| | i [ [ 5é zy — | ni agriculture, nature rem —— se CENSUS — ) ‘ =| and food quality cues: ea GFMARINE FE. = NLLWA-— El Seamounts, deep-sea corals and fisheries Census of Marine Life on Seamounts (CenSeam) Data Analysis Working Group SZ 0EaS Digitized by the Internet Archive in 2010 with funding from UNEP-WCMC, Cambridge http://www.archive.org/details/seamountsdeepsea06clar \/ p) MiNGrhar 24) Seamounts, deep-sea corals and fisheries i ] iis” | a | Vulnerability of deep-sea corals to fishing on seamounts beyond areas of national jurisdiction Sty UNEP WCMC ~o!8aS ® Regional wo eg {oa yy, (69) ——=_ UNEP WCMC UNEP World Conservation Monitoring Centre 219 Huntingdon Road, Cambridge CB3 ODL, United Kingdom Tel: +44 (0) 1223 277314 Fax: +44 (0) 1223 277136 Email: info@unep-wemc.org Website: www.unep-wcmc.org @UNEP-WCMC/UNEP 2006 ISBN: 978-92-807-2778-4 A Banson production Design and layout J-P Shirreffs Printed in the UK by The Lavenham Press Photos Front cover: Left, Cold-water coral (Lophelia pertusa), André Freiwald, IPAL-Erlangen; Centre, Multibeam image of Ely seamount [Alaska] with the caldera clearly visible at the apex. Jason Chaytor, NOOA Ocean Explorer (http://oceanexplorer.noaa.gov/ explorations/O4alaska); Right, Orange roughy haul. Image courtesy of M Clark {NIWA]. Back: Multibeam image Brothers NW, NIWA We acknowledge the photographs (pp 16, 21, 25, 26, 29, 31, 47] from the UK Department of Trade and Industry's offshore energy Strategic Environmental Assessment {SEA] programme care of Dr Bhavani Narayanaswamy, Scottish Association for Marine Science and Mr Colin Jacobs, National Oceanography Centre, Southampton. DISCLAIMER The contents of this report are the views of the authors alone. They are not an agreed statement from the wider science community, the organizations the authors belong to, or the CoML. The contents of this report do not necessarily reflect the views or policies of the United Nations Environment Programme, the UNEP World Conservation Monitoring Centre, or the supporting organizations. The designations employed and the presentations do not imply the expressions of any opinion whatsoever on the part of these organizations concerning the legal status of any country, territory, city or area or its authority, or concerning the delimitation of its frontiers or boundaries. Moreover, the views expressed do not necessarily represent the decision or the stated policy of UNEP or contributory organizations, nor does citing of trade names or commercial processes constitute endorsement ACKNOWLEDGEMENTS The authors would like to thank Matt Gianni Advisor, Deep Sea Conservation Coalition and Dr Stefan Hain, Head, UNEP Coral Reef Unit for promoting the undertaking of this work and for many constructive comments and suggestions. Adrian Kitchingman of the Sea Around Us Project, Fisheries Centre, University of British Columbia, kindly provided seamount location data, and lan May at UNEP-WCMC re-drew most of the original maps and graphics. Derek Tittensor would like to thank Prof. Ransom A Myers and Dr Jana McPherson, Department of Biological Sciences, Dalhousie University and Dr Andrew Dickson, Scripps Institute of Oceanography for discussions, comments, and advice on modelling of coral distributions and ocean chemistry; Dr Alex Rogers would like to thank Prof. Georgina Mace, Director of the Institute of Zoology and Ralph Armond, Director-General of the Zoological Society of London for provision of facilities for undertaking work for this report. Dr Amy Baco-Taylor (Woods Hole Oceanographic Institution) provided very useful comments ona draft of the manuscript. Three reviewers provided comments on the final manuscript. Dr Christian Wild, GeoBiocenter LMU Muenchen reviewed and guided the report on behalf of UNESCO- IOC. The authors gratefully acknowledge the support of the Census of Marine Life programmes FMAP (Future of Marine Animal Populations} and CenSeam (a global census of marine life on seamounts). The Netherlands’ Department of Nature, Ministry of Agriculture, Nature and Food Quality are gratefully acknowledged for funding the original workshop and NIWA for hosting the meeting. Hans Nieuwenhuis is additionally acknowledged for help in bringing this report to fruition. The UNEP Regional Seas Programme and the International Oceanographic Commission of UNESCO kindly provided funding for the publication of this report. Citation: Clark MR, Tittensor D, Rogers AD, Brewin P, Schlacher T, Rowden A, Stocks K, Consalvey M (2006). Seamounts, deep-sea corals and fisheries: vulnerability of deep-sea corals to fishing on seamounts beyond areas of national jurisdiction. UNEP- WCMC, Cambridge, UK. URL: www.unep-wemc.org/resources/publications/ UNEP_ WCMC_bio_series/ and www.unep.org/regionalseas/Publications/Reports/ Series_Reports/Reports_and_Studies The work was initiated with data compilation and analysis in 2005 supported by CenSeam, followed by a workshop funded by the Department of Nature, Ministry of Agriculture, Nature and Food Quality, Netherlands, which was held at the National Institute of Water and Atmospheric Research (NIWA) in Wellington, New Zealand, from 8 to 10 February 2006 CENSUS OF MARINE LIFE AND CENSEAM The Census of Marine Life [CoML] is an international science research programme with the goal of assessing and explaining the diversity, distribution and abundance of marine life - past, present and future. It involves researchers in more than 70 countries working on a range of poorly understood habitats. In 2005 a CoML field project was established to research and sample seamounts [Stocks et al. 2004; censeam.niwa.co.nz]. This project, termed CenSeam [a Global Census of Marine Life on Seamounts), provides a framework to integrate, guide and expand seamount research efforts on a global scale. It has established a seamount researcher network of almost 200 people around the world, and Is collating existing seamount information and expanding a data- base of seamount biodiversity. Its Steering Committee comprises people who are at the forefront of seamount research, and can therefore contribute a wealth of knowledge and experience to issues of seamount biodiversity, fisheries and conservation. One of the key themes of CenSeam is to assess the impacts of fisheries on seamounts, and to this end, it has established a Data Analysis Working Group (DAWG) that includes people with a wide range of expertise on seamount datasets and analysis and modelling techniques. Seamounts, deep-sea corals and fisheries AUTHORS Malcolm R Clark* National Institute of Water and Atmospheric Research PO Box 14-901, Kilbirnie, Wellington, New Zealand Derek Tittensor* Department of Biological Sciences Life Sciences Centre 1355 Oxford Street Dalhousie University Halifax, Nova Scotia, B3H 4J1, Canada Alex D Rogers* Institute of Zoology Zoological Society of London Regent's Park, London NW1 4RY, United Kingdom Paul Brewin San Diego Supercomputer Center University of California San Diego 9500 Gilman Drive, La Jolla, CA 92093-0505, USA Thomas Schlacher Faculty of Science, Health and Education University of the Sunshine Coast Maroochydore DC Qld 4558, Queensland, Australia Ashley Rowden National Institute of Water and Atmospheric Research PO Box 14-901, Kilbirnie, Wellington, New Zealand Karen Stocks San Diego Supercomputer Center University of California San Diego 9500 Gilman Drive, La Jolla, CA 92093-0505, USA Mireille Consalvey National Institute of Water and Atmospheric Research PO Box 14-901, Kilbirnie, Wellington, New Zealand *These authors made an equal contribution to this report and are therefore joint first authors. Supporting organizations Regional UNEP’s Regional Seas Programme aims to address the accelerating degradation of the world’s oceans and coastal areas through the sustainable management and use of the marine and coastal environment, by engaging neighbouring countries in comprehensive and specific actions to protect their shared marine environment. Cm ~ 0) UNEP WCMC The UNEP World Conservation Monitoring Centre (UNEP-WCMC) is the biodiversity assessment and policy implementation arm of the United Nations Environmental Programme (UNEP), the world’s foremost intergovernmental environment 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 programmes of action The Intergovernmental Oceanographic Commission (IOC) of the United Nations Educational, Scientific and Cultural Organization {UNESCO} provides Member States of the United Nations with an essential mechanism for global cooperation in the study of the ocean. The IOC assists governments to address their individual and collective ocean and coastal problems through the sharing of knowledge, information and technology and through the coordination of national programmes agriculture, nature and food quality all Department of Nature, Ministry of Agriculture, Nature and Food Quality, Netherlands. CENS i oe OF MARINE LIFE The Census of Marine Life (CoML) is a global network of researchers in more than 70 nations engaged in a ten-year initiative to assess and explain the diversity, distribution and abundance of marine life in the oceans - past, present and future. es N. - — Taihoro Nukurangi The National Institute of Water and Atmospheric Research (NIWA) is a research organization based in New Zealand, and is an independent provider of environmental research and consultancy services. Foreword Foreword ‘How inappropriate to call this planet Earth, when it is quite clearly Ocean’ look at a map of the world shows how true this Acoeres is. Approximately two-thirds of our planet is covered by the oceans. The volume of living space provided by the seas is 168 times larger than that of terrestrial habitats and harbours more than 90 per cent of the planet's living biomass. The way most world maps depict the oceans is deceiving: while the land is shown in great detail with colours ranging from greens, yellows and browns, the sea is nearly always indicated in subtle shades of pale blue. This belies the true structure of the seafloor, which is as complex and varied as that of the continents - or even more so. Some of the largest geological features on Earth are found on the bottom of the oceans. The mid-ocean ridge system spans around 64000 km, four times longer than the Andes, the Rocky Mountains and the Himalayas combined. The