosystem levels®: e fisheries operations; e chemical pollution and eutrophication; e alteration of physical habitat; e invasions of exotic species; e global climate change. WORLD MARINE CAPTURE FISHERIES World marine capture fisheries have grown fivefold in the past half-century, with annual landings increasing from nearly 18 million metric tons between 1948 and 1952 to around 87 million metric tons during the period 1994- 97, with a fall caused mainly by El Nino in 1998 and a subsequent recovery in 1999. Marine capture fisheries made up just over 70 percent of current recorded global production of aquatic resources in the late 1990s, the remainder being accounted for by inland capture fisheries {see Chapter 7] and aquaculture“. With capture fisheries appar- ently remaining more or less stable, the increase in total marine production during the 1990s was due to a continuing increase in aquaculture production. Composition of marine fisheries Marine fisheries encompass a wide range of organisms, including algae, invertebrate ani- mals in various phyla and vertebrates including fishes (often termed finfishes in fisheries analy- sis), reptiles, mammals and birds [although by convention the last of these groups is not normally considered in fisheries analysis). The Food and Agriculture Organization of the United Nations (FAO) recognizes in total just under 1000 ‘species items’ (species, genera or families} that feature at least periodically in national catch statistics. However, globally important marine fisheries are confined to relatively few groups, with over 80 percent of landings by weight being finfishes and virtually all the remainder mollusks and crustaceans. In terms of major species groups, by far the most important are the herrings and anchovies in the order Clupeiformes, which in 1998 accounted for over 22 million metric tons, or around 25 percent of marine landings. These are followed by cod, hake and haddock (Gadiformes], and jacks and mullets (some Per- ciformes and Mugiliformes], with production of more than ? million and nearly 8 million metric tons respectively. The most important inver- tebrate group overall is cephalopod mollusks (squid, cuttlefish and octopus) of which some 3.4 million metric tons were reported landed. In terms of individual species, for several years during the 1990s just five species of finfish, anchoveta Engraulis ringens, Alaska pollock Theragra chalcogramma, Chilean jack mackerel Trachurus murphyi, Atlantic herring Clupea harengus and chub mackerel Scomber japonicus, together made up around one quarter of marine landings. Each accounted for over 2 million metric tons annually, and among several others around this level, Japanese anchovy Engraulis japonicus and Skipjack tuna Katsuwonus pelamis were increasingly important. In most years by far the most im- portant single species is the anchoveta Engraulis ringens, whose fishery loff the west coast of South America] was nearly 13 million metric tons in 1994, constituting by far the largest single-species fishery the world has ever seen; but just under 8 million in 1997 and less than 2 million in 1998. Distribution of marine fisheries The geographical distribution of marine fisheries is determined both by the distri- bution of harvestable fish stocks and by a range of complex socioeconomic factors. The former is largely determined by variations in productivity, which, as noted above, are them- selves largely determined by nutrient availa- bility, so that overall the most productive fisheries areas are on continental shelves at higher latitudes and in upwelling zones at lower latitudes. As a generalization, the latter are associated with pelagic fish stocks and the former more with demersal or semi- demersal (deep-water or bottom-dwelling] stocks, although pelagic stocks play an in- creasingly important role even here. As might be expected purely on the basis of its size, the Pacific Ocean is by far the most important major fisheries area, accounting for over 60 percent of marine landings. The northwest Pacific alone - an area with extensive continental shelf development - accounts for nearly half this total. The various upwelling zones are not all of equal importance in fisheries. That associated with the Humboldt current off Peru and Chile is the single most productive, while those associated with the California, Benguela and Canary currents are of somewhat lesser importance, although each is still a major fisheries area. The Arabian Sea upwelling appears to be anomalous, in that it evidently supports major populations of mesopelagic (i.e. middle-depth) rather than epipelagic species. Not only are the former generally considered of low value, with an identified market only as animal feed, but capture and Processing requires expensive, advanced technology. They thus remain virtually un- exploited at present and are considered along with the Antarctic krill stocks to be the major unexploited fisheries resource left. Trends in marine fisheries National fishery statistics are collated by the FAO. These data are the principal source of information on global fishery trends, although it is widely acknowledged that they are variable in quality. During the 1950s and 1960s, total landings increased steadily as 1996 Hy 1998 Million metric tons 0 2 4 6 8 10 T T T T 1 | Alaska pollock (Theragra chalcogramma] 4.5 Atlantic herring (Clupea harengus] 2.3 Japanese anchovy (Engraulis japonicus] 1.3 Chilean jack mackerel (Trachurus murphyi) 44 a 20 Skipjack tuna /Katsuwonus pelamis) Chub mackerel (Scomber japonicus] 2D. Anchoveta [Engraulis ringens] 8.9 1.7 Largehead hairtail (Trichiurus lepturus] 1683 ae Atlantic cod (Gadus morhua] 183 . Yellowfin tuna (Thunnus albacares) 11 | Blue whiting (Micromesistius poutassou] 0.6 2 new stocks were discovered, while improved fishing technology and an expansion of fishing effort enabled fuller exploitation of existing stocks of both pelagic and demersal species. Long-range fleets increased in size during this period and, as traditional fishing grounds in the North Atlantic and North Pacific became fully exploited, moved into new fish- ing grounds closer to the tropics and in the Marine biodiversity 145 Mi Figure 6.1 Species contributing most to global marine fisheries Source: FAO”? 1446 WORLD ATLAS OF BIODIVERSITY In EE Figure 6.2 Marine fisheries landings by major group Note: 1984-99 data are for capture fisheries only; pre-1984 data include aquaculture Source: FAO“ 7° 50 - —— Pelagic —— Demersal 40- = Crustaceans ——— Cephalopods w [= 2 30+ 2 . a) E c (ODF al = 20 = 10}- Species not elsewhere included ——— Mollusks (excl. cephalopods) southern hemisphere. By concentrating their efforts in the richest ocean areas, these fleets were largely responsible for the rapid increase in world catches. At the beginning of the 1970s, the Peruvian anchoveta fishery alone contributed some 20 percent of marine fisheries production. These stocks collapsed around 1972, at the same time as the important South African pilchard fishery in the Atlantic, seemingly in asso- ciation with an ENSO [El Nino Southern Oscillation] event. There was a sharp drop in overall marine fisheries production, after which the global catch increased more slowly than before, reaching the early 1970s level by the end of the decade. Landings of most demersal fish stocks remained relatively constant, however, implying that they were close to full exploitation. Long-range fleets continued to expand in importance. The 1980s once again saw a period of continuous growth [averaging 3.8 percent a year] in reported world landings. As in the 1970s landings of demersal stocks were generally static or declining so that shoaling pelagic species provided most of the increase in fish production. In fact, just three pelagic species (Peruvian anchoveta, South American sardine Sardinops sagax, and Japanese 1 1999 sardine Sardinops melanostictus) and one semi-demersal species (Alaska pollock] accounted for 50 percent of the increase in world landings during the 1980s‘. Most of this increase appears to have been because of favorable climatic effects on stock sizes rather than new fishery developments or im- proved management practices”. Following a sharp decline at the end of the 1980s, FAO data indicate slow net growth in marine capture fisheries through the 1990s. Four of the five most important fishes in fisheries in the late 1990s are pelagic, the exception being the Alaska pollock. This dominance of pelagic over demersal species is reflected in overall fisheries figures, with pelagic landings well over twice demersal landings globally. This contrasts sharply with the situation in the early 1950s when pelagic landings were only some 30 percent greater in volume than demersal landings (Figure 6.2). This increasing dependency on pelagic fish stocks is symptomatic of a major crisis in global marine fisheries. In general demersal fishes are more valuable per unit weight than pelagic species so that all else being equal the former are preferentially harvested. The increased importance of the latter in the past 40 years is indicative of the growing overexploitation of fisheries stocks worldwide - as valuable demersal stocks have been depleted so attention has turned to the intrinsically less valuable pelagic stocks. Of some 441 fishery stocks for which status data were available in 1999, FAO considered that only 4 percent were underexploited, and with a further 21 percent assessed as moderately exploited, around 25 percent of stocks analyzed were above the level of abundance thought to correspond to maxi- mum sustainable yield (MSY) level (or have a fishing capacity below this level). The remain- ing 75 percent of stocks were considered to require strict control of fishing capacity and fishing effort in order for them to recover to MSY biomass”. The proportion of stocks in this condition has increased between 1974, when first systematically reviewed, and 1999 (Figure 6.3). In terms of ocean regions, the situation worsened steadily in the North Atlantic and North Pacific until the early 1990s, when there were signs of possible stabilization, mainly in the former. Stocks appear still to be in decline in the tropical and southern parts of these oceans, with some possible stabilization in the tropical Atlantic. Recent analysis of fishery data suggests that the widespread overexploitation of marine fisheries is probably more serious than the global catch statistics indicate because mis- reporting by countries with particularly large fisheries, coupled with wide fluctuation in Peruvian anchoveta stocks (linked with El Nino events], can produce spurious trends at global level. When more realistic estimates of the catch in China, for example, are substituted for reported catch figures believed to be incorrect, the global catch appears to have declined by 0.36 million metric tons annually since 1988, rather than increased by 0.33 million metric tons (Figure 6.4). The declining trend is much steeper if the pelagic Peruvian anchoveta are excluded”. There are three major reasons for the declining state of many marine fisheries. First, and most fundamental, most fisheries have traditionally been regarded as an open access’ resource, so that, in effect, it pays any one fisher to harvest as much as possible at any given time because, if they do not, somebody else will. Secondly, technological innovations have made fishing much more efficient. Thirdly, there has been high investment in the world’s commercial fishing fleet (partly a consequence of the nature of fisheries as an open access resource but also for complex socioeconomic and political reasons). Bycatches and discards The effects of overfishing are compounded by the wastefulness of many marine capture fisheries. FAO estimated in 1994 that global marine fisheries bycatch and discards amounted to 18-40 million metric tons {mean 27 million) (Map 6.7). This represented just over 25 percent of the annual estimated total catch [i.e. landings represent around 75 percent of actual catch). Although figures are not available, it is generally assumed that the great majority of discards die. Further losses are caused by the mortality of animals which escape from fishing gear during fishery Marine biodiversity 147 | | 60 | % | Fully 50 — exploited | 40 } Underexploited + moderately exploited 30 + | 20 |- Overexploited + 10 - depleted + recovering 0 i a T T al T T T 1 1970 1975 1980 1985 1990 1995 2000 2005 operations, but it is impossible at present to estimate the importance of this. Shrimp fishing produces the largest volume of discards (around 9 million metric tons annually). Bycatches include non-target, often low- | pauls A02Pted Irom Figure 40 in value or ‘trash’ species, as well as undersized fish of target species. Non-target species may include marine mammals, reptiles (sea turtles] and seabirds, as well as finfishes and invertebrates. Of particular concern in recent years has been mortality of marine mammals, especially dolphins, in pelagic drift nets, of sea turtles in shrimp trawls and more recently of diving seabirds, especially Sata ee etl rama Wetsement albatrosses, in long-line fisheries. Discarding Pauly Figure 6.3 Global trends in the state of world stocks since 1974 Figure 6.4 Trends in global fisheries catch since 1970 85 - Uncorrected foe} i=) li ~ oa T ~ l=} T Corrected, no achoveta Global catch (million metric tons] 65 |- 60 |- Si = 50;- a El Nino El Nino El Nino event event event 0 inv | Gane) Fe a il | T ih Ira ore al ial T T Taal T T T 1970 1975 1980 1985 1990 1995 2000 (\ Map 6.7 Marine fisheries catch and discards The location of the fishery areas recognized by FAO for statistical purposes is shown on this map, with symbols representing the approximate late 1990s yield from capture fisheries and the volume of discarded catch, most of which is presumed not to survive. Each symbol represents approximately 1 million metric tons. Source: Data from FAO“® 1448 WORLD ATLAS OF BIODIVERSITY a EE onthe Fisheries catch <> 1 million metric tons landed O oHita4< 1 million metric tons discarded may be a side-effect of management systems intended to regulate fisheries {e.g. non- transferable quotas may cause discarding of over-quota catch; species-specific licensing may cause discard of non-licensed but still commercially valuable species). Solutions to bycatches and discards will be found essentially through improvement in the selectivity of fishing gear and fishing methods. Much of the research in this has been carried out in higher latitudes and is not readily transferable to multispecies tropical fisheries, where the tropical shrimp trawls still produce high rates of bycatch. Improved use of bycatch either as fishmeal or human food is also a possibility; however, this does not address the problem of mortality of potentially threatened species (sea turtles, seabirds, cetaceans], nor the wasteful capture of immature specimens of harvestable species. A further problem in the efficient use of marine resources is post-harvest loss. It is almost impossible to estimate this accurately, but FAO believes it to exceed 5 million metric tons per year [i.e. around 5 percent of harvest). Most significant are physical losses of dried fish to insect infestations and loss of fresh fish through spoilage. These problems are partic- ularly significant in developing countries. AQUACULTURE One major response to the growing crisis in marine capture fisheries has been the rapid rise in various forms of aquaculture (Figure 6.5). The latter may be defined as the rearing in water of organisms (animals, plants and algae) in a process in which at least one Pht phase of growth is controlled or enhanced by human action. The animals used are gen- erally finfishes, mollusks and crustaceans, although a number of other groups such as sea squirts (Tunicata), sponges (Porifera} and sea turtles are cultured in small quantities. Seaweeds of various kinds are also cultured, some in large amounts. Most of the species grown in any quantity are low in the food chain, being either primary producers, filter- feeders or finfishes that in their adult stages are either herbivores or omnivores. FAO notes that aquaculture is the world’s fastest growing food production sector, annual output having increased at an average rate of some 10 percent in the period 1984-98 (compared with less than 2 percent for capture fisheries) (see Figure 6.5). In 1999 aquaculture provided around one quarter of recorded global fisheries production. Of the total 32.9 million metric tons recorded in 1999, almost 20 million originated inland, and nearly 13 million were produced in marine and brackish environments”. In 1996, some 7.7 million metric tons of algae and plants were produced, almost all of this seaweed, chiefly Japanese kelp Laminaria japonica, nori Porphyra tenera and wakame Undaria pinnatifida. The first of these was, in terms of volume, the most important of all aquaculture species, with around 4.4 million metric tons produced. In marine and brackish (usually estuarine] environments, by far the most important animal group in terms of volume is the moll- usks, whose 1997 recorded production of some 8.6 million metric tons made up more than 75 percent of all animal production in Marine biodiversity 149 t [ 1509 WORLD ATLAS OF BIODIVERSITY SE ee na err eee ee 10 5 —— Seaweeds == Mollusks 8 =—— Diadromous fishes =—— Marine fishes a =—— Crustaceans ‘cS Miscellaneous 0 6 aquatic animals 2 E Cc g4- = 2 he T 1984 Figure 6.5 Marine aquaculture production Source: FAO“””° 71 1998 these environments. Around 50 mollusk species are produced in significant quantity, almost all bivalves. As with most culture systems, production is heavily skewed to a small number of species, with 65 percent of production composed of just three: the Pacific cupped oyster Crassostrea gigas, Japanese carpet shell Ruditapes philippinarus and Yesso scallop Pecten yessoensis. The Far East dominates production, with around 75 percent of that recorded taking place in China and most of the remainder in Japan. Although production of marine crusta- ceans accounts for only 10 percent or so by volume of marine and brackish water animal aquaculture, it has disproportionately high economic importance, and is also the sector that has given rise to most environmental concerns. Between 1984 and 1998 annual production grew nearly sixfold, from less than 200 000 to over 1 million metric tons. The great majority of production takes place in tropical and subtropical Asia and is dominated by Penaeus species; globally this genus pro- duces over 90 percent of aquaculture crusta- cean supply by weight. Three species of Penaeus account for around three quarters of crustacean production. The giant tiger prawn P. monodon is the most widely cultivated and accounts for nearly half; the whiteleg shrimp P. vannamei is cultured in the Americas and accounts for around 15 percent of estimated global supply {around 70 percent of this originating in Ecuador); and the fleshy prawn P. chinensis is cultured in China and currently accounts for around 10 percent of production, having declined considerably since the early 1990s when around 200 000 metric tons were produced annually. Other marine crustaceans cultivated include other Penaeus species, some Metapenaeus, and spiny lobsters Panulirus. These groups, however, make an insignificant contribution to global supply. Growth in crustacean aquaculture has been fuelled by the high value of the product: the market in 1996 was estimated to be worth nearly US$7.5 billion, or around one quarter of the total value of marine and brackish water aquaculture“. The great majority of production takes place in low-income countries - the five countries producing over 100 000 metric tons annually being China, Thailand, Indonesia, Ecuador and Bangladesh - and is aimed mainly at the export market (primarily to Europe, the United States and Japan) and to a lesser extent at the domestic luxury market. Pressure is high to produce maximum returns on invest- ment so that increasingly intensive farming methods are used. These are widely acknow- ledged to be having adverse social and environmental impacts in the countries of production, as well as leading to increasing difficulties in maintaining supply, owing to the spread of major diseases. The last of these accounts for the major decline in the Chinese fleshy prawn industry during the 1990s. Impacts include: e loss of mangrove habitat; e abstraction of freshwater; e introduction of pathogens and other damaging non-native species; e escape of cultured non-native species; ¢ pollution; e diversion of low-quality or cheap fish food resources (may lead to more efficient use of bycatches and trash fish but also to more indiscriminate catch fisheries); e diversion of effort from other forms of aqua- culture (notably milkfish Chanos chanos). The aquarium trade Up to 2 million people worldwide [about half in the United States and a quarter in Europe) are thought to keep marine aquariums, most of which are stocked with wild-caught species. In 1997 a total of 1 200 metric tons of coral was traded internationally, with 56 percent impor- ted by the United States and 15 percent by the European Union. Approximately half of this was live coral for aquariums, a tenfold increase on the amount of live coral traded in the late 1980s”. Qualitative estimates of trade suggest that 14-30 million fish may be traded per year, representing some 1 200 species, about two thirds of which are from coral reefs. Aquarium species are typically gathered by local fishers using live capture techniques or chemical stupefactants [such as sodium cyanide) which are non-selective and adversely affect the health of specimens as well as killing non- target organisms. Inappropriate shipping methods and poor husbandry along the supply chain often cause high mortality among the fish and invertebrates collected. While the current impacts of the aquarium trade remain poorly known, the industry has considerable potential to contribute to sus- tainable development. It is relatively low in volume but very high in value - a kilo of aquarium fish from one island country was valued at almost US$500 in 2000, whereas reef fish harvested for food were worth only US$6™. Aquarium species are a high-value source of income in many coastal com- munities with limited resources, with the actual value to the fishers determined largely by market access. In Fiji many collectors pay an access fee to the villages to collect on their reefs, but by selling directly to exporters they can have incomes many times the national average. By contrast, in the Philippines there are many middlemen, and collectors them- selves typically earn only around US$50 per month. Targeting mostly non-food species, aquarium fisheries could in principle provide an alternative economic activity for low- income coastal populations and an important source of foreign exchange for national economies, as well as an economic incentive for the sustainable management of reefs. The application of international certification schemes may provide an important tool for achieving this. OTHER MAJOR IMPACTS ON THE MARINE BIOSPHERE Alteration of physical habitat Physical alteration of habitats through human action chiefly affects coastal and inshore areas. Impacts here can be severe, although few attempts have been made to quantify them on a global basis. Major causes include coastal development, particularly landfilling and construction of groynes and jetties, aqua- culture, dredging of channels for navigational purposes, extraction of materials such as sand and coral, stabilization of shorelines, and destructive fishing methods such as beam-trawling, use of explosives and muro- ami (using rocks on ropes to drive fishes into nets). Upstream activities, such as defores- tation and dam construction can greatly alter sediment loads in rivers, affecting patterns of sediment deposition in estuarine areas. Chemical pollution and eutrophication Human activities have increased inputs of a huge range of organic and inorganic chem- icals into marine ecosystems. Such inputs may enter by direct discharge [e.g. in sewage outflow pipes], via river and stream outflow, as land runoff, through the atmosphere or from seagoing vessels. Because virtually all such input originates on land, as with most other human impacts on marine ecosystems, areas most affected are coastal and inshore regions, particularly enclosed or semi- enclosed water bodies. In oceanic regions, Marine biodiversity 151 SSS SSS SSS SSE SS The aquarium trade targets mostly non-food species and many invertebrates such as crabs, anemones and shrimp. 152 WORLD ATLAS OF BIODIVERSITY ee ee Worldwide, human activities have increased inputs of nitrogen and phosphorus in rivers and coastal waters fourfold. mixing of the enormous volume of sea water generally ensures that inputs become rapidly diluted. Major categories of input include nutrients of various kinds [e.g. nitrates and nitrites, phosphates, dissolved organic matter], per- sistent organic pollutants (POPs), including a range of chlorinated hydrocarbons, and heavy metals such as cadmium (Cd], copper (Cul, mercury {Hg}, lead [Pb], nickel {Ni} and zinc (Zn). Quantifying these inputs and assessing their impact is problematic, particularly because many occur naturally in sea water. Many POPs and heavy metals can act as toxins above certain concentrations, inducing mortality or morbidity or impairing repro- ductive success, particularly in cases where they become increasingly concentrated to- wards the top of food chains. Their overall impact on marine ecosystems remains un- certain. More easily observable is the impact of eutrophication resulting from the increased input of organic and inorganic nutrients (particularly nitrogen and phosphorus] into coastal waters, mainly through fertilizer runoff and sewage disposal. It is believed that human intervention has increased river inputs of nitrogen and phosphorus worldwide into coastal areas by more than fourfold over background levels. These inputs lead to increases in productivity in coastal waters, often in the form of algal blooms. These blooms may themselves be noxious; they also typically cause the euphotic zone to reduce in vertical extent and are implicated in the development of hypoxic (Low dissolved oxygen concentration] and anoxic (zero dissolved oxygen) zones. A shallowing of the euphotic zone may cause die-off of photosynthesizing benthic algae in shallow-water areas. This has occurred, for example, in the Black Sea where the euphotic zone had decreased from 50-60 m vertical extent in the early 1960s to around 35 m by 1990, leading to a decrease of up to 95 percent in living biomass of benthic macrophytic algae such as Phyllophora, formerly an important harvested resource. Hypoxia and anoxia result from the activities of oxygen-respiring bacteria below the euphotic zone feeding on accumulated dead algae and other organisms and waste matter raining down from above. Hypoxia re- sults in the emigration of mobile aerobic species and mortality of sedentary ones. This may have catastrophic impact on local fisheries. Most hypoxic zones vary in extent through the year and from year to year and some are only seasonal, disappearing when winter mixing causes re-oxygenation of bottom waters. They may be very extensive - the hypoxic zone to the west of the Mississippi delta covered some 16 000 km’ in 1997, having covered some 9 000 km’ in 1989. Over 50 such zones have been identified worldwide to date; some appear to be at least in part induced by natural phenomena while others are believed entirely anthropogenic. Invasions of exotic species As on land, the breakdown of biogeographic barriers in the sea appears to be having a major, and increasing, impact on marine ecosystems. The chief source is the deliberate or accidental translocation of organisms, but the construction of marine corridors between previously isolated areas has had a major impact on geographically restricted areas. These factors have resulted in a rapid spread of alien species in all the world’s oceans. Translocation can result from deliberate introduction of harvestable species, acciden- tal escapes from aquaculture and aquarium operations, transport in ballast water of ships and release of fouling organisms that adhere to the hulls of ships and boats. The extent of introductions of this kind has yet to be fully assessed but is certainly large. In many cases (see Box 6.2] the introduced species appear to be having a major impact on native biota, although in general it is difficult to separate the effects of a particular species from general ecosystem deterioration. Deliberate introductions include the plant- ing of mangroves and Nipa palms along coastlines, and the cultivation of fish, crus- taceans, mollusks and algae in many coastal regions. Atlantic salmon that have escaped from aquaculture are reportedly affecting wild stocks in the northeast Pacific, and their pathogens have themselves moved into the wild populations of closely related species. Ballast water is commonly pumped into the hold of ships as a means of controlling balance and position in the water, and is liable to be flushed out far distant from where it was taken on. It has been estimated that on any one day the ballast water of the world’s ocean fleets contains around 10 000 different species”. The principal artificial corridors are the Suez Canal (opened in 1869) and the Panama Canal (opened in 1914]. The 165-km long Suez Canal is a continuous seawater channel with a water level at the Red Sea end some 1.2 m higher than at the Mediterranean end, leading to a constant northward flow of water. It is estimated that to date some 400-500 marine species have migrated through the canal (so- called Lessepsian migrants”, after Ferdinand de Lesseps who planned the canal) and established themselves in the Mediterranean, while a far smaller number have moved in the other direction. New species are believed to arrive in the Mediterranean at the rate of four to five annually. Because the Panama Canal has a separate freshwater section, some 25 m above sea level, migration through it has been limited to date. Global climate change Human-induced climate change is liable to impact directly on marine and coastal areas by warming (particularly of the surface layers), by sea-level rise (associated both with thermal expansion and the melting of terres- trial ice caps and glaciers}, and through change in the gases dissolved in surface waters. These impacts are well understood, Marine biodiversity 153 SS I I I IE EL ED and measurable changes are already appar- ent. A more complex array of secondary changes may also occur, including changes to ocean stratification and surface mixing, changes to patterns of surface current, and perhaps to global systems such as the El Nino Southern Oscillation. Tropical coral reefs appear particularly sensitive to temperature change. Reef- building corals are adapted to stable thermal conditions and in most areas appear to be growing close to their upper temperature limit. Temperatures little more than 1°C above the normal maximum for a period of a few weeks are sufficient to drive a stress response known as coral bleaching. During a particularly strong El Nino event in 1998, warmer waters around the Seychelles and the Maldives induced a bleaching event in which 60-90 percent of all corals in the area died, equivalent to 5 percent of the world coral reef area. Although this event was linked to an extreme climatic perturbation, it occurred at a time of rising global temperatures and provides an indication of the impact of potential future climate change™. More subtle changes associated with the gradually changing background conditions, particularly temperatures, have been recorded in the Leidy’s comb jelly Menopsis leidyi was introduced from the American Atlantic into the waters of the Black Sea in 1982, presumably in ballast water. Unchallenged by natural predators, this species proliferated to a 1988 peak estimated at around 1 000 million metric tons wet weight (about 95 percent of the entire wet weight biomass in the Black Sea]. The species depleted the natural zooplankton stocks, with subsequent algal blooms and decline of the fishing industry in the Black Sea®. The Asian clam Potamocorbula amurensis spread through the northern San Francisco estuary (United States) following its introduction, possibly in ballast water, in 1986. The species reaches high density, up to 2 000 individuals per m’, and has caused sharp declines in the abundance and extent of several plankton species; its impact on fisheries is not yet clear. With more than 200 introduced species, this bay may be the most invaded aquatic habitat in North America. The green alga Caulerpa taxifolia is thought to have escaped from an aquarium in the western Mediterranean and is spreading rapidly in the coastal waters of Spain, France and Italy, with severe impact on the native seagrass beds and on coastal fisheries™. It has recently been reported in the coastal waters of California”. 