largest ocean trench dwarfs the Grand Canyon, and is deep enough for Mount Everest to fit in with room to spare. Only in the last decades, advanced technology has revealed that there are also countless smaller features - seamounts - arising in every shape and form from the sea floor of the deep sea, often in marine areas beyond national jurisdiction. Observations with submersibles and remote controlled cameras have documented that seamounts provide habitat for a large variety of marine animals and unique ecosystems, many of which are still to be discovered Veerle Vandeweerd, Head, United Nations Environment Programme (UNEP] Regional Seas Programme, Coordinator, GPA Monitoring Centre Jon Hutton, Director, UNEP World Conservation attributed to Arthur C Clarke and described. However, the same observations also provided alarming evidence that seamount habitats are increasingly threatened by human activities, especially from the rapid increase of deep-sea fishing. The United Nations General Assembly has repeatedly called upon States and international organizations to urgently take action to address destructive practices, such as bottom trawling, and their adverse impacts on the marine biodiversity and vulnerable ecosystems, especially cold-water corals on seamounts. This report, compiled by an international group of leading experts working under the Census of Marine Life programme, responds to these calls. It provides a fascinating insight into what we know about seamounts, deep-sea corals and fisheries, and uses the latest facts and figures to predict the existence and vulnerability of seamount communities in areas for which we have no or only insufficient information. The deep waters and high seas are the Earth's final frontiers for exploration. Conservation, management and sustainable use of the resources they provide are among the most critical and pressing ocean issues today. Seamounts and their associated ecosystems are important and precious for life in the oceans, and for humankind. We hope that this report provides inspiration to take concerted action to prevent their further degradation, before it is too late. ee Patricio Bernal, Executive Secretary, Intergovernmental Oceanographic Commission (IOC) of the United Nations Educational, Scientific and Cultural Organization (UNESCO) Seamounts, deep-sea corals and fisheries Executive summary three-quarters (71 per cent) of the surface of the Earth. The overwhelming majority (95 per cent) of the ocean area 1s deeper than 130 m, and nearly two-thirds [64 per cent] are located in areas beyond national jurisdiction. Recent advances in science and technology have provided an unprecedented insight into the deep sea, the largest realm on Earth and the final frontier for exploration. Satellite and shipborne remote sensors have charted the sea floor, revealing a complexity of morphological features such as trenches, ridges and seamounts which rival those on land. Submersibles and remotely operated vehicles have documented rich and diverse ecosystems and communities, which has changed how we view life in the oceans. The same advances in technology have also documented the increasing footprint of human activities in the remote and little-known waters and sea floor of the deep and high seas. A large number of video observations have not only documented the rich biodiversity of deep-sea ecosystems such as cold-water coral reefs, but also gathered evidence that many of these biological communities had been impacted or destroyed by human activities, especially by fishing such as bottom trawling. In light of the concerns raised by the scientific community, the UN General Assembly has discussed vulnerable marine ecosystems and biodiversity in areas beyond national jurisdiction at its sessions over the last four years (2003-2006), and called, inter alia, for urgent consideration of ways to integrate and improve, on a scientific basis, the management of risks to the marine biodiversity of seamounts, cold-water coral reefs and certain other underwater features’. This report, produced by the Data Analysis Working Group of the global census of marine life on seamounts (CenSeam], is a contribution to the international response to this call. It reveals, for the first time, the global scale of the likely vulnerability of habitat-forming stony (scleractinian) corals, and by proxy a diverse assemblage of other species, to the impacts of trawling on seamounts in areas beyond national jurisdiction. In order to support, focus and guide the ongoing international discussions, and the emerging activities for the conservation and sustainable management of cold-water coral ecosystems on seamounts, the report: 1. compiles and/or summarizes data and information on the global distribution of seamounts, deep-sea corals on seamounts and deep-water seamount fisheries; 2. predicts the global occurrence of environmental conditions suitable for stony corals from existing records Ts oceans cover 361 million square kilometres, almost on seamounts and identifies the seamounts on which they are most likely to occur globally; 3. compares the predicted distribution of stony corals on seamounts with that of deep-water fishing on seamounts worldwide; 4. qualitatively assesses the vulnerability of communities living on seamounts to putative impacts by deep-water fishing activities; 5. highlights critical information gaps in the development of risk assessments to Seamount biota globally. SEAMOUNT CHARACTERISTICS AND DISTRIBUTION A seamount is an elevation of the seabed with a summit of limited extent that does not reach the surface. Seamounts are prominent and ubiquitous geological features, which occur most commonly in chains or clusters, often along the mid-ocean ridges, or arise as isolated features from the sea floor. Generally volcanic in origin, seamounts are often conical in shape when young, becoming less regular with geological time as a result of erosion. Seamounts often have a complex topography of terraces, canyons, pinnacles, crevices and craters — telltale signs of the geological processes which formed them and of the scouring over time by the currents which flow around and over them. As seamounts protrude into the water column, they are subject to, and interact with, the water currents surrounding them. Seamounts can modify major currents, increasing the velocity of water masses that pass around them. This often leads to complex vortices and current patterns that can erode the seamount sediments and expose hard substrata. The effects of seamounts on the surrounding water masses can include the formation of Taylor’ caps or columns, whereby a rotating body of water is retained over the summit of a seamount. In the present study the global position of only large seamounts (>1 000 m elevation) were taken into account due to methodological constraints. Based on an analysis of updated satellite data, the location of 14 287 large sea- mounts has been predicted. This ts likely an underestimate. Extrapolations from other satellite measurements estimate that there may be up to 100 000 large seamounts worldwide. Numbers of predicted seamounts peak between 30°N and 30°S, with a rapid decline above 50°N and below 60°S. The majority of large seamounts (8 955) occur in the Pacific Ocean area (63 per cent), with 2 704 (19 per cent) in the Atlantic Ocean and 1 658 (12 per cent) in the Indian Ocean. A small proportion of seamounts are distributed between the Southern Ocean (898; 6 per cent), the Mediterranean/Black Seas (59) and Arctic Ocean (13) (both less than 1 per cent). An analysis of the occurrence of these seamounts inside and outside of Exclusive Economic Zones (EEZs} indicates that just over half (52 per cent] of the world’s large seamounts are located beyond areas of national jurisdiction. The majority of these seamounts (10 223; 72 per cent) have summits shallower than 3 000 m water depth. DEEP-SEA CORALS AND BIODIVERSITY Compared to the surrounding deep-sea environment, seamounts may form biological hotspots with a distinct, abundant and diverse fauna, and sometimes contain many Species new to science. The distribution of organisms on seamounts Is strongly influenced by the interaction between the seamount topography and currents. The occurrence of hard substrata means that, in contrast to the mostly soft sediments of the surrounding deep sea, seamount communities are often dominated by sessile, permanently attached organisms that feed on particles of food suspended in the water. Corals are a prominent component of the suspension-feeding fauna on many seamounts, accompanied by barnacles, bryozoans, polychaete worms, molluscs, sponges, sea squirts and crinoids [which include sea lilies and feather stars). Most deep-sea corals belong to the Hexacorallia, including stony corals (scleractinians) and black corals (antipatharians}], or the Octocorallia, which include soft corals such as gorgonians. Three-dimensional structures rising above the sea floor in the form of reefs created by some species of stony coral, as well as coral ‘beds’ formed by black corals and octocorals, are common features on seamounts and continental shelves, slopes, banks and ridges. Coral frameworks add habitat complexity to seamounts and other deep-water environments. They offer refugia for a great variety of invertebrates and fish [including commercially important species) within, or in association with, the living and dead coral framework. Cold-water corals are frequently concentrated in areas of the strongest currents near ridges and pinnacles, providing hard substrata for colonization by other encrusting organisms and allowing them better access to food brought by prevailing currents. Although the co-existence between coral and non-coral species is in most cases still unknown, recent research is showing that some coral/non-coral relationships may show different levels of dependency. A review of direct dependencies on cold- water corals globally, including those on seamounts, has shown that of the 983 coral-associated species studied, 114 were characterized as mutually dependent, of which 36 were exclusively dependent on cnidarians (group of Executive Summary animals that contains the corals, hydroids, jellyfishes and sea anemones). A recent study recorded more than 1 300 species