156 WORLD ATLAS OF BIODIVERSITY Tee er ee ere errr Much less readily observed, marine species are in general more difficult to monitor and assess than terrestrial ones. distribution of pelagic seabirds along the Californian coast”; the faunal composition of intertidal communities”; and penguin distri- bution in the Antarctic”’. The mean sea level has risen 18 cm during the past 100 years and further increases could have a severe impact on coastal communities. Rising sea levels will lead to the inundation of some coastal lands, whilst in many other areas they will alter patterns of coastal erosion, and they may increase groundwater salinization. While many inter- tidal habitats are highly adaptable, the growing human presence in most coastal areas will prevent the natural migration of these habitats, leading to overall losses of saltmarsh, mangrove, or even beach and rocky shore habitats. The biomass of the world’s oceans is very low compared with terrestrial environments, but because of the rapid turnover in oceanic carbon cycles, marine phytoplankton (cyano- bacteria and algae) play an important role in removing dissolved carbon dioxide from solution, and are intermediates in the trans- port of organic carbon to the deep ocean. Once in the deep ocean, this carbon is effec- tively removed from exchange with the atmosphere for millennia. In this way, marine photosynthesizers help to buffer the rising concentration of carbon dioxide in the atmosphere, but this service, and marine primary productivity, may be affected if ocean temperatures rise (warmer waters hold less carbon dioxide in solution] and if the broad patterns of ocean circulation change significantly*. It is thought that increasing atmospheric temperatures may affect the generation of cold, oxygen-rich bottom waters beneath Arctic and Antarctic ice sheets, with major implications for deep-sea biota and for global patterns of seawater circulation, in particular the Great Conveyor, driven by bottom water generated in the North Atlantic. THE CURRENT STATUS OF MARINE BIODIVERSITY Because they are usually much less readily observed, marine species are in general much more difficult to monitor and assess than terrestrial ones. Assessment is based on sampling and, in the case of harvested species, often on the basis of catch rates, although the latter may vary in response to a wide range of factors in addition to changes in the population of the species concerned. An exception lies with those groups such as pinnipeds, sea turtles and seabirds that nest or breed on land. Because many of these tend to be colonial species and because they tend to breed in open habitats (beaches, cliff tops, ice sheets}, they may be easier to monitor than many other species, either terrestrial or aquatic. In the case of large, commercially valuable fish stocks, monitoring at large scale has in some cases been carried out for many years, so that estimates of the stock level are obtainable. Threatened and extinct species The only major marine species groups (classes or above) that have been com- prehensively assessed in terms of threatened species status to date are mammals and birds. In addition, sea turtles and a number of fish families and genera (e.g. the sturgeons in the order Acipenseriformes and the sea- horses Hippocampus in the order Syngnathi- formes] have also been assessed. Other threatened marine species have been identi- fied on more of an ad hoc basis. Data are summarized in Table 6.12. Relatively speaking, far fewer marine species are known to have become extinct since 1600 than either terrestrial or fresh- water ones. Cataloged extinctions comprise two marine mammals (the Caribbean monk seal Monachus tropicalis and Steller’s sea cow Hydrodamalis gigas) and five seabirds (three island petrels, Pallas’s cormorant Phalacrocorax perspicillatus and the great auk Alca impennis). In addition five coastal or island duck species have disappeared at various times from the late 17th century onwards; however there is in most cases insufficient information to determine whether these species were predominantly marine or terrestrial. As a gross generalization, marine species appear to be somewhat less extinction prone as a result of mankind's activities than fresh- water or terrestrial ones. There are arguably two main reasons for this. First, because of the size of the world ocean and the fact that people do not actually live in it, the marine biosphere remains as a whole considerably more buffered from human intervention than terrestrial and inland water areas. Second, marine species on the whole appear to be more widespread than terrestrial or inland water ones. In the open ocean, there are vast areas with apparently similar habitat con- ditions and there are few barriers to dispersal so that many species have circumglobal distributions. In addition, many forms that as adults are sessile (e.g. sponges and corals) or sedentary (many mollusks and crustaceans) have planktonic larvae that are often widely dispersed in water currents. For this reason, many coral reef species, for example, are found in suitable habitat throughout the Indo- Pacific region. In addition, many of the most heavily exploited fish species have high fecundity [in the case of some tunas amoun- ting to several million eggs in a single spawning}, so that they have at least poten- tially high population growth rates unparallel- ed in terrestrial vertebrates. There are of course significant exceptions to all these. Coastal regions in many parts of the world, and enclosed or semi-enclosed marine areas such as the Baltic, Black and Yellow Seas, are often under intense pressure from a range of human activities. A number of marine species do appear to have restricted ranges (e.g. the Hawaiian coral reefs are relatively rich in species found nowhere else while many southern hemisphere seabirds are apparently confined to a small number of breeding sites) and significant numbers have low or very low reproductive rates [many chondrichthyine fishes, marine mammals and seabirds). Until recently by far the most important human activity affecting marine species was uncontrolled exploitation. Where species are either easily exploitable or are highly sought- after [i.e. have high unit value], or both, they may suffer catastrophic declines. This is the case with sea turtles and a number of marine mammals and birds that are or have been harvested principally at their terrestrial breeding sites [which are often colonial], as well as with the great whales and the dugong, which although strictly marine are air- breathing and therefore spend some time at the sea surface [when they may be spotted). Most of these species have relatively low reproductive rates, so that even if they are ultimately afforded protection population recovery rates may be very slow. A recent synthesis of a wide range of information, including paleoecological and archeological data relating to early human Marine biodiversity 155 Ey HN Marine species which are easily exploitable or have high economic value may suffer catastrophic declines if exploitation is not strictly controlled. Table 6.12 Taxonomic distribution and status of threatened marine animals Note: Only the birds and mammals have been comprehensively assessed for species at risk; numbers refer to species (units such as subspecies and geographic populations appear in the Red List database but these are not tabulated here). Source: Status categories from 2000 Red List database, www.redlist org’ (accessec February 2002) 156 WORLD ATLAS OF BIODIVERSITY EEE communities, suggests that overfishing of larger vertebrates and mollusks is charac- teristic of indigenous and colonial human use of coastal ecosystems, and the first of what is typically a series of impacts”. Massive losses in biomass and abundance appear to have occurred, on a scale largely unsuspected, and seemingly amounting to the loss of entire trophic levels of consumer organisms, with radical consequences for ecosystem status. Overfishing is likely to be followed by impacts of pollution, mechanical habitat loss, introduced species and climate change. Loss of filter-feeding organisms that maintain water quality is liable to be followed by eutrophication, hypoxia and disease, as exemplified by conditions in Chesapeake Bay following the collapse of the oyster fishery in the early 20th century. Synergistic interactions of this kind are making the effective, long-term manage- ment of marine resources one of the major — and most intractable - problems currently facing humankind. Land-breeding species may also be sus- ceptible to other threats, such as predation, Cnidaria Anthozoa Stony corals 2 Mollusca Bivalvia Bivalves Gastropoda Gastropods 1 2 Craniata - fishes Carcharhiniformes Ground sharks 1 3 5 Hexanchiformes Cow sharks 1 Lamniformes Mackerel sharks 3 Orectolobiformes Carpet sharks 2 Pristiformes Sawfishes 2 5 Rajiformes Rays 2 1 Rhinobatiformes Guitar fishes i 1 Squaliformes Dogfishes and sleeper sharks 1 Squatiniformes Angel sharks 1 2 Coelacanthiformes Coelacanths 1 Acipenseriformes Sturgeons 2 10 6 Batrachoidiformes Toadfishes 5 Clupeiformes Herrings and anchovies i 1 Gadiformes Cods, hakes, rattails 1 2 Gasterosteiformes Sticklebacks 1 Lophiiformes Anglerfishes, etc. 1 Ophidiiformes Pearlfishes, cusk-eels, brotulas 1 Perciformes Perches, etc. 7 5 33 Pleuronectiformes Plaice, flounders, soles 1 ] Salmoniformes Salmonids 1 4 6 Scorpaeniformes Gurnards, scorpionfishes, etc. 1 2 1 Siluriformes Catfishes 1 Syngnathiformes Pipefishes, seahorses, etc. 1 i 35 Tetraodontiformes Triggerfishes, etc. 3 coastal development and pollution. It is note- worthy in this context that the family Procellariidae contains nearly three times as many threatened species (36 out of 115 species, or 28 percent] as the average bird family, in which 11 percent of species are threatened, and nearly six times as many critically endangered species as would be expected at random. It is almost certainly the tendency of these birds to nest on islands, whose biotas have in general suffered enor- mously more from mankind's influence in the past few centuries (see Chapter 4), rather than their seagoing habits that has led to this. For truly marine species (chiefly finfishes and invertebrate animals] the situation appears somewhat different. Even when these have been exploited to the point of stock collapse, as has occurred for example with the cod Gadus morhua stocks off Newfoundland in the North Atlantic, the species concerned do not appear to have become imminently threatened with biological extinction. This is in part because once stocks are reduced below a certain level it is often no longer economically Craniata - Reptilia Squamata Iguanidae Iguanas 1 Chelonia Dermochelyidae Leathery turtle 1 Cheloniidae Sea turtles 2 3 1 Craniata - Aves = Anseriformes Anatidae Ducks i i 1 Charadriiformes Alcidae Auks, puffins 4 Charadriidae Plovers i 2 1 Laridae Gulls, terns, skuas, auks, skimmers 1 | 5 Ciconiiformes Ardeidae Egrets, herons {| Pelecaniformes Fregatidae Frigatebirds i 1 Pelecanidae Pelicans 1 Phalacrocoracidae Cormorants and shags 8 Sulidae Gannets and boobies 1 i Procellariiformes Diomedeidae Albatross 2 2 12 Hydrobatidae Strom petrels i 1 Pelecanoididae Diving petrel 1 Procellariidae Petrels, shearwaters 10 6 20 Sphenisciformes Spheniscidae Penguins 3 7 Craniata - Mammalia Carnivora Mustelidae Otters, etc. 2 Otariidae Eared seals 1 5 Cetacea Phocidae Earless seals 1 1 1 Balaenidae Right whales 1 Balaenopteridae Rorquals 3 1 Delphinidae Dolphins 1 1 Monodontidae Beluga 1 Phocoenidae Porpoises 1 1 Physeteridae Sperm whales 1 Sirenia Dugongidae Dugong 1 Trichechidae Manatees 3 Marine biodiversity 157 qu SSS ESSE SSS SSS ne te yt ¥ Ni 1588 WORLD ATLAS OF BIODIVERSITY ea RR SS SSS 140 120 viable to continue harvesting them. Generally, the residual population at this stage is still large enough to allow recovery if harvesting ceases, particularly in the case of species with high fecundity and therefore high potential intrinsic rates of increase. Exceptions to this are species that have low fecundity, partic- ularly if they also have a long period to maturity, with limited ranges, and which may either have high unit value or be caught as bycatches. In the case of bycatches, because the fishery is not directed at the species concerned, its intensity will not decrease as population levels decrease so that it may theoretically be possible at least locally to extirpate species, particularly if they are habitually caught before they reach maturity. Examples include several sawfish species 100 60 40 7 T 1970 1975 Figure 6.6 Marine population trends Note: A simplified representation of the average population change ina sample of 217 marine species, see text Source: Loh”” T 1980 T 1995 if T i] 1985 1990 1999 (family Pristidae]. These are large, slow- growing, predominantly inshore species that give birth to relatively small numbers of live young. Population densities appear to be naturally low and animals are widely caught as bycatch in inshore fisheries before they are large enough to reproduce. As a result five species are classified as ‘endangered’ and two as ‘critically endangered’. In addition, it is possible that trophic shifts may occur when populations of some species are severely reduced, inhibiting recovery of these popu- lations when exploitation ceases. This has been suggested in the case of some great whale populations that have not apparently recovered as rapidly as projected following the cessation of their harvest. The marine living planet index An impression of the overall trend in a large sample of species for which population indicators are available can be derived from the WWF living planet index”. This approach is designed to represent the change in the ‘average species’ in the sample from one five- year interval to the next, starting in 1970. The marine sample represents 217 aquatic and coastal species of mammals, birds, reptiles and fishes, and the overall trend is for a significant decline in population levels over the last three decades of the 20th century (Figure 6.6). The sample is dominated by the stocks and species that humans have an interest in monitoring, most of the fishes among these being of commercial importance as a fisheries resource. These should also be stocks that humans have an interest in managing as well as possible. That the index has declined in every five-year interval since 1970 is evidence that such management is failing, as confirmed by the picture painted above of global marine capture fisheries. Assessing the status of marine and coastal ecosystems Threatened species inventories and the marine living planet index can give a very general overall impression of the status of marine biodiversity. Assessing marine ecosystem ‘health’ is much more problematic. However, snapshots can be obtained from examining particular ecosystems, such as mangroves and coral reefs*. In the former, an overall assessment can be made on the basis of the area destroyed or severely degraded. In the latter areal measures are more problematic, in part because reef extent is much more difficult to measure than mangrove extent and, of greater importance, because the vast majority of a reef is composed of non-living calcareous deposits. Measures of the change in extent of these give little insight into the state of the living component of the reef. For this reason other measures, such as estimates of inci- dence of coral disease, may be feasible. Marine biodiversity 159 i li ll ————————___, REFERENCES 1 Longhurst, A.R. and Pauly, D. 1987. Ecology of tropical oceans. |CLARM [International Center for Living Aquatic Resources Management] Contribution No. 389. Academic Press Inc., California. 2 Longhurst, A.R. 1995. Seasonal cycles of pelagic production and consumption. Progress in Oceanography 36: 77-167. 3 Longhurst, A.R. 1998. Ecological geography of the sea. Academic Press, San Diego. 4 Sherman, K. and Busch, D.A. 1995. Assessment and monitoring of large marine ecosystems. In: Rapport, D.J., Guadet, C.L. and Calow, P. [eds]. Evaluating and | monitoring the health of large-scale ecosystems. Springer-Verlag, Berlin, published in cooperation with NATO Scientific Affairs Division. NATO Advanced Science Institutes Series |: Global Environmental Change, 28: 385-430. 5 Available online at http://www.edc.uri.edu/lme/default.htm [accessed February 2002). 6 Olson, D.M. and Dinerstein, E. 1998. The Global 200: A representation approach to conserving the earth's distinctive ecoregions. Conservation Biology 12: 502-515. 7 Vaulot, D. 1995. Marine biodiversity at the micron scale. International Marine Science Newsletter 75/76. United Nations Educational, Scientific and Cultural Organization, Paris. 8 Rex, M.A., Stuart, C.T. and Coyne, G. 2000. Latitudinal gradients of species richness in the deep-sea benthos of the North Atlantic. Proceedings of the National Academy of Sciences 97: 4082-4085. 9 Adey, W.H. and Steneck, R.S. 2001. Thermogeography over time creates biogeographic regions: A temperature/space/time-integrated model and an abundance-weighted test for benthic marine algae. Journal of Phycology 37: 677-698. 10 Roberts, C.M. et al. 2002. Marine biodiversity hotspots and conservation priorities for tropical reefs. Science 295: 1280-1284. 11 Nybakken, J. 1993. Marine biology: An ecological approach. 3rd edition. HarperCollins College Publishers, New York. 12 World Conservation Monitoring Centre 1992. Groombridge, B. (ed.) Global biodiversity: Status of the Earth's living resources. Chapman and Hall, London. 13 Finlayson, M. and Moser, M. (eds) 1991. Wetlands. Facts on File Ltd, Oxford. 14 Woodroffe, C.D. and Grindrod, J. 1991. Mangrove biogeography: The role of Quarternary environmental sea-level change. Journal of Biogeography 18: 479-492. 15 Tomlinson, P.B. 1986. The botany of mangroves. Cambridge University Press, Cambridge. 16 Spalding, M.D., Blasco, F. and Field, C.D. (eds) 1997. World mangrove atlas. International Society for Mangrove Ecosystems, Okinawa. 17 Lasserre, P. 1995. Coastal and marine biodiversity. /nternational Marine Science Newsletter 75/76: 13-14. 18 Risser, P.G. 1986. Spatial and temporal variability of biospheric and geospheric processes: Research needed to determine interactions with global environmental change. Report of a workshop. International Council of Scientific Unions Press, Paris. 19 Ducrotoy, J.P., Elliott, M. and de Jonge, V.N. 2000. The North Sea. In: Sheppard, C. (ed.) Seas at the millennium: An environmental evaluation, Vol. |, pp. 43-64. Elsevier Science, Amsterdam. 20 Lewis, J.R. 1977. Rocky foreshores. In: Barnes, R.S.K. (ed.) The coastline, pp. 147-158. Wiley, London. 21 Wilson, S.C. 2000. Northwest Arabian Sea and Gulf of Oman. In: Sheppard, C. [ed.] Seas at the millennium: An environmental evaluation, Vol. II, pp. 17-34. Elsevier Science, Amsterdam. 22 Phillips, R.C. and Menez, E.G. 1988. Seagrasses. Smithsonian Contributions to the Marine Sciences 34: 104. 160 WORLD ATLAS OF BIODIVERSITY a EEE EE EE ee—n0Vwe 23 Hemminga, M.A. and Duarte, C.M. 2000. Seagrass ecology. Cambridge University Press, Cambridge. 24 Short, F.T. and Wyllie-Echeverria, S. 2000. Global seagrass declines and effects of climate change. In: Sheppard, C. (ed.] Seas at the millennium: An environmental evaluation, Vol. Ill: Global issues and processes. Elsevier Science, Amsterdam. 25 Wilkinson, C.R. and Buddemeier, R.W. 1994. Global climate change and coral reefs: Implications for people and reefs. Report of the UNEP-IOC-ASPEI-IUCN Global Task Team on the Implications of Climate Change on Coral Reefs. 26 Hulm, P. and Pernetta, J.C. 1993. Reefs at risk: Coral reefs, human use and global climate change. A programme of action. |UCN-the World Conservation Union, Gland. 27 Veron, J. 2000. Corals of the world. 3 vols. Australian Institute of Marine Science, Queensland, Australia. 28 Cairns, S.D., Hoeksema, S.B. and van der Land, J. 1998. List of extant stony corals. Available online at http://www.nodc.noaa.gov/col/projects/coral/hardcoral/ Hardcoralmain.html (accessed February 2002). 29 Mortensen, P.B. et al. 1995. Deep water bioherms of the scleractinian coral Lophelia pertusa at 64°N on the Norwegian shelf: Structure and associated megafauna. Sarsia 80: 145-158. 30 Rogers, A.D. 1999. The biology of Lophelia pertusa and other deep-water reef-forming corals and impacts from human activities. International Review of Hydrobiology 84(4): 315-406. 31 Fossa, J.H. et al. 1999. Effects of bottom trawling on Lophelia deep water corals in Norway. Poster presented to the ICES (International Council for Exploration of the Seas) workshop on the ecosystem effects of fishing, February 1999. 32 Koslow, J.A. and Gowlett-Holmes, K. 1998. The seamount fauna of southern Australia: Benthic communities, their conservation and impacts of trawling. Report to Environment Australia and the Fisheries Research Development Corporation, FRDC Project 95/058. 33 Gjosaeter, J. and Kawaguchi, K. 1980. A review of the world resources of mesopelagic fish. FAO Fisheries Technical Paper 193: 1-151. 34 Rowe, G.T. 1983. Biomass and production of the deep-sea macrobenthos. In: Rowe, G.T. (ed.) Deep-sea biology, Vol. 8 The sea, pp. 453-472. John Wiley and Sons Inc., New York. 35 Hessler, R.R. and Sanders, H.L. 1967. Faunal diversity in the deep-sea. Deep-Sea Research 14: 65-78. 36 Angel, M.V. 1991. Biodiversity in the deep ocean. A working document for the UK Overseas Development Administration. Unpublished ms. 37 Grassle, J.F. 1991. Deep-sea benthic biodiversity. Bioscience 41(7). | 38 Grassle, J.F. and Maciolek, N.J. 1992. Deep-sea species richness: Regional and local estimates from quantitative bottom samples. American Naturalist 139: 313-341. 39 May, R.M. 1992. Bottoms up for the oceans. Nature 357: 278-279. | 40 Angel, M.V. 1982. Ocean trench conservation. Commission on Ecology Papers No. 1. IUCN-the World Conservation Union, Gland. 41 Gage, J.G. and Tyler, PA. 1991. Deep-sea biology: A natural history of organisms at the deep-sea floor. Cambridge University Press, Cambridge. 42 Grassle, J.F. 1986. The ecology of deep-sea hydrothermal vent communities. Advances in Marine Biology Vol. 23. Academic Press. 43 Hecker, B. 1985. Fauna from a cold sulphur-seep in the Gulf of Mexico: Comparison with hydrothermal vent communities and evolutionary implications. Bulletin of the Biological Society of Washington 6: 465-473. | 44 Grassle, J.F. 1989. Species diversity in deep-sea communities. TREE (Trends in Ecology and Evolution) 4(1). Marine biodiversity 161 || 45 UNCED 1992. The global partnership for environment and development. A guide to Agenda 21. United Nations Conference on Environment and Development, Geneva. 46 Committee on Biological Diversity in Marine Systems 1995. Understanding marine biodiversity: A research agenda for the nation. National Academy Press, Washington DC. 47 FAO 1990. Review of the state of world fishery resources. Marine Resources Service, Food and Agriculture Organization of the United Nations, Rome. 48 FAO 1999. The state of world fisheries and aquaculture 1998. Food and Agriculture Organization of the United Nations, Rome. 49 FAO 2000. The state of world fisheries and aquaculture 2000. Fisheries Department, Food and Agriculture Organization of the United Nations, Rome. Available online at http://www.fao.org/DOCREP/003/X8002E/X8002E00.htm (accessed April 2002). 50 Watson, R. and Pauly, D. 2001. Systematic distortions in world fisheries catch trends. Nature 414: 534-536. 51 Green, E. and Shirley, F. 1999. The global trade in coral. World Conservation Press, Cambridge. 52 Edwards, A.J. and Shepherd, A.D. 1992. Environmental implications of aquarium fish collection in the Maldives, with proposals for regulation. Environmental Conservation 19: 61-72. 53 Carlton, J.T. 1999. Invasions in the sea: Six centuries of re-organising the Earth's marine life. In: Sandlund, O.T., Schei, P.J. and Viken, A. (eds). Invasive species and biodiversity management, pp. 195-212. Kluwer Academic Publishers, Netherlands. 54 Goldschmid, A. 1999. Essay about the phenomenon of Lessepsian migration. Colloquial Meeting of Marine Biology |, Salzburg 1999. Available online at http://www.sbg.ac.at/ipk/avstudio/pierofun/lm/Lesseps.htm (accessed April 2002). 55 GESAMP 1997. Opportunistic settlers andthe problem of the ctenophore Mnemiopsis leidyi invasion in the Black Sea. IMO/FAO/UNESCO-IOC/WMO/WHO/IAEA/UN/UNEP Joint Group of Experts on the Scientific Aspects of Marine Environmental Protection. Rep. Stud. GESAMP, 58: 84. 56 Anon 1997. Medwaves 34: 11-13. News Bulletin of the Mediterranean Action Plan, Athens. 57 Dalton, R. 2000. Researchers criticize response to killer algae. Nature 406: 447. 58 Spalding, M., Teleki, K. and Spencer, T. 2001. Climate change and coral bleaching. In: Green, E. et al. (eds). Impacts of climate change on wildlife, pp. 40-41. 59 Veit, R.R. et al. 1997. Apex marine predator declines 90% in association with changing oceanic climate. Global Change Biology 3: 23-28. 60 Sagarin, R.D. et al. 1999. Climate-related changes in an intertidal community over short and long time scales. Ecological Monographs 69: 465-490. 61 Ainley, D., Wilson, P. and Fraser, W.R. 2001. Effects of climate change on Antarctic sea ice and penguins. In: Green, E. et al. (eds). Impacts of climate change on wildlife, pp. 26-27. 62 Balino, B.M., Fasham, M.J.R. and Bowles, M.C. 2001. Ocean biogeochemistry and global change: JGOFS research highlights 1988-2000. \nternational Geosphere-Biosphere Programme Science No. 2. International Council of Scientific Unions, Stockholm. 63 Jackson, J.B. et al. 2001. Historical overfishing and the recent collapse of coastal ecosystems. Science 293: 629-638. 64 Loh, J. (ed.) 2000. Living planet report 2000. WWF - World Wide Fund for Nature, Gland. 65 Couper, A.D. 1993. Times atlas of the oceans. Times Books. 66 Nelson, J.S. 1994. Fishes of the world. John Wiley and Sons Inc., New York. 67 Croxall, J.P., Evans, P.G.H. and Schreiber, R.W. 1994. Status and conservation of the world’s seabirds. ICBP Technical Publication No. 2. ICBP {now BirdLife International), Cambridge. 68 Sibley, C.G. and Monroe, Jr, B.L. 1990. Distribution and taxonomy of birds of the world. Yale University Press, New Haven and London. 69 Duke, N.C. 1992. Mangrove floristics and biogeography. In: Robertson, A.I. and Alongi, D.M. leds]. Tropical mangrove ecosystems. Coastal and estuarine Studies 41, p. 329. American Geophysical Union, Washington DC. 162 WORLD ATLAS OF BIODIVERSITY a TE eee 70 71 72 73 74 75 Spalding, M.D., Ravilious, C. and Green, E. 2001. World atlas of coral reefs. University of California Press, Berkeley. IUCN 2000. 2000 IUCN Red List database available online at www.redlist.org {accessed February 2002). Compagno, L.J.V. 1984. Sharks of the world. An annotated and illustrated catalogue of shark species known to date. FAO Fisheries Synopsis No. 125 Vol. 4 (Parts 1-2]. Spalding, M.D. 1998. Biodiversity patterns in coral reefs and mangrove forests: Global and local scales. Unpublished PhD thesis, University of Cambridge. Veron, J.E.N. 1995. Corals in space and time: The biogeography and evolution of the Scleractinia. University of New South Wales Press. FAO 2002. http://www.fao.org/DOCREP/003/X8002E/x8002e04.htm#P1_6 (accessed April 2002). Inland water biodiversity 163 Inland water biodiversity 1 percent) of the world’s water resource. Despite this, they encompass a wide range ’ NLAND WATERS MAKE UP A MINUTE PROPORTION (much less than a hundredth of of habitat types and contain a disproportionately high fraction of the world’s biodiversity. Freshwater is also a vital resource for human survival. In consequence, inland water ecosystems are placed under many, often conflicting, pressures, with increasingly adverse consequences for their biodiversity. There are indications that, overall, a higher proportion of inland water species are in decline than marine or terrestrial forms. INLAND WATERS The hydrosphere is estimated to contain about 1 386 million cubic kilometers (km*) of water, almost all of which (97.5 percent) is saline water making up the world ocean, leaving some 35 million km* (2.5 percent) of fresh- water. A major proportion (about 69 percent) of this freshwater is locked up in the form of ice and permanent snow, where it is unavail- able to living organisms. The Earth’s liquid freshwater is mostly in subterranean ground- waters, with a small proportion in soils and wetlands, and the smallest proportion of all - about 0.01 million km’ or 0.3 percent of all freshwater —- makes up the world’s lakes and rivers on which inland water biodiversity depends (Table 7.1). There are very large regional differences in the concentrations of water in all its forms (e.g. about twice as much atmospheric water in equatorial as in temperate latitudes], and in the occurrence of different types of inland waters [e.g. South America has an enormous concentration of river water but few large lakes, while the converse is true for Africa). Over the oceans, evaporation exceeds input from rivers and rainfall, and over land precipi- tation exceeds evaporation. This excess on land amounts to about 43 000 km? annually and this represents the global runoff that replenishes the world’s rivers, lakes and marine waters, and which humans draw on, together with groundwater, to meet their domestic, agricultural and industrial needs. Chapter 5, on terrestrial ecosystems, outlines some of the principal differences between terrestrial and aquatic environments as they determine the living conditions of organisms in them. In general, freshwater sys- tems and the organisms within them are far more strongly affected by daily and seasonal changes as a result of weather patterns and climate conditions than are marine aquatic environments, in parallel with change in the broader terrestrial environment. For example, upland streams may receive an influx of cold meltwater in spring, and be subject to strong insolation during midsummer; the volume, speed and transparency of river water is liable to change radically following rainfall in Earth's surface 510 Land 149 29 World ocean 361 vA 1351 Freshwater - - 34.65 Ice and permanent snow 16 - 23.8 Groundwater = - 10.4 Wetlands, soil water, permafrost 2.6 - 03 Lakes and rivers 1.5 - 0.01 Table 7.1 Components of the hydrosphere 97.5 2.5 1.76 0.77 0.02 0.0007 4 55 Source: Anon’; Shiklomanov eae mpemes ry 164 WORLD ATLAS OF BIODIVERSITY Inland water habitats can be divided into running or lotic and standing or lentic sys- tems. They may also be divided into per- manent water bodies, periodically {usually seasonally] inundated, and ephemeral or transient. Each of these has its own distinct set of ecological characteristics. There is not necessarily a rigid dividing line between an inland aquatic habitat on the one hand and a terrestrial or marine habitat on the other. Any temporarily inundated area, such as a river floodplain, is effectively a hybrid or transitional system, being at some times essentially aquatic, at other times terrestrial. Similarly there are many areas that consist of shifting mosaics of land and shallow water, or areas of saturated vege- tation, such as sphagnum moss bogs that are strictly neither land nor water. These trans- itional areas are often collectively termed Some coastal or high altitude lakes and lagoons combine high concentrations of dissolved chemicals with high water temperatures, sometimes as high as 70°C. Few kinds of species thrive in these harsh alkaline or saline conditions. Communities typically include cyanobacteria, diatoms, a few small invertebrate animals such as brine shrimps (Artemial, with flamingos often being the only large organisms present. Flamingos, of which five living species are recognized, are filter-feeders, with the bill and mouthparts specialized to extract small particles from water. All share the same basic morphology, but tend to extract prey of different sizes: e.g. in sub-Saharan Africa the lesser flamingo feeds almost exclusively on cyanobacteria and other microorganisms, while the greater flamingo (often found in the same lakes], feeds mainly on small macroscopic invertebrates, such as brine shrimps and other crustaceans. Fishes are often absent from lakes where flamingos are abundant, and flamingos are rarely abundant where fishes are common. This may be because the two compete with each other for their food supply but, in the inhospitable lakes where they abound, flamingos have no real competitors. Their early adaptation to an extreme environment that no other large animal was capable of exploiting may explain why they have survived relatively unchanged in morphology for so long [fossils are known from the Eocene, about 50 million years before the present). the catchment basin. The soil and surface geology of watershed areas have a strong influence on the chemical composition of both river and lake waters and, for example, buffer or reinforce the effect of acid rain. Inland water habitats Although the terms ‘inland water’ and ‘freshwater’ are often used more or less interchangeably they are not equivalent. A considerable number of inland waters are saline, some much more so than sea water. Conversely, waters of the deltaic regions of some major river systems (most notably the Amazon] may be fresh a considerable dis- tance out to sea. Despite their vastly smaller extent, inland aquatic habitats show far more variety in their physical and chemical characteristics than marine habitats. They encompass systems as varied as the world’s great lakes and rivers, small streams and ponds, temporary puddles, thermal springs and even the minute pools of water that collect in the leaf axils of certain plants, such as bromeliads. Chemically they range from almost pure water to highly concentrated solutions of mineral salts, toxic to all but a few specialized organisms (see Box 7.1). ‘wetlands’. Similarly, estuarine areas are transitional areas between inland and marine systems. Lotic systems: rivers and catchment basins A river system is a complex but essentially linear body of water draining under the influence of gravity from elevated areas of land toward sea level. The typical drainage system consists of a large number of smaller channels (streams, rills, etc.} at higher ele- vation merging as altitude falls into pro- gressively fewer but larger channels, which in simplest form discharge by a single large watercourse. Most such systems discharge into the coastal marine environment. Some discharge into lakes within enclosed inland basins; a few watercourses in arid regions enter inland basins where no permanent lake exists. The source area of all the water passing through any given point in the drainage system is the catchment area for that part of the system. In parallel with the hierarchical aggregation of tributaries of the major river system, sub-catchments aggregate into a single major catchment basin; this is the entire area from which all water at the final discharge point of the system - i.e. usually the sea - is derived. Strictly, the watershed is the line of higher elevation dividing one catchment basin from another, but this term is increasingly used as a synonym of catchment. The speed and internal motion of river water depends largely on water volume and the shape of its channel. These factors typi- cally differ greatly through the river system, from narrow, steep and fast upland feeder streams to broad, level and slow downstream reaches. In virtually all river systems water volume also varies seasonally. Some rivers in arid or semi-arid catchments flow for only part of the year, or in extreme cases only once every several years. Large rivers may span many degrees of latitude and pass through a wide range of climatic conditions within their catchments. Variations in water flow and underlying geo- logy also create a wide range of habitats within any river and often within a short distance. Different organisms are typically adapted to different parts of any given river system. River systems can change course radically as a result of deposition and erosion of their channel, and the uplift and erosion of watershed uplands. Despite the dynamic physical state of these systems, large rivers rarely disappear, and although direct evi- dence is scarce, indications are that some have been in continuous existence for tens of millions of years. This is consistent with the fact that running waters include represen- tatives of almost all taxonomic groups found in freshwaters, and that several invertebrate taxa occur only in running waters or attain greatest diversity there. By far the largest river catchment in the world is the Amazon (with the Ucayali} in South America which covers just under 6 million km’ and is nearly 60 percent larger than the next largest, the Congo in central Africa. Unsurprisingly, the former is the major repository of the world’s freshwater biodiversity. Between them the 20 largest river catchments cover around 45 million km’, or about one third of the world’s ice-free land surface’. Lentic systems Lakes and ponds The great majority of existing lakes, of which around 10000 exceed 1 km’ in extent, were formed as a result of glacial activity, with most of the rest a result of tectonic activity’. Tectonic lakes are formed either as a result of faults caused by deep crustal movements or by volcanism. In the case of the former, a lake may form in a depression caused by a single fault, or in a depressed area between two or more faults - these being graben lakes - or in a rift valley. Most volcanic lakes form in craters or calderas of volcanoes while a few (usually short-lived) may form behind dams caused by lava flows. Glacial lakes occupy basins caused by the scouring action of ice masses. Most of the world’s existing lakes are glacial and geologically very young, dating from the retreat of continental ice sheets at the start of the Holocene, around 11500 years before present. All such lakes are expected to Inland water biodiversity 165 SS | My The majority of lakes were formed as a result of glacial activity. 166 WORLD ATLAS OF BIODIVERSITY od Table 7.2 Physical and biodiversity features of major long- lived lakes Notes: Lakes ordered by volume. A few other lakes have notable endemism among fishes, mollusks, crustaceans or other groups. Among these are lakes Inle (Myanmar), Lanao (Philippines), Malili (Indonesia) and the Cuatro Cienegas basin (Mexico) - but their ages are not yet firmly established. Qualitative remarks (e.g. ‘very high’, ‘low’) in the Biodiversity column are related to long- lived lakes, not to lake systems in general. All biodiversity data are approximate and subject to change with new survey data or different taxonomic opinion. i Evidence indicates that the lake dried out completely, or nearly so, around the late Pleistocene, 10-12 000 years ago" Source: Collated from data in Martens et al.°’. Fish estimates for East African Lakes from Snoeks™. Baikal Largest, deepest, oldest extant freshwater lake (20% of all liquid fresh surface water on Earth] Tanganyika Malawi Victoria World's second largest freshwater lake (area) Titicaca One of world’s highest altitude lakes Biwa Ohrid Fed mainly by subterranean karst waters Russia 25-30 Burundi 20 Tanzania Zambia DR Congo (former Zaire) Malawi >2 Mozambique Tanzania Kenya >4?! Tanzania Uganda Bolivia 3 Peru Japan 4 Albania 3 Macedonia (FYR) 1 637 1470 780 70 280 104 295 23 000 18 880 8 400 2760 890 674 50 Very high sp. richness; exceptional endemism in fishes and several invertebrate groups Total animal spp.: 1 825, endemic: 982 Fishes: 56 spp., 27 endemic Very high sp. richness; high endemism, especially high among cichlid fishes Total animal spp.: 1 470, endemic: 632 Fishes: 325 spp., including 250 cichlids of which 98% endemic Very high sp. richness; high endemism, especially high among cichlid fishes Fishes: 845 spp., including 800 cichlids of which 99% endemic High sp. richness, especially of fishes; exceptional endemism among cichlid fishes - many fish endemics depleted or extirpated following introduction of Nile perch Fishes: 545 spp., including 500 cichlids of which 99% endemic Moderate sp. richness and endemism (highest among fishes} Total animal spp.: 533, endemic: 61 Fishes: 29 spp., 23 endemic Moderate sp. richness and endemism (highest in gastropod mollusks and fishes) Total animal spp.: 595, endemic: 54 Fishes: 57 spp., 4 endemic Moderate sp. richness; exceptional endemism in several groups (planarians, oligochaetes, gastropod mollusks, ostracod crustaceans} Fishes: 17 spp., up to 10 endemic fill slowly with sediment and plant biomass, and to disappear within perhaps the next 100 000 years, along with any isolated biota. Lakes may also be caused by the dissolution of soluble rocks, most notably limestone in karst regions which is gradually dissolved by dilute acids in water running through it, and by changes in the course of rivers in floodplain regions, which result in ox-bow and scroll lakes. Only about ten existing lakes are known with certainty to have origins much before the Holocene (Table 7.2)’, and most of these occupy basins formed by large-scale sub- sidence of the Earth's crust, dating back to at most 20 million (Lake Tanganyika) or 30 million (Lake Baikal) years before the present. There is good evidence that some extinct lake systems in the geological past were very large and very long-lived under different climatic and tectonic conditions. In general, the long-lived lakes are of particular interest in terms of biodiversity because they tend to be rich in species of several major groups of animals and many of these species are restricted to a single lake basin. Wetlands As indicated above, the distinction between a wetland, an aquatic system and a terrestrial system may be essentially arbitrary. However, a number of mixed shallow-water and terres- trial habitat types share several characteristics and are habitually grouped as wetlands. Wetlands in this sense are typically hetero- geneous habitats of permanent or seasonal shallow water dominated by large aquatic plants and broken into diverse microhabitats’. The four major broad habitat types are: Bogs Bogs are peat-producing wetlands in moist climates where organic matter has accum- ulated over long periods. Water and nutrient input is entirely through precipitation. Bogs are typically acid and deficient in nutrients and are often dominated by sphagnum moss. Fens Fens are peat-producing wetlands that are influenced by soil nutrients flowing through Inland water biodiversity 167 Ce a the system and that are typically supplied by mineral-rich groundwater. Grasses and sedges, with mosses, are the dominant vege- tation. Fens are typically more productive and less acidic than bogs. Marshes Marshes are inundated areas with herb- aceous emergent vegetation, commonly dom- inated by grasses, sedges or reeds. They may be either permanent or seasonal and are fed by ground or river water, or both. Swamps Swamps are forested freshwater wetlands on waterlogged or inundated soils where little or no peat accumulation occurs. As with marshes, they may be either permanent or seasonal. Biogeography and important areas Freshwater lineages that originated in contin- ental water systems may show general patterns of distribution similar to terrestrial groups, corresponding more or less to broad biogeographic realms. Lineages of marine origin may remain restricted to peripheral systems corresponding to the area where the ancestral forms moved into freshwater. Unlike many terrestrial species, which can disperse widely in suitable habitat, the spatial extent of the range of strictly freshwater species tends to correspond to present or formerly continuous river basins or lakes. These species include fishes and most moll- usks and crustaceans. Watersheds between river basins are the principal barriers to their dispersal between systems, and their ranges are extended mainly by physical changes to the drainage pattern le.g. river capture following erosion or uplift can allow species formerly restricted to one system to move into another], or by accidental transport of eggs by waterbirds, or by flooding. In many instances, the range within a system will also be restricted by particular habitat requirements (variations in water turbulence or speed, shelter, substrate, etc.). These frequently differ at different stages in the life cycle {for example in fishes the conditions and sites required for egg depos- ition and development, for early growth of fry, way 168 WORLD ATLAS OF BIODIVERSITY Ce ene nn nnn a e there is also a strong relationship between species richness and the latitude of the basin. Africa L. Malawi Fishes Mollusks Recent analysis suggests that latitude may be Africa L. Tanganyika Fishes Mollusks Crabs a surrogate measure for energy availability Africa L. Victoria Fishes Mollusks and productivity within the basin‘*, factors Africa Madagascar Fishes Mollusks Crabs known to be well correlated with variation in Africa Niger-Gabon Fishes Crabs terrestrial diversity (Chapter 5). No taxonomic Africa Upper Guinea Fishes Mollusks Crabs class restricted to inland waters has yet been Africa Lower Congo Fishes Mollusks Crabs mapped globally at species level but, at a Australia SE Australia & Tasmania Fishes Mollusks Crayfish higher taxonomic level, a density surface of Australia SW Australia Fishes Fairy shrimp freshwater fish families has been developed Eurasia SE Asia and lower {see Map 7.1) with a view to providing an Mekong River Fishes Mollusks Crabs indication of global variation in inland water Eurasia Balkans (southwest } Fishes Mollusks diversity analogous to those available for Eurasia L. Baikal Fishes Mollusks terrestrial groups (Chapter 5). Eurasia L. Biwa Fishes Mollusks A recent analysis of areas important for the Eurasia L. Inle Fishes Mollusks maintenance of global freshwater biodiversity’ Eurasia L. Poso Fishes Mollusks was based on the expert view of a number of Eurasia Malili Lakes Fishes Mollusks regional and taxonomic specialists. The analy- Eurasia Sri Lanka Fishes Crabs sis was designed to make effective use of Eurasia Western Ghats Fishes Mollusks Crabs readily available information and, although preliminary, yielded the first global overview of freshwater biodiversity hotspots’. Maps 7.2, 7.3 North America East Mississippi drainage (Ohio, Cumberland, Tennessee rivers} Fishes Mollusks Crayfish and 7.4 show, respectively, important areas for North America Mobile Bay drainage Fishes Mollusks Crayfish freshwater fishes, mollusks and selected crus- North America Western USA Fishes Mollusks Fairy shrimp tacean groups. Further details of all these South America _L. Titicaca Fishes Mollusks areas can be found in Appendix 6. Table 7.3 South America La Plata drainage Fishes Mollusks lists the sites and areas that have been South America Amazon basin Fishes 2 Crabs identified as of special importance for more Table 7.3 Partial list of global hotspots of freshwater biodiversity Notes: This table lists areas of special importance for diversity in fishes and either mollusks or crustaceans or both. See text and Appendix 6. Six of the seven long-lived lakes in Table 7.2 also appear here Source; See sources cited in Appendix 6 and for feeding and breeding of adults are often different). Many cave or subterranean freshwater aquatic species [e.g. of fishes, amphibians and crustaceans] have restricted ranges, perhaps consisting of a single cave or aquifer, and limited opportunities for dispersal, depending on the surrounding geology and the consequent morphology of the water system occupied. Analysis of data from some 151 river basins indicates that there is a strong cor- relation between the spatial extent of a river catchment and the number of fish species therein. The ‘size’ of a river can be represen- ted by the area of the basin or by the volume of water flowing through the river system in any given period; the latter is a better predictor of fish species richness than is basin area. When area is taken into account, than one of the above groups. It is not a comprehensive global listing because it omits several large but imprecisely defined areas of known high diversity, and it omits diverse taxa not covered in the assessment (e.g. amphi- pods, copepods); nor does it mention sites of key importance mainly for one group of animals. Although the Amazon basin is a vast region rather than an identifiable site, it has such an exceptional diversity of fishes that it could not reasonably be excluded from a list of globally important areas. More detailed continental reviews are now also available for Asia, including discussion of taxonomy, hotspots and policy’, for Latin America’ and North America’. In order to help prioritize investment, the conservation organi- zation WWF has selected 53 freshwater eco- regions, based on a combination of biogeo- graphic region, water body type, biodiversity and representativeness’’. Perhaps more so than in other biomes, these freshwater ecoregions TE TE EI IEEE BESTS SP RET have a firm objective basis because they correspond broadly with catchment basins. BIOLOGICAL DIVERSITY IN INLAND WATERS At high taxonomic levels the diversity of freshwater organisms is considerably lower than on land or in the sea. Only one extant eukaryote phylum (Gamophyta - green conju- gating algae] is apparently confined to fresh- water habitats. The number of species overall is low in absolute terms in comparison with marine and terrestrial groups, but species richness in relation to habitat extent is relatively high. For example, about 10 000 (40 percent) of the 25 000 known fish species are freshwater forms’. Given the distribution of water on the Earth's surface this is equivalent to one fish species for every 15 km? in fresh- waters compared with one for every 100 000 km® of sea water. This high diversity of freshwater fishes relative to habitat extent is undoubtedly promoted by the extent of isol- ation between freshwater systems. Many lineages of fishes and invertebrates have evolved high diversity in certain water sys- tems, and in some cases, species richness and endemism tend to be positively correlated between different taxonomic groups”. As is the case with terrestrial habitats, species richness increases strongly toward the equator, so that in most groups of organisms, there are many more species in the tropics than in temperate regions, although in a few specific cases (e.g. fresh- water crayfish] this appears to be reversed. Protoctists The larger algae comprise some 5 000 species in three major groups [the green, brown and red algae], the great majority of which are marine or brackish water forms (‘seaweeds’). The green algae Chlorophyta includes one order of around 80 species (Ulotrichales) that is mainly freshwater. However, one major group usually associated with the green algae - the stoneworts (charophytes) - is almost entirely freshwater. The stoneworts include some 440 species, most of which are endemic at continent level or below; they tend to be very sensitive to nutrient enrichment and have declined in many areas”. Fungi There are more than 600 species of fresh- water fungi known, currently more from temperate regions than from the tropics, although probably only a small fraction of existing species have been described, and the tropics have been little sampled”. Virtually all described freshwater fungi are ascomycotes with few basidiomycotes and zygomycotes having been identified. They occur wherever vascular plant material is available as a substrate. They appear to be important as parasites, endotrophs and saprotrophs of emergent aquatic macrophytes, as decom- posers of submerged allochthonous woody debris, and as a food resource for inver- tebrates’*”. Most are very small, with sporomes (fruiting bodies) less than 0.5 mm in diameter. Plants Wetland or aquatic species occur with some frequency in the non-vascular plant phyla, which generally prefer moist habitats, and among the ferns and allies. Mosses in the order Sphagnales [a single family Sphag- naceae and genus Sphagnum) often grow submerged, and are key components of peat bogs. Many groups of damp-loving (hygrophilous) terrestrial mosses (e.g. Thamnium, Bryum, Mnium) have aquatic forms. Several genera of Bryales are aquatic or have aquatic species. A number of Inland water biodiversity 169 : = mT Species richness in inland waters, in relation to habitat extent, is relatively high. 177 WORLD ATLAS OF BIODIVERSITY y f SR SS ES Map 7.1 Freshwater fish family diversity Family richness of typica bony fishes [Actinopterygii in inland waters, plotted as a world density surface. It is based on generalized range maps of 157 families. Color depth represents the number of families, up to a maximum of 44, potentially present at any point. Two families of cartilaginous fishes (Elasmobranchii) that together have a very few inland water species are omitted. About ten or so families of bony fish that occur in coastal and estuarine waters, but do not extend significantly into inland freshwaters, are omitted. Several families range more or less widely in inland waters and also occur in coastal and estuarine waters around the continents, but this peripheral part of the range is in most cases not represented. Source: Produced by UNEP-WCMC using range maps prepared from information in Berra Density High Low liverwort species growing otherwise in wet terrestrial situations may also live sub- merged, sometimes at considerable depths. Truly aquatic liverworts include several species of Riccia and Ricciocarpus natans (Ricciaceae; Marchantiales) that live free- floating on the surface of eutrophic lakes. At least 16 species of Riella in the Riellaceae (Jungermanniales) are aquatic forms charac- teristic of temporary waters in semiarid regions, reaching highest diversity in north- ern and southern Africa. Among the lyco- phytes, most of the 60 or so species of Isoetes (family Isoeteacae} are aquatic, some of great limnological importance, and Stylites is an endemic member of the littoral community of Andean lakes. Sphenophytes (horsetails) often occur in moist situations, including around water margins. Equisetum fluviatile, for example, is a notable emergent littoral form of north temperate lakes. The Filicinophyta include several aquatic forms. The genus Ceratopteris (four species, family Pteridaceae]) has the only truly aquatic (floating) homosporous ferns; some are culti- vated ornamentals, others are edible. A few other species, e.g. Microsorum pteropus (Polypodiaceae] and Microlepia speluncae (Dennstaedtiaceae) can grow in water. Among heterosporous ferns, the family Marseliaceae comprises three genera and 55-75 species which are either amphibious or fully aquatic. All members of the families Salviniaceae (one genus and about ten species] and Azollaceae {one genus and around six species) are floating aquatic ferns. The latter supports the symbiotic nitrogen-fixing Anabaena azollae (phylum Cyanobacteria). Vascular plants are essentially terrestrial forms, and existing aquatic species are derived from terrestrial ancestors; several different lineages include aquatic species and this transition has therefore occurred several times. It has been estimated that at most 1 percent of angiosperms, i.e. up to 2 700 species, are aquatic’. Around 14 angiosperm families consist largely or exclusively of in- land water forms (Ceratophyllaceae, Hippuri- daceae, Hydrostachydaceae, Nymphaeaceae, Podostemaceae, Trapaceae, Butomaceae, Hydrocharitaceae, Lemnaceae, Limnochari- taceae, Najadaceae, Pontederiaceae, Potamo- getonaceae, Zannichelliaceae). Most inland water plant species are relatively widespread, ranging over more than one continental land mass; many are cosmo- politan, occurring around the world and on Inland water biodiversity 11 remote islands. Of the widespread forms, some are essentially northern temperate species extending to a great or lesser extent into the tropics; some are mainly tropical”. The Podostemaceae is particularly note- worthy for its many monotypic genera, and a large number of narrowly endemic species, in at least one instance with several forms restricted to different stretches of a single river. Tropical South America, Madagascar, Sri Lanka, India, Myanmar and Indonesia hold such localized species”. Inland water animals Animal species are considerably more diverse and numerous in inland waters than plants. Most of the major groups include terrestrial or marine species as well as freshwater forms. Apart from fishes, important groups IE SoS, mal 172 WORLD ATLAS Table 7.4 Insects of inland waters Notes: Data refer to number of species; all estimates approximate i Partially aquatic as adult and sometimes as nymph. ii Number in parentheses refers to fully aquatic Nepomorpha. iii All these species are parasitoids as larvae Source: Collated from data in 5 Hutchinson Ephemeroptera 84 Odonata 302 Plecoptera 196 Orthoptera’ = Blattodea’ = Hemiptera 236 Megaloptera 26 Neuroptera 58 Coleoptera 730 Hymenoptera” = Diptera 1 300 Trichoptera 478 Lepidoptera - OF BIODIVERSITY go TE with inland water species include crus- taceans (crabs, crayfishes, shrimps, as well as planktonic forms such as filter-feeding Cladocera and filter-feeding or predatory Copepoda}, mollusks [including mussels Bivalvia, and snails Gastropoda], insects {including stoneflies Plecoptera, caddisflies Trichoptera, mayflies Ephemoptera], sponges, flatworms, polychaete worms, oligochaete worms, numerous parasitic species in various groups, and numerous microscopic forms. Information is incomplete for many groups, but crustaceans and mollusks have speciated profusely in certain freshwater systems, with a tendency to form local endemic species. Because of the feeding mode - attached bottom-living filter-feeders - bivalves can help maintain water quality but tend to be susceptible to pollution (their larvae are parasitic on fishes]. The diversity and eco- logical role of microorganisms and micro- invertebrates in freshwater sediments have been reviewed”. Insects with an aquatic larval phase but a winged adult phase are often restricted to particular river basins [even if adults disperse widely, they may not find suitable habitat), but in general are much less restricted in this way than entirely aquatic species. A relatively large number of species, particularly of crustaceans, occupy temporary pools and have a stage that is desiccation-resistant and can undergo long-range passive dispersal 614 224 2 250 415 127 4875 578 387 2 140 ca 20 - ca 20 0 - ca 10 404 129 (81)! 3 200 43 6 300 6 9 ca 100 1 655 1072 5 000 55 74 ca 100 5 547 4050 >20 000 1340 895 7000 - 5 ca 100 between drainage basins; some such species are thus widely distributed. Insects As on land, insects [phylum Mandibulata] are as far as is known by far the most diverse group of organisms in inland waters. The true number of aquatic insects remains unknown; data for three relatively well known areas (Europe, Australia and North America} and extrapolations for possible global totals are included in Table 7.4. In contrast to terrestrial faunas, where beetles (order Coleoptera) are the most diverse, flies and their relatives (order Diptera) appear to be by far the most abundant group in inland water habitats, although also one of the less fully known. In terms of life histories, there are two main groups of aquatic insects: those in which the adult stage and the active immature stages are passed in water [in some cases with a terrestrial pupal stage); and those in which, after a nymphal or larval stage in water, the adult stage is spent on land or in the air. The great majority of Diptera, and therefore most aquatic insects, form part of the latter group. Included amongst their number are several of enormous economic significance to humans, of which the most significant are almost certainly mosquitoes of the genus Anopheles, intermediary hosts of the malaria parasite. Most aquatic insects are benthic, living in or on the bottom; a small number are planktonic and live suspended in the water column. Around half of the aquatic Hemiptera and a few other insects and non-insect invertebrates live on the water surface (epipleuston). Fishes Around 40 percent of known fish species occur in freshwater: almost exactly 10 000 species are confined to freshwater, and a further 1 100 or so occur in freshwater but are not confined to it (Table 7.5). These last include species, such as many salmonids, that grow in the seas but ascend rivers to spawn (anadromous), and others, such as eels, that grow in inland waters but spawn at sea (catadromous]. Freshwater fishes are taxonomically diverse, although not as diverse as marine ones. Thirty-four of the 57 or so extant orders of fishes have at least one strictly freshwater species, while a further two, the sawfishes [(Pristiophoriformes] and Inland water biodiversity 173 amma || tarpons (Elopiformes] have species that occur in freshwater but are not confined to it". This compares with 38 orders that have at least one marine species. Of the orders of fishes with freshwater species, ten are entirely freshwater and Petromyzontiformes Carcharhiniformes Pristiophoriformes Rajiformes Ceratodontiformes Lepidosireniformes Acipensiformes Amiiformes Anguiliformes Atheriniformes Batrachoidiformes Beloniformes Characiformes Clupeiformes Cypriniformes Cyprinodontiformes Elopiformes Esociformes Gadiformes Gasterosteiformes Gonorhynchiformes Gymnotiformes Mugiliformes Ophidiiformes Osmeriformes Osteoglossiformes Perciformes Percopsiformes Pleuronectiformes Polypteriformes Salmoniformes Scorpaeniformes Semionotiformes Siluriformes Synbranchiformes Tetraodontiformes Lampreys | 6 4) 32 Ground sharks 7 47 208 1 Sawfishes | 2 5 0 Rays 12 62 456 24 Australian lungfish ] i 1 1 Lungfishes 2 2 5 5 Sturgeons 2 6 26 14 Bowfin 1 1 1 1 Eels = 15 141 738 6 Silversides 8 47 285 146 Toadfishes 1 19 69 5 Needlefishes, sauries, flyingfishes, halfbeaks 5 38 191 51 Characins 10 237 1 343 1 343 Herrings and anchovies 5 83 357 72 Carp, minnows, loaches 5 279 2 662 2 662 Rivulines, killifishes, pupfishes, poeciliids, goodeids 8 88 807 794 Ladyfishes and tarpons 2 2 8 0 Pikes and mudminnows 2 4 10 10 Cods, hakes, rattails 12 85 482 i Pipefishes, sticklebacks, sandeels, etc. 1 71 257 19 Milkfish and beaked sandfishes 4 7 35 28 Knifefishes 6 23 62 62 Mullets 1 17 66 1 Pearlfishes, cusk-eels, brotulas 5 92 355 5 Smelts 13 74, 236 42 Bonytongues 6 29 217 217 Perches, basses, sunfishes, whitings, etc. 148 1496 9 293 1922 Trout-perches, pirate perch, cavefishes 3 6 9 9 Plaice, flounders, soles 1 123 570 6 Bichirs 1 2 10 10 Salmonids 1 1 66 45 Gurnards, scorpionfishes, velvetfishes, etc. 25 266 1271 52 Gars i 2 HW 6 Catfishes 34 412 2 405 2 280 Swamp-eels 3 12 87 84 Triggerfishes, puffers, boxfishes, filefishes, molas 9 100 339 12 Table 7.5 Fish diversity in inland waters, by order Source: Nelson” (differs in detail from the later taxonomy of Eschmeyer 4 78 8 0 1 0 28 5 i 100 5 100 26 54 1 100 26 1 171 51 6 7 56 27 1 343 100 80 20 2 662 100 805 98 7 0 10 100 2 0 4 7 29 80 62 100 7 2 6 1 71 18 217 100 2815 21 9 100 20 i 10 100 66 68 62 4 7 86 2 287 95 87 97 20 4 5I 8 ) 1742 WORLD ATLAS OF BIODIVERSITY p) Serene epee ee ee Virtually all reptiles of inland waters return to land, at least to nest. another five are very largely so (with more than 80 percent of their known species in fresh- waters]. A further 13 are very largely marine with a small proportion of freshwater species {<10 percent) while the remainder have significant numbers of both marine and freshwater species. Over 80 percent of fresh- water species are confined to just four orders: the carps and their relatives (Cypriniformes); the characins (Characiniformes]; the catfishes (Siluriformes); and the perches and their relatives (Perciformes). The first three of these are wholly or almost entirely freshwater, while the last, the largest order of fishes with nearly 40 percent of known species, is unusual in having significant numbers of both marine and freshwater species. Amphibians The great majority of the 5000 or so living amphibian species have aquatic larval stages and, as none is known to occur in sea water, all these are dependent on inland waters of various kinds for continued survival of populations. In some cases such water bodies may be temporary pools or puddles, or water in the leaf axils of plants. Relatively few species are fully aquatic (Table 7.6). Although the number of fully aquatic species in each of the three extant orders is roughly similar {ca 20-30 in each], these represent very different proportions of each order, being less than 1 percent of anurans, around 5 percent of caudate amphibians and more than 10 percent of caecilians. Aquatic caudate amphibians are neotenic, that is retain features of the larval stage, most notably external gills. In addition to the fully aquatic amphibians (several of which can survive for short periods in damp conditions out of water), many other species may lead largely aquatic lives or may, as in the case of the Mexican axolotl Ambyostoma mexicanum, have completely aquatic neotenic populations. Reptiles Very few completely aquatic inland water reptiles are known. The three file-snakes in the family Acrochordidae are live-bearing and may pass their entire lives in water, often in coastal and estuarine areas as well as Sm SS SL freshwaters. Virtually all other reptiles of inland waters are egg-laying and return to land at least to nest; most also spend a proportion of their time on land, often basking on banks or logs. The two most aquatic orders are the Crocodilia and the Chelonia. All the 22 or so extant species of the former are predominantly aquatic and occur in fresh- waters, although one or two may also be found in marine areas. Of the latter, some two thirds of the 250 or so extant species are largely or predominantly aquatic, and a further 30 or so species may be considered amphibious. Aside from the file-snakes and the homal- opsine snakes, a number of other snake species are at least semi-aquatic. These in- clude several genera of natricine snakes, including the North American Nerodia and the Asiatic Sinonatrix, as well as the anacondas Eunectes (family Boidae) and the water cobra Boulengeria annulata (family Elapidae). Amongst lizards, no wholly aquatic species is known. However, many can swim proficiently, often using water as a means of escape from predators, and a number are semi-aquatic. These last include several Australian and Old World monitors Varanus spp. (family Varanidae], water dragons Hydrosaurus and Physignathus (Agamidae], the crocodile lizard Shinisaurus from south China (Xenosauridae], the Bornean earless monitor Lanthanotus {Lanthanotidae], and a number of New World teiids (Teiidae]. AMPHIBIA Caudata Anura Gymnophiona REPTILIA Chelonia Crocodilia Squamata AVES Anseriformes Gruiformes Ciconiiformes Passeriformes MAMMALIA Monotremata Didelphiomorpha Insectivora Rodentia Cetacea Sirenia Carnivora Artiodactyla Amphiumidae Cryptobranchidae Plethodontidae Proteidae Sirenidae Pseudidae Pipidae Typhlonectidae Carettochelidae Trionychidae Platysternidae Chelydridae Dermatemydidae Chelidae Kinosternidae Pelomedusidae Emydidae Alligatoridae Crocodylidae Gavialidae Acrochordidae Colubridae Anseranatidae Dendrocygnidae Anatidae Heliornithidae Rallidae Jacanidae Laridae Podicipedidae Anhingidae Phalacrocoracidae Phoenicopteridae Pelecanidae Gaviidae Cinclidae Ornithorhynchidae Didelphidae Tenrecidae Soricidae Talpidae Castoridae Muridae Hydrochaeridae Platanistidae Trichechidae Mustelidae Viverridae Phocidae Hippopotamidae Congo eels Giant salamanders and hellbenders Lungless salamanders Mudpuppies and olm Sirens Paradox frogs Clawed frogs and pipid toads Typhlonectid caecilians Pig-nosed soft-shelled turtle Soft-shelled turtles Big-headed turtles Snapping turtles Central American river turtle Austro-american side-necked turtles Mud and musk turtles Side-necked turtles Pond and river turtles Caimans and alligators Crocodiles Gharial and false gharial File-snakes Colubrid snakes Magpie goose Whistling-ducks Ducks, geese and swans Limpkin and sungrebes Rails, gallinules and coots Jacanas Gulls, terns, skuas, auks, skimmers Grebes Anhingas Cormorants and shags Flamingos Pelicans and shoebill Divers Dippers Platypus Opossums Tenrecs and otter shrews Shrews Moles and desmans Beavers Voles and mice Capybara River dolphins Manatees Mustelids, otters Viverrids Earless seals Hippopotamus 141 3 (li 8 22 (13) 21 4 5 (2)" 9 6 5 5 = wo Hnrwow on ON Www Ff —= — Inland water biodiversity 175 SS SS SSS TIS SESS ST SS 0 eT Sn Table 7.6 Tetrapod diversity in inland waters Notes: Entirely freshwater families in bold. Taxonomy based on the same vertebrate sources cited in Table 2.1 Bird groups here differ to some extent from Table 7.9 which uses a more traditional arrangement of bird higher taxa i Genera Leurognathus, Haideotriton, Typhlomolge only. ii Sometimes included in the Caecilidae. iii Subfamily Homalopsinae iv Genus Fulicula {coots) only. v Figures in parentheses indicate those species of the total that breed largely or entirely inland also included as seabirds in Table 6.6. 