associated with the stony coral Lophelia pertusa on the European continental slope or shelf. Thus some cold- water corals may be regarded as ‘ecosystem engineers’ because they create, modify and maintain habitat for other organisms, similar to trees in a forest. Cold-water corals can form a significant component of the species diversity on seamounts and play a key ecological role in their biological communities. The assess- ment of the potential impacts of bottom trawling on corals is therefore a useful proxy for gauging the effects of these activities on seamount benthic biodiversity as a whole. A comprehensive assessment of biodiversity is currently impossible because of the lack of data for many faunal groups living on seamounts. DISTRIBUTION OF CORALS ON SEAMOUNTS One of the data sources utilized for this report was a database of 3 235 records of known occurrences of five major coral groups found on seamounts, including some shallower features <1 000 m elevation. Existing records show that the stony corals (scleractinians) were the most diverse and commonly observed coral group on seamounts (249 species, 1715 records) followed by Octocorallia (161 species, 959 records}, Stylasterida (68 species, 374 records), Antipatharia (34 species, 159 records) and Zoanthidea (14 species, 28 records]. These records included all species of corals, including those that were reef-forming, contributed to reef formation, or occur as isolated colonies. The most evident finding in analysing the coral database is that sampling of seamounts has not taken place evenly across the world’s oceans, and that there are significant geographic gaps in the distribution of studied seamounts. For some regions, such as the Indian Ocean, very few seamount samples are available. In total, less than 300 seamounts have been sampled for corals, representing only 2.1 per cent of the identified number of seamounts in the oceans globally (or 0.03 per cent when assuming there are 100000 large seamounts). Only a relatively small number of coral species have wide geographic distributions, and very few have near cosmopolitan distributions. Many of the widely distributed species are the primary reef, habitat or framework-building stony corals such as Lophelia pertusa, Madrepora oculata and Solenosmilia variabilis. In most parts of the world, stony corals were the most diverse group, followed by the octocorals. However, in the northeastern Pacific, octocorals are markedly more diverse than stony corals. Most stony corals and stylasterid species occur in the upper 1000-1500 m depth range. Antipatharians also occurred in the upper 1 000 m, although a higher proportion of species occurs in deeper waters than Seamounts, deep-sea corals and fisheries the two previous groups. Octocorals were distributed to greater depths, with most species in the upper 2 000 m. Very little sampling has occurred below 2 000 m. There are a number of reasons for the differences in the depth and regional distribution of the coral groups, including species-related preferences of the nature of substrates available for attachment, quantity, quality and abundance of food at different depths, the depth of the aragonite saturation horizon, temperature and the availability of essential elements and nutrients. PREDICTING GLOBAL DISTRIBUTION OF STONY CORALS ON SEAMOUNTS The dataset for corals on seamounts revealed significant areas of weakness in our knowledge of coral diversity and distribution on seamounts, especially the lack of sampling on seamounts at equatorial latitudes. Thus, to make a reasonable assessment of the vulnerability of seamount corals to bottom trawling (and, by proxy, determine the potential impacts of this activity on non-coral assemblages), it was necessary to fill the sampling gaps by predicting the global occurrence of suitable coral habitat by modelling coral distribution. An environmental niche factor analysis (ENFA) was used to model the global distribution of deep-sea stony corals on seamounts and to predict habitat suitability for unsampled regions. Other groups of coral, such as octocorals, for example, can also form important habitats such as coral beds. These corals may have very different distributions to stony corals, which would also be useful to appreciate in the context of determining the vulnerability of seamount communities to bottom trawling. The available data for octocorals are, unfortunately, currently too limited to enable appropriate modelling. ENFA compares the observed distribution of a species to the background distribution of a variety of environmental factors. In this way, the model assesses the environmental niche of a taxonomic group - i.e. how narrow or wide this niche is — identifies the relative difference between the niche and the mean background environment, and reveals those environmental factors that are important in determining the distribution of the studied group. The model used and combined: (i). the location data of 14 287 predicted large seamounts; lii]. the location records of stony corals (Scleractinia] on seamounts; and lili). physical, biological and chemical oceanographic data from a variety of sources for 12 environmental parameters (temperature; salinity; depth of coral occurrence; surface chlorophyll; dissolved oxygen; per cent oxygen saturation; overlying water productivity; export primary productivity; regional current velocity; total alkalinity; total dissolved inorganic carbon; aragonite saturation state). The model predictions were as follows: in near-surface waters (0-250 m), habitat predicted to be suitable for stony corals lies in the southern North Atlantic, the South Atlantic, much of the Pacific, and the southern Indian Ocean. The Southern Ocean and the northern North Atlantic are, however, unsuitable. Below 250 m depth, the suitability patterns for coral habitat change substantially. In depths of 250-750 m, a narrow band occurs around 30°N + 10°, and a broader band of suitable habitat occurs around 40°S + 20°. In depths of 750-1 250 m, the North Pacific and northern Indian Ocean are unsuitable for stony corals. The circum-global band of suitable habitat at around 40°S narrows with increasing depth [to + 10°). Suitable habitat areas also occur in the North Atlantic and tropical western Atlantic. These areas remain suitable for stony corals with increasing depth (1 250-1750 m; 1750-2 250 m; 2250 m-2500 mj, whereas the band at 40°S breaks up into smaller suitable habitat areas around the southeast coast of South America and the tip of South Africa. The global extent of habitat suitability for seamount stony corals was predicted to be at its maximum between around 250 m and 750 m. The majority of the suitable habitat for stony corals on seamounts occurs in areas beyond national jurisdiction. However, suitable habitats are also predicted in deeper waters under national jurisdiction, especially in the EEZs of countries: 1. between 20°S and 60°S off Southern Africa, South America and in the Australia/New Zealand region; 2. off Northwest Africa; and 3. around 30°N in the Caribbean. Combining the predicted habitat suitability with the summit depth of predicted seamounts indicates that the majority of seamounts that may provide suitable habitat for stony corals on their summits are located in the Atlantic Ocean. The rest are mostly clustered in a band between 15°S and 50°S. A few seamounts elsewhere, such as in the South Pacific, with summits in the depth range between 0 m and 250 m, are highly suitable. In the Atlantic, a large proportion of suitable seamount summit habitat is beyond national jurisdiction, whereas in the Pacific, most of this seamount habitat lies within EEZs. In the southern Indian Ocean, suitable habitat appears both within and outside of EEZs. When analysing habitat suitability on the basis of summit depth, it should be noted that suitable habitat for stony corals might also occur on the slopes of seamounts, |.e. at depths greater than the summit. The analysis found the following environmental factors were important for determining suitable habitat for stony corals: high levels of aragonite saturation, dissolved oxygen, per cent oxygen saturation, and low values of total dissolved inorganic carbon. Neither surface chlorophyll nor regional current velocity appears to be important for the global distribution of stony corals on seamounts. Nevertheless, these factors may be important for the distribution of corals at smaller spatial scales, such as on an individual seamount. The strong dependency of coral distribution on the availability of aragonite {a form of calcium carbonate) is noteworthy. Stony corals use aragonite to form their hard skeletons. A reduction in the availability of aragonite, for example through anthropogenically induced acidification of the oceans due to rising CQ2 levels, will limit the amount of suitable habitat for stony corals. SEAMOUNT FISH AND FISHERIES Seamounts support a large and diverse fish fauna. Recent reviews indicate that up to 798 species are found on and around seamounts. Most of these fish species are not exclusive to seamounts, and occur widely on continental shelf and slope habitats. Seamounts can be an important habitat for commercially valuable species, which may form dense aggregations for spawning or feeding targeted by large-scale fisheries. For the purpose of this report, the distribution and depth ranges of commercial fish species were compiled from a number of Internet and literature sources, including seamount fisheries catch data of Soviet, Russian and Ukrainian operations since the 1960s; published data on Japanese, New Zealand, Australian, European Union (EU) and Southern African fisheries; Food and Agriculture Organization of the United Nations (FAO) catch statistics; and unpublished sources. Although known to be incomplete, this is the most comprehensive compilation attempted to date for seamount fisheries, and is believed to give a reasonable indication of the general distribution of seamount catch over the last four decades. Deep-water trawl fisheries occur in areas beyond national jurisdiction for around 20 major species. These include alfonsino (Beryx splendens/, black cardinalfish (Epigonus telescopus], orange roughy (Hoplostethus atlanticus], armourhead and southern boarfish (Pseudopentaceros spp.J, redfishes (Sebastes spp.], macrourid rattails (primarily roundnose grenadier Coryphaenoides rupestris), oreos [including smooth oreo Pseudocyttus maculatus, black oreo Allocyttus niger) and Patagonian toothfish [Dissostichus eleginoides), and in some areas Antarctic toothfish (Dissostichus mawsonil, which has a restricted southern distribution. Many of these fisheries use bottom-trawl gear. Other fisheries occur over seamounts, such as those for pelagic species (mainly tunas] Executive Summary and target species for smaller-scale line fisheries (e.g. black scabbardfish Aphanopus carbo). The distribution of four of the most important seamount fish species [for either their abundance or commercial value] is as follows: 1. ORANGE ROUGHY is widely distributed throughout the Northern and Southern Atlantic Oceans, the mid- southern Indian Ocean and the South Pacific. It does not extend into the North Pacific. It is frequently associated with seamounts for spawning or feeding, although it is also widespread over the general continental slope. 