174 WORLD ATL Orreere ant Se Map 7.2 Major areas of diversity of inland water fish This map represents an informal synthesis of documented expert opinion on globally important areas for freshwater fish diversity, taking into account species richness and endemism. Two categories are shown: discrete areas and systems known to be of high diversity, and areas where diversity is globally important but less concentrated. Note: For numbered locations see Appendix 6. Source: Compiled with the help of members of IUCN/SSC specialist groups and other ichthyologists; first published in WCMC'. AS (0'F BO DIMER Sis ¥ Fish diversity dl Key areas ieee Other important areas Birds Unlike mammals, there are no wholly aquatic birds, because all species lay eggs that cannot survive prolonged immersion in water; however, a much higher proportion of bird than mammal species is associated with inland water ecosystems. As with other tetrapod groups, it is impossible to separate rigidly inland water species from primarily terrestrial forms or from seabirds. Table 7.6 includes bird species that are highly adapted to aquatic ecosystems and largely or ex- clusively inhabit inland waters, rather than marine or coastal areas. Nearly 60 percent of the roughly 250 species belong to one family - the Anatidae - all of whose members are more or less associated with aquatic habitats, although some {such as most goose species] feed very largely on land. It is noteworthy that, among the Passeriformes or perching- birds - by far the most species-rich order {accounting for over half of all bird species) - only one small family, the dippers (Cinclidae], can be considered truly aquatic in habits. In addition to these, there are a large number of wading bird species associated with inland or coastal wetland and littoral habitats. These include all or most members of the following families: Anhimidae (scream- ers - three species]; Eurypygidae (sunbittern - one species}; Gruidae [cranes - 15 species); Rallidae (rails, gallinules and coots - 142 species]; Scolopacidae (sandpipers and their relatives - 88 species); Rostratulidae (painted- snipe - two species}; Charadriidae (plovers and their relatives - 89 species); Ardeidae (herons, bitterns and egrets - 65 species); Scopidae (hammerkop - one species); Threskiornithidae [ibises and spoonbills - 34 species); and Ciconiidae (storks - 26 species). There are also a considerably smaller number of non-wading birds that feed largely on fishes and other aquatic animals and are adapted to diving or surface-snatching. Among these are many kingfishers [families Alcedinidae, Dacelonidae and Cerylidae}, the fish-owls (Ketupa and Scotopelia spp., family Strigidae], fish-eagles (Haliaeetus and /Ichthyophagus spp., family Accipitridae) and a few other raptors. Mammals Wholly aquatic inland water mammals are confined to two orders, Cetacea and Sirenia. In the former, four of the five species of the family Platanistidae (the river dolphins) are confined to river systems and the fifth occurs Inland water biodiversity 177 in estuarine and coastal waters. A number of other cetaceans may enter the lower reaches of river systems but all are predominantly marine. Two of the four living species of Sirenia - the Amazonian manatee Trichechus inunguis and the West African manatee T. senegalensis - are wholly or very largely confined to freshwaters, and a third, the Caribbean manatee 7. manatus, is found in both inland and marine waters. Amongst other groups, a number of species may lead more or less aquatic lives but all these are effectively amphibious, in that at the very least they produce young on land. For most of these, a predominantly aquatic life is evident from direct observations, but for some (chiefly the Muridae) it is inferred from morphological adaptations. In direct contrast to terrestrial systems, where the majority of S ainnaiamesiimeameneed 172 WORLD ATLAS OF BIODIVERSITY Freshwater is essential to human survival Tee mammal species are herbivores, a high proportion of these amphibious species {around 50 percent) are carnivores, with only eight true herbivores and most of the aquatic murids believed to be omnivores [though predominantly insectivorous or piscivorous]. It is also noteworthy that at least four of these species - the two beavers Castor spp., the hippopotamus Hippopotamus amphibius and the capybara Hydrochaeris hydrochaeris - feed very largely or entirely on land plants. In addition to the species included in Table 7.6, a number of other mammals are largely or wholly confined to wetland habitats (marshes, floodplains and swamps]. These include three African antelopes - the Nile lechwe Kobus megaceros, red lechwe Kobus leche and sitatunga Tragelaphus spekei - and the South American marsh deer Blastocerus dichotomus. HUMAN USE OF AND IMPACT ON INLAND WATERS Freshwater — as precipitation, groundwater or in inland water ecosystems - is essential for human survival, chiefly because humans must drink and also because it is needed, in far greater quantity, to produce food. It also has a wide range of subsidiary uses - for transport, industrial production, cleaning, waste disposal, generation of hydroelectric power, recreation, esthetic purposes and in the form of inland water ecosystems as sites for the production of food. Many of these demands conflict with each other, so that for example the use of water for disposal of noxious wastes is incompatible with the provision of safe drinking water. Moreover, while the amount of freshwater available is limited, demands on it continue to grow steadily as the global human population con- tinues to expand. This problem is exacerbated by the fact that freshwater is unevenly distributed around the world, so that it is often not available where and when needed, nor in the appropriate amounts, nor with the necessary quality. The two last are particularly important for the maintenance of freshwater biodiversity. Freshwater systems are therefore under growing pressure, as flow patterns are disrupted and the load of waste substances increases. Inevitably, per capita shares of water for human use are decreasing and water stress is becoming more widespread”. Agriculture consumes around 70 percent of all water withdrawn from the world’s riv- ers, lakes and groundwater”. In places, more than half the water diverted or pumped for irrigation does not actually reach the crop, and problems of waterlogging and salin- ization (deposition in soil of salts left by evaporation of pumped groundwater] are increasing. However, irrigated agriculture produces nearly 40 percent of world food and other agricultural commodities on only 17 percent of the total agricultural land area, and is thus disproportionately important to global food security”. The principal use of freshwater species, not considering the properties of aquatic sys- tems themselves, is as food. Subsidiary uses include the aquarium trade, materials for medicinal or ornamental use, and as fertilizer. Inland water fishery production has two com- ponents: capture fisheries and aquaculture, although as discussed below the distinction between the two is becoming increasingly blurred. For many human communities, par- ticularly in countries less developed indus- trially, capture fisheries provide a major portion of the diet. Although it appears to be under-reported, inland water production has usually been regarded as far less important than marine fisheries and, with few excep- tions where countries have access to both marine and inland aquatic resources, repor- ted yield from inland waters is a small fraction of marine yield. Even in landlocked countries, the recorded inland harvest is often low both in absolute size and in relation to consumption of meat and other agricultural produce. INLAND WATER FISHERIES Capture fisheries and aquaculture Globally, the reported inland water capture fishery for 1999 amounted to 8.2 million metric tons, with 19.8 million metric tons of aquaculture production recorded’; over 85 percent of the former and about 98 percent of the latter comprised finfishes, with virtually all the remainder being freshwater crus- taceans and mollusks””'. The crustaceans are mainly crayfishes and freshwater shrimps, both exploited for food, and most of the mollusks are bivalve, taken for pearls and for food. These reported totals compare with reported marine capture fisheries of some 84 million metric tons, and marine and brackish water aquaculture animal production of around 13 million metric tons {see Chapter 6). The reported global inland water capture fishery has increased slowly in the period 1984-99, by nearly 2 percent per year, al- though this masks considerable regional variation, with declines in some areas [e.g. Europe and North America) and more marked increases elsewhere (notably Asia]”. Reported inland aquaculture production has been rising at a higher rate, and was well over twice the reported production of inland capture fisheries in 1999 (Figure 7.1). A major proportion of global inland aquaculture is produced by countries in Asia. China alone reportedly gen- erates more than one quarter of the global total [Table 7.7], and has been responsible for most of the recent increase in this sector. In this particular instance the dividing line between aquaculture and capture fisheries is indistinct; no husbandry is involved beyond release of hatchery stock, and the fishery operates as a capture fishery”. However, national statistics do not ad- equately reflect the actual magnitude, location or importance of inland fisheries. The repor- ted inland capture production is certainly a gross underestimate because much of the catch is made far from recognized landing Inland water biodiversity 179 = nN N wo o i=) ou oO T T T 71 = oO T Million metric tons places where catches are monitored, and is consumed directly by fishers or marketed locally without ever being reported. The evi- dence suggests that actual capture fisheries catch may be twice or conceivably even three times the reported total, i.e. around 15-23 million metric tons per year”. Because a far higher proportion of inland fisheries than marine fisheries harvest is apparently used directly for human consumption (rather than production of oils and meals, often used for livestock feed), and because discards are believed to be negligible, it has been argued by some that the provision of foodfish from inland waters is not that much less than that from recorded marine catch”. Inland water capture fisheries, particularly in countries less developed industrially, certainly provide a staple part of the diet for many human communities. This is the case in West Africa generally, locally in East Africa, Aquaculture 2nd: 1998 Figure 7.1 Reported global inland fisheries production Source: FAO”! Table 7.7 Major inland fishery countries Note: The top ten producing countries in 1998. Source: FAO” China 2 280 000 India 650 000 Bangladesh 538 000 Indonesia 315 000 Tanzania 300 000 Russia 271 000 Egypt 253 000 Uganda 220 000 Thailand 191 000 Brazil 180 000 1909 WORLD ATLAS OF BIODIVERSITY Map 7.3 Major areas of diversity of inland water mollusks This map illustrates the location of areas regarded as globally important for diversity in the bivalve and gastropod mollusks of inland waters, taking into account species richness and local endemism Note: For numbered locations see Appendix 6 Source: Compiled using information and expertise provided by the IUCN/SSC Mollusc Specialist Group; first published in WeMC! Mollusk diversity eed Important areas and in parts of Asia and Amazonia. In some landlocked countries inland fisheries are of crucial importance, providing more than 50 percent of animal protein consumed by humans in Zambia” and nearly 75 percent in Malawi’. In the low-income food-deficit countries fish protein may be particularly important in times of food scarcity. It is impossible at the global level to carry out any meaningful analysis of the relative contribution of different species or species groups to inland capture fisheries because of the inadequacy of reporting. In FAO (Food and Agriculture Organization of the United Nations) statistics, by far the largest group recorded is ‘freshwater fishes not elsewhere included’, that is those that are completely unclassified other than being identified as finfishes. These make up just under half of all reported landing 105 ©134 by weight with a further 15 percent consisting of mollusks and crustaceans similarly classi- fied. The majority of the remaining catch is classified into broad species groups [e.g. cyprinids, characins, siluroids}, with only three individual fish species having annual reported global landings of more than 100 000 metric tons. These are the Nile perch Lates niloticus {ca 330 000 metric tons reported in 1997], Nile tilapia Oreochromis niloticus (226 000 metric tons in 1997] and the common carp Cyprinus carpio (100 000 metric tons in 1997). It is, however, evident that the importance of different species of freshwater finfishes varies considerably between different areas. In terms of food security for local subsistence or mixed market/subsistence communities, particularly in the tropics, there is an increas- ing amount of evidence that the diversity of 136 135 > 30 =) species harvested is in itself a major factor in ensuring a continuous food supply. Many of the species that contribute to these fisheries are often small and would be considered ‘trash’ fishes in orthodox fisheries, with much lower market value than larger [often non- native) species that might therefore be considered for introduction. However, these small species are easy to preserve and keep under local conditions and moreover are eaten whole, providing a valuable source of calcium and other minerals. Larger species, such as the introduced Nite perch in Lake Victoria, cannot be easily preserved locally and are in any case not eaten whole, leading to a danger of calcium deficiency. Fisheries for such species tend to become indus- trialized or semi-industrialized, producing fish products for commercial high-value Inland water biodiversity 181 2) 85 45 markets, often for export. While these may improve the balance of payments for the countries concerned, they may ultimately worsen the nutritional status of local people. Additionally, there are some indications that fish populations in mixed species fisheries are more stable over time, less susceptible to ‘boom and bust’ than those based on a small number of often introduced species. It is difficult rigorously to assess the condition of inland fish stocks because they appear able to respond rapidly to changing environmental conditions. However, there is a consensus that, regionally, most stocks are fully exploited and in some cases over- exploited. Exploitation has become more efficient because of new technologies, and developing infrastructure has allowed easier access to freshwater resources. Some stocks, 1822 WORLD ATLAS OF BIODIVERSITY a Le 2 a a Map 7.4 Major areas of diversity of selected inland water crustacean groups This map represents a preliminary assessment of areas believed to support high diversity among three of the several major crustacean groups that occur in inland waters, taking into account species richness and local endemism Note: For numbered locations see Appendix 6 Source: Compiled using information and expertise provided by the IUCN/SSC Inland Water Crustacean Specialist Group; first published in wemc! Species group eal Crabs Crayfish Fairy shrimps especially in river fisheries, appear to be in decline, but this is seemingly a result mainly of anthropogenic changes to the freshwater environment. In many parts of the world fishing has high recreational value, as well as being a means of food gathering. Locally, notably in the Amazon basin and in parts of Southeast Asia, capture for the ornamental fish trade may be an important source of income with potential impact on wild populations. Increasingly it is becoming difficult to distinguish between truly wild fish stocks and those that are artificially managed or enhanced in some way. Other harvested species Other exploited animal groups in inland waters are far less important globally than finfishes, but may still be highly significant. 93 92 ‘7a : Apart from crustaceans and mollusks, men- tioned above, these include: frogs (chiefly family Ranidae], exploited for food; crocodil- ians, hunted mainly for leather; freshwater chelonians, taken for food and to a lesser extent for medicinal purposes, particularly in eastern Asia; waterfowl which are hunted for recreation and for food; fur-bearing mam- mals, such as beavers Castor spp., otters (subfamily Lutrinae] and muskrats (Ondatra zibethicus and Neofiber alleni), taken for their skins; and manatees (family Trichechidae), taken mostly for food although also used non- consumptively on a small scale for biological control of weeds. Rice is the principal cultivated wetland plant of global importance to food security. Most of the relatively few plants associated with inland waters that are heavily exploited in the wild state are also marginal or wetland species. Some species (e.g. Aponogeton, in Madagascar) are collected for use as orna- mentals; reeds are used as building materials (e.g. thatch]; and some are collected for food or as medicines (e.g. Spirulina algae). Rhizomes, tubers and seeds (rarely leaves) of aquatic and wetland plants are used as a food source, mainly in less-developed regions where they can be important to food security in times of shortage, but globally they make a relatively minor contribution to human nutrition. Most important are some forms of edible aroid (Araceae), notably some cultivars of Colocasia (taro) and the giant swamp taro Cyrtosperma chamissonis that grow in flood- ed conditions and are important food crops in the Caribbean, West Africa and the Pacific islands. Conservation and collection of wild Inland water biodiversity 183 35 34 forms of these is considered a high priority. Sago palms Metroxylon spp. in Southeast Asia and the Pacific and watercress Rorippa nasturtium-aquaticum in Europe are other examples of cultivated aquatic plants, the wild relatives of which merit conservation. Aquatic plants have been widely used for medicinal purposes, documented for at least two millennia, but such use appears at present to be minor and probably of real significance in few areas. However, interest in ornamental or aquarium water plants is widespread and of some economic importance. OTHER MAJOR IMPACTS Physical alteration and destruction of habitat Destruction of inland water ecosystems is most simply effected by the removal of water. Although humans have always made use of | tome 184 WORLD ATLAS OF BilO:DIVER SITY freshwater systems, the last 200 years (spanning the Industrial Revolution, the growth of cities, the spread of high-input agriculture) have brought about transformations on an unprecedented scale. The global rate of water withdrawal rose steeply at the start of the 20th century, and further after mid-century. Major changes in the distribution of water have resulted mainly from withdrawals for irrigation and secondarily from domestic and industrial use. It has been estimated that humans use 26 percent of the total evapotranspiration from Aral Sea” Until the mid-20th century the Aral Sea was the world’s fourth largest inland water body {after the Caspian Sea, Lake Superior and Lake Victoria). Located within a catchment area of some 1.9 million km’ extending over six countries, the Aral is fed by two major rivers, the Amu Darya, rising in the Pamir, and the Syr Darya, rising in the Tien Shan. Starting in the 1960s, excess water withdrawal from these rivers, primarily for cotton irrigation, has severely affected the Aral Sea. Its area had reduced from more than 65 000 km’ to about 28 500 km’ in 1998, with volume falling by 75 percent, water level falling by around 20 meters {m), and salinity greatly increasing. In consequence, problems with drinking water quality and availability, and with dustborne pollutants, have severely affected the health status of the human population; the commercial fishery has collapsed; waterlogging and salinization have degraded agricultural lands; and the deltaic marshlands of the two feeder rivers have largely been replaced by sandy drylands. Mesopotamia”! Serial satellite images confirm a loss of around 90 percent of the lakes and marshlands in the lower Mesopotamian wetlands during the last three decades. The only significant permanent marshland remaining is in the Al-Hawizeh region. The large number of dams now present on upstream parts of the Tigris-Euphrates system may have contributed to this loss, but it appears to be primarily the result of major hydrological engineering works in southern Iraq, notably the completion of the major outfall drain (or ‘third river’) which diverts water to the head of the Gulf. This loss has placed further pressure on the Ma'dan (Marsh Arabs], now largely displaced within Iraq or in refugee camps in Iran. Recent information is scarce, but biodiversity in the region will inevitably have been affected, probably including the endemic form of smooth-coated otter (Lutrogale perspicillata, an otherwise oriental species, assessed as globally threatened). Azraq oasis” Groundwater extraction for urban needs in Jordan rose from about 2 million m’ in 1979 to about 25 million m’ in 1993, with an additional 25 million m’ per year used for agricultural irrigation. The important Azraq wetlands natural reserve, formerly extending over some 12 000 hectares, and a vital staging site for bird migrants, now supplies around one quarter of Amman’s water needs, and as a consequence has lost most of its marshland and migrant bird populations. 5 SS land surfaces and 54 percent of the accessible runoff’. Unregulated withdrawal can lead to the wholesale destruction of inland water eco- systems, as has occurred with the Aral Sea in central Asia (Box 7.2). Similarly, many wetlands have been completely destroyed by drainage, often for conversion to agriculture. Other fac- tors can modify or destroy particular habitats within inland water ecosystems. For example, canalization, usually undertaken to improve navigability, generally destroys riparian (shore- line] habitats while flood-control systems drastically alter regimes in floodplains”. Dams and reservoirs Dams, particularly large dams, have a major impact on the rivers on which they are built. They affect flow regimes, often dram- atically, destroy large areas of existing habitat (while at the same time creating new ones} and can catastrophically disrupt the life cycles of species that migrate up and down rivers”. Large dams are unevenly distributed across the world’s major catchments, with a partic- ularly high concentration in North America, especially within the contiguous states of the United States, where at least eight catch- ments have in them more than 100 large dams each. In contrast, small and medium- sized dams, which may cumulatively have as substantial an impact, are concentrated in eastern Asia, particularly China”. Dams may be primarily for the generation of hydro- electric power, or to create reservoirs for the storage of water, or both. The size of a dam is not necessarily directly related to the area or volume of the impoundment created or to its downstream impact. Pollution and water quality Assessment of anthropogenic changes in water quality is not always easy as such changes are invariably superimposed on natural background variations. Historically, a similar sequence of water quality issues has became apparent in both Europe and North America during rapid socioeconomic develop- ment over the past 150 years. Problems of fecal and other organic pollution were evi- dent in the mid-19th century, followed by salinization, metal pollution and eutrophi- cation in the first half of the 20th century, with radioactive waste, nitrates and other organic micropollutants, and acid rain most promin- ent in recent decades. Newly industrializing countries are likely to face these problems over a much more compressed period, and typically without the capacity to monitor and analyze water quality, or manage water use appropriately”. Different kinds of pollutants appear to affect different classes of water system to differing extent”. With regard to quality for human use, contamination by pathogens of fecal origin is the major problem in river systems, and eutrophication probably the most widespread problem affecting lake and reservoir waters”. Acid deposition through precipitation has been recognized as a regional transboundary phenomenon since the 1960s. Industrial emissions of sulfur and nitrogen oxides (SO,, NO,), mainly a result of fossil fuel combustion, are the principal source of acid rain. Most evidence of acid rain and its effects relates to North America and Europe, but emission rates are rising steeply in rapidly industrializing countries elsewhere. Acid rain in one country may be a consequence of compounds released into the atmosphere by industry in another country hundreds of kilometers distant. The geology, soil and vegetation of drainage basins strongly influence the acidification process. Acid rain has been shown to decrease species diversity in lakes and streams but has not been implicated in any recorded species extinction or any major species decline. It has not yet been shown to be a significant issue in tropical freshwaters, where global freshwater diversity is concentrated’. Sedimentation Removal or extension of forest cover, or any anthropogenic interference with soils and landcover (e.g. agriculture, urbanization, road construction, mining), modifies the rate of runoff from catchment slopes and also the density of particles carried in the drainage system. All moving waters carry some mass of suspended material, and there is consider- able natural variation in this in space and time, but logging can increase sediment load Inland water biodiversity 185 by up to 100 percent for a short period, and 20-50 percent over the longer term. Sediment reaching lakes will be deposited and in effect enter long-term storage; depending on water velocity, sediment in rivers will settle out on floodplains or other parts of the course, or be carried into the coastal marine environment. Increased sedimentation can have several effects on aquatic biodiversity: deposition can radically change the physical environment of species restricted to particular conditions of depth, light penetration and velocity; it is a major carrier of heavy metals, organic pollu- tants, pathogens and nutrients; and it can interfere mechanically with respiration in gill-breathing organisms’. The endemic cichlid fishes of the African Great Lakes rely on complex visual signals in breeding, and reduced water clarity because of sedimen- tation is suspected to be affecting their breeding success. Introduced species Unplanned or poorly planned introduction of non-native species and genetic stocks is a major threat to freshwater biodiversity. Such introductions can have negative or positive effects on fishery production; it is a reasonable assumption that all successful introductions will have an impact on existing population size and community structure, and many changes are likely to be undesirable”. The incomplete a NE rece ee Although some factories have introduced cleaner production procedures, industrial emissions are still the principal source of acid rain. rae Smee 18% WORLD ATLAS OF BIODIVERSITY - a a 1 000 800 }+- 2) cc 2 3) 3 600 £ | Running total c | zs) & 400 a E =) Zz 200 - | Decade total = 0 camel ees T EE Inn ee Ie aa oO 2) 2) w Ww w Ww Ww Ww 1) w wn Ww Ww 2) c V9) (=) (=) oO (=) i=) i=) o fo So (=) (=) f=} =} (=) vo feo) vp) Se) i= feo} on (f=) 7 N ise) = ie) Se) i oc (3) = foo) foe) co foe} co o o oa oa o o~ lon o lon eile (2 eee Sc ages os a Figure 7.2 information available suggests that although a Inland water fish introductions Source: Welcomme™” Figure 7.3 Freshwater population trends Note: A simplified representation of the average population change ina sample of 194 inland water species, see text Source: Loh”” 140 - 120 F significant number of fish introductions took place during the 19th century, the three dec- ades starting with the 1950s were particularly important (Figure 7.2]. A classic example of the effect of introduced species is the impact of the Nile perch Lates niloticus on the haplochromine cichlids of Lake Victoria discussed further below. Several species of aquatic plant, particular free-floating species that are able to spread rapidly by vegetative growth (most notoriously the South American water hyacinth Eichhornia crassipes}), but also other forms, have dispersed widely over the globe and become major pest species. They block in 40 1970 1975 =I I 1980 T T T 1 1985 1990 Is) 1999 drainage channels, sluices and hydroelectric installations, impede boat traffic and hinder fishing. In recent decades the question of how best to control or eradicate pest species has been the foremost issue in conservation and management of aquatic plants’. THE CURRENT STATUS OF INLAND WATER BIODIVERSITY As with marine species, assessment of the status of wholly aquatic inland water species is hampered by difficulties of direct obser- vation. However, because these species also in general have far more restricted ranges than marine species, it is easier to infer their status from assessment of habitat condition and from sampling efforts. Amphibious or surface-dwelling species may be relatively easier to monitor. Where such species are of economic importance - as for example with those European and North American water- fowl that play a role in the recreational hunting industry - they may be among the best monitored of all wild species. Threatened and extinct species In the few cases where elements of inland water faunas - usually fishes - have been studied in any detail, it has generally been found that more species than suspected are threatened or cannot be re-recorded”. Among the 20 or so countries where the entire inland water fish fauna has been evaluated, an average of 17 percent of the species are regarded as globally threatened (categories ‘critically endangered’, ‘endangered’ or ‘vulnerable’ in the IUCN (World Conservation Union) threat categorization system“) {Table 7.8). A far larger proportion is likely to be in local decline, although not in danger of global extinction. The proportion of inland water chelonians that is believed threatened is even higher: 99 such species were categorized as threatened in 2000, equivalent to about 60 percent of the number of inland water chelon- ians listed in Table 7.6. Amongst mammals and birds the pro- portions are considerably lower, probably because many semi-aquatic species are able to disperse from one inland water body to another relatively easily. Nevertheless, some- what more species than average for the groups as a whole are regarded as threatened. Table 7.9 shows the category and taxonomic distribution of threatened inland water vertebrates. Two groups of species, the Lake Victoria cichlid fishes and the Mobile Bay drainage gastropod mollusks, serve as exemplary case studies illustrating the major threats faced by inland water biota worldwide. Lake Victoria, the largest tropical lake in the world, provides a classic example of the potential negative impacts of species intro- ductions. Until some 30 years ago, when the large top predator, the Nile perch Lates niloticus, was introduced, the lake supported an exceptional ‘species flock’ of more than 300 species of haplochromine cichlid fishes, as well as smaller numbers from other families. The cichlids are of enormous interest in the study of evolutionary biology. Not all the species have yet been formally described; many of these are known among aquarists and others only by informal common names. At least half of the native species are believed to be extinct or so severely depleted that too few individuals exist for the species to be harvested or recorded by scientists. Predation by the Nile perch is believed to be the major cause of this decline, but important additional factors include increas- ing pollution and sediment load, excess fishing pressure, and possible competition from intro- duced tilapiine cichlids. The lake itself has now become depleted of oxygen, and a shrimp tolerant of oxygen-poor waters provides a major food source for the Nile perch. In recent years the Nile perch and one of the introduced tilapiines have formed the basis of a high- yielding fishery and an important national and export trade. However, it is thought unlikely that such high yields will be maintained because of continued overfishing and the suspected instability of the already highly disturbed lake ecosystem’. Dam construction is the prime cause of extinction in the gastropod fauna of the Mobile Bay drainage in Alabama, United States. Historically, the freshwater snail fauna of Mobile Bay basin was probably the most diverse in the world, followed by that of the Inland water biodiversity 187 ” or RE I ge SE ee Mekong River. Nine families and about 118 species were known at the turn of the century to occur in the Mobile Bay drainage. Several genera and many species were endemic, particularly in the Pleuroceridae. Recent surveys suggest at least 38 species are extinct (32 percent); decline in species richness ranges between 33 percent and 84 percent in the main river systems. The richest fauna was in the Coosa River and this system has undergone the greatest decline [from 82 to 30 species]. Almost all the snail species presumed extinct were members of the Pleuroceridae and grazed on plants growing on rocks in shallow oxygen-rich riffle and shoal zones. The system has 33 major hydro- electric dams and many smaller impound- ments, as well as locks and flood control structures. A combination of siltation behind dams and submergence of shallow water shoals has removed the snails’ former habitat. Where habitat remains it has dimin- ished in area and become fragmented”. The inland water living planet index An impression of the overall trend in a large sample of species for which indicators of USA 822 Mexico 384 Australia 216 South Africa 94 Croatia 64 Turkey 174 Greece 98 Madagascar 4) Canada 177 Papua New Guinea 195 Romania 87 Italy 45 Bulgaria 72 Hungary 79 Spain 50 Moldova 82 Portugal 28 Sri Lanka 90 Slovakia 62 Japan 150 iif Table 7.8 Numbers of threatened freshwater fishes in selected countries Notes: These are the 20 countries whose fish faunas have been evaluated completely, or nearly so, and which have the greatest number of globally threatened freshwater fish species. The estimates of total fish species present are all approximations. Source: Total species estimates from UNEP-WCMC database; threatened species data from online Red List http://www.redlist.org {accessed March 2002) 120 15 82 21 27 13 24 26 22 34 22 13 19 19 13 32 12 7 11 6 11 13 11 24 11 15 10 13 10 20 9 11 9 32 9 10 9 15 9 6 Table 7.9 Taxonomic distribution and status of threatened inland water vertebrates Notes: Family selection based on Table 7.6 Numbers refer to species (not subspecies and geographic populations} recorded in the Red List database as occurring in freshwater; only the birds and mammals have been comprehensively assessed for species at risk. i Emydidae here includes batagurine turtles sometimes treated asa separate family (several of the species are primarily terrestrial not aquatic] Source: Status categories from 2000 Red List database, www.redlist org“ (accessed February 2002) 188 WORLD ATLAS OF BIODIVERSITY og a lum and Craniata - fishes Family Common name Petromyzontiformes Lampreys 1 2 Carchariniformes Ground sharks 1 i Pristiformes Sawfish 2 3 Myliobatiformes Rays 5 2 Acipenseriformes Sturgeons 6 10 8 Atheriniformes Silversides 6 5 31 Beloniformes Needlefishes, sauries, etc. 2 3 8 Characiformes Characins i 1 Clupeiformes Herrings and anchovies 2 3 Cypriniformes Carp, minnow, loaches 42 37 117 Cyprinodontiformes Rivulines, killifish, etc. 18 20 26 Gasterostiformes Sticklebacks i Ophidiiformes Pearlfishes, etc. 6 Osteoglossiformes Bonytongues 1 Perciformes Perches, etc. 51 25 104 Percopsiformes Trout-perches, cavefishes 1 3 Salmoniformes Salmonids 8 8 22 Scorpaeniformes Gurnards, scorpionfishes, etc. 2 4 Siluriformes Catfishes 8 7 22 Synbranchiformes Swamp eels 1 Syngnathiformes Pipefishes, seahorses, etc. 1 Craniata - Amphibia Caudata Cryptobranchidae Giant salamanders and hellbenders 1 Proteidae Mudpuppies and olm 1 Anura Pipidae Clawed frogs and pipid toads 1 Craniata - Reptilia Chelonia Carettochelidae Pig-nosed soft-shelled turtle i Chelidae Austro-american side-necked turtles 3 4 7 Chelydridae Snapping turtles 1 Dermatemydidae Central American river turtle 1 Emydidae’ Pond and river turtles 13 22 16 population change are available can be derived from the WWF living planet index. This method is designed to represent the change in the ‘average species’ in the sample from one five-year interval to the next, starting in 1970. The inland waters sample represents 194 species of mammals, birds, reptiles and fishes, and the index suggests a significant declining trend over the last three decades of the 20th century (Figure 7.3)”. The sample includes a large number of wetland and water margin species in addition to truly aquatic forms. The declining trend in the inland water sample is a little steeper than the equivalent marine index, and sub- stantially steeper than the terrestrial index. Assessment of the status of inland water ecosystems Indicators of habitat condition in river catchments While the living planet index methodology provides an indication of global trends in inland water biodiversity, an alternative approach aims to assess the overall condition of inland water ecosystems. In one approach to a global assessment’ two high-order indicators of likely habitat condition in different river catchments were combined. First, a wilderness measure for each river catchment was calculated using the wilderness index methodology developed by the Australian Heritage Commission‘”“ (see Chapter 4 and particularly Map 4.5). This Inland water biodiversity 189 Kinosternidae Mud and musk turtles 4 Pelomedusidae Side-necked turtles 2 6 Trionychidae Soft-shelled turtles 4 5 6 Crocodilia Alligatoridae Caimans and alligators i Crocodylidae Crocodiles 3 2 3 Gavialidae Gharial and false gharial i Craniata - Aves Anseriformes Dendrocygnidae Whistling-ducks 1 Anatidae Ducks, swans and geese 5 7 12 Charadriiformes Charadriidae Plovers, etc. i 1 2 Laridae Gulls, terns, skua, auks 1 Rhynchopidae Skimmers 1 Scolopacidae Curlews, etc. 1 Cicontiformes Ardeidae Herons, egrets 3 5 Ciconiidae Storks 3 2 Threskiornithidae — Ibis, spoonbill 2 2 Phoenicopteridae Flamingos 2 Gruiformes Heliornithidae Limpkin and sungrebes i Rallidae Rails, gallinules and coots i i 6 Figure 7.4 Pelicaniformes Pelecanidae Pelicans and shoebill 1 River basin richness and Passeriformes Cinclidae Dippers. 1 vulnerability Podicipediformes Podicipedidae Grebes 2 2 A F Notes: Each symbol Craniata = Mammalia : ; , represents a a basin Artiodactyla Hippopotamidae © Hippopotamus 1 scoredion fish family Carnivora Mustelidae Mustelids, otters 2 3 aie neceeadibeein Phocidae Earless seals l { vulnerability; the darker Cetacea Platanistidae River dolphins 1 2 1 symbols score high on both Insectivora Soricidae Shrews 2 i counts and may be regarded Talpidae Moles and desmans 2 as high priority. See Map 7.5 Tenrecidae Tenrecs and otter shrews 4 and Table 7.10. Rodentia Muridae Mice, voles, etc. 1 5 3 Sirenia Trichechidae Manatees 3 ; Source: WCMC provides a measure of the spatial extent of 30 human impacts in the land area of the catchment as a whole and does not directly reflect the condition of riverine ecosystems. 20 ee Secondly, a national water resource vulnera- ® e bility index (WRVI)” developed on the basis of “ S three water resource stress indices (reliability, use-to-resource and coping capacity) was resolved at catchment level {by measuring that proportion of each catchment that lies within any given country and weighting this proportion i by the national WRVI). This provides an indirect measure of vulnerability for each catchment. By normalizing and combining these two measures, a single value representing vulner- 0.8 06 04 02 0 02 04 06 08 ability or stress level for each major river Overall vulnerability Family richness ' = oO Map 7.5 Priority river basins A possible scheme for prioritizing river basins at global level. Major catchment basins are assessed for fish diversity {at family level) and for vulnerability {a combined indication of disturbance and potential water stress). Systems with both high diversity and vulnerability are proposed as high priority for investment and management action; those with low diversity and little disturbance are lower priority. See text and source. Source: WCMC'. Table 7.10 Thirty high-priority river basins Notes: These are 30 river basins that support high biodiversity {assessed as fish family richness] and are most vulnerable to future pressures (have a low wilderness score, and are high on the water resource vulnerability index). See text for further explanation and Figure 7.4. Basins are listed in alphabetical sequence. Ca Irrawaddy Cauvery Krishna Chao Phraya Ma Gambia Magdalena Ganges- Mahanadi Brahmaputra Mekong Godavari Narmada Indus Niger 199 WORLD ATLAS OF BIODIVERSITY Priority catchment can be calculated. Overall, the pattern mapped agrees well with what might intuitively be expected, with few evident anomalies. Globally, the most stressed catch- ments are to be found in South Asia (the Indian subcontinent], the Middle East and western and northcentral Europe. The least stressed Nile Senegal Pahang Sittang Parana Song Hong (Red] Parnaiba Tapti Penner Tembesi-Hari Perak Uruguay Salween Volta Sao Francisco are those in the northwestern part of North America. Further refinements of this analy- sis would involve applying water resource vulnerability measures for individual catch- ments” rather than to countries and incor- porating measures of direct impacts on inland waters, in particular water quality and the number and kind of dams. The number of freshwater fish families present in each basin can be calculated from the family density surface (Map 7.1) in order to provide an indication of biodiversity value. Plotting family number against vulnerability allows the basins with high diversity and high vulnerability to be identified, and these can reasonably be regarded as global priorities for management intervention designed to minimize biodiversity loss (Figure 7.4, Table 7.10, Map 7.5). EE §6=—hr Habitat condition in lakes One semi-quantitative study attempted to evaluate the changing condition of freshwater lakes during the last three decades of the 20th century”. A baseline was provided by Project Aqua, a project initiated by the Societas Internationalis Limnologiae in 1959, which collated and later published” information provided by national and regional specialists in relation to more than 600 water bodies. Many of these lakes were treated in later infor- mation sources relating to the 1980s and 1990s", and in some 93 cases it was possible to make an assessment that, although imprecise, is likely to be indicative of changing conditions. Each lake was scored according to whether its condition appeared to have deteriorated (or impacts increased), or to have improved, or whether no change [or no new Inland water biodiversity 19 information) was reported. Improvement was reported in a very small number of lakes, but the overwhelming trend was for a deterior- ation in conditions (see Figure 7.5). MG Better Not known 100 Asia (24) Africa (20) America (9) Central & South Australia Figure 7.5 Changes in condition of a sample of freshwater lakes between 1950s and 1980s Note: Number following area name is number of lakes in sample. 52, 53 Source: Various sources 3 I) Worse TOTAL (93) Europe (3) (37) 1972 WORLD ATLAS OF BIODIVERSITY SS TS RR TER St PR a REFERENCES 1 WCMC 1998. Freshwater biodiversity: A preliminary global assessment. By Groombridge, | B. and Jenkins, M. WCMC-World Conservation Press, Cambridge. 2 Gorthner, A. 1994. What is an ancient lake? Speciation in ancient lakes. In: Martens, K., Goddeeris, B. and Coulter, G. (eds). Archiv ftir Hydrobiologie. Ergebnisse der Limnologie 44: 97-100. | 3 Horne, J.A. and Goldman, C.R. 1994. Limnology. McGraw-Hill Inc., New York. | 4 Guégan, J.-F., Lek, S. and Oberdorff, T. 1998. Energy availability and habitat heterogeneity predict global riverine fish diversity. Nature 391: 382-384. 5 Oberdorff, T., Guégan, J.-F. and Hugueny, B. 1995. Global scale patterns of fish species richness in rivers. Ecography 18: 345-352. 6 Kottelat, M. and Whitten, T. 1996. Freshwater biodiversity in Asia with special reference to fish. World Bank Technical Paper No. 343. 7 Olson, D.M. et al. (eds) 1997. Freshwater biodiversity of Latin America and the Caribbean: A conservation assessment. Proceedings of a workshop. World Wildlife Fund, Washington DC. 8 Abell, R. et al. 1998. A conservation assessment of the freshwater ecoregions of North America. Final report submitted to the US EPA, April. World Wildlife Fund-US. 9 Olson, D.M., and Dinerstein, E. 1998. The Global 200: A representation approach to conserving the earth's distinctive ecoregions. Conservation Biology 12: 502-515. 10 WWF. Online at http://www.panda.org/resources/programmes/global200/pages/list.htm {accessed March 2002). 11 Watter, G.T. 1992. Unionids, fishes, and the species-area curve. Journal of Biogeography 19: 481-490. 12 Tittley, |. 1992. Contribution to chapter 7, Lower plant diversity. In: World Conservation Monitoring Centre. Global biodiversity. Groombridge, B. [ed.} Chapman and Hall, London. 13 Goh, T.K. and Hyde, K.D. 1996. Biodiversity of freshwater fungi. Journal of Industrial Microbiology 17: 97-100. 14 Shearer, C.A. 1993. The freshwater ascomycetes. Nova Hedwigia 56: 1-33. 15 Shearer, C.A. Freshwater ascomycetes and their anamorphs. Available online at http://fm5web.life.uiuc.edu:23523/ascomycete/ (accessed March 2002). 16 Sculthorpe, C.D. 1967. The biology of aquatic vascular plants. Edward Arnold, London. 7 Palmer, M. et al. 1997. Biodiversity and ecosystem processes in freshwater sediments. Ambio 26(8): 571-577. 18 Nelson, J.S. 1994. Fishes of the world. 3rd edition. John Wiley and Sons, Inc., New York. 19 UN (CSD] 1997. Comprehensive assessment of the freshwater resources of the world. Economic and Social Council. Fifth Session, 5-25 April. E/CN.17/1997/9. Available online at http://www.un.org/esa/sustdev/csd.htm {accessed April 2002). 20 FAO 1996. Food production: The critical role of water. Technical background document. World Food Summit. Available online in pdf at http://www.fao.org (accessed April 2002). 21 FAO 2000. The state of world fisheries and aquaculture 2000. Fisheries Department, Food and Agriculture Organization of the United Nations, Rome. Available online at http://www. fao.org/DOCREP/003/X8002E/X8002E00.htm (accessed April 2002). 22 FAO 1999. Review of the state of world fishery resources: Inland fisheries. FAO Fisheries Circular No. 942. FIRI/C942. 23 Coates, D. 1995. Inland capture fisheries and enhancement: Status, constraints and prospects for food security. Paper presented at international conference on sustainable contribution of fisheries to food security. Kyoto, Japan, 4-9 December 1995. Food and Agriculture Organization of the United Nations, Rome. KC/FI/95/Tech/3. 24 Borgstrom, R. 1994. Freshwater ecology and fisheries. In: Balakrishnam, M., Borgstrom, R. | and Bie, S.W. Tropical ecosystems, pp. 41-69. International Science Publishers, Oxford and | IBH, New Dehli. a a IEE I SOO SEE 25 Scudder, T. and Conelly, T. 1985. Management systems for riverine fisheries. FAO Fisheries Technical Paper No. 263. Food and Agriculture Organization of the United Nations, Rome. 26 Munthali, S.M. 1997. Dwindling food-fish species and fishers’ preference: Problems of conserving Lake Malawi's biodiversity. Biodiversity and Conservation 6: 253-261. 27 Postel, S.L., Daily, G.C. and Ehrlich, P.R. 1995. Human appropriation of renewable fresh water. Science 271: 785-788. 28 L'Vovich, M.I. et al. 1990. Use and transformation of terrestrial water systems. In: Turner ll, B.L. The earth as transformed by human action. Global and regional changes in the biosphere over the past 300 years, pp. 235-252. Cambridge University Press with Clark University. 29 FAO AQUASTAT. Online at http://www.fao.org/landandwater/aglw/aquastat/ regions/fussr/main8.htm (accessed March 2002). 30 Evans, M.I. 2001. The ecosystem. Draft paper for conference, Iraqi Marshlands: Prospects, London, 21 May 2001. Available online at http://www.amarappeal.com/documents/ Draft_Report.pdf (accessed April 2002). 31 UNEP 2001. UN study sounds alarm about the disappearance of the Mesopotamian marshlands. 18 May. Available online at http://www.grida.no/inf/news/news01/ news42.htm, and also see http://www.grid.unep.ch/./activities/sustainable/ tigris/marshlands/marshlands.pdf {accessed April 2002). 32 Fariz, G.H. and Hatough-Bouran, A. 1998. Population dynamics in arid regions: The experience of the Azraq Oasis Conservation Project (Case study: Jordan). In: de Sherbinin, A. and Dompka, V. [eds]. Water and population dynamics: Case studies and policy implications. Available online at http://www.aaas.org/international/ehn/waterpop/ front.htm [accessed April 2002). 33 Dynesius, M. and Nilsson, C. 1994. Fragmentation and flow regulation of river systems in the northern third of the world. Science 266: 735-762. 34 Avakyan, A.B. and lakovleva, V.B. 1998. Status of global reservoirs: The position in the late twentieth century. Lakes and Reservoirs: Research and Management 3: 45-52. 35 Meybeck, M. and Helmer, R. 1989. The quality of rivers: From pristine stage to global pollution. Palaeogeography, palaeoclimatology, palaeoecology 75: 283-309. 36 Chapman, D.V. 1992. Water quality assessment. Chapman and Hall, London. 37 UNEP 1991. United Nations Environment Programme. Freshwater pollution. UNEP/GEMS Environment Library No. 6. 38 UNEP 1995. United Nations Environment Programme. Water quality of world river basins. UNEP Environment Library No. 14. 39 Moyle, P.B. 1996. Effects of invading species on freshwater and estuarine ecosystems. In: Sandlund, O.T. et al. (eds). Proceedings of the Norway/UN Conference on Alien Species, Trondheim, 1-5 July 1996, pp. 86-92. Directorate for Nature Management and Norwegian Institute for Nature Research. Trondheim. 40 Moyle, P.B. and Leidy, R.A. 1992. Loss of biodiversity in aquatic ecosystems: Evidence from fish faunas. In: Fielder, P.L. (eds). Conservation biology, the theory and practice of nature conservation, preservation and management, pp. 129-169. Chapman and Hall, New York and London. 41 Stiassny, M. 1996. An overview of freshwater biodiversity with some lessons learned from African fishes. Fisheries 21(9}: 7-13. 42 Reinthal, P.N. and Stiassny, M.L.J. 1991. The freshwater fishes of Madagascar: A study of an endangered fauna with recommendations for a conservation strategy. Conservation Biology 5(2): 231-243. 43 Kirchhofer, A. and Hefti, D. (eds) 1996. Conservation of endangered freshwater fish in Europe. Advances in Life Sciences. Birkhauser Verlag, Basel. Inland water biodiversity 193 SECA Wh 19%8 WORLD ATLAS OF BIODIVERSITY a 44 Hilton-Taylor, C. (compiler) 2000. 2000 IUCN Red List of threatened species. |UCN-the World Conservation Union, Gland and Cambridge. Also available online at http://www.redlist.org/ {accessed April 2002). 45 Bogan, A.E., Pierson, J.M. and Hartfield, P. 1995. Decline in the freshwater gastropod fauna in the Mobile Bay basin. In: LaRoe, E.T. et al. (eds). Our living resources: A report to the nation on the distribution, abundance, and health of U.S. plants, animals, and ecosystems, pp. 240-252. US Department of the Interior, National Biological Service, Washington DC. 46 Loh, J. (ed.] 2000. Living planet report 2000. WWF - World Wide Fund for Nature, Gland. 47 Lesslie, R., in litt., 30 May 1998. 48 Lesslie, R. and Maslen, M., 1995. National wilderness inventory, handbook and procedures, content and usage. 2nd edition. Commonwealth Government Printer, Canberra. 49 Raskin, P. et al. 1997. Water futures: Assessment of long-range patterns and problems. Background Report No. 3 of Comprehensive Assessment of the Freshwater Resources of | the World. Stockholm Environment Institute. | 50 Alcamo, J., Henrichs, T. and Rosch, T. 2000. World water in 2025. Global modeling and scenario analysis for the World Commission on Water for the 21st Century. Kassel World Water Series. Report No. 2. University of Kassel, Centre for Environmental Systems Research. 51 Loh, J. et al. 1998. Living planet report 1998. WWF-World Wide Fund for Nature, Gland. 52 Luther, H. and Rzoska, J. 1971. Project Aqua: A source book of inland waters proposed for conservation. IBP Handbook No. 21. IUCN Occasional Paper No. 2. International Biological Programme. Blackwell Scientific Publications, Oxford and Edinburgh. 53 For example, Scott, D.A. {ed.} 1989. A directory of Asian wetlands. |\UCN, Gland and Cambridge; Scott, D.A. and Carbonell, M. 1986. A directory of neotropical wetlands. |UCN, Gland and Cambridge; International Lake Environment Committee (ILEC}, World lakes database, available online at http://www. ilec.or.jp/database/database.html {accessed March 2002). 94 Anon 1978. Centre for Natural Resources, Energy and Transport of the Department of Economic and Social Affairs, United Nations. Register of International Rivers. Water Supply and Management 2: 1-58. {Special issue). Pergamon Press. 55 Shiklomanoy, |.A. 1998. World water resources at the beginning of the XXIst century. Report to UNESCO (Division of Water Sciences). Available online in summary form at http://webworld.unesco.org/water/ihp/db/shiklomanov/summary/htmlU/summary.html {accessed March 2002). 56 Martens, K., Goddeeris, B. and Coulter, G. (eds) 1994. Speciation in ancient lakes. Archiv fur Hydrobiologie. Ergebnisse der Limnologie 44. 57 Hutchinson, G.E. 1993. A treatise on limnology. Vol. IV. The zoobenthos. John Wiley and Sons, Inc., New York. 58 Eschmeyer, W.N. et al. 1998. A catalog of the species of fishes. Vols 1-3. California Academy of Sciences, San Francisco. Available online in searchable format at http://www.calacademy.org/research/ichthyology/catalog/fishcatsearch.html {accessed January 2001). 59 Welcomme, R.L. [compiler] 1988. International introductions of inland aquatic species. FAO Fisheries Technical Paper No. 294. Food and Agriculture Organization, Rome. 60 Berra, T.M. 2001. Freshwater fish distribution. Academic Press, San Diego and London. | 61 Johnson, T.C. et al. 1996. Late Pleistocene desiccation of Lake Victoria and rapid evolution of cichlid fishes. Science 273: 1091-1093. 62 Snoeks, J. 2000. How well known is the ichthyofauna of the large East African Lakes? Advances in Ecological Research 31: 17-38. Global biodiversity: responding to change 195 ES AE LEE EF EE NS PS NS AI TE ETE LE s. Global biodiversity: responding to change of larger animals and in the naturalness of landscapes, dates primarily from the turn of the 20th century. Although some early measures were taken, prompted by pioneering conservation organizations, concerted international effort did not develop until mid-century. Since this period, actions have tended to focus on conservation of individual species or of large areas of habitat, as national parks and other protected areas. During the 1970s several international agreements aimed at conserving wild species and habitats were agreed and entered into force. During the 1980s the word ‘biodiversity’ was coined, and a new paradigm formulated, aiming to integrate biodiversity conservation with sustainable human development. In 1992 the pivotal Earth Summit was held in Rio, culminating in agreement on the Convention on Biological Diversity, and Agenda 21 - a plan of action for sustainable human development. After a decade of planning and implementation, the status of some species and the condition of some ecosystems have improved. In some areas, this has been achieved by restoration of degraded ecosystems, an approach likely to be of increasing importance in the future. However, given future trends in human population growth and development, it seems probable that pressures on biodiversity will continue to intensify in coming decades. In many parts of the world, the most significant challenge to conservation will be to minimize losses of biodiversity while improving the livelihood of human populations, particularly those experiencing severe poverty. C ONCERN OVER THE RATE OF CHANGE IN THE BIOSPHERE, particularly in populations BIODIVERSITY CHANGE Change is a dominant theme in the biosphere; species have diversified and become extinct throughout the history of life, and habitats too have expanded and declined, along with changes in community composition, climate and landforms. However, the recent rate of change in global biodiversity appears higher than that prevailing over most periods of geological time. Earlier chapters have indi- cated how in well-assessed groups of organisms, such as birds and mammals, a significant proportion of species now appears to be threatened with global extinction. Countless other species exist in reduced numbers and as fragmented populations, many of which are threatened with extinction at national or more local scale. The rate of extinction over the past few centuries, so far as this can be reliably estimated, appears to be much higher than the average background rate estimated from the fossil record. Whatever proximate causes may be impli- cated, the increasing numbers and material aspirations of the human species, the in- creasing burden of waste and continuing inequities in the distribution of wealth and resources, together appear to drive most con- temporary biodiversity loss. Despite abundant evidence of change in biodiversity, often involving radical modifi- cation of landcover or water bodies, much of the relevant information on the status of species or populations is qualitative or anec- dotal in nature. At the level of individual species, a qualitative assessment of trends in numbers or range may be adequate evidence of the need for management intervention, or a 196 WiOiTRIE DT ATIEAS OF IB NOIDIMiERSiaiy Concern about the decline of species became the focus of sustained international attention only in the last 30 years guide to its effectiveness, but it remains diffi- cult to develop a comprehensive and quantit- ative view of global species trends. The relatively few large-scale population monitor- ing programs that exist have tended to concentrate on marine fishery stocks, on par- ticular groups [e.g. farmland birds) or the larger threatened species of animals. The res- ulting series of data are not always directly comparable, and often not amenable to pres- entation in a manner appropriate for assisting global and regional planning. Much recent discussion’ has focused on the design of biodiversity indicators that could serve this purpose, but few operational systems yet exist. This chapter first provides an overview of current approaches to conservation, focusing on species, areas and ecosystems, and the particular role of protected areas, as well as the potential for ecological restoration. The international dimension is then considered, with specific reference to multilateral environ- mental agreements and conventions relating to biodiversity. Finally, changes in global biodiversity that might be anticipated in future are explored through description of recently developed scenarios. Such changes highlight the challenges facing current and future efforts aimed at the conservation and sus- tainable use of global biodiversity. RESPONSES TO BIODIVERSITY CHANGE Although concern about the decline of species and loss of undisturbed habitats arose within developed industrialized societies around the start of the 20th century, they became subject to sustained international attention only during the 1970s. In general, major sectors that used biodiversity, such as forestry and fisheries, failed to incorporate consideration of bio- diversity into their planning and regulations. The issue was widely perceived as marginal to other concerns, and conservation of nature was commonly seen as an impediment to human development. The past three decades have been marked by the emergence of concerted responses to the crisis in biodiversity at all levels from the local to the global. Civil society, largely in the form of a hugely diverse and increasingly sophisticated network of non-governmental organizations [NGOs], has undoubtedly been the driving force for most of this change [see Box 8.1). Response to pressure from civil society can be seen at both governmental and intergovernmental levels. Much of the progress that has been made in recent decades can be attributed to the vision articulated in the World Conservation Strategy by the International Union for Conservation of Nature and Natural Resources, the United Nations Environment Programme and the World Wildlife Fund in 1980. This document helped set the conservation agenda during the period leading up to the UN Conference on Environment and Development [the Earth Summit) in Rio de Janeiro in 1992. It em- phasized conservation for development, and embraced the notion of the sustainable use of natural resources as a means of achieving this. Maintenance of biological diversity as part of a functioning biosphere was presented as a fundamental prerequisite of sustainable hu- man development, not an impediment to it. On the ground, this was manifested in increasing numbers of integrated conservation and de- velopment projects in developing countries. The Earth Summit was undoubtedly the major environmental milestone of the 1990s, and from it emerged the Convention on Biological Diversity, which entered into force in December 1993. At the start of the 21st century concerns about loss of species and habitats have intensified. Additional issues have emerged, such as climate change, with its inevitable effects on coastal, montane and other eco- systems, and the ability of humans directly to modify gene structure and expression in living organisms, with debate over the risks involved when genetically modified organisms are re- leased into the biosphere. Protecting species Maintaining biodiversity requires that viable populations of diverse organisms are main- tained in viable ecosystems. This may involve activities carried out on site or off site. While the latter (ex situ) has an important role, as with the seed banks and germplasm collec- tions for agricultural plants, it can only have very limited application. The former (in situ) is essential for the vast majority of organisms, and ecosystem conservation self-evidently depends entirely on maintaining environ- ments in which communities of organisms can interact and evolve. Because the vast majority of the world’s biodiversity exists within the territorial boun- daries established by nations, most conser- vation action is carried out within the policy and legal systems established by national governments (or in a few instances, by regional or provincial governments). A wide range of national measures exists, varying to some extent from country to country depen- ding on the social, political and economic environment. Despite this variety, most such measures involve regulation of the taking, possession and trade in a set of species, typically named and listed in legislation. The protection of wild fauna has generally been given much more attention than the pro- tection of wild flora, but most modern examples of species-specific legislation cover both flora and fauna, as exemplified by the US Endangered Species Act, passed in 1973. The species approach has been supported at the international level by the IUCN {World Conservation Union) Red Data Book program of activities and information tools. The focus is on documenting and disseminating infor- mation on species at risk of extinction, and on Global biodiversity: responding to change Some of the earliest established non-governmental organizations remain influential today. The Sierra Club was founded in the United States in 1892, with a focus on maintaining wilderness areas in North America. The Society for the Preservation of the Wild Fauna of the Empire - now Fauna and Flora International (FFI] - was founded in 1903 in order to safeguard southern Africa's declining large mammal populations and is now worldwide in scope. Initiated in 1962, Operation Oryx was a landmark project of FFI, and averted extinction of the Arabian oryx by means of a captive breeding and reintroduction program. The International Council for Bird Preservation - now BirdLife International (BLI) - was founded in 1922. It has played a lead role in advancing standards for conservation information and has promoted grassroots involvement in conservation through its worldwide membership structure. The International Union for the Protection of Nature - now IUCN-The World Conservation Union - was created in Switzerland in 1948. Sir Peter Scott became the chair of two of IUCN’s key commissions: on protected areas and on species survival (SSC), and later of FFI, and in 1966 he initiated the Red Data Book approach to documenting species at risk. The World Wildlife Fund {WWF} - now known as WWF-World Wide Fund for Nature outside North America - was designed originally to generate public contributions to support IUCN’s work, and was launched in 1961 with a campaign on black rhino. These two NGOs have been driving forces in global policy development. detailing the management steps - often in- cluding preservation of important areas or reduction in exploitation levels - needed to reduce that risk. In some cases, targets and indicators of progress may also be defined. In practice, conservation attention has tended to focus on species that are large, charismatic and possibly also ecologically important or highly threatened, or both. The tiger Panthera tigris, Arabian oryx Oryx leucoryx, white rhinoceros Ceratotherium simum and the blue whale Balaenoptera musculus are familiar mammalian examples. Considerable success has been achieved with the species named above and the many others {mainly animals) subject to similar conser- vation efforts. However, the species approach could never be extended to cover more than a minute fraction of the approximately 300 000 larger organisms (plants and vertebrates) in the biosphere, let alone the other several million that probably exist. In fact, the primary benefit may be that large organisrns, and terrestrial vertebrates in particular, generally require large areas of suitable habitat, and if such areas can be managed to minimize risk, other species may be safeguarded. 