2. ALFONSINO has a global distribution, being found in all the major oceans. It is a shallower species than orange roughy, occurring mainly at depths of 400-600 m. It is associated with seamount and bank habitat. 3. ROUNDNOSE GRENADIER is restricted to the North Atlantic, where it occurs on both sides, as well as on the Mid-Atlantic Ridge, where aggregations occur over peaks of the ridge. 4. PATAGONIAN TOOTHFISH has a very wide depth range and is sometimes associated with seamounts, but it is also found on general slope and large bank features. The distribution of historical seamount fisheries includes heavy fishing on seamounts in the North Pacific Ocean around Hawaii for armourhead and alfonsino; in the South Pacific for alfonsino, orange roughy and oreos; in the southern Indian Ocean for orange roughy and alfonsino; in the North Atlantic for roundnose grenadier, alfonsino, orange roughy, redfish and cardinalfish; and in the South Atlantic for alfonsino and orange roughy. Antarctic waters have been fished for toothfish, icefish and notothenioid cods. The total historical catch from seamounts has been estimated at over 2 million tonnes. Many seamount fish stocks have been overexploited, and without proper and sustainable management, they have followed a ‘boom and bust’ cycle. After very high initial catches per unit effort, the stocks were depleted rapidly over short time scales (<5 years) and are now closed to fishing or no longer support commercial fisheries. The life history character- istics of many deep-water fish species (e.g. slow growth rate, late age of sexual maturity) make the recovery and recolonization of previously fished seamounts slow. Over the last decade, exploratory fishing for deep- water species in many areas beyond national jurisdiction has focussed on alfonsino and orange roughy. The depth distribution of the two main target fisheries for alfonsino and orange roughy differ. The former is primarily fished between 250 and 750 m, and includes associated commercial species like black cardinalfish and southern boarfish. The orange roughy fisheries on seamounts, Seamounts, deep-sea corals and fisheries _ between 750 and 1 250 m depth (deeper fishing can occur on the continental slope), include black and smooth oreos as bycatch. Seamount summit depth data was used to indicate where such suitable fisheries habitat might occur in areas beyond national jurisdiction. Combined with information on the geographical distribution of the commercial species, various areas where fishing could occur were broadly identified. Many of these areas are in the southern Indian Ocean, South Atlantic and North Atlantic. The South Pacific Ocean also has a number of ridge structures with seamounts that could host stocks of alfonsino and orange roughy. Many of these areas have already been fished and some are known to have been explored, but commercial fisheries have not developed. ASSESSING THE VULNERABILITY OF STONY CORALS ON SEAMOUNTS In order to assess the likely vulnerability of corals and the biodiversity of benthic animals on seamounts to the impact of fishing, the report examines the overlap and interaction between: 1. the predicted global distribution of suitable habitat for stony corals; 2. the location of predicted large seamounts with summits in depth ranges of alfonsino and orange roughy fisheries; and 3. the distribution of the fishing activity on seamounts for these two species, and combines this with information on the known effects of trawling. Many long-lived epibenthic animals such as corals have an important structural role within sea floor communities, providing essential habitat for a large number of species. Consequently, the loss of such animals lowers survivorship and recolonization of the associated fauna, and has spawned analogies with forest clear-felling on land. A considerable body of evidence on the ecological impacts of trawling is available for shallow waters, but scientific information on the effects of fishing on deep-sea seamount ecosystems is much more limited to studies from seas off northern Europe, Australia and New Zealand. These studies suggested that trawling had largely removed the habitats and ecosystems formed by the corals, and thereby negatively affected the diversity, abundance, biomass and composition of the overall benthic invertebrate community. The intensity of trawling on seamounts can be very high. From several hundred to several thousand trawls have been carried out on small seamount features in the orange roughy fisheries around Australia and New Zealand. Such intense fishing means that the same area of the sea floor may be trawled repeatedly, causing long-term damage to the coral communities by preventing any significant recovery 10 or recolonization. Trawling’s impact on sea floor biota differs depending on the gear type used. The most severe damage has been reported from the use of bottom trawls in the orange roughy fisheries on seamounts. Information is currently lacking about the potential impact of trawling practices for alfonsino, where mid-water trawls are often used on seamounts. These may have only a small impact if they are deployed well above the sea floor. However, in many cases the gear is most effective when fished very close to, or even lightly touching, the bottom. Thus, it is likely that the effects of the alfonsino fisheries on the benthic fauna would be similar to that of the orange roughy fisheries. The comparison between the distributions of commercially exploited fish, fishing effort and coral habitat on seamounts highlighted a broad band of the southern Atlantic, Pacific and Indian Oceans between about 30°S and 50°S, where there are numerous seamounts at fishable depths, and high habitat suitability for corals at depths between 250 m and 750 m (the preferred alfonsino fisheries depth range), and again - but somewhat narrower - between 750 m and 1 250 m depth [the preferred orange roughy fisheries depth range). This spatial concordance suggests there could be further commercial exploration for alfonsino and orange roughy fisheries on large seamounts in the central- eastern southern Indian Ocean, the southern portions of the Mid-Atlantic Ridge in the South Atlantic, and some regions of the southern-central Pacific Ocean. Importantly, since these areas also contain habitat suitable for stony coral, impacts on deep-water corals and seamount ecosystems in general are likely to arise in such a scenario. However, it is uncertain whether fisheries exploration will result in economic fisheries. A WAY FORWARD This report has identified sizeable geographical areas with large seamounts, which are suitable for stony corals and are vulnerable to the impacts of expanding deep-sea fishing activities. To establish and implement adequate and effective management plans and protection measures for these areas beyond national jurisdiction will present major challenges for international cooperation. In addition, the report has identified that there are large gaps in the current knowledge of the distribution of seamounts and the biodiversity they harbour. In light of these findings, the report recommends a number of activities to be carried out collaboratively by all stakeholders under the following headings: How can the impacts of fishing on seamounts be managed in areas beyond national jurisdiction? Management initiatives for seamount fisheries within national EEZs have increased in recent years. Several countries have closed seamounts to fisheries, established habitat exclusion areas and stipulated method restrictions, depth limits, individual seamount catch quotas and bycatch quotas. In comparison, fisheries beyond areas of national jurisdiction have often been entirely unregulated. There are 12 Regional Fisheries Management Organizations (RFMOs) with responsibility to agree on binding measures that cover areas beyond national jurisdiction, including some of the geographical areas identified in this report that might see further expansion of exploratory fishing for alfonsino and orange roughy on seamounts. An RFMO covers parts of the eastern South Atlantic where exploratory fishing has occurred in recent decades, and where further trawling could occur. However, the western side of the South Atlantic is not similarly covered by an international management organization. There have been recent efforts to improve cooperative management of fisheries in the Indian Ocean, although there are no areas covered by an RFMO. In addition, efforts are underway - in the South Pacific, for example - to establish a new regional fisheries convention and body, which would fill a large gap in global fisheries management. However, it should be noted that only the five RFMOs for the Southern Ocean, Northwest Atlantic, Northeast Atlantic, Southeast Atlantic and the Mediterranean currently have the legal competence to manage most or all fisheries resources within their areas of application, including the management of deep-sea stocks beyond national jurisdiction. The other RFMOs have competence only with respect to particular target species like tuna or salmon. In the light of the recent international dialogues concerning the conservation and sustainable management and use of biodiversity in areas beyond national jurisdiction held within and outside the United Nations system, various fisheries bodies are more actively updating their mandates and including benthic protection measures as part of their fisheries management portfolios. It appears that a growing legislation and policy framework, including an expanding RFMO network, particularly in the southern hemisphere, could enable the adequate protection and management of the risks to vulnerable seamount ecosystems and resources identified in this report. In order to be successful, a number of challenges will have to be overcome, including: 1. Establishing adequate data reporting requirements for commercial fishing fleets. Some unregulated and unreported fishing activities take place, even in areas where there are well-defined fishery codes of practice and allowable catch limits (e.g. Patagonian toothfish fishery). Some countries require vessels registered to Executive Summary them to report detailed catch and effort data, but many do not. Therefore it is difficult at times to know where certain landings have been taken. 