197 pa ED SS SSE EP I a ES EE EE I OE STE I EEL, f i) 19g WORLD ATLAS OF B ESSE OE Figure 8.1 Development of the global network of protected areas Source: UNEP-WCMC database, maintainted in collaboration with IUCN World Commission on Protected Areas IODIVERSITY Protecting areas The regulation of access to and use of particular areas has always been seen as complementary to the focus on individual species. Protected areas are in many ways the most important form of legislative measure for the conservation of biodiversity. Whereas the initial purpose of many such areas was to protect spectacular scenery and provide recreational facilities, the concept evolved to encompass habitats of threatened species and ecosystems rich in biodiversity. By the beginning of the 20th century many countries had either already established protected areas or were contemplating doing so. The concept, however, was slow to be put into practice and it was not until the 1940s that protected areas were being established in any significant number. The rate at which land was being incorporated into the system did not increase markedly until the early 1960s. The first World Parks Congress held in Seattle, in the United States, in 1962 was an important stimulus for the increase. This meeting signified the emergence of the mod- ern protected area network with over 80 percent of the world’s protected areas being established since then®. The cumulative and periodic growth of world protected areas are plotted in Figure 8.1 (the creation of the Greenland National Park in 1974, covering some 97 million hectares, and the Great 2500 -- LINE: + 14 000 Cumulative protected area (right-hand scale] BARS: ~ 12 000 2000 |. Number of sites (over 1 000 ha, IUCN categories | to VI) gazetted over a five-year period 110 000 | {left-hand scale] 1 500 }- 8 000 > Cc wn om a 1000 +6 000 = 3 Ls} ~ 4 000 500 + ~ 2000 0 1970 1975 1980 1985 1990 1995 2000 Barrier Reef Marine Park in the 1980s, extending over around 34 million hectares, have had marked individual effects on the global totals]. The location of protected areas in the IUCN/WCPA (World Commission on Protected Areas) categories I-VI greater than 100 000 hectares Map 8.1. Many protected areas, particularly those where secure funding allows management to be maintained, are effective in conserving species, habitats and landscapes of value. An extensive questionnaire study involving 93 protected areas in the tropics, in 22 countries, showed that areas where basic management activities are in place tend to avoid wholesale land clearance (but often still suffer some degree of disturbance, from hunting or logging, for example)‘. An inte- grated measure of their effectiveness is not yet available, and it may be that firm manage- ment within park boundaries can have the effect of increasing pressure on outside land that is less controlled; on the other hand, many protected areas are badly under- resourced, reducing their effectiveness. Wise management of areas outside the protected areas network, by means of planning controls and voluntary agreements and by incorporating conservation principles in landuse planning, also plays an essential role in conservation of biodiversity. Indeed, in many countries, management of landuse outside the national network of protected areas - in agricultural landscapes, for example - will play as important a role in the maintenance of national biodiversity as will the network itself. in extent is shown in Maintaining ecosystems Increasingly in recent years a more holistic approach has emerged in which ecosystems themselves have become the focus of con- servation efforts. Many would argue that the primary goal of conservation action is the long-term maintenance of ecosystem pro- cesses at the global scale and over foreseeable human generations. This is essentially equivalent to achieving sustain- ability in human development and use of biodiversity. The Convention on Biological Diversity adopted ‘the ecosystem approach’ as the guiding framework for actions taken in pursuit of its goals and, in elaborating on the meaning of this term, stressed the need to conserve ecosystem structure and function in order to maintain ecosystem services. However, making an assessment of the organization, vigor and resilience of eco- systems - the key components of ‘ecosystem health’ - is fraught with immense practical difficulties. Much conservation activity there- fore is quite properly based on a strong form of the precautionary principle (Box 8.2], and focuses on maintaining so far as possible Global biodiversity: responding to change qu ES SO ee benefit will accrue from ensuring the integrity of areas of higher biodiversity value than areas of lesser value. One early area-based approach identified some 12 ‘megadiversity’ countries, which between them include a large proportion of global biodiversity in selected major groups’. A concise definition of the precautionary principle is provided by Principle 15 of the Rio Declaration, made at the 1992 Earth Summit: the prominent elements of ecosystems, i.e. ‘In order to protect the environment, the precautionary approach shall be widely applied by species and populations and their physical States according to their capabilities. Where there are threats of serious or irreversible environment, in anticipation that the system damage, lack of full scientific certainty shall not be used as a reason for postponing cost- will thus be perpetuated. Defining priority areas for protection Management action aimed at biosphere con- servation demands financial resources, but these are limited, while human numbers and impacts continue to increase. This implies that choices must be made between possible actions, whether by design or by default. Rational and informed decision-making should seek to increase the efficiency with which conservation funds are used. Several studies have been undertaken that aim to identify and sometimes to rank areas of high biodiversity value. The approach is generally based on the premise that ‘more is better’. That is, an area with greater biodiversity value is more worth conserving than an area with lower value. At its simplest biodiversity may be equated with species number overall or per unit area, but other selection criteria are possible. Higher priority may be accorded to an area with populations of threatened species, or one rich in endemic species {especially if there are endemics in several different groups), or in species of commercial or cultural importance; or one which is par- ticularly representative of an ecosystem (perhaps one that is widely degraded else- where]. While the results of such assess- ments are usually of considerable biological interest, they may also have direct application as a basis for choice between different courses of action, on the grounds that greater effective measures to prevent environmental degradation.” Another early study delineated 18 endemic- rich botanical ‘hotspots’, which between them included around 20 percent of the world’s known plant species in less than 1 percent of the land surface, and were also undergoing rapid habitat conversion®’. A strength of the former approach is that it is focused at country level, and it is at this administrative level that most conservation action is undertaken; a weakness is that by evaluating species rich- ness alone it was not able to address unique- ness, and adjacent countries rich in species are likely to have many species in common. A strength of the latter approach is that it focus- ed on areas rich in restricted-range endemic species and such areas by definition make a large contribution to global biodiversity. The country-based approach has been extended in a study® using a database of estimates of richness and endemism in land vertebrates and vascular plants for all coun- tries of the world (Appendix 5). Indices of overall diversity (weighting richness and endemism equally) and diversity adjusted for country area have been produced. Within the relatively wide margin of error associated with species inventory [see Chapter 2], these indices can yield a useful view of variation in diversity in geopolitical terms or, if the data are treated as geographic samples, a view of general global variation in diversity (see Map 200 WORLD ATLAS Sn ea Cee ea Map 8.1 World protected areas An overview of the world’s surface nominally subject to protection and appropriate management. The location of protected areas in IUCN/WCPA categories I-VI greater than 100 000 hectares in area Is shown, represented by a point symbol. Protected areas greater than 1 million hectares in extent are represented by polygons instead of point symbols whenever boundary data are available. Source: UNEP-WCMC database (data extracted March 2002), maintained in collaboration with IUCN World Commission on Protected Areas. OF BIODIVERSITY Protected areas to 2 1 million hectares - 2 100 000 hectares 5.4). The hotspots approach has been further developed by Conservation International in a recent synthesis, and adopted as the basis for its conservation planning and action’. Twenty- five terrestrial hotspot regions were identi- fied, combining high species endemism, particularly among plants, and high rates of habitat loss, with at least 70 percent of origi- nal natural vegetation having been lost. Coral reef areas of the world have now also been evaluated in this way [see Chapter 6 and Map 6.1), and freshwater ecosystems are currently being evaluated. The term ‘hotspot’ is now often applied loosely to any area that has a concentration of diversity, whether by a measure of simple richness, or endemism or number of threatened species, and whether undergoing habitat conversion or not. Of studies based on biogeographic rather than geopolitical areas, the Centres of Plant Diversity project’® remains one of the largest. It relied heavily on extensive consultation among botanists to identify several hundred important sites worldwide, defined semi- quantitatively on the basis of a general combination of richness and endemism (Map 8.2). A broadly similar but less structured approach, also based on expert knowledge, has been taken to identify 43 areas of special importance for amphibian diversity, again using species richness and endemism as criteria’ (Map 8.3). A preliminary selection of areas of importance for inland water bio- diversity has also been made on the basis of expert knowledge of richness and endemism among fishes, mollusks and crustaceans [see Chapter 7, and Maps 7.2-7.4). However, the most systematic and com- plete global level assessment to date has involved bird species. Birds are by far the best-known major group of organisms on the planet, with a relative wealth of distribution and population data available, and global analyses by BirdLife International and its partners have set a standard yet to be matched for other taxonomic groups. The dis- tributions of all restricted-range bird species (defined as those in which the area encom- passing all distribution records is less than 50 000 km’), amounting to 25 percent of all birds, have been mapped in digital format. The co-occurrence of restricted-range species defines a set of 218 endemic bird areas ({EBAs]"*; these are shown, ranked in three categories according to biodiversity impor- tance, in Map 8.4. ‘Importance’ here takes account of the number of restricted-range Global biodiversity: responding to change a ae : eras Se — zs = _—— -~ ~ 7 te + "s oe us * - ’ gar cee *s = ial “+ i = ™ eos Le i. Sur ie Wee = ¥ Loe species in the EBA and the number of EBAs in which they are present, taxonomic unique- ness and EBA area. There are a number of limitations to area- based methods for establishing priorities among possible conservation actions. First, there is no single unequivocal way of com- paring value in different categories: how many vulnerable species is a single critically endangered species worth? Or how many endemic beetles is a single endemic bird worth? Second, information is always incom- plete, in that it never covers all taxa at all sites of interest. Resolving the former requires more or less arbitrary assigning of value. Attempts to resolve the latter entail the search for indicators, that is groups of species or other variables that can act as surrogates for wider measures of biodiversity. 201 a EE = Map 8.2 Centers of plant diversity This map shows the location of the sites and areas identified as important centers of plant diversity at regional and global levels, using expert knowledge, and mixed criteria emphasizing Species richness and endemism. See the three- volume source cited below for further details and extended documentation. Source: WWF and IUCN™” 2022 WORLD ATLAS OF BIODIVERSITY Plant diversity ie » Areas and centers The search for biological indicators of this kind has generated a great deal of research and discussion. Findings to date have gen- erally been equivocal, and depend in part on spatial scale. In general, though, it seems that at coarse scales there may be quite close agreement between different taxa so that, for example, many EBAs [which may be up to 600 000 km’ in extent) also hold significant numbers of other restricted-range species”. At this scale, birds may serve as indicators of high biodiversity value more generally. In other instances, and perhaps more generally at finer resolution, the relationship appears to break down. Studies in areas as disparate as North America, South Africa’, Cameroon” and the British Isles’ indicate that areas important for rare species in different groups often do not coincide {and these may be 1. = negatively correlated with areas of high species richness]. Also, richness levels in any one group do not necessarily provide an indication of species richness in others. A slightly different approach that avoids some of the limitations faced by methods based on species-specific information is to identify areas on the basis of their biological communities, and on biogeographic criteria, with reduced emphasis on richness or endemism in particular taxa, and none on current habitat condition. The conservation organization WWF's ‘ecoregions’ system, mentioned in Chapters 5-7 above, is the Principal example of this approach, and currently covers much of the world’s marine and inland waters, together with the entire land surface. It has been adopted as the basis for conservation planning and action by WWF. Whilst a number of different groups and organizations have explored the use of spatial biodiversity information in suggesting con- servation priorities, these generally remain disconnected and to some extent duplicative”. Recent developments suggest that a more integrated approach is feasible. Firstly, the conceptual and methodological framework of a more systematic approach to priority setting has been elaborated” (see Box 8.3]. Secondly, sustained effort to map species distribution in a globally consistent manner is beginning to allow robust analysis of large-scale multi- taxon patterns of biodiversity’. Combining a systematic approach with a geographically extensive set of spatial biodiversity data has the potential to allow priorities to be defined in a way that is flexible, transparent, and robust. Global biodiversity: responding to change Key stages in systematic planning for conservation by means of adding to or modifying protected area systems: 1. Compile data on the biodiversity of the planning region. 2. Identify conservation goals for the planning region. 3. Review existing conservation areas. 4. Select additional conservation areas. 5. Implement conservation actions. 6. Maintain the required values of conservation areas. Source: Margules and Pressey’, 203 2046 WORLD ATLAS O FE BilO DIMER Simpy: caanm ere eees reser errr eee eee Map 8.3 Major areas of amphibian diversity The location of areas identified on the basis of expert knowledge as globally important for amphibian diversity, using data on species richness and endemism. Areas are here shown classified in five categories according to species richness. See source cited below for further details. Source: Adapted from Duellman'! Number of species 160 - 334 77 - 159 37 - 76 21 - 36 4-20 It should, however, be emphasized that many priority-setting exercises are likely to remain only theoretical. This is because the opportunities to establish entirely new terres- trial protected areas, or redesign existing protected area networks in the light of priorities identified, are in general extremely limited. The designation and management of protected areas, aS So much other conservation action, is generally driven and constrained far more by socio-political and economic considerations than it is by conservation biology. Management of inland waters To a greater extent than is typical for terrestrial ecosystems, inland waters are often subject to management and use decisions made within many different sectors, including agriculture and forestry, navigation, public utility supply (power, water and waste disposal) and recre- ation. Most inland waters or their catchments also typically intersect several subnational administration units (counties, provinces, etc.) Although it has been recognized for some time that the catchment basin is the fundamental unit within which management must be formulated, it has proved difficult to reconcile the many different interests concerned and coordinate action. This is particularly problem- atic because sources of adverse impacts on such systems often originate far distant from where they are felt: thus pollutants and other inputs that enter the top reaches of a river system may have an impact in all downstream parts of the system as far as the river mouth and beyond. Similarly, water abstraction may have little effect in upper reaches, but be a serious constraint downstream. Biodiversity maintenance and conser- vation of inland water capture fisheries (other than some lucrative sports fisheries] have not ranked highly among these competing interests, so that it has been difficult to impose catchment-wide regulations or remedial measures for their benefit. This problem is exacerbated by the fact that it is often difficult to pinpoint a distant source of a problem or to unequivocally demonstrate that actions in one place are having an adverse effect somewhere else [e.g. to convince farmers that application of large doses of nitrogenous fertilizer on up- stream agricultural land is causing deleter- ious eutrophication of estuarine wetlands). Although they cannot solve catchment-wide problems, inland water protected areas may play a valuable role in safeguarding particular sites or populations of species from immediate Global biodiversity: threats. Protection may be most effective where sites are relatively small and thus manageable, and have a relatively low level of allochthonous inputs. Wetlands, with their often abundant and highly conspicuous avi- fauna, have in general received most attention in this regard. Notable wetland protected areas include the Moremi game reserve in the Okavango delta (Botswana), Camargue national reserve (France], Keoladeo (Bharatpur) national park (India], Donana national park (Spain) and Everglades national park (United States). Inland water ecosystems are unusual in that an international convention is dedicated specifically to them: the Convention on Wetlands of International Importance espec- ially as Waterfowl Habitat (the Ramsar Convention, see below). responding to chang ees aera e 205 206 WORLD Map 8.4 Endemic bird areas More than one quarter (2 561) of the world’s bird species, including more than 70% of the threatened birds, have a range restricted to less than 50 000 km’. Virtually all occur within the 218 endemic bird areas (EBAs) defined by BirdLife International”. The world’s EBAs are shown on this map categorized 1,2 o0r3 according to increasing biodiversity importance (based on the number of restricted range species, whether shared between EBAs, taxonomic uniqueness and EBA size) Source: Data provided by BirdLife International, and see Stattersfield'? AREAS, OF SBiOIDIMER SIN, a ee tg a SS SS 2S we Category Transboundary inland waters Waters that delineate or cross international boundaries present a special class of Management issue. Such waters and the living resources they contain are shared by one or more countries, and require positive international collaboration for effective use and management. Available water in any given country within an international basin (or other administrative unit within a basin more generally) can be divided into endogenous, i.e. locally generated runoff available in national aquifers and surface water systems, and exogenous, i.e. remotely generated runoff imported in flow from upstream. Some countries (e.g. Canada and Norway] have an abundance of water from endogenous sources; others (e.g. Egypt and Iraq) have a small endogenous supply but large exogenous volumes [others have small supplies from both sources]. Use of exogen- ous water carries an increasing risk because of dependence on sufficient supply from upstream countries. There are well over 200 major inter- national rivers and a host of smaller ones”. As demands on inland water resources continue to grow through the 21st century, as they undoubtedly will, management of these and the biological resources they contain will also grow ever more challenging. Management of marine ecosystems Management of the terrestrial environment is typically carried out alongside more or less severe anthropogenic disturbance. Although the particular nature of the marine biosphere has to some extent buffered it from the impacts of humans, it also imposes its own set of difficulties and constraints for rational management. Firstly, because almost all of it is generally out of sight, impacts are not immediately apparent, so that extreme deter- ioration may take place before anyone is aware of the fact. There may also be less incentive to take action than with terrestrial ecosystems where such deterioration has a direct impact on people. Secondly, with some exceptions [such as communal property resources on reefs and other inshore areas in parts of the South Pacific}, living marine resources have been widely considered open- access resources, particularly those outside territorial waters (usually up to 12 nautical miles (nm) from shore). There is thus, quite simply, an incentive for any given individual to exploit a resource as fast and as intensively Global biodiversity: as possible before someone else does. Thirdly, the ability of water to transport large amounts of dissolved and suspended mater- ials, including living organisms, means that it is extremely difficult to manage limited areas of marine habitat in isolation. With the introduction of the exclusive economic zone [EEZ) under the United Nations Convention on the Law of the Sea (UNCLOS}, which allows nations control over resources in an area up to 200 nm offshore, a far greater proportion of the world’s seas now come within the control of individual nations. At present 99 percent of world fisheries catch is taken within EEZs. Although this should theoretically allow more rational manage- ment of marine resources, and more effective enforcement of management measures, progress in both has in practice been limited responding to change 208 WOR LD ATL Map 8.5 Marine protected areas The map shows the location of protected areas in l|UCN categories I-VI that are entirely or in part marine, with map symbols graded according to protected area size [including any land present). Note: For presentation purposes it has been necessary to use symbols that in most cases at this map scale greatly exceed the size of the protected area represented, giving the impression that much more of the world’s coastal waters are protected than is in fact the case Source: UNEP-WCMC database [data extracted March 2002), maintained in collaboration with IUCN World Commission on Protected Areas. AS OF IBiONDIIM ER Sit ¥ Protected area (km’) ie >5 000 e > 1000 ° > 100 C < 100 to date, as evidenced by the increasing proportion of world’s fisheries that are over- exploited. This is because fisheries manage- ment regimes are frequently subject to political pressure, so that in many countries quotas are habitually set higher than those recommended by fisheries biologists, and also because the active enforcement of regu- lations is difficult and expensive. Many countries lack the resources or the political will, or both, to enforce such regulations adequately. It is also increasingly apparent that individ- ual marine resources cannot be effectively managed in isolation from each other. Complex interactions between populations of different organisms, when combined with perturbations in the environment and vari- ations in human impact (e.g. changes in os = : nea 5 eee Cora oe™ = = a = ou a # J , ate 6 ® s P Aaa Fi COR ns Go Fa e 3 - Yen Ve “A OUTS age coe Fe - . . Fy ™ 5 Be Nar Ce e s —_— . at . & 4 u SA ‘ a io ae 7") Sa 43 2 fisheries technology or fishing effort) create responses that may be far from intuitively predictable. Recognizing that the under- standing of such responses will require modeling and management of large-scale ecosystem processes, a number of large marine ecosystem (LME) units have been identified’, based on the world’s coastal and continental shelf waters, which are regarded as central to such analysis. Over 95 percent of the usable annual global biomass yield of fishes and other living marine resources is produced within 64 identified LMEs, nearly all of which lie within and immediately adjacent to the boundaries of EEZs of coastal nations. Many LMEs include the coastal waters of more than one state. In these cases, it will be effectively impossible for individual nations to assess whether their use of marine resources Global biodiversity: responding to change Se peer 2 ; ~ * = fc Py, * a? a ey al i, Abe CS) re | Ro P @ : ; . ~) $ ey Pe ad ¢ Tit As; $5 er. Sa hyp th y et 7 . 8 ¥ we “ \ a ee ifse = ! ~ ~ a Dot "a ‘e o® q \ 2 4 SOP) ° ee ‘ Ff Sal } e F a . ‘ie r% 2 yf c ~ Ss ny £> ‘ <3 w/e aD Re: 4 Se (/ Soe. 3 { : & Nee te) Ne q g e. z ~~ apeng as & ONE . ee ° > J { a ye > ae a 2 Ki oe . +8 e € . \ P Ce +2 s ‘ @ e A. A a ‘e one ge ¥¢ Jt e € ~, ° cane > . . ig e : ta Reunion Early 19th C (1800) Cylindraspis indica Réunion Early 19th C (1800) — Cylindraspis inepta Mauritius Early 18th C Cylindraspis peltastes Rodrigues Early 19th C (1800) Cylindraspis triserrata Mauritius Early 18th C Cylindraspis vosmaen Rodrigues Early 19th C (1800) Class AMPHIBIA nf Order ANURA 3 Family Discoglossidae otal Discoglossus nigriventer Israel painted frog Israel Mid 20th C (1940) Family Myobatrachidae A 2 Uperoleia marmorata Marbled toadlet Australia Late 19th C 4 Family Ranidae : a i Arthroleptides dutoiti Kenya Early 20thC | Rana fisheri Relict leopard frag USA Mid 20th C (1960) Rana tlaloci Mexico Late 20th (1990s) Class ACTINOPTERYGII Order CYPRINIFORMES. Family Catostomidae od Hi i nal | | Family Cyprinidae : Acanthobrama hulensis LakeHuleh Late 20thC? H (Israel) eae, | Anabarilius alburnops Lake Dianchi 20th C- od (China) == = = oh ela Anabarilius polylepis Lake Dianchi 20the 43 (China) a Barbus microbarbis Lake Luhondo Mid 20thC [Rwanda] 2 ae APPENDIX 4 289 wh EE TE AAS lybopsis amecae Ameca shiner Mexico Late 20th C (1969) lybopsis aulidion Durango shiner Mexico Late 20th C Notropis: orca Phantom shiner Mexico, USA Late 20th C ? xinellus egridiri = Yag baligi Lake Egridir Mid 20th C (1955) (Turkey) Lake Egridir Mid 20th C (1955) (Turkey) ellus handlirschi ' Cicek _ _ Pait Lake Lanao Early 20th C (1910) P ; 3 (Philippines) | 5 baoulan ; Baolan Lake Lanao Early 20th C (1926) eho ; (Philippines) Bagangan Lake Lanao Early 20th C {1921} a epee [Philippines] Ro Disa etn Lake Lanao Early 20th C [1932] ‘ = i (Philippines) 4 Katapa-tapa Lake Lanao Early 20th C (1921) : [Philippines] Lake Lanao Early 20th C (1908) : (Philippines} See ikanday ; Lake Lanao Early 20th C (1922 gee ___ [Philippines) ; : ~ Manalak S ~ Lake Lanao ~ Early 20th C (1924) 3 Sent a oe —___[Philippines} ipeen ie re Siang’ Lake Lanao Early 20th C (1932) a al was (Philippines) ed . Early 20th C (1925) Lake Lanao [Philippines] , 20 WORLD ATLAS OF BIODIVERSITY a Order CHARACIFORMES | Family Characidae / : Brycyon acuminatus Brazil 20th C Order OSMERIFORMES Family Retropinnidae Order SILURIFORMES Family Schilbeidae Platytropius siamensis Thailand Family Trichomycteridae ea Rhizosomichthys totae Lake Tota Mid 20th C (1957) {Colombia} ae Order SALMONIFORMES - Family Salmonidae i Coregonus alpenae Longjaw cisco Great Lakes Late 20th C (1975) (Canada, USA] Coregonus confusus Parrig Lake Morat ~ Mid 20th C- (Switzerland) Rea : Coregonus fera Féra ; Lake Geneva Mid 20th C : (Switzerland) ‘ Coregonus gutturosus Kilch ; Lake Constance Mid 20th C (Switzerland) ee Coregonus hiemalis Gravenche Lake Geneva — Mid 20thC (Switzerland] Lake Morat (Switzerland) Coregonus restrictus ~ Salvelinus inframundis Orkney char = ; Hoy |. {UK} Early 20th C (19 a | Salvelinus scharffi Scharff’s char Lough Owel Early 20th | 908) (Ireland) Order ATHERINIFORMES = Family Atherinidae : : 2 Chirostoma compressum Cuitzeo silverside Lake Cuitzeo (Mexico) Rheocles sikorae Zona ; Madagascar ify Order CYPRINODONTIFORMES — : Family Cyprinodontidae APPENDIX 4 a Late 20th C e Utah Lake sculpin Utah Lake - Early 20th C (1928) : (USA) Israel : Late 20th C (1997) Haplochromis altigenis Lake Victoria Mid/late 20th C ? Haplochromis apogonoides Lake Victoria Mid/late 20th C ? Haplochromis arcanus Lake Victoria Mid/late 20th C ? Haplochromis argenteus Lake Victoria Mid/late 20th C ? Haplochromis artaxerxes Lake Victoria Mid/late 20th C ? Haplochromis barbarae Lake Victoria Mid/late 20th C ? Haplochromis bareli Lake Victoria Mid/late 20th C ? Haplochromis bartoni Lake Victoria Mid/late 20th C ? Haplochromis bayoni Lake Victoria Mid/late 20th C 2 Haplochromis boops Lake Victoria Mid/late 20th C ? 2922 WORLD ATLAS OF BIODIVERSITY Haplochromis cassius : Lake Victoria Mid/late 20th C ? Haplochromis cinctus Lake Victoria Mid/late 20th C ? Haplochromis cnester Lake Victoria Mid/late 20th C ? Haplochromis decticostoma Lake Victoria Mid/late 20th C ? Haplochromis dentex group Lake Victoria Mid/late 20th C ? Haplochromis diplotaenia Lake Victoria Mid/late 20th C ? Haplochromis estor Lake Victoria Mid/late 20th C ? Haplochromis flavipinnis Lake Victoria Mid/late 20th C ? Haplochromis gilberti Lake Victoria Mid/late 20th C 2 Haplochromis gowersi Lake Victoria Mid/late 20th C ? Haplochromis guiarti Lake Victoria Mid/late 20th C ? Haplochromis heusinkveldi Lake Victoria Mid/late 20th C ? Haplochromis hiatus Lake Victoria Mid/late 20th C ? Haplochromis iris Lake Victoria Mid/late 20th C ? Haplochromis longirostris Lake Victoria Mid/late 20th C ? Haplochromis macrognathus Lake Victoria Mid/late 20th C ? Haplochromis maculipinna Lake Victoria Mid/late 20th C ? Haplochromis mandibularis Lake Victoria Mid/late 20th C ? Haplochromis martini Lake Victoria Mid/late 20th C ? Haplochromis megalops Lake Victoria Mid/late 20th C ? Haplochromis michaeli Lake Victoria Mid/late 20th C ? Haplochromis microdon Lake Victoria Mid/late 20th C ? Haplochromis mylergates Lake Victoria Mid/late 20th C ? Haplochromis nanoserranus Lake Victoria Mid/late 20th C ? Haplochromis nigrescens Lake Victoria Mid/late 20th C ? Haplochromis nyanzae Lake Victoria Mid/late 20th C ? Haplochromis obtusidens Lake Victoria Mid/late 20th C ? Haplochromis pachycephalus Lake Victoria Mid/late 20th C ? Haplochromis paraguiarti Lake Victoria Mid/late 20th C ? Haplochromis paraplagiostoma Lake Victoria Mid/late 20th C ? Haplochromis parorthostoma Lake Victoria Mid/late 20th C ? Haplochromis percoides Lake Victoria Mid/late 20th C ? Haplochromis pharyngomylus Lake Victoria Mid/late 20th C ? Haplochromis prognathus Lake Victoria Mid/late 20th C ? Haplochromis pseudopellegrini Lake Victoria Mid/late 20th C ? Haplochromis pyrrhopteryx Lake Victoria Mid/late 20th C ? Haplochromis spekii Lake Victoria Mid/late 20th C ? Haplochromis teegelaari Lake Victoria Mid/late 20th C ? Haplochromis thuragnathus Lake Victoria Mid/late 20th C ? Haplochromis tridens Lake Victoria _ Mid/late 20th C ? Haplochromis victorianus Lake Victoria Mid/late 20th C ? Haplochromis xenostoma Lake Victoria Mid/late 20th C ? Haplochromis bartoni-like’ Lake Victoria Mid/late 20th C ? Haplochromis ‘bicolor’ Lake Victoria Mid/late 20th C ? Haplochromis ‘big teeth’ Lake Victoria Mid/late 20th C ? Haplochromis ‘black cryptodon’ Lake Victoria Mid/late 20th C ? Haplochromis ‘back pectoral’ Lake Victoria Mid/late 20th C ? APPENDIX 4 293 SS SS SE ES Haplochromis ‘chlorocephalus Lake Victoria Mid/late 20th C ? Haplochromis ‘citrus Lake Victoria Mid/late 20th C ? Haplochromis ‘coop’ Lake Victoria Mid/late 20th C ? Haplochromis elongate rockpicker’ Lake Victoria Mid/late 20th C ? Haplochromis filamentus’ Lake Victoria Mid/late 20th C ? Haplochromis fleshy lips’ Lake Victoria Mid/late 20th C ? Haplochromis ‘gray pseudo-nigricans Lake Victoria Mid/late 20th C ? Haplochromis large eye guiarti’ Lake Victoria Mid/late 20th C ? Haplochromis \lividus-frels’ Lake Victoria Mid/late 20th C ? Haplochromis ‘longurius Lake Victoria Mid/late 20th C ? Haplochromis macrops like’ Lake Victoria Mid/late 20th C ? Haplochromis micro-obesus Lake Victoria Mid/late 20th C ? Haplochromis morsei’ Lake Victoria Mid/late 20th C ? Haplochromis orange cinereus’ Lake Victoria Mid/late 20th C ? Haplochromis orange macula’ Lake Victoria Mid/late 20th C ? Haplochromis ‘orange yellow big teeth’ Lake Victoria Mid/late 20th C ? Haplochromis orange yellow small teeth’ Lake Victoria Mid/late 20th C ? Haplochromis paropius-like’ Lake Victoria Mid/late 20th C ? Haplochromis ‘pink paedophage’ Lake Victoria Mid/late 20th C ? Haplochromis pseudo-morsei’ Lake Victoria Mid/late 20th € ? Haplochromis purple head’ Lake Victoria Mid/late 20th C ? Haplochromis ‘purple miller’ Lake Victoria Mid/late 20th C ? Haplochromis (2) ‘purple rocker’ Lake Victoria Mid/late 20th C ? Haplochromis red empodisma’ Lake Victoria Mid/late 20th C ? Haplochromis ‘red eye scraper’ Lake Victoria Mid/late 20th C ? Haplochromis reginus Lake Victoria Mid/late 20th C ? Haplochromis ‘regius Lake Victoria Mid/late 20th C ? Haplochromis short supramacrops Lake Victoria Mid/late 20th C ? Haplochromis ‘small blue zebra’ Lake Victoria Mid/late 20th C ? Haplochromis ‘small empodisma’ Lake Victoria Mid/late 20th C ? Haplochromis ‘smoke’ Lake Victoria Mid/late 20th C ? Haplochromis ‘soft gray’ Lake Victoria Mid/late 20th C ? Haplochromis ‘stripmac’ Lake Victoria Mid/late 20th C ? Haplochromis supramacrops Lake Victoria Mid/late 20th C ? Haplochromis theliodon-like Lake Victoria Mid/late 20th C ? Haplochromis ‘tigrus Lake Victoria Mid/late 20th C ? Haplochromis too small Lake Victoria Mid/late 20th C ? Haplochromis twenty’ Lake Victoria Mid/late 20th C ? Haplochromis two stripe white lip’ Lake Victoria Mid/late 20th C ? Haplochromis wyber’ Lake Victoria Mid/late 20th C ? Haplochromis xenognathus-like Lake Victoria Mid/late 20th C ? Haplochromis ‘yellow’ Lake Victoria Mid/late 20th C ? Haplochromis yellow-blue’ Lake Victoria Mid/late 20th C ? Hoplotilapia retrodens Lake Victoria Mid/late 20th C ? Psammochromis cryptogramma group take Victoria Mid/late 20th C ? 2% WORLD ATLAS OF BIODIVERSITY gy SSS REFERENCES 1 BirdLife International 2000. Threatened birds of the world. Lynx Edicions and BirdLife International, Barcelona and Cambridge. 2 CREO. Committee on Recently Extinct Organisms (CREO) website: http://creo.amnh.org/ {accessed January 2002). 3 Hilton-Taylor, C. [compiler] 2000. 2000 IUCN Red List of threatened species. IUCN, Gland and Cambridge. Online at http://www.redlist.org/ {accessed April 2002). APPENDIX 5: BIODIVERSITY AT COUNTRY LEVEL This table includes estimates of the number of mammals, breeding birds and vascular plants in each country of the world, together with estimates of the number of these endemic to each country. The threat status of mammals and birds has been comprehensively assessed, and for these classes the number and percentage of globally threatened species present in each country are given. In the mammals, most estimates of the total number present relate to native non-marine species only, but the threatened species counts for many countries include marine species (a few island countries have a small number of non-marine mammals but a higher number of threatened species - the percentage figure is omitted in such cases}. The richness and endemism data should Methodology notes Numerical indices representing national biodiversity have been derived from available estimates of species total and species endemism. Taxonomic groups covered are: mammals, birds, reptiles, amphibians and vascular plants [ferns to flowering plants). For birds alone, the database includes estimates of both the total number recorded (i.e. including non-breeding migrants and accidental visitors] and the number of breeding bird species; the latter has been used in the analysis. All countries below 5 000 km’ in area have been excluded from the analysis, leaving 169 countries. Four arbitrary assumptions are made: the four vertebrate classes included are of equal importance; plants are of equal importance to the vertebrates combined; richness and endemism are reasonably well correlated at country level across the taxonomic groups covered; vertebrates plus plants provide a valid surrogate for biodiversity in general. Because interest lies in relative biodiversity rather than absolute values, data in each column are first normalized. Each estimate Ni for each country /is divided by Nmax where Nmax is the highest value for be taken as provisional; they are subject to change with new taxonomic treatments and new surveys. The columns headed DI and Al, respectively, contain the unweighted national diversity indices, and the diversity indices adjusted for area [see methodology notes below for details]. The index is based on data for the groups shown here, plus reptiles and amphibians, not included here. The Al column is the basis for Map 5.4. Source: WCMC database; data derived from a large number of published and unpublished sources, including country reports and regional checklists. Numbers of threatened species retrieved from the online version of the 2000 IUCN Red List of threatened species, at http://www.redlist.org/ {accessed March 2002). that parameter. This transforms the data to within the range 0-1, with the most important country having the value Nmax/Nmax = 1, and the least important having the value closest to zero. Estimates for mean vertebrate richness [VR] and mean vertebrate endemism (VE) are derived by averaging the figures for all classes, and estimates for combined richness (R) and combined endemism (E) are derived for each country by averaging figures for vertebrates and plants. Inspection of the data shows that PR and VR correlate quite closely, while PE tends to be approximately half VE; where estimates of PR and PE are missing the appropriate vertebrate-based value has been inserted before calculating Rand E. An overall diversity index (Di) is calculated for each country as the mean of Rand E. This treats richness and endemism equally and so makes fewest assumptions about their relative significance in terms of overall biodiversity, but the D/ could be weighted to give greater importance to either. Because species richness tends to increase with area, and with proximity to the humid tropics, Di is strongly affected by country area APPENDIX 5 295 Renee es eee I Ie 29% WORLD ATLAS OF BIODIVERSITY RS and geographical position, but it also takes account of relative levels of biodiversity per unit area, i.e. how levels of endemism, which are shaped by several much more or less rich in species is any given area factors, including topography, geographic isolation (or country}. This may be addressed by the Arrhenius and tectonic history. equation describing the species-area relationship Because area is an important determinant of (log S = g + zlog A], where S = number of species, species number, there is much interest in evaluating A= area, zis the slope of the line, and g another Afghanistan 652 225 0.063 -0.296 119 2 Albania 28 750 0.035 -0.019 68 0 Algeria 2 381 745 0.045 - 1.003 2 2 American Samoa 197 3 0 Andorra 465 44 0 Angola 1 246 700 0.176 0.544 276 if Anguilla 91 3 0 Antigua and Barbuda 442 7 0 Argentina 2777 815 0.196 0.423 320 49 Armenia 29 800 0.042 0.153 84 3 Aruba 193 = 0 Australia 7 682 300 0.608 1.268 252 206 Austria 83 855 0.036 -0.293 83 0 Azerbaijan 86 600 0.05 0.027 99 0 Bahamas 13 865 0.017 -0.503 12 3 Bahrain 661 17 0 Bangladesh 144 000 0.059 0.058 125 0 Barbados 430 6 0 Belarus 207 600 0.029 -0.771 74 0 Belgium 30 520 0.023 -0.441 58 0 Belize 22 965 0.056 0.526 125 0 Benin 112 620 0.08 0.437 188 0 Bermuda 54 3 0 Bhutan 46 620 0.058 0.366 160 0 Bolivia 1 098 575 0.239 0.882 316 16 Bosnia and Herzegovina 51129 0.034 -0.200 72 0 Botswana 575 000 0.062 -0.287 164 0 Brazil 8511 965 0.74 1.436 394 119 British Ind. Oc. Terr. ; - 0 Brunei 5 765 0.071 1.145 157 0 Bulgaria 110 910 0.044 -0.167 81 0 Burkina Faso 274 122 0.068 0.011 147 0 Burundi 27 835 0.072 0.723 107 0 Cambodia 181 000 0.059 0.001 123 0 Cameroon 475 500 0.167 0.762 409 14 Canada 9 922 385 0.067 -1.014 193 7 Cape Verde 4 035 5 0 Cayman Islands 259 8 0 Central African Republic 624 975 0.08 — 0.058 209 2 APPENDIX 5 297 eee constant. If it is assumed, given its area dependence, gives a measure (Al] of how much more (+ve] or less that Di is scaled in the same way as S, a regression (-ve] diverse is a given country than expected. A/ analysis of log D/ and log A allows the constants gandz attempts to assess diversity per unit area, rather than to be calculated. The regression line establishes the overall biodiversity value per country, and it is noticeable expected biodiversity value of each country for its area, that several smaller mainland countries and island and the distance of each country point from the line states rnove up in rank order, as would be expected. 0 = zero; - = no data. 235 0 11 5 4 000 800 230 0 3 1 3 031 24 192 1 6 3 3 164 250 34 0 2 6 47) 15 113 0 0 0 1 350 = 765 12 15 2 5 185 1 260 = 0 0 321 1 49 0 1 2 1 158 22 897 19 38 4 9 372 1 100 242 0 4 2 3 553 = 48 0 0 0 460 25 649 350 32 5 15 638 14 074 213 0 3 1 3 100 35 248 0 8 3 4 300 240 88 3 4 5 1111 118 28 0 6 21 195 = 295 0 23 8 5 000 = 24 0 1 4 572 3 221 0 3 1 2 100 - 180 0 2 1 1 550 1 356 0 2 1 2 894 150 307 0 2 1 2 500 - 8 1 2 25 167 15 448 0 12 S 5 468 75 = 18 27 17 367 4 000 218 0 3 1 = = 386 1 7 2 2 ASI 17 1492 185 113 8 56 215 = 14 0 0 0 101 - 359 0 15 4 6 000 7 240 0 10 4 3572 320 335 0 2 1 1 100 # 451 0 7 2 2 500 - 307 0 19 6 = = 690 8 15 2 8 260 156 426 5 8 2 3 270 147 38 4 2 5 774 86 0 45 0 1 2 539 19 «© 537 1 3 1 3 602 100 29 WORLD ATLAS OF BIODIVERSITY a Chad 1284000- 0.049 -0.739 134 1 17 Chile 751 625 0.112 0.229 91 16 21 China 9 597 000 0.392 0.767 394 83 76 Colombia 1) 138 1S 0.538 1.685 359 34 36 Comoros 1 860 12 2 2 Congo, Dem. Rep. 2 345 410 0.218 0.579 450 28 40 Congo, Republic 342 000 0.128 0.589 200 2 12 Cook Islands 233 1 0 1 Costa Rica 50 900 0.162 1.358 205 7 14 Cote d'Ivoire 322 465 0.116 0.507 230 0 17 Croatia 56 538 0.036 -0.169 76 0 9 Cuba 114 525 0.12 0.829 31 12 11 Cyprus 9 250 0.017 -0.429 21 1 3 Czech Republic 78 864 0.033 -0.356 81 0 8 Denmark 43 075 0.021 -0.643 43 0 5 Djibouti 23 000 0.02 = 0.528 61 0 4 Dominica 751 12 0 1 Dominican Republic 48 440 0.076 0.625 20 0 5 Ecuador 461 475 0.353 1.519 302 25 31 Egypt 1 000 250 0.038 -0.936 98 7 2 El Salvador 21 395 0.048 0.393 135 0 2 Equatorial Guinea 28 050 0.084 0.869 184 1 15 Eritrea 117 600 0.057 0.088 112 0 12° Estonia 45 100 0.025 -0.483 65 0 5 Ethiopia 1 104 300 0.145 0.383 277 31 Ca Falkland Islands 15 931 0.004 -2.040 0 0 a Faroe Islands 0 3 Fiji 18 330 0.028 -0.100 4 i "Bae Finland 337 030 0.023 -1.145 60 0 6 France 543 965 0.051 =0.473 93 0 18 French Guiana 91 000 0.079 0.483 150 3 9 French Polynesia 3 940 0 0 3 French S. and Antarctic Terr. 7 241 0.001 -3.261 - 0 je Gabon 267 665 0.116 0.56 190 3 15 _ Gambia 10 690 0.036 0.308 117 0 Sie Georgia 69 700 0.051 0.111 107 2 14 Germany 356 840 0.033 -0.770 76 0 12 Ghana 238305 0.114 0.571 222 1 13 Gibraltar il 7 0 0 Greece 131985 0.062 0.129 oe) 3 14 Greenland 2 175 600 0.007 -2.821 9 0 Ta Grenada 345 15 0 on Guadeloupe 1 780 11 4 5a Guam 450 2 0 26 Guatemala 108 890 0.142 1.014 250 3 6 Guinea 245 855 0.094 0.373 190 1 =. Jee Guinea-Bissau 36 125 0.05 0.289 108 0 2 APPENDIX 5 299 SS | 13 370 0 5 1 1 600 : 23 296 16 15 5 5 284 2 698 19 1 100 70 73 7 32 200 18 000 10 1 695 67 77 5 51 220 15 000 17 50 14 9 18 721 136 9 929 24 28 3 11 007 1 100 6 449 0 3 1 6 000 1 200 100 27 6 7 26 284 3 7 600 6 13 2 12119 950 7 535 2 12 2 3 660 62 12 224 0 4 2 4 288 2 35 37 21 18 13 6 522 3 229 14 79 2 3 ih 1 682 = | 10 99 0 2 1 1 900 : | 12 196 0 1 1 1 450 1 | 7 126 1 5 4 826 6 . 8 52 2 3 6 1 228 1 | 25 36 0 15 1 5 657 1 800 10 1 388 37 60 4 19 362 4 000 2 153 0 7 5 2076 70 1 251 0 0 0 2911 17 8 273 3 5 2 3 250 66 1 319 0 7 2 : 8 213 0 3 1 1 630 Z 12 626 28 16 3 6 603 1 000 64 4 3 5 165 14 vA 0 0 0 236 1 7h 24 12 16 1518 760 10 248 0 3 1 1 102 : 19 269 1 5 2 4 630 133 6 5 1 0 5 625 144 60 25 22 37 959 560 48 3 3 6 - = 8 466 1 5 1 6 651 : 3 280 0 2 1 974 : 13 2 0 3 4350 380 16 239 0 5 2 2 682 6 6 529 0 8 2 3 725 43 0 34 0 1 3 600 k 15 251 0 7 3 4992 742 78 62 0 0 0 529 15 0 50 1 1 2 1 068 4 45 52 D 1 2 1 400 26 100 18 2 2 11 330 69 2 458 1 6 1 8 681 1171 6 409 0 10 2 3 000 88 2 243 0 0 0 1 000 12 300 WORLD ATLAS OF BIODIVERSITY SSF ES Guyana 214 970 0.133 0.758 193 1 9 Haiti 27 750 0.071 0.71 20 0 4 Honduras 112 085 0.094 0.597 173 2 9 Hungary 93 030 0.031 -0.457 83 0 9 Iceland 102 820 0.006 - 2.080 11 0 6 India 3 166 830 0.326 0.896 390 44 86 Indonesia 1919 445 0.731 1.844 515 222 140 Iran 1 648 000 0.091 -0.194 140 6 23 Iraq 438 445 0.041 -0.629 81 2D 10 Ireland 68 895 0.013 - 1.248 25 0 5 Israel 20 770 0.043 0.285 116 4 14 Italy 301 245 0.065 -0.056 90 3 14 Jamaica 11 425 0.051 0.619 24 2 5 Japan 369 700 0.124 0.536 188 42 37 Jordan 96 000 0.036 -0.310 71 0 8 Kazakhstan 2 717 300 0.071 -0.581 178 4 18 Kenya 582 645 0.145 0.56 359 23 51 Kiribati 684 - 0 ‘Oe Korea, DPR 122 310 0.025 -0.775 - 0 13 Korea, Republic 98 445 0.03 -— 0.518 49 0 13 Kuwait 24 280 0.007 -1.564 21 0 1 Kyrgyzstan 198 500 0.036 -0.537 83 1 7 Lao PDR 236 725 0.081 0.229 172 0 27 Latvia 63 700 0.025 -0.553 83 0 5 Lebanon 10 400 0.031 0.145 57 0 6 Lesotho 30 345 0.025 -0.354 33 0 3 Liberia 111 370 0.059 0.132 193 0 16 Libya 1 759 540 0.029 - 1.343 76 25) 9 Liechtenstein 160 64 0 3 Lithuania 65 200 0.026 -0.544 68 0 5 Luxembourg 2 585 55 0 6 Macedonia, FYR 25 713 0.037 0.077 78 0 iil Madagascar 594 180 0.298 1.277 141 93 50 Malawi 94 080 0.079 0.473 195 0 ys Malaysia 332 965 OSA eeeeZ8 300 36 47 3 Maldives 298 3 0 0 Mali 1 240 140 0.053 -0.658 137 0 ie} Malta 316 22 0 Sm Marshall Islands 181 0 0 ie Martinique 1079 9 0 (0. Mauritania 1 030 700 0.041 -0.856 61 1 10 Mauritius 1 865 4 1 3 Mayotte 376 - 0 Os Mexico 1972 545 0.589 1.621 491 140 69 Micronesia, Fed. States 702 6 3 on Moldova 5 33 700 0.025 -0.396 68 0 om Monaco 2 - 0 So APPENDIX 5 301 a H 678 0 2 0 6 409 : 75 1 14 19 5 242 1 623 422 1 5 1 5 680 148 205 0 8 4 2214 38 88 0 0 0 377 1 923 58 70 8 18 664 5 000 1519 408 113 v7) 29 375 17 500 323 1 13 4 8 000 2 172 1 1 6 “ : 142 0 1 1 950 2 180 0 12 7 2317 Z 234 0 5 2 5599 12 113 26 12 1 3 308 923 250 21 32 13 5 565 2 000 141 0 8 6 2 100 = 396 0 15 4 6 000 = 844 9 24 3 6 506 265 26 1 3 12 60 2 115 1 19 17 2 898 107 112 0 =25 22 2 898 224 z 20 0 7 35 234 z é 0 4 4 500 = 487 1 19 4 8 286 a 217 0 3 1 1153 = 154 0 7 5 3 000 = 58 0 7 12 1591 2 372 1 1 3 2 200 103 9 0 1 1 1 825 134 124 0 1 1 1410 = 202 0 4 2 1796 = 126 0 1 1 1 246 = 210 0 3 1 3 500 2 202 105 27 13 9 505 6 500 521 0 1 2 3765 49 501 18 37 7 15 500 3 600 23 0 1 4 583 z 397 0 4 1 1741 i 26 0 1 4 914 5 17 0 1 6 100 5 52 1 2 4 1 287 30 4 273 0 2 1 1 100 = 27 8 9 33 750 325 27 2 3 1 500 A 769 92 38 5 26 071 12 500 40 18 5 12 1194 293 177 0 5 3 1752 = 0 3022 WORLD ATLAS OF BIODIVERSITY ee ee ee eee Mongolia 1 565 000 0.051 -0.767 133 0 Montserrat 104 7 0 Morocco 458 730 0.057 -0.304 105 4 Mozambique 784 755 0.09 0.005 179 2 Myanmar 678 030 0.141 0.493 300 6 Namibia 824 295 0.102 0.116 250 3 Nauru - 0 Nepal 141 415 0.096 0.549 181 2 Netherlands 41 160 0.022 -0.599 55 0 Netherlands Antilles 800 - 0 New Caledonia 19 105 0.078 0.904 11 3 New Zealand 265 150 0.065 -0.017 2 2 Nicaragua 148 000 0.098 0.555 200 2 Niger 1 186 410 0.061 -0.512 131 0 Nigeria 923 850 0.107 0.131 274 4 Niue 259 1 0 Northern Marianas 477 - 0 Norway 386 325 0.024 - 1.107 54 0 Oman 271 950 0.03 — 0.812 56 2 Pakistan 803 940 0.08 = 0.121 188 4 Palau 492 2 0 Panama 78 515 0.162 1.236 218 16 Papua New Guinea 462 840 0.271 1.254 214 65 Paraguay 406 750 0.115 0.429 305 2 Peru 1 285 215 0.396 1.344 460 49 Philippines 300 000 0.225 1.188 153 102 Pitcairn Islands 0 0 Poland 312 685 0.032 -0.761 84 0 Portugal 92 390 0.045 -0.088 63 1 Puerto Rico 8 960 0.033 0.259 16 0 Qatar 11 435 0.005 -1.770 11 0 Réunion 2510 2 0 Romania 237 500 0.039 -0.490 84 0 Russia 17 075 400 Oe SW) 269 22. Rwanda 26 328 0.087 0.925 151 0 San Marino 13 0 Sao Tomé and Principe 964 8 4 Saudi Arabia 2 400 900 0.04 - 1.129 77 0 Senegal 196 720 0.057 -0.065 192 0 Seychelles ; 404 ‘ Q 2 Sierra Leone 72 325 0.083 0.588 147 0 Singapore 616 85 1 Slovakia 14 035 0.037 0.252 85 0 Slovenia 20 251 0.036 0.106 75 0 Solomon Islands 29 790 0.049 0.316 53 21 Somalia 630 000 0.087 0.025 171 2 i South Africa 1 184 825 0.252 0.915 247 35 APPENDIX 5 303 TT TEEETEEEEESeeeSSEne EE ,, !) "_|"qX|qg==x&&22: 9 426 0 16 4 2 823 229 14 37 1 2 5 671 2 15 210 0 9 4 3 675 625 8 498 0 16 3 5 692 219 12 867 4 35 4 7 000 1071 6 469 3 9 2 3174 687 9 1 2 22 50 1 15 611 2 26 4 6 973 315 20 191 0 4 2 1 221 = 77 0 1 = - 55 107 22 9 8 3 250 3 200 150 74 49 63 2 382 1942 iS 482 0 5 1 7590 40 8 299 0 3 1 460 = 9 681 Z 9 4715 205 0 15 0 1 7 178 1 28 2 8 29 315 81 19 243 0 2 1 Wi i 16 107 0 10 9 1 204 73 10 375 0 17 5 4950 372 45 10 2 4 s = 9 732 7 16 2 ON 1 222 Zi 644 94, 32 5 11 544 = 3 556 0 26 5 7 851 = 10 1 538 112 7\ 3) 17 144 5 356 Be) 196 186 67 34 8931 3 500 19 5 8 42 76 14 18 227 0 4 2 2 450 3 27 207 2 7 3 5 050 150 13 105 12 8 8 2493 235 0 23 0 6 26 358) - 18 4 5 28 546 165 20 247 0 8 3 3 400 4) 16 628 13 38 6 11 400 - 5 513 0 7 2 2 288 26 8 - 0 0 = = 38 63 25 9 14 895 134 9 155 0 15 10 2 028 - 6 384 0 4 1 2 086 26 67 38 11 10 26 250 182 7 466 1 10 2 2090 74 4 118 0 7 6 2 282 2 11 209 0 4 2 3 124 92 2 207 0 1 0 3 200 22 40 163 43 23 14 3172 30 11 422 11 10 2 3 028 500 17 596 8 20 3 23 420 = 304 WORLD ATLAS OF BIODIVERSITY rn Spain 504 880 0.067 -0.172 82 4 Sri Lanka 65 610 0.082 0.606 88 15 St Helena and dep. 411 2 0 St Kitts and Nevis 261 7 0 St Lucia 619 9 0 St Vincent 389 8 1 Sudan 2 505 815 0.137 0.093 267 11 Suriname 163 820 0.092 0.471 180 2 Swaziland 17 365 0.044 0.353 47 0 Sweden 440 940 0.026 - 1.067 60 0 Switzerland 41 285 0.033 -0.173 75 0 Syria 185 680 0.046 -0.265 63 2 Taiwan 36 960 0.058 0.418 63 11 Tajikistan 143 100 0.0383 -0.536 84 1 Tanzania 939 760 0.189 0.693 316 15 Thailand 514 000 0.162 0.709 265 7 Togo 56 785 0.094 0.781 196 0 Tokelau 0 0 Tonga 699 2 0 Trinidad and Tobago 5 130 0.045 0.729 100 1 Tunisia 164 150 0.033 -0.572 78 1 Turkey 779 450 0.114 0.237 116 2 Turkmenistan 488 100 0.044 -0.572 103 0 Turks and Caicos Islands 430 - 0 Tuvalu 25 - 0 Uganda 236 580 0.12 0.624 345 6 Ukraine 603 700 0.05 - 0.509 108 1 United Arab Emirates 75 150 0.019 -0.883 25 0 United Kingdom 244 880 0.024 - 1.003 50 0 United States of America 9 372 614 0.342 0.638 428 105 Uruguay 186 925 0.05 — 0.186 81 1 US Pacific Islands 658 - 0 Uzbekistan 447 400 0.051 -0.413 97 0 Vanuatu 14 765 0.014 -0.728 11 2 Venezuela 912 045 0.379 1.398 323 19 Viet Nam 329 565 0.147 0.737 213 9 Virgin Islands (British) 153 3 0 Virgin Islands (US) 352 - 0 Wallis and Futuna 255 1 0 Western Sahara 32 1 Western Samoa 2 840 3 0 Yemen 477 530 0.041 -0.654 66 1 Yugoslavia 102 173 0.046 -0.086 96 0 Zambia 752 615 0.096 0.074 233 3 Zimbabwe 390 310 0.099 0.298 270 0 APPENDIX 5 305 eae A 278 5 7 3 5 050 941 250 24 14 é 3314 890 53 9 9 17 165 50 32 0 1 3 659 1 50 4 5 10 1028 1 108 2 2 2 1166 680 1 6 1 3 137 50 603 0 1 0 5018 = 364 0 5 1 2715 4 249 0 2 1 1750 1 193 0 2 1 3 030 1 204 0 8 4 3 000 Z 160 14 20 13 3.568 és 2 0 7 5 000 Z 822 24 33 4 10 008 1122 616 2 37 6 11 625 : 391 0 0 0 3.085 z 5 0 1 20 26 Z 37 2 3 8 463 25 260 1 = 0 2 259 236 zs 173 0 5 3 2196 2 302 0 in 4 8 650 2 675 = 0 6 = 42 0 3 7 448 9 9 0 1 11 E Z 830 3 13 2 4900 : 263 0 8 3 5 100 E 67 0 8 12 : : 230 1 2 1 1 623 16 650 67 54 8 19 473 4,036 237 0 1 5 2 278 40 5 0 1 ss a 2 0 9 4800 400 76 9 9 870 150 1340 40 24, 2 21 073 8 000 535 10 35 7 10 500 1 260 70 0 2 3 : Z 70 0 2 3 Z 25 0 1 4 475 7 60 0 0 0 330 : ‘ 40 8 6 15 737 - 143 8 2 8 1 650 135 224, 0 5 2 4082 E 605 2 1 2 4 747 211 532 0 10 2 4.440 95 30 WORLD ATLAS OF BIODIVERSITY eae APPENDIX 6: IMPORTANT AREAS FOR FRESHWATER BIODIVERSITY This table presents information on areas identified as of special importance for diversity (species richness, and/or endemism) in the inland water groups treated (fishes, mollusks [‘Moll.’ in table], crabs, crayfish, fairy shrimps). This is a preliminary synthesis, designed to represent expert opinion on relative levels of Source: This information was prepared for WCMC”™ on the basis of data kindly provided by a number of expert ichthyologists and members of the IUCN/SSC Specialist Groups for Inland Water Crustaceans and Molluscs. The particular source of information is indicated by letters in square brackets in the Remarks text. Please note that numerical estimates and other information may have been superceded by later survey and taxonomic work. GA Gerald Allen, jn litt. March 1998. MK Maurice Kottelat, report compiled for wemc”™. SK Sven Kullander, report compiled for WcMc”™. MSG Adapted from a report by IUCN/SSC Mollusc Specialist Group, primarily by Philippe Bouchet and Olivier Gargominy, and also by Arthur Bogan and Winston Ponder. KC Keith Crandall, information on distribution of crayfish genera and species. DB Denton Belk, summary of fairy shrimp distribution patterns. Neil Cumberlidge and R. von Sternberg, report compiled for wemce”™. NC/RvS diversity for each taxon at continent level. In the absence of global criteria for relative importance, areas on different continents do not represent strictly equivalent levels of diversity. The rows of data in this list are sorted first by continent, and secondly by the taxon concerned. For North American and African fishes, where source is UNEP-WCMC, information has been extracted from available literature, with additional data and advice for Africa from Christian Levéque and Guy Teugels. In most instances (MK, SK, MSG, NC/RvS, DB) contributors indicated the approximate location of the important areas concerned on a series of A3-sized base maps provided. These areas, and those identified from literature by UNEP-WCMC, were digitized for purposes of presentation (Maps 7.2, 7.3, 7.4) and analysis. 1 Remarks L. Tanganyika Lower Congo Madagascar Niger-Gabon Upper Congo Upper Guinea Southern Africa Cape rivers Crabs Crabs Crabs Crabs Crabs Crabs Fairy shrimp Fishes L. Tanganyika is the only East African great lake where endemic species of freshwater crabs occur: of the 9 species and 2 genera present, | genus and 7 species are endemic. [NC/RvS] Diversity is marked in the Congo R, basin, but appears highest in 2 areas, the lower parts of the basin {including Congo, Cabinda and DR Congo (former Zaire}} and the upper reaches (including Rwanda/Burundi and parts of DR Congo). [NC/RvS] 4 genera and 10 species of freshwater crabs, all endemic, occur in Madagascar. [NC/RvS] Southeast Nigeria, southern Cameroon and Gabon: 3 endemic genera and more than 10 endemic species of freshwater crabs'. [NC/RvS] Diversity is marked in the Congo R. basin, but appears highest in 2 areas, the lower parts of the basin {including Congo, Cabinda and DR Congo (former Zaire}} and the upper reaches including Rwanda/Burundi and parts of DR Congo). (NC/RvS] Upper Guinean rainforest, centered on Guinea, Sierra Leone, Liberia, and western Cote d'Ivoire [including Mount Nimbal: 2 endemic genera and 5 endemic species of gecarcinucids**"“. [NC/RvS] 2 endemic genera, 45 species, 38 endemic. South Africa proper: 34 species, 22 endemic. [DB] With 4 families and 33 species the fish fauna of southern Africa is rather poor in comparison with most other parts of the continent; most species are cyprinids. However, there is marked local endemism; most rivers in the southern Cape region have 3 or 4 native endemics [several species are threatened). [UNEP-WCMC] APPENDIX 6 307 AFRICA 308 WORLD ATLAS OF BIODIVERSITY Papen SE AT meee ee TT = ST I DT ET EE AFRICA 9 Congo (Zaire) basin 10 Congo ‘Cuvette Centrale’ 11 Congo rapids 12 Cross R. 13 L. Barombi-Mbo 14 L. Bermin 15 L. Malawi Fishes Fishes Fishes Fishes Fishes Fishes Fishes General region of very high richness; second only to the Amazon basin in species richness. 25 families and 686 species have been reliably reported from the Congo/Zaire basin, excluding L. Tanganyika and L. Moero’. Around 548 of the species present (ca 70%) are endemic to this basin. The basin can be divided into 4 sections: Upper Lualaba, Cuvette Centrale, Luapula-Mweru and the rapids. [UNEP-WCMC] High richness plus marked endemism. Around 690 species occur in the Congo system; the Cuvette Centrale section possibly has the highest species richness owing to the great diversity of freshwater habitats available. [UNEP-WCMC] High richness plus marked endemism. The rapids between Kinshasa and the sea have a high concentration of fish species (150 species), 34 of which are endemic to this section. The caves near Thysville are fed by the Congo system and support one of Africa's few true hypogean fishes Caecobarbus geertsi. Caecomastacembelus brichardi and Gymnanallabes tihoni, not strictly cave fishes, have been collected in the Stanley Pool in rifles under flagstones or in crevices. [UNEP-WCMC] Nigeria-Cameroon. 42 families, 166 species’. Very high species diversity compared to the relatively modest catchment area, and marked endemism. Transitional ichthyofauna between the Nile-Sudan province and the Lower Guinea province. (UNEP-WCMC] This small (ca 4.5 km’) crater lake in Cameroon has 15 species (plus another 2 present in the inflow stream, not the lake proper]. At least 12 of the species are endemic, notably the 11 cichlids that form 1 of the 2 recorded ‘species flocks’ in West Africa. 4 of the 5 cichlid genera are endemic: Konia, Myaka, Pungu and Stomatepia. This very important site is at risk from overfishing, the effects of introduced crustaceans and fishes, siltation from local deforestation and water pollution. (UNEP-WCMC] A small {ca 0.5 km’) crater lake in southwest Cameroon with 2 non-endemic fishes and a remarkable species flock of 9 tilapiine cichlids. The cichlids are very small and not exploited; they are at some risk because of the small distribution and deforestation in the surrounding area. [UNEP-WCMC] 30 800 km’. 12 families, more than 845 species, most of them endemic to the lake. Rich species flocks among Cichlidae, and a small species flock of Clariidae. [UNEP-WCMC] 16 L. Tana 17 L. Tanganyika 8 L. Turkana 19 _L. Victoria 20 Madagascar 21 Niger basin 22 Ntem R. Fishes Fishes Fishes Fishes Fishes Fishes Fishes APPENDIX 6 309 The fish fauna of this large (3 150 km’) lake includes 21 species in 4 families and is dominated by lake endemic cyprinids. The large Barbus cyprinids form 1 of 2 recorded cyprinid species flocks (the other being that of L. Lanao in the Philippines, many species of which are severely threatened]. [UNEP-WCMC] 32 000 km’. In the lake itself, 16 families, around 250 cichlid species and 72 non-cichlid species. Several species flocks are present not only in Cichlidae, but also among Clariidae, Bagridae, Mochokidae, Centropomidae, Mastacembelidae. 6 750 km’. 51 species, 35 genera, 17 families. High family and generic diversity, many of the species are lake endemic; cyprinids form the most diverse family. [UNEP-WCMC] 68 800 km’. 12 families, around 545 species [many undescribed]. High species diversity dominated by cichlids. The majority of species are lake endemic. [UNEP-WCMC] Around 140 fish species have been recorded from the brackish and freshwaters of Madagascar’; although species richness is not remarkable, endemism is high. Two endemic families (Bedotiidae and Anchariidae) have been recognized in Madagascar, as well as 13 endemic genera and 43 endemic species. Most endemic species are restricted to freshwater habitats, mainly in eastern forested regions. About one quarter of endemic species are known only from the type of locality, Blind cave fishes have been described from Madagascar: the gobiid Glossogobius ankaranensis and the elotrids Typheleotris madagascarensis and T. pauliani. [UNEP-WCMC] General region of high richness. 36 families, ca 243 species, with 225 primary freshwater species’. Endemism moderate: 20 species endemic to Niger. The basin includes 11 of the 13 primary freshwater families that are endemic to Africa. 164 primary freshwater fishes reported’ from the Niger delta in Nigeria, based on reference specimens for each species; the high diversity (73% of the freshwater species in the entire basin} in this area is seriously threatened by oil pollution. [(UNEP-WCMC] Cameroon. High richness for area, plus marked endemism. 16 families, 94 species, 8 endemic. [UNEP-WCMC] AFRICA AFRICA 23 Ogooue ([Ogowe] R. 24 Sanaga R. 25 Upper Guinea rivers i) 6 Volta basin NS 7 L. Malawi ine) 8 L. Tanganyika 29 L. Victoria Fishes Fishes Fishes Fishes Moll. Moll. Moll. Gabon. High richness for area, plus marked endemism. 23 families, 185 species, 48 species endemic to Ogowe. A relatively small drainage basin with a very high concentration of species. Many of the families represented are endemic to Africa. Available data certainly underestimate actual diversity (several new species are now being described, resulting from a project of Tervuren Museum, the American Museum of Natural History and Cornell University). [UNEP-WCMC] Cameroon. 