2. Ensuring compliance with measures, especially in areas that are far offshore and where vessels are difficult to detect. Compliance monitoring is also acute in southern hemisphere high seas areas, where there are no quotas for offshore fisheries. 3. Facilitating RFMOs, where necessary, to undertake ecosystem-based management of fisheries on the high seas. 4. Establishing, where appropriate, dialogue to ensure free exchange of information between RFMOs, governments, conservation bodies, the fishing industry and scientists working on benthic ecosystems. The experiences gained by countries in the protection of seamount environments in their EEZs and in the management of their national deep-water fisheries can provide useful case examples for the approach to be taken under RFMOs. Other regional bodies, such as Regional Sea Conventions and Action Plans, might be able to provide lessons learned from regional cooperation to conserve, protect and use coastal marine ecosystems and resources sustainably, including the implementation of an ecosystem approach in oceans management and the establishment of networks of marine protected areas (MPAs]. Regional Sea Conventions and Action Plans also provide a framework for raising awareness of coral habitats in deep water areas under national jurisdiction, and coordinating and supporting the efforts of individual countries to conserve and manage these ecosystems and resources sustainably. In calling for urgent action to address the impact of destructive fishing practices on vulnerable marine ecosystems, Paragraph 66 of UN General Assembly Resolution 59/25 places a strong emphasis on the need to consider the question of bottom-trawl fishing on seamounts and other vulnerable marine ecosystems on a scientific and precautionary basis, consistent with international law. The UN Fish Stocks Agreement (FSA) Articles 5 and 6 - General principles’ and the Application of the precautionary approach’ - also establish clear obligations for fisheries conservation and the protection of marine biodiversity and the marine environment from destructive fishing practices. The Articles also establish that the use of science is essential to meeting these objectives and obligations. At the same time, the FSA recognizes that scientific understanding may not be complete or comprehensive, and in such circumstances, caution must be exercised. The absence of adequate scientific information shall not be used as a reason for postponing or failing to take conservation and management measures. 11 Seamounts, deep-sea corals and fisheries A precautionary approach, consistent with the general principles for fisheries conservation contained in the FSA, as well as the UN FAO Code of Conduct for Responsible Fisheries and the principles and obligations for biodiversity conservation in the Convention on Biological Diversity (CBD), would require the exercise of considerable caution in relation to permitting or regulating bottom-trawl fishing on the high seas on seamounts. This is because of the widespread distribution of stony corals and associated assemblages on seamounts in many high seas regions, and the likelihood that seamounts at fishable depths may also contain other species vulnerable to deep-sea bottom trawling even in the absence of stony corals. In this regard, a prudent approach to the management of bottom-trawl fisheries on seamounts on the high seas would be to ascertain whether vulnerable species and ecosystems are associated with a particular area of seamounts of potential interest for fishing, and only then permitting well-regulated fishing activity provided that no vulnerable ecosystems would be adversely impacted. Further and improved seamount research The conclusions of this report apply only to the association of stony corals with large seamounts. In order to consider other taxonomic groups on a wider range of seamounts, further sampling and research is required. Spatial coverage of sampling of seamounts is poor and data gaps currently impede a comprehensive assessment of biodiversity and species distributions. Only 80 of the 300 biologically surveyed seamounts have had at least a moderate level of sampling. Existing surveys have tended to concentrate on a few geographic areas, thus the existing data on seamount biota are highly patchy on a global scale, and the biological communities of tropical seamounts remain poorly documented for large parts of the oceans. Most biological surveys on seamounts have been relatively shallow and thus the great majority of deeper seamounts remain largely unexplored. Very few individual seamounts have been comprehensively surveyed to determine the variability of faunal assemblages within a single seamount. In addition to the previous spatial gaps in sampling coverage, there are a number of technical issues that make direct comparisons of seamount data sometimes problematic. These issues relate to the availability of non- aggregated data, differences in collection methods and taxonomic resolution. In order to expand the type of analyses conducted for this report to other faunal groups common on seamounts, and to work at the level of individual species, certain steps should be taken. These include the adoption of a minimum set of standardized seamount sampling protocols; more funding for existing taxonomic experts and training of new 12 taxonomists; increased accessibility of full [non-aggregated) datasets from seamount expeditions through searchable databases; and the further development of integrated, Internet-based information systems such as Seamounts Online and the Ocean Biogeographic Information System. It should be noted that the activities under the two headings above are closely interrelated and linked. Increased research and collaboration between scientists and fishing companies will not only improve the amount and quality of data, it will also expand the scientific foundation for reviewing existing measures (e.g. those which were taken on a precautionary basis in the light of information gaps), and for developing new, focussed management strategies to mitigate against negative human impacts on seamounts and their associated ecosystems and biodiversity. Requirements in this context include: 1. obtaining better seamount location information; addressing geographic data gaps [including the sampling of other deep-sea habitats); 2. assessing the spatial scale of variability on and between seamounts; increasing the amount and scope of genetic studies; 3. undertaking better studies to assess trawling impacts; assessing recovery from trawling impacts; undertaking a range of studies to improve functional understanding of seamount ecosystems; and 4. implementing the means to obtain better fisheries information. Without a concerted effort by a number of organizations, institutions, consortia and individuals to attend to the previously identified gaps in data and understanding, the ability of any body to effectively and responsibly manage and mitigate the impact of fishing on seamount ecosystems will be severely constrained. Considering what this report has revealed about the vulnerability of seamount biota - particularly deep-sea corals — to fishing, now is the time for this collaborative effort to begin in earnest. Contents PNCKMOW EUG EMG MS cues tere cu snes te Sepa ov eseacaa te ates cease oes geners cc! Un basen pnsee tne ase ae ons te rena dara ens eenetnnes Ueenciea tae setecs divals Denervarreet 2 SUPPOGUMGLONGa MZatl OM Sr erceccesecescsnssceccscsterasesesereateanceree tera a sez tatcesecece setae cence eaterentena scatsessee natin era aa sacacasets tote setae saree 4 FO OWO Gece rinse cacy oct eet race eves setts: Cas ease Sea Ne og ate ae oa ne eatian tee npascetee Sivan coats t ose cRaaue Ueescadery 2 cosas sas eae 5 EXO GUUIVERS UMMM A Gy acccecscsccsccesscecvsersce caccecuvoseneeeeeeceoec ere cteeeee nearte aceasta ae sone eae Sena eal Sav adze sae seaet ner eet tree Tacetee since 6 1. INTRODUCTION 15 StU yi | CCLIVE Siete ern et serena ott tee ate acy ta te ea meen nS cee Yibe sey nn eva P ee sus /Dus ty eeeeenke epee Me ete est 7 2. SEAMOUNT CHARACTERISTICS AND DISTRIBUTION Simall’anidblangetSeanmOl mts eoeeerscrnccsccscsscseccsrertr sees ears ta eras asec n sean Sacto eentee sate ant Sarr RSE Rae HlowimanyilangerSeanmOulmeSssane nthe Gein. -cceereceeenceeteneseeesete rere tema emec ter ener eet ne eee eee WihereranentihellangesseanmOUmtsil@ Gate cli? semserenenerete te eren ster tases atte vaseeese nee eee ate erect see neat eh reer sects cee ane The origin and physical environment of seamounts What environmental conditions influence life ON SEAMOUNTS? .........eeeceeceececcseseeeeeeesesesescscsesescsesceeeteescesserersecisicaceeeeseees 21 3. DEEP-SEA CORALS AND SEAMOUNT BIODIVERSITY 23 WARE CEMETRSTINY GT IMSS ON) SSIEV AOU US ¢ecccccococosanenceccancecoccencoocanauchaesbcontacctedocase. cacocaceKoce caserenenoon/sconceseeeeeacreskeenocanazaaenocencaceh 23 ihiemnelationshipibetweenicoralSrancvothie ralifesereessees aces eeseeesec eect eeeaneee cater neee nee ene none eee 25 Attempting to determine the global Seamount FAUNA oo... cece este es eeseseseesesesssseesssssesesnesescesescererseesssseeeseeessnerascensueenenss 27 Clolsell Chewilouiniorn OF Se Ynnyellinya) Ola SEM NOUS oocccocococcnsapaopocccer2soveacaccasosbosbeacecaceanc ec enosanasAvoecEdecécoaracreeanccstosoaasoxcecscacocacath 27 A preliminary assessment of global seamount faunal diversity.........cccccccccsesessesesseseseesreseeseseeeesesesssrssesrersesiseeenereeesess 30 How to alternatively assess seamount faunal diversity 4. DISTRIBUTION OF CORALS ON SEAMOUNTS 31 iihemeedtoyassessithexdisthi bUtioMOf COmAl Sp ssescceees.2svecsssrsnaseeseeseaessea ss seancesssaeeraersses-ccvssesensreetstaceernssusesnceereeecenreeere 31 Wine tess ii Govny OLUIAG) UISENUL CNG) coset ce cecencacceccoccesoce: re 9ac AS G05O: cc Poe Con eNRO oP EECA BR ESCO ROCAEAALEEL CIOS SOOOOSCOSHECD “OSCE COECSDGOE 31 Globalidistrbutionefinecondsiformseann OUMi Comal See cestseocomerseceseeerterecnerssscenese stereos seeecee eee meerene ster arnstereceseseesrees 32 Patttehimszo feeornalltli vers tty. es te cre st cs es Tae ee, a ahr east stan, Save chacetee oantad eee eee eons asete 33 The relative occurrence and depth distribution of the main Coral QrOUPS 0.0.0.0... cece tet ee tes eeeeeteereseetetsteenesenesenes oo Getting a better understanding of coral distribUtion OM SEAMOUNKS......... ee eee ec eesecsess estes eeseceeereetesesseesessestesteseeteeeerees 35 5. PREDICTING GLOBAL DISTRIBUTION OF STONY CORALS ON SEAMOUNTS 37 IXTOMIN) CLONE CST OWUO Ln) cececooconoaaeccccsooeceecpeoccacoccreccng scarey scae daa obca arte sco acetone 93060: CaaC PACCEEC EC EDR OS COCATT SeneeBCADH aL CCE CCoEOCE 37 Using habitat suitability modelling to predict stony Coral GIStriDUtION......... ce cccceceeeseeteceeteseeterereerestetsieeestensenenseeeeses 38 Stony coral distribution and environmental data.. LR Be ee ee Se be ee Re ee 38 Predictedihabitatstitabilityitorm Stomyico lal Sieeteteeesteeererete sence oats neee sete atereeeareeseee menace stars ene ertsedrercetencee nena 39 What environmental factors are important in determining stony coral diStriDUtiON? «2... cec cece eee eee eee eeneeeneees 40 6. SEAMOUNT FISH AND FISHERIES 46 FFI IOOCEVETETS sicccossascocceece, oxtpnacccecoooasecceceréctsescdecendsposécereebeoveeececoacneconceschnceescAcris oecSoatcocce pcorctce63 soosaseqanendasoacaovacctcctccOse Pen 46 DISSDAWENGIF [HEINAITES CEN) -coceocosssceenccccaectsosce-ehc05 se 35660 00 EOS Ae 0920350 2S 928TCOCENTON GCC IE ICE CECC REA IEC NEE AOR SEAT INL AG CEEE OC 46 Globalldistribtitionto fie ee p= Werte atic eee eeemee eens etert ercereeteee eteseseneee yes teres tetsu enrateces cw eare soe nreeeneee ened rars terrane 47 Global distribution of major Seamount trawl fISMEPIES .........c. ees ec cscs ecseseeseseseseseeseresscscassessececnecsenenecesnersccarereesereneaterarse 49 Areas of exploratory fishing in areas beyond national JUrISGICtION.......... eee eee ecsceseeseeeceeceereetectectereetetteseeesteteseeeceeey 50 7. ASSESSING THE VULNERABILITY OF STONY CORALS ON SEAMOUNTS 55 RELMONELS reccoensidec ch zansaceenicsctidococeet aoocc cedoacnaodonceRereecce6e. cpt SeReeG0U02730009-6 7-500 ada cc GpeGeeEOReaACE ESE cuca, coaaie Osc 0: eco acca GoCec See ceRCcerec 55 13 Seamounts, deep-sea corals and fisheries Overlapibetweenistomy:coralstamaifiSMeGleS: esses. .eseseecaeuecert satis sesenecsraeuseuceer onsees aver ee aera eee ker eer Vulnerability of corals on seamounts to bottom trawling bi Winenejaneithe:malmraneasiotiniSkama come iiueeese seers. cect cetacean rae tener =e 60 8. A WAY FORWARD 61 How can the impact of fishing on seamounts be managed in areas beyond national jurisdiction? ............cccceee 61 FunthemandhimnprovediseamoUmtsnhe Sean hier see -seeestys..gtcescceccteesete renee, maaan cece em nena re 63 AGRO MY GINS ies sesseciys one Zags evesunseaeansearedeashs aes aeey hans eas ea peateeg es ioe onset eee ent concen ime ce 66 GlOSSANY ssasaecsserciaeses as scvecdssasstussovenersstoeucanscraistagiaderastawee ss veved iv stelar Rarities arene, gn a ines Ie ene 67 FREFEREM CES sssocsysccesceccavonsiesaveyuen sas stsasaten eras deresieses setinto. veces oa recente snutre este Aiaeste teane tetova eee ee ee ee rr 69 Selection of institutions and researchers working on seamount and cold-water coral ECOLOGY .........ccccecceeccescesseseees 76 SelectioniofaconaliamaliseamountineSOUnCe Se o-cxoneeeceere ter tester ceca ee sneer eee ete 78 APPENDICES AppendixilsiPMySiald ate yes. 19 cheese tecerchenscttuecase sae Gea esac cso aes oe ee ee hy Appendix Il: The habitat suitability model 14 il, Wendrexe UIC orn eamounts are prominent and ubiquitous features found on the sea floor of all ocean basins, both within and outside marine areas under national jurisdiction. With food availability on and above seamounts often higher than that of the surrounding waters and ocean floors, seamounts may function as biological hotspots, which attract a rich fauna. Pelagic predators such as sharks, tuna, billfish, turtles, seabirds and marine mammals can aggregate in the vicinity of seamounts [Worm et al., 2003}. Deep-sea fish species such as orange roughy (Pankhurst, 1988; Clark, 1999; Lack et al., 2003] and eels (Tsukamoto, 2006) form spawning aggregations around seamounts. The bottom fauna on seamounts can also be highly diverse and abundant, and they sometimes contain many species new to science [Parin et al., 1997; Richer de Forges et al., 2000; Koslow et al., 2001). Suspension-feeding organisms, such as deep-sea corals, are frequently prolific on seamounts, mainly because the topographic relief creates fast-flowing currents and rocky substrata, providing suspension feeders with a good food supply and attachment sites (Rogers, 1994]. Corals are recognized as an important functional group of seamount ecosystems, as they can form extensive, complex and fragile three-dimensional structures. These may take the form of deep-water reefs built by stony corals (scleractinians] (Rogers, 1999; Freiwald et al., 2004; Roberts et al., 2006), or coral gardens or beds formed by black corals and octocorals (e.g. Stone, 2006). All Introduction can provide important habitat for a great variety of associated invertebrates and fish, which use the coral as food, attachment sites and/or for protection and shelter. Deep-water corals can support a rich fauna of closely associated animals with, for example, greater than 1 300 species reported living on Lophelia pertusa reefs in the northeastern Atlantic alone (Roberts et al., 2006). Many fish species, including several of commercial significance, show spatial associations with deep-water corals {e.g. Stone, 2006), and fish catches have been found to be higher in, and around, deep-water coral reefs {Husebg et al., 2002). The fragility of cold-water corals makes them highly vulnerable to fishing impacts, particularly from bottom trawling (Koslow et al., 2001; Fossa et al., 2002; Hall-Spencer et al., 2002), but also from gill nets and long-lining gear (Freiwald et al., 2004; ICES, 2005, 2006). Ground-fishing gear can completely devastate coral colonies (Fossa et al., 2002), and such direct human impacts can be extensive. For example, coral bycatch in the first year of the orange roughy fishery on the South Tasman Rise was estimated at 1 750 tonnes, but this fell rapidly to 100 tonnes by the third year of the fishery as attached organisms on the seabed were progressively removed by repeated trawling [Anderson and Clark, 2003). Because corals provide critical habitat for many other seamount species, destruction of corals has knock- on’ effects, resulting in markedly lower species diversity and biomass of bottom-living fauna (Clark et al., 1999; Koslow et 15 M. Clark (NIWA) Seamounts, deep-sea corals andfisheries Benthoctopus sp. and crinoid, Davidson Seamount, 2 422 m. (NOAA/MBARI) al., 2001; Smith, 2001; Clark and O'Driscoll, 2003). Importantly, recovery of cold-water coral ecosystems from fishing impacts is likely to be extremely slow or even impossible, because corals are long lived and grow extremely slowly {in the order of a few millimetres per year]. Individual octocorals can reach ages of several hundred (Andrews et al., 2002; Risk et al., 2002; Sherwood et al., 2006) or even more than a thousand years old (Druffel et al., 1995), and larger reef complexes, formed by stony corals, may be more 8 000 years old (Freiwald et al., 2004; Roberts et al., 2006). Corals also have specific habitat requirements and may be sensitive to alteration of the character of the seabed by fishing gear, or to increased sedimentation resulting from trawling (Commonwealth of Australia, 2002; ICES, 2006). Such effects may prevent recovery of cold-water coral reefs or octocoral gardens permanently (Rogers, 1999; ICES, 2006). There has been a dramatic expansion of fishing over the last 50 years (Royal Commission on Environmental Pollution, 2004) and the exploitation of deep-sea species of fish in the last 25 years (Lack et al., 2003). The expansion of deep-sea fisheries has been driven by the depletion of shallow fisheries based on the continental shelf, the establishment of the 200 nautical mile economic exclusion zones by states under the UN Convention on Law of the Sea (UNCLOS}, overcapacity of fishing fleets, technological advances in fishing — including developments in navigation, acoustics and capture gear and in the power of vessels - and the availability of subsidies for building new fishing vessels equipped for deep-sea fishing (Lack et al., 2003; Royal Commission on Environmental Pollution, 2004). It is estimated that 40 per cent of the world’s trawling grounds are now located in waters deeper than the continental shelf (Roberts, 2002). The catch of commercial fish species beyond areas of national jurisdiction by bottom trawling has been estimated at about 200 000 tonnes annually (Gianni, 16 Brisingid sea star, Hatton Bank. (DTI SEA Programme, ©/o Bhavani Narayanaswamy) 2004). Most of this is taken from shelf and slope areas of the Northwest Atlantic, but outside this region fishing effort tends to focus on deep-water species from seamounts. Over 77 fish species have been commercially harvested from seamounts (Rogers, 1994), including major fisheries for orange roughy [Hoplostethus atlanticus], pelagic armour- head [Pseudopentaceros spp.) and alfonsino (Beryx splendens]. Most of these fisheries have not been managed in a sustainable manner, with many examples