21 families; high concentration of species in a small river basin; probably at least 135 (and this figure is believed to be a significant underestimate}’. Between 10 and 18 species endemic to the Sanaga. [UNEP-WCMC] High richness for area, plus marked endemism. The Upper Guinea province includes coastal rivers from south of the Kogon R. in Guinea to Liberia, and has faunal affinities with the lower Guinea province and the Congo/Zaire. The fauna includes many taxa endemic to the area’, Many small river basins, many of them still poorly investigated. Konkoure R. (Guinea): 19 families, 85 species, at least 10 endemic. Kolente or Great Scarcies R. (Guinea-Sierra Leone): 19 families, 68 species. Jong R. (Sierra Leone}: 20 families, 94 species. Saint-Paul R, [Liberia]: 19 families, 76 species. Cess-Nipoué R. (Liberia-Céte d'Ivoire): 20 families, 61 species. [UNEP-WCMC] General region of high richness. 27 families, about 139 species, 8 endemic to Volta basin. High species richness, with 9 of the 13 African endemic primary freshwater fish families represented”. [UNEP-WCMC] Gastropods: 28 species, 16 endemic. Bivalves: 9 species, 1 endemic. [MSG] Gastropods: 68 species, 45 endemic. Bivalves: 15 species, 8 endemic. [MSG] Gastropods: 28 species, 13 endemic. Bivalves: 18 species, ? endemic. [MSG] 30 31 32 33 34 38 36 37 38 39 40 4) Lower Congo basin Madagascar Western lowland forest and Volta basin The Mollucas, New Guinea and northerm Australia Southeast Australia Southwest Australia Fly R., Papua New Guinea Kikori R., L. Kutubu, Papua New Guinea Kimberley District, Western Australia Aikwa {Iwaka} R., Irian Jaya Southeast Australia Southwest Western Australia APPENDIX 6 311 Moll. Moll. Crabs Crayfish Fairy shrimp Fishes Fishes Fishes Fishes Fishes Fishes The region downstream of Kinshasa in Congo and DR Congo (former Zaire]. Gastropods: 96 species, 24 endemic. Endemic gastropods are almost all prosobranchs; 5 endemic ‘rheophilous’ (specialized for life in the rapids] genera, belonging to the Bithyniidae [Congodoma, Liminitesta) and Assimineidae (Pseudogibbula, Septariellina, Valvatorbis). Bivalves: no data. [MSG] Gastropods: 30 species, 12 endemic. Genus Melanatria endemic. Bivalves: no data. [MSG] Upper Guinea region in Ghana, Cote d'Ivoire, Sierra Leone, Liberia, Guinea. Around 28 gastropod species of which 19 endemic (and 9 near-endemic]. Bivalves: no data. [MSG] More than 30 species of freshwater crabs belonging to genera, all in Parathelphusidae. [NC/RvS] Large area of high richness and endemism, centered on Victoria, 35 species, and Tasmania, 19 species. [KC] 19 species, 12 endemic. [DB] High species richness, 103 species in Fly proper, and high local endemism, 12 endemics in system. [GA] Headwaters of Kikori and Purari systems, with L. Kutubu. High richness, 103 species and high endemism, 16 species in Kikori; plus 14 species in L. Kutubu. [GA] 14 endemic species (a density second only in Australia to Tasmania and equal to southwest Western Australia], including 5 species within Prince Regent Reserve and 4 in the Drysdale R. area; and 47 species in total. [GA] Near Timiki, Irian Jaya. High species richness: ca 78 species. {GA] 11 endemic species occur in coastal southeast Australia, a lower count per area than the other 3 areas cited here, and 42 species in total. [GA] 9 endemic species (i.e. density similar to the Kimberleys], and 14 species in total. [GA] AFRICA AUSTRALASIA AUSTRALASIA EURASIA 312 WORLD ATLAS OF BIODIVERSITY ——— EE See re 42 Tasmania 43 Vogelkop, Irian Jaya 44 Great Artesian basin, Australia 45 New Caledonia 46 Western Tasmania, Australia 47 \ndonesia 48 Myanmar-Malaysia 49 South China 50 South India 51 Sri Lanka Fishes Fishes Moll. Moll. Moll. Crabs Crabs Crabs Crabs Crabs 12 endemic species, a greater number per area than anywhere else in Australia, including 6 concentrated in the Central Plateau area; and 24 species in total. [GA] Moderate richness with high local endemism, ca 14 endemic species, including Triton and Etna Bay lakes. [GA} Springs and underground aquifers. Important area of gastropod diversity. Bivalves: no data. [MSG] Springs and underground aquifers. Gastropods: 81 species, 65 endemic. Bivalves: no data. [MSG] Springs and underground aquifers. Important area of gastropod diversity. Bivalves: no data. [MSG] The area containing Sumatra, Java, Borneo, Sulawesi and the southern Philippines has the greatest freshwater crab diversity in Indo-Australia, with representatives of the Parathelphusidae (10 genera and 71 species) and the Gecarcinucidae (5 genera and 21 species). [NC/RvS] Northeast India (Assam), Myanmar, Thailand, the Mekong basin in southern Indochina, to the Malaysian peninsula and Singapore. In this region there are an estimated 30 genera and more than 100 species of freshwater crabs in 3 families, the Potamidae, the Parathelphusidae and the Gecarcinucidae”". [NC/RvS] Only the Potamidae occur in China, but more than 160 species and subspecies in 22 genera are present, most of which are endemic. The southern provinces of China represent the hotspot of biodiversity for this country". [NC/RvS] The freshwater crabs of the Indian peninsula south of the Ganges basin are all endemic to the subcontinent and belong to 2 families, the Gecarcinucidae and the Parathelphusidae™"’. The west coast of the peninsula and the south show most diversity: an estimated 7 endemic genera and about 20 endemic species in 2 families (the Parathelphusidae and Gecarcinucidae). A third freshwater crab family, the Potamidae, is found only in northern India but is not represented in the Indian peninsula. [NC/RvS] Sri Lanka has some 16 endemic species of freshwater crabs belonging to 3 genera, 1 of which (Spiralothelphusa) is endemic to the island". [NC/RvS] 54 55 96 37 38 APPENDIX 6 313 Group Italy Fairy shrimp Borneo highlands Fishes Caspian Sea Fishes Central Anatolia Fishes Coastal peat swamps and Fishes swamp forests of Malaysia, Sumatra and Borneo Coastal rainforest Fishes of Southeast Asia High Asia Fishes 16Species, 7 endemic. [DB] The fish fauna of the highlands of Borneo seems to be poor in absolute number of species, but many of them have developed specialization for hill-stream habitats and are endemic to single basins, The area is still largely unsurveyed. About 50 known endemic species, but actual figure might be over 200%. [MK] Moderate species richness. Although many species are shared with the Black Sea region, and/or the Aral basin, there is marked endemism, including the monotypic lamprey Caspiomyzon, ca 12 gobies, including monotypic genera Asra and Anatirostrum, also 3 Alosa. [UNEP-WCMC] An arid plateau with several endorheic lakes. About 20 endemic species, apparently underestimated by inadequate taxonomy. Adjacent areas also have a number of endemics. In urgent need of critical reassessment; probably one of the most poorly known fish faunas in Eurasia. [MK] Includes Bangka island. Extent along eastern coast of Borneo not known. Probably formerly present on Java but apparently cleared. About 100 endemic species in peat swamp forests, a habitat type often restricted to a narrow fringe along the coasts, still largely unsurveyed. Although peat swamps are traditionally considered as a habitat with poor diversity, good data for limited areas in Malay peninsula and Borneo indicate that up to 50 species may be found within a small area [less than 1 km‘), about half of them endemic and stenotypic. Most species have small distribution ranges [some possibly only a few km‘]®”. [MK] Thailand, Cambodia and southern Viet Nam. Southern extent not known accurately. This habitat is largely destroyed in Thailand, and virtually unsurveyed in Cambodia and Viet Nam. Endemic species expected in peat swamp forests”. [MK] Boundaries not known with accuracy; includes the Tibetan plateau and probably parts of Chinese Turkestan. Distribution and ecological data are sparse outside the Chinese literature. About 150 known fish species, about half of them endemic to this area’. Survey probably still superficial as a result of difficulties of access. [MK] EURASIA EURASIA on 9 o 0 o 1 o~ 2 o~ (ee) o 4 o 5 66 oO 7 Karstic basins of Yunnan, Guizhou and Guangxi L. Baikal, Siberia L. Biwa, Japan L. El’'gygytayn, Siberia L. Inle, Myanmar L. Lindu, Sulawesi L. Poso, Sulawesi L. Thingvalla, Lakes of Isles Fishes Fishes Fishes Fishes Fishes Fishes Fishes Fishes Fishes Boundary not known with accuracy. About 14 known species of cave fishes. Survey Is still superficial and numerous additional species are expected”. [MK] A species flock of 36 species of the family Cottidae {sculpins] including the endemic family Comephoridael, 4 ‘ecologically differentiated stocks’ [many probably endemic species using Western concepts) of Coregonus, 2 of Thymallus, 2 of Lota”. Endemic mollusks, gammarids, sponges and Baikal seal. {MK] Reportedly 4 endemic species”. [MK] An old lake formed on the site of a meteorite crater. 113 km’. Total fish diversity: 5 species, including an endemic genus and species (Salvethymus svetovidovil, an endemic species (Salvelinus elgyticus), and 1 species endemic to eastern Siberia (Salvelinus boganidae}. Endemic diatom species and apparently endemic invertebrate(s]*~. [MK] About 25 native fish species, ca 10 of them endemic, including 3 endemic genera®™. [MK] Very limited information. One native and endemic species; others might be expected”. [MK] 10 native and endemic species, 2 endemic genera (both extinct?) and with L. Lindu comprises the entire known distribution of the subfamily Adrianichthyinae””. Additional species might still be expected, [MK] 5 native fish species, including 3 endemic Salvelinus (recent summary in Kottelat“!). [MK] A number of lakes host 1 or 2 species of British Salvelinus, although information on individual lakes is usually inadequate. At the beginning of the century up to 14 species were recognized; although generally not accepted under later systematic concepts, recent work suggests that this figure may be underestimated. Also at least 5 endemic Coregonus, 1 endemic Clupeidae and potential for endemic Salmo {recent summary in Kottelat‘'), [MK] 68 Lakes of Central Yunnan, China 69 Lough Melvin, Ireland 70 Lower Danube 7\ Mainland Southeast Asian hills Fishes Fishes Fishes Fishes Lakes Dianchi, Fuxian, Er Hai, Yangling, Yangzong, Xingyun, etc., have a distinctive fauna; despite the lakes being now in different river basins (Mekong, Yangtze, Nanpangjiang], they have similar fauna, characterized by numerous endemic species in the genera Cyprinus, Schizothorax, Anabarilius and Yunnanilus. Exact up-to-date figures of the number of species are difficult to extract from the Chinese literature, but we have the following data: Dianchi: 25 native species, 11 endemic of which apparently all but 2 are extinct. The lake basin has 2 other endemics”; Fuxian: 25 native species, 12 endemic plus 2 endemic shared only with Xingyun*; Er Hai: 17 native species, 9 endemic, several apparently extinct; Yangzong has (had) at least 2 endemics; Yangling and Xingyun at least 1 each. [MK] Three endemic species of Salmo {recent summary in Kottelat“}. [MK] The lower Danube basin has a relatively richer fauna (especially more diverse communities] than any other European river. Endemics: about 6, possibly underestimated {counted in Kottelat 1997"'). [MK] Northern boundary not clear as published data on fish distribution (and ground surveys) in southern China are too scanty, Could be subdivided into a] upper Song Hong {includes hills of Hainan and southern Nanpang Jiang); b) Annamite cordillera; c) upper Mekong, ChaoPhraya and Mae Khlong basins; d) Salween, upper Irrawaddy and southeastern Assam including Tenasserim). Recorded fish fauna estimated to be over 1 000 species [with an estimated 200-500 species still awaiting discovery], 500 endemic to this area. Includes ca 400 known species endemic to headwaters of individual sub-basins. The fauna of the lower reaches of the main rivers (excluded from this polygon) is richer {in terms of the number of species that can be observed at a given locality) but most have wide distributions crossing several river basins”““. [MK] APPENDIX 6 315 316 WORLD ATLAS OF BIODIVERSITY EURASIA 73 Maros karst, Sulawesi 74 Mindanao, Philippines 75 Northwest Mediterranean drainage 76 Palawan, Philippines Fishes Fishes Fishes Fishes Most important single site for aquatic biodiversity in Asia. A complex of 5 lakes (Towuti, Matano, Mahalona, Wawontoa, Masapi) with endemic radiations of fishes of the families Telmatherinidae (3 genera, 15 species, all but 1 endemic}, Hemiramphidae [3 endemic species), Oryziidae [3 endemic species), Gobiidae (at least 8, all but 1 endemic], prawns (ca 12 species?], crabs [4 species?], mollusks (ca 60 endemic species}, etc. The distribution of the fishes is not uniform within the lakes, all but 1 of the species of L. Matano are endemic, while the others {and 2 genera} are endemic to Towuti, Mahalona, Wawontoa. Masapi has not yet been surveyed. Only 2 species of the Telmatherinidae are known outside this area. [MK] One endemic genus (possibly an artifact of limited collection; more surveys might show it to be present outside this area) and about 6 endemic species, including a cave species”. [MK] About 30 endemic species of cyprinid fishes, including about 18 endemic species of Puntius in L. Lanao [all but 2 or 3 reportedly extinct}. Cyprinids are fishes which live only in freshwater and cannot disperse in marine environments; several other families also occur in the island's freshwaters, but all are able to disperse through the seas. [MK] Includes Spain, Portugal, southern France and northern Italy. The total diversity in the whole area is quite low, the communities are quite poor, but this area holds 55 endemics, many with small distribution ranges. 3 of the Rhone endemics extend almost to the northern extremity of the basin. Endemics:1 Petromyzonidae, 1 Acipenseridae, 1 Clupeidae, 34 Cyprinidae, 5 Cobitidae, 6 Salmonidae, 1 Valenciidae, 1 Cyprinodontidae, 2 Cottidae, 1 Percidae and 4 Gobiidae (counted in Kottelat‘'). [MK] About 10 recorded species of cyprinid fishes, actual figure probably higher. [MK] APPENDIX 6 317 Group Remarks 77 78 79 80 81 82 83 fos) 4, Southwest Balkans Southwest Sri Lanka Subalpine lakes Sundaic foothills and floodplains Western Ghats, India Balkans region (Former Yugoslavia, Austria, Bulgaria, Greece] Chilka L. L. Baikal Fishes Fishes Fishes Fishes Fishes Moll. Moll. Moll. The total diversity in the whole area is quite low, the communities are quite poor, but the area holds 84 endemics, most of them with restricted or very restricted distribution ranges: 1 Petromyzonidae, 2 Clupeidae, 48 Cyprinidae, 8 Cobitidae, | Balitoridae, 1 Siluridae, 13 Salmonidae, 1 Valenciidae, 1 Gasterosteidae, 1 Percidae and 7 Gobiidae. The systematics of many groups is still very poorly known and more species will be recognized or even discovered in the future (possibly 10-20). Noteworthy are L. Ohrid with apparently 4 endemic Salmo, L. Prespa with apparently 7 endemic species and the Vardar basin with at least 8 endemic species“. [MK] 28 of the 91 native fish species of Sri Lanka are endemic to this area. Several of the species traditionally given the same name as Indian species are being revised and turn out to be specifically distinct, so that the figure will rise. [MK] Stretches from L. Bourget in the west to Traunsee in the east. Numerous endemic Coregonus (possibly 27, several already extinct], at least 2 endemic Salvelinus and possibly some endemic Salmo. Some lakes have more complex communities, e.g. L. Konstanz with 4 Coregonus, 2 Salvelinus, 1 Salmo and several other species [recent summary in Kottelat*. [MK] About 400 known species. Most of the floodplain species are widely distributed over the whole area, while those of foothill streams have more localized distributions and are of greater interest in terms of endemicity. Northern limit: Tapi basin in peninsular Thailand *”*". [MK] About 100 endemic fish species estimated from Talwar and Jhingran®; Pethiyagoda and Kottelat®). Difficult to give accurate figures. Many wide-ranging species’ of fishes in South Asia are in fact complexes of species, so that the actual number of species is likely to increase significantly after adequate systematic revision. [MK] Springs and underground aquifers. Gastropods: ca 190 species, some 180 endemic. Bivalves: no data. [MSG] Brackish water. Gastropods: 28 species, ca 11 endemic. Bivalves: 43 species, 25 endemic. [MSG] Gastropods: 147 species, 114 endemic. Bivalves: 13 species, 3 endemic. [MSG] EURASIA 318 WORLD ATLAS OF BIODIVERSITY EURASIA 85 L. Biwa Moll. Gastropods: 38 species, 19 endemic. Bivalves: 16 species, 9 endemic. [MSG] 86 L. Inle Moll. Gastropods: 25 species, 9 endemic. Bivalves: 4 species, 2 endemic. [MSG] [oe] 7 L. Ohrid and Ohrid basin Moll. Gastropods: 72 species, 55 endemic. Bivalves: no data. [MSG] fos] 8 L. Poso and Malili Moll. Sulawesi. Gastropods: ca 50 species, ca 40 endemic. Lakes system Bivalves: 5 species, 2 endemic. [MSG] 89 Lower MekongR., Thailand, Moll. River habitat. Only ca 500 km of the lower Laos, Cambodia Mekong main course [with the tributary Mun R.) has been well studied, Gastropods: 121 species, 111 endemic. Two rissoacean groups dominate this entirely prosobranch assemblage of 120 plus species, the pomatiopsid Triculinae (92 endemic species, 11 endemic genera] and the Stenothyridae (19 endemic species). Bivalves: 39 species, 5 endemic. [MSG] 90 North Western Ghats Moll. River habitat. Gastropods: ca 60 species, 10 endemic. 2 endemic genera Turbinicola, Cremnoconchus. The succineid genus Lithotis is known from 2 species: L. tumida not collected since its description in 1870, and L. rupicola only known from a single locality. The highly localized genus Cremnoconchus is the only littorinid living in a freshwater/terrestrial environment. Bivalves: 11 species, 5 endemic. [MSG] 91 Zrmanja R., Croatia Moll. Gastropods: all are hydrobioid snails, 11 species, 5 endemic. Bivalves: no data. [MSG] NORTH AMERICA 92 Southeast USA Crayfish Large area of high richness and endemism at generic and species level; including the eastern and southern Mississippi drainage (Ohio R., Tennessee R., to Ozark and Ouachita mountains]; 72 species in Alabama, 71 in Tennessee. [KC] 93 Western USA Fairy shrimp — 26 species, 13 endemic. [DB] 94 BearL. Fishes This lake is part of the Bonneville R. basin and contains 1 local endemic [Prosopium gemmiferum) and 2 species that are now restricted to this site [Prosopium spilonotus and P. abyssicola|*. [UNEP-WCMC] APPENDIX 6 319 NORTH AMERICA 95 Colorado basin Fishes The largest basin of the western USA, this has high species richness and endemism, including 5 endemic genera of which only Plagopterus is monotypic. About one third of the ichthyofauna of the Colorado is threatened, endangered or extinct due to dams and introduced species”. [UNEP-WCMC] 96 Cumberland Plateau Fishes This area has the highest species richness and local (Cumberland and endemism in North America. It is part of the highly diverse Tennessee rivers] Mississippi basin, with ca 240 species in total, 160 present in both the Tennessee and Cumberland drainages, 14 endemic to the 2 basins. Of these 14, 10 are darters, 3 are minnows and 1 is a topminnow, The Tennessee has the greatest species diversity with 224 species including 25 endemics {as well as 64 not found in the Cumberland). The Cumberland has 176 native species, including 9 endemics and 16 species not shared with the Tennessee”. [UNEP-WCMC] 97 Death Valley region Fishes There is a high level of local endemism associated with the dispersed pattern of springs and marshes. 4 families are present (Cyprinidae, Catostomidae, Cyprinodontidae and Goodeidae} with 9 species including an endemic species of Catostomidae™. Several are globally threatened, including 2 of the 5 Cyprinodon species (Cyprinodon radiosus and C. diabolis}. [UNEP-WCMC] 98 Eastern USA Fishes This is a general area of high species richness and endemism which, with the possible exception of the incompletely known East Asian fish species, represents the most diverse of all the freshwater faunas of the temperate zone™. This includes a} the Ozark Plateau, b] the Quachita Mountains, c] the South Atlantic Central Plain and d} the Tennessee-Cumberland Plateau. [UNEP-WCMC] 99 Klamath-upper Sacramento Fishes The Klamath R. basin contains 28 species in total with relatively high endemism. The 6 endemic species include 2 Catostomus, 1 Chasmistes and 1 Gila*. The ichthyofauna of the Sacramento differs from that of the Klamath and contains 4 genera that are confined to this river and a few neighboring drainages®.[UNEP-WCMC] 100 Ouachita Mountains Fishes This area includes parts of the lower Red and Ouachita rivers, each containing 133 species. The Ouachita and the Red river systems both contain 18 endemic species®. [UNEP-WCMC] 320 WORLD ATLAS OF BIODIVERSITY NORTH AMERICA Area name Group — 101 Ozark Plateau Fishes The Ozark Plateau is an area of high species diversity and particularly high local endemism in the southeast USA; it represents a concentration of the species-rich southwestern Mississippi drainage (more than 30 endemic fish species]*. [UNEP-WCMC] 102 Rio Grande-Pecos confluence Fishes The Rio Grande basin overall has more than 60 endemic species” and the Pecos, a tributary, has 5”. Many of the endemics occur at the confluence of the 2 rivers, and many are globally threatened. [UNEP-WCMC] 103 Southern Oregon-California © Fishes These rivers share few family similarities with the eastern rivers USA and have about 25% of the number of species, but the region is high in local endemism. [UNEP-WCMC] 104 Southern Atlantic Fishes This includes the Alabama-Tombigbee basin with a species- coastal plain rich fauna, including about 40 endemic taxa”. This region also contains the Pearl R., with 106 species” and the species- rich lower Mississippi. [UNEP-WCMC] 105 Arid/semi-arid western USA Moll. Springs and underground aquifers. Gastropods: all are hydrobioid snails, ca 100 species, at least 58 endemic. Great radiation in genus Pyrgulopsis. 3 extinct species, and all others are candidates for listing by US Fish and Wildlife Service. Bivalves: no data. [MSG] 106 Cuatro Cienegas basin, Moll. Springs and underground aquifers. Gastropods: all are Mexico hydrobiids; 12 species, more than 9 endemic. 5 genera’ (Nymphophilus, Coahuilix, Paludiscala, Mexithauma, Mexipyrgus) are endemic to this small area of 30 x 40 km. Bivalves: no data. [MSG] 107 Florida, USA Moll. Springs and underground aquifers. Gastropods: mostly hydrobiid snails. 84 species, ca 43 endemic. No bivalves. [MSG] 108 Mobile Bay basin Moll. Tombigbee-Alabama rivers. River habitat. Gastropods: 118 species, 110 endemic; 6 endemic genera; greatest species richness (76 species) in Pleurocercidae. 38 of the gastropod species believed extinct, 70 candidates for listing by US Fish and Wildlife Service. Bivalves: 74 species, 40 endemic, 25 extinct. [MSG] 109 Ohio-Tennessee rivers Moll. Eastern Mississippi drainage. River habitat. High species richness and endemism. [MSG] APPENDIX 6 321 pare safflower seed 260 Sahara desert 109 Sahul, colonization 37 Saiga tatarica 104 saline lakes 164 salmon, Atlantic 153 salmonid fish 126, 172, 788 salt marshes 133 Salviniaceae 170 satellite remote sensing 73, 74-5 savannahs 102-5, 108-9 wooded 96 sawfishes 126, 158, 173, 188 scenarios 217 GLOBIO 219-21 RIVM IMAGE 217-19 Schoningen, Germany 37 Scleractinia 138 sclerophyllous forests 84, 92 Scott, Sir Peter 197 sea krait 127 sea level rises 154 sea otter 130 sea snakes 126 sea turtles 127-8, 130, 147 seabirds 128-9, 147, 155, 157 seagrasses 134-6, 136-7 seahorse 135 seals 131 seas see marine biosphere; oceans and named seas seawater 118 seaweeds 122-3, 124-5, 133, 138, 149, 150 sedge-moss communities 102 sedimentation 119, 136, 185 sediments, marine 38, 179 semi-deserts 109-11 Sequoiadendron giganteum 82 sesame seed 260 Shannon-Wiener function 78 sharks 125, 128-9 shea nut 260 sheep 274 shifting cultivation 91 shorelines, rocky 133-4 shrimps aquaculture 150 fishing 147, 148 see also fairy shrimp shrublands 73 biodiversity 105-9 global distribution 76-7, 108-9 Sierra Club 797 Sierra Leone, marine sediments 38 Silurian period 25, 26 Siluriformes 173, 174, 188 Sirenia 130, 135, 775, 177 sloths, ground 53 snails 64 snakes brown tree 64-5 inland waters 174 marine 126-7 snow, permanent 163 soala 16 soda lakes 164 soil protection 91 soil water 73-4 soils grasslands 104 temperate forests 83, 84, 85 tropical forests 84, 87, 88, 89,94 tundra 102 Solenosmilia variabilis reefs 139 Somateria fischeri 102 sorghum 246 South Africa 105, 106-7, 108 South America alpine tundra 101 forests 84, 85 inland water biodiversity 321-4 see also individual countries Soviet Union, former 97, 103, 104-5 soybean 252 sparse trees 95-6, 106-7 speciation 13 Species 2000 17 species concepts and definition 14- 16 dispersal 79-80 lifespan in fossil record 28 living number estimates 16-17, 18 new discoveries 16 nomenclature 14 protection 197 rediscovered 57 reintroductions 212 species-area relationship 76-8 species-energy relationship 50 species-latitude relationship 78-9, 123, 129, 131, 168 Sphagnales 169 Sphenophytes 170 spices 267-8 spinach 267 Spix’s macaw 93 squash 40, 262 Steller’s eider 102 Steller’s sea cow 130 stone tools 37, 39 stoneworts (charophytes) 169 Straddling Fish Stocks Agreement 215 strawberry 262 streams, upland 163 Strix occidentalis caurina 82 stromatolites 24 Stylites 170 sub-Saharan Africa 38, 42 subspecies 15 Suez Canal 153 INDEX 339 SS sugar beet 269 sugar cane 254 sugar crops 254, 269 sunflower seed 253 swamps 167 Swartkrans cave 38 sweet orange 262 sweet pepper 268 sweet potato 257 Sypheotides indica 104 systematic conservation planning 203 systematics data sources 17 groups and names 14 roles of 13-14 species concepts 14-16 T taiga 95-6 tangerine 262 tannia 257 taro 183, 257 tarpons 173 taxonomy defined 13-14 global initiatives 17 species recognition 14-16 tea 266 teak 92, 94 Tectona grandis 92, 94 tectonic activity 142-3 tectonic lakes 165 temperate broadleaf forests 83-6 biodiversity 84-5 carbon cycle 84, 85 distribution, types and taxa 83-4, 106-7 human use and ecosystem services 86 losses 97 structure and ecology 84 temperate needleleaf forests 81-3 biodiversity 82-3 carbon cycle 83, 84 distribution, types and taxa 81, 106-7 human use and ecosystem services 83 structure and ecology 82 termite mounds 92 termites 89 terrestrial biodiversity fossil record 26-8 global variations 75-80 threatened species 63 vertebrate diversity 100-1 see also individual ecosystems terrestrial biosphere 71-2 habitat variations 72-5 primary productivity 70 see also individual ecosystems Tertiary period climate 34 tetrapod diversity fossil record 27-8 inland waters 174-8 marine 125, 126-31 see also individual groups and species Thailand 93, 94, 150 Theragra chalcogramma 145, 146 Thiomicrospira spp. 142 thorn forest 92 threatened species arid regions 110 assessment 60-1 by country 295-305 causes of recent declines 64-5 forests 82, 84-5, 89, 93 grasslands 104 inland waters 788-9 major biomes 63 mammals 58-9 marine biosphere 154-8 plants 67 recent population declines 61-4 tundra 102 tilapia, Nile 180 timber production 80, 98 forest plantations 96 temperate forests 83, 85, 86 tropical forests 90, 94 toad, golden 99 tomato 266 tools, hominid/early human 37, 39 topography global 4-5 and species diversity 79 tortoise, Madagascar flat- tailed 93 trade 65 aquarium/ornamental fish 150-1, 182 timber 85, 90, 94, 98 traditional agriculture 45 traditional medicine 42, 183 trawl fishing 139 treaties, multilateral 212-16 tree nuts 269-70 trees factors in distribution 73-5 threatened species 85 see also named species; forests Triassic period 28 Triassic-Jurassic boundary 29, 30 Trichechus spp. 130, 175, 177 Trimmatom nanus 125 tropical dry forests biodiversity 93-4 carbon cycle 84, 94 distribution, types and taxa 91-2, 106-7 human use and ecosystem services 94-5 losses 97 structure and ecology 92 tropical moist forests 86-91 biodiversity 60, 78-9, 87-90 carbon cycle 84, 89-90 distribution, types and taxa 86, 106-7 ecosystem services 91 human use 90 structure and ecology 86-7 Truong Son muntjac 76 tuber crops 248-9 tundra 74, 99-102, 108-9 turtles intand waters 174, 175, 186 marine 125, 127-8, 130-1, 147 U UNCCD see United Nations Convention to Combat Desertification UNCLOS see United Nations Convention on the Law of the Sea undernourishment 44 undescribed species 16-17 UNEP see United Nations Environment Programme UNESCO see United Nations Educational, Scientific and Cultural Organization UNFCCC see United Nations Framework Convention on Climate Change Note: Page numbers in bold refer to figures or maps in the text; those in italics to boxed material or tables in the text or Appendices. ¢ 340 WORLD ATLAS OF BIODIVERSITY Ve a SSS Note: Page numbers in bold refer to figures or maps in the text; those in italics to boxed material or tables in the text or Appendices. United Nations Conference on Environment and Development (Earth Summit) 1, 195, 196, 199 United Nations Convention to Combat Desertification (UNCCD) 111, 275 United Nations Convention on the Law of the Sea (UNCLOS) 207, 213, 215 Straddling Fish Stocks Agreement 215 United Nations Educational, Scientific and Cultural Organization (UNESCO) forest classification 80, 81 Man and the Biosphere (MAB) Programme 210-11, 216 United Nations Environment Programme (UNEP], Global Environment Outlook (GEO) 217-21 United Nations Framework Convention on Climate Change (UNFCCC) 275, 217 United States forests 84 inland water biodiversity 318-20 see also North America upwelling zones 118-19, 121, 131, 140-1, 145 V vascular plants alpine communities 100-2 extinctions 29 global diversity 88-9 inland waters 171 see also named groups and species vegetable crops 249, 260-1 vegetation mapping 73, 74-5 Vendian period 28 Venezuela 101, 103 vent communities 4, 142-3 Verreaux’s sifaka 93 vertebrates 229 boreal needleleaf forests 82 fossil record 27-8 global distribution of families 79 global terrestrial diversity 100-1 marine diversity 723 recent extinctions 59, 278- 93 threatened species 61-4 see also individual groups and species Vicugna vicugna 104 vicuna 104 Viet Nam 16 viruses 19 volcanic lakes 165 Vu Quang Nature Reserve, Viet Nam 16 W wading birds 102, 128, 175, 176-7 wakame 149 walnut 269 water abstraction 178, 184 availability and vegetation 72, 73-4 as environment for life 72 global resources 163 see also freshwater; inland waters; seawater water buffalo 273 water resource vulnerability index (WRVI) 189-90 watercraft, human 37 watermelon 265 watersheds see catchment basins Weddel Sea, Greenland 118 wetlands 167 exploited species 182-3 losses 184 mammals 178 protected areas 205, 214, 215-16 whale shark 125 wheat 247 wild ox 49 wilderness measures 47, 54- 5, 188-9 wolf domestication 39-40 Mexican gray 212 wood see fuelwood; timber production woodlands global distribution 76-7 open 95-6 see also forests World Conservation Strategy 196 World Conservation Union (IUCN) 60-1, 196, 197, 211 World Food Summit 44 World Heritage Convention 214, 216 World Heritage sites 210-11 World Parks Congress (1962) 198 WRVI see water resource vulnerability index WWF-World Wide Fund for Nature 197 ecoregions system 80, 120, 168-9, 202 Forests for Life program 211 living planet index 85, 104, 158, 186, 187-8 if yak 272 yams 248 Yellow River 119 Yellowstone National Park 82 Z Zambia 180 Zhoukoudien, China 38 BRIAN GROOMBRIDGE MARTIN D. JENKINS “This is an exciting piece of work—well written, well researched, and authoritative.” = =... Editor-in-Chief of Encyclopedia of Biodiversity “A thorough and up-to-the-minute account of the variety of life on Earth and how fast it is shrinking. One cannot plan strategically to prevent the loss of that variety without the information and maps that this excellent and timely book provides.” =) ))5)) 3 Su). Columbia University “The book amounts to a detailed report card on the status of the world’s biodiversity resources at the beginning of the new millen- ium. It brings together, and integrates to substantial degrees, huge amounts of information that is widely scattered in both technical literature and in governmental and NGO reports.” 5-092) © 5)0)9). coauthor of Invasions of the Land: The Transitions of Organisms from Aquatic to Terrestrial Life “Tt is an amazingly diverse collection of data on global biological diversity—very effectively analyzed and dis- played.” 5 Ss Wildlife Research Group, University of Cambridge “An indispensable resource on information about Earth’s biological variety and why its conservation is crucially important for human survival and well being. This vol- ume should be in the hands of biologists, policymakers, educators, and the general public concerned with our global environ- 12 OEE CRACKAR Ts Gi: cso ts Charge, Department of Ornithology, Amer- ican Museum of Natural History Printed in the United Kingdom UNIVERSITY OF CALIFORNIA PRESS Berkeley 94720 www.ucpress.edu