of ‘boom and bust’ fisheries, which rapidly developed and then declined sharply within a decade (Koslow et al., 2000; Clark, 2001; Lack et al, 2003). In most cases there is insufficient infor- mation on the target fish species, let alone the seamount ecosystem, to provide an adequate basis for good manage- ment {Lack et al. 2003). Furthermore, the life-history characteristics of many exploited deep-sea fish are unlike those of shallow-water species, rendering some fisheries management practices inappropriate (Lack et al., 2003). In the light of the evidence found in numerous in situ observations, the scientific community raised concern about the damage that trawling can have on the bottom-dwelling (benthic) communities in deep-waters and on seamounts (MCBI, 2003 et seq.). Taking into account that most of the potential areas affected by the expanding deep-sea fishing activities are in areas beyond national jurisdiction, the United Nations General Assembly (UNGA) addressed the issue in its 58th (2004), 59th (2005) and 60th sessions (2006), both in its discussions on ‘Oceans and the Law of the Sea’ and ‘Sustainable Fisheries’. Seamounts and cold-water corals/ reefs were specifically mentioned in the following resolutions: UN resolutions on oceans and the law of the sea (UN General Assembly, 2003, 2004a, 2005a, 2006) Reaffirms the need for States and competent international organizations to urgently consider ways to integrate and improve, based on the best available scientific information and in accordance with the Convention [UN Convention on Oceans and the Law of the Sea, 1982] and related agreements and instruments, the management of risks to the marine biodiversity of seamounts, cold-water corals, hydrothermal vents and certain other underwater features; (Resolution 60/30, Paragraph 73, following similar text in the previous resolutions 59/24, 58-240 and 57-141) Calls upon States and international organizations to urgently take action to address, in accordance with international law, destructive practices that have adverse impacts on marine biodiversity and ecosystems, including seamounts, hydrothermal vents and cold-water corals; (Resolutions 60/30, Paragraph 77 and 59/24) UN resolutions on sustainable fisheries (UN General Assembly, 2004b, 2005b) Requests the Secretary-General, in close cooperation with the Food and Agriculture Organization of the United Nations [FAO], and in consultation with States, regional and subregional fisheries management organizations and arrangements and other relevant organizations, in his next report concerning fisheries to include a section outlining current risks to the marine biodiversity of vulnerable marine ecosystems including, but not limited to, seamounts, coral reefs, including cold-water reefs and certain other sensitive underwater features related to fishing activities, as well as detailing any conservation and management measures in place at the global, regional, subregional or national levels addressing these issues; [Resolution 58/14, Paragraph 46). Calls upon States, either by themselves or through regional fisheries management organizations or arrangements, where these are competent to do so, to take action urgently, and consider on a case-by-case basis and on a scientific basis, including the application of the precautionary approach, the interim prohibition of destructive fishing practices, including bottom trawling that has adverse impacts on vulnerable marine ecosystems, including seamounts, hydrothermal vents and cold-water corals located beyond national jurisdiction, until such time as appropriate conservation and management measures have been adopted in accordance with international law; (Resolution 59/25, Paragraph 66) In 2003, the UNGA requested the Secretary General to prepare a report on vulnerable marine ecosystems and biodiversity in areas beyond national jurisdiction (cf. Introduction paragraph 52 of Resolution 58/240). Following the examination of this report in 2004, the UNGA decided to establish an Ad Hoc Open-ended Informal Working Group to study issues relating to the conservation and sustainable use of marine biological diversity beyond areas of national jurisdiction (cf. Paragraph 73 in Resolution 59/24). The outcome of their first meeting [New York, 13-17 February 2006) will be presented to the 61st session of the UNGA. Furthermore, the UNGA requested in 2005 the Secretary General, in cooperation with the FAO, to include in his next report concerning fisheries a section on the actions taken by States and regional fisheries management organizations and arrangements to give effect to Paragraphs 66 to 69 of Resolution 59/25, in order to facilitate discussion of the matters covered in those paragraphs. The UNGA also agreed to review, within two years, progress on action taken in response to the requests made in these paragraphs, with a view to further recommendations, where necessary, in areas where arrangements are inadequate. From the above, it is apparent that the UNGA discussions on: li]. conservation and sustainable management of vulnerable marine biodiversity and ecosystems (including seamount communities} in areas beyond national jurisdiction, and [ii]. the role of regional fisheries management organizations or arrangements in regulating bottom fisheries and the impacts of fishing on vulnerable marine ecosystems are set to continue. Itis hoped that the scientific findings presented in this report by members of the Census of Marine Life programme CenSeam will help and guide policy and decision makers to make progress on these issues. STUDY OBJECTIVES The study presented here aimed to: 1. compile and/or summarize data for the distribution of large seamounts, deep-sea corals on seamounts and deep-water seamount fisheries; 2. predict the global occurrence of environmental conditions suitable for stony corals from existing records on seamounts and identify the seamounts on which they are most likely to occur globally; 3. compare the predicted distribution of stony corals on seamounts with that of deep-water fishing on seamounts worldwide; 4. qualitatively assess the vulnerability of communities living on seamounts to putative impacts by deep-water fishing activities; and 5. highlight critical information gaps in the development of risk assessments to seamount biota globally. 17 Seamounts, deep-sea corals and fisheries 2S Caniounimehiaraeeanisuies AAG CISA SCH em SMALL AND LARGE SEAMOUNTS eamounts are submarine elevations with a limited Goes across the summit and have a variety of shapes, but are generally conical with a circular, elliptical or more elongate base (Rogers, 1994]. The slopes of seamounts can be extremely steep, with some showing gradients of up to 60° (e.g. Sagalevitch et al., 1992], although, in general, slopes are less steep (generally less than 20° in the New Zealand region; Rowden et al., 2005). Younger seamounts tend to be more conical and regular in shape, whereas older seamounts that have been subject to scouring and erosion by currents are less regular. Geophysical definitions distinguish between [i] hills, with summits lower than 500 m; [ii] knolls, with summits between 500 m and 1000 m; and [iii] seamounts, with summits over 1000 m. However, the size component of seamount definitions has become more flexible with the growing appreciation of the abundance of elevated sea floor features of similar morphology but with smaller vertical extent (greater than 50 m; Smith and Cann, 1990); the observation that such features may represent similar habitat and faunistic characteristics as their larger counterparts (e.g. Epp and Smoot, 1989; Rogers, 1994; Rowden et al., 2005); and because small features are targeted by commercial fisheries {greater than 100 m, Brodie and Clark, 2004). Differences in the methodologies available to determine the number and distribution of seamounts have led to a distinction between ‘small’ and ‘large’ seamounts. Generally, large seamounts are those with a vertical height of greater than 1 000 m (e.g. Wessel, 2001) or 1 500 m (ICES, 2006). For global and regional studies of seamounts and aspects of the present report, methodological and practical constraints mean that examinations have been restricted to large seamounts only (e.g. ICES, 2006). HOW MANY LARGE SEAMOUNTS ARE THERE? The deep oceans are the largest ecosystem on Earth. This vast area of seabed has been only partially mapped; therefore itis not possible to give a figure for the number of {both small and large] seamounts globally. Attempts at estimating the numbers of seamounts globally have been made by extrapolation of the known numbers of seamounts in a geographic region (e.g. Smith and Jordan, 1988 for the Pacific Ocean]. Recently, satellite sensors have been used to estimate the position and size of large seamounts. 18 Seamounts are masses of rock and give rise to anomalies in the usual straight-down force of gravity. These minute variations in the Earth's gravitational pull cause seawater to be attracted to seamounts. This means that the sea surface is pitched up over a seamount with a shape that reflects the underlying topographic feature (Wessel, 1997 and 2001). Satellite sensors can detect the anomalies in the Earth's gravitational field (e.g. Seasat gravity sensor) or the small differences in sea-surface height (e.g. Geosat/ERS1 alti- meter] (Stone et al., 2004). Efforts to estimate the number of seamounts worldwide using satellite altimetry and gravitational gradient data have indicated that there are between 5 000 and 16 000 features with an elevation greater than 1000 m [reviewed in Stone et al., 2004). However, the available satellite datasets are limited in terms of resolution because of defence policy, and there are limitations in the methods employed by researchers. This has led to analyses that suggest (after extrapolation) that globally there may be as many as 100 000 seamounts with an elevation of more than 1000 m (Wessel, 2001]. The most recent [non-extra- polative} estimate of the global number of large seamounts is 14 287 (Kitchingman and Lai, 2004). This number originated from the Sea Around Us Project {SAUP], which used depth difference algorithms applied to a digital global elevation Fig. 2.1 Distribution of predicted large seamounts by latitude. Source: Kitchingman and Lai (2004) 2000 ;- 1500 |- 1000 |- 500 |- Number of seamounts -90 -70 -50 -30 -10 10 30 50 70 90 Latitude (degrees) Seamount characteristics and distribution Fig. 2.2 Global distribution and summit depths of predicted large seamounts. Source: Kitchingman and Lai (2004) Key Summit depth (m) Summit depth (m) fe 0-500 ME .000-4 500 Pa 500-1 000 ME s¢.500-5 000 Ms 000-1500 M5 000-5 500 Msi 500-2000 ME 5 500-6 000 M2 000-2500 M4, 000-6 500 M2 500-3 000 M6, 500-7 000 MS s-3:000-3 500 M7 000-7500 Mi s3:500-4. 000 map and a more generalized definition to detect seamounts that fit into ecological and management contexts. Kitchingman and Lai (2004) used the National Oceanic and Atmospheric Administration’s (NOAA) ETOPO2 dataset as the source for all analyses to estimate the global number and location of large seamounts. The dataset was supplied at a 2-minute cell resolution (13.7 km? at the equator}, which allowed for a generalized global analysis, but certainly missed many seamounts. Thus the estimated number is an underestimate of the global number of large seamounts. Two methods were used to identify possible seamounts. The first method involves isolating peaks that have significant rise from the ocean floor. The second method isolates peaks with a circular or elliptical base in an effort to eliminate peaks found along ridges. The two methods produced different numbers of predicted seamounts (30314 and 15962, respectively). The over- lapping seamounts (14 287} found by using both these methodologies were used as the SAUP seamount dataset. Characteristics of the second method could mean that some real’ seamounts that occur on ridges could be eliminated from the dataset, as well as possibly including some features such as semi-circular banks. The SAUP data not only provide information on the location and elevation of predicted seamounts but also, usefully, on the depth of the seamount summit. WHERE ARE THE LARGE SEAMOUNTS LOCATED? The distribution by latitude of the large seamounts estimated from an analysis of global digital elevation data generated by SAUP [Kitchingman and Lai, 2004) is shown in Figure 2.1. The location of some seamounts will be in error because the combining of the results from the two methods used by Kitchinman and Lai (2004) will reduce the location of seamounts with a double peak to a single location at a mid- point between the two, maintaining the shallower depth value of the pair. The error in real-world location is enhanced by a misregistration of the underlying ETOPO2 bathymetry dataset. However, ground truthing performed on a dataset of known seamounts produced from a combination of data from the US Department of Defense Gazetteer of Undersea Features [1989] and SeamountsOnline revealed that approximately 60 per cent of the known seamounts were within 30 arc minutes of predicted seamounts. Numbers of identified seamounts peak between 30°S and 30°N, with a rapid decline above 50°N and below 60°S. Available surface [ocean] area by latitude probably drives this pattern. Figure 2.2 shows the global distribution and summit depths of the large seamounts identified by 19 Seamounts, deep-sea corals andfisheries Fig. 2.3 Summit depths of predicted large seamounts and current depth range of bottom trawling on seamounts. Source: Kitchingman and Lai (2004) Depth range of bottom trawling on seamounts 250m-1500m v Summit depth ['000 metres} B 1 =I 1 ! ==) 0 500 1000 1500 2000 2500 Number of seamounts Table 2.1: Number of predicted large seamounts in major FAO fishing areas and in areas beyond national Kitchingman and Lai (2004), many of which are located along plate boundaries. Table 2.1 shows the distribution of large seamounts in the United Nation's Food and Agriculture Organization (FAO) major fishing areas, and identifies the number of large seamounts that fall outside the EEZs of countries, i.e. are in areas beyond national jurisdiction ([Kitchingham et al., in press). Although FAO areas do not exactly fit oceanic boundaries, their use allows broad and more specific comparison with other studies and allows an appreciation of seamounts in a global and regional fishery management context. The majority of large seamounts occur in the Pacific Ocean area (63 per cent), with 19 per cent and 12 per cent of seamounts occurring in the Atlantic and Indian Ocean areas, respectively. A small overall proportion of seamounts are distributed between the Southern Ocean [6 per cent}, Mediterranean/Black Seas and Arctic Ocean (both less than 1 per cent] areas. The occurrence of large seamounts inside and outside EEZs shows that just over half (52 per cent) of the world’s large jurisdiction Ocean Areas FAO area Number of predicted Number of predicted large large seamounts seamounts in areas beyond national jurisdiction Pacific All 8 955 3 540 Eastern Central 77 P2 TKS) 967 Northeast 67 265 176 Northwest 61 1 350 630 Southeast 87 939 700 Southwest 81 996 643 Western Central 71 2 670 424 Atlantic All 2704 1959 Eastern Central 34 536 433 Northeast 27 325 211 Northwest 21 83 77 Southeast 47 639 512 Southwest 4l 452 301 Western Central 31 669 425 Indian All 1 658 1 082 Eastern 57 588 426 Western 51 1.070 656 Mediterranean and Black Seas 37 59 59 Southern Ocean All 898 713 Atlantic, Antarctic 48 498 371 Indian Ocean, Antarctic 58 212 154 Pacific, Antarctic 88 188 188 Arctic 18 13 13 Totals : 14 287 7366 Source: Kitchingham et al. in press} 20 seamounts are located in marine areas beyond national jurisdiction. Figure 2.3 shows that there are many large seamounts with summits at less than 500 m depth, and another peak between 1500 m and 3000 m. Thus, most large seamounts have summits shallower than 3000 m water depth. The current depth range of bottom trawling for commercially valuable fish (250-1 500 m) encompasses about 18 per cent of the summits of large seamounts. THE ORIGIN AND PHYSICAL ENVIRONMENT OF SEAMOUNTS Seamounts are generally volcanic in origin and may be associated with the continental margin or located on the abyssal plains, either as isolated features, clusters or chains. Most commonly, however, seamounts occur along the mid-ocean ridges. These are areas where new oceanic crust is formed by lava welling up from magma chambers below the sea floor, generating enormous ranges of seamounts. As the oceanic crust is formed and moves away from the mid-ocean ridge, the associated seamounts move with it, becoming older and subsiding, causing decreased elevation. Seamounts are also associated with areas where oceanic plates meet and one plate is subducted under the other. The enormous pressures associated with this process melt the subducted plate, resulting in an arc of volcanic activity giving rise to islands and seamounts lying adjacent to an oceanic trench. Examples include the Scotia Arc in the Southern Ocean and the islands of Tonga and associated seamounts in the southwestern Pacific. Seamounts are also generated by ocean hotspots, areas where plumes of magma well up from the Earth's mantle and form volcanoes on the sea floor. In geological time scales, as oceanic plate passes over the hotspot, a chain of seamounts and islands is formed. Examples include the Hawaiian Islands and Emperor Seamount Chain in the North Pacific, and the Louisville Seamount Chain in the southwestern Pacific. Seamounts on or close to the continental margin can have different origins, arising from rifting margin volcanoes or rifted continental blocks. As a result of the volcanic origin of seamounts they may be associated with high temperature [e.g. Marianas Seamounts or Brothers Seamount, Kermadec Ridge] or low temperature (e.g. Loihi Seamount, Hawaiian Ridge] hydrothermal venting, though the majority of seamounts are no longer geologically active and are not venting. The bases of seamounts associated with continental margins tend to be shallower and have an overall elevation lower than those located away from continents (e.g. Rowden et al., 2005). In some cases, for example the Rosemary Bank in the northeastern Atlantic, such features may be termed banks, as definitions of the two types of features can overlap (ICES, 2006). ___ Seamount characteristics and distribution WHAT ENVIRONMENTAL CONDITIONS INFLUENCE LIFE ON SEAMOUNTS? The geographical location, depth and elevation of the seamount determine the interactions of the seamount with the water masses and currents that impinge on it. Water masses have different environmental characteristics such as flow velocity, temperature, salinity, nutrient availability and pH. The environmental characteristics of the waters that overly seamounts can influence the spatial and temporal patterns of supply of organic material to a seamount benthic (seabed) community in terms of phytoplankton, zooplankton and organic detritus (dead organisms, faecal pellets, and so on). Pelagic communities and supply of larvae will also largely reflect the dominant oceanographic influences on a seamount. Within the immediate vicinity of a seamount, complex current-topography interactions can take place at all scales. At the largest scale, seamount chains can divert major currents [e.g. the Emperor Seamount chain deflects the Kuroshio and subarctic currents; Roden et al., 1982; Roden and Taft, 1985; Vastano et al., 1985]. At smaller scales, the interactions of seamounts with the surrounding currents are complex and difficult to measure, although in some cases such responses can be modelled. For example, models predict the formation of a rotating body of water retained over the summit of a seamount {known as a Taylor’ column). Observations have demonstrated the existence of such columns above many seamounts (Meincke, 1971; Vastano and Warren, 1976; Cheney et al., 1980; Genin et al., 1989; Roden, 1991; Dower et al., 1992], although the stratification of water layers above a seamount often reduces the column to a cap. Seamounts may also interact with tides, amplifying them and accelerating currents to greater than 40 cm s! (Chapman, 1989; Genin et al., 1989; Noble and Mullineaux, 1989]. The seamounts themselves may also generate internal tides {Noble et al., 1988] and generate or interact Side scan sonar image of Anton Dohrn Seamount, Northeast Atlantic. (DT) SEA Programme, S/o Colin Jacobs) 21 Seamounts, deep-sea corals and fisheries with internal waves [e.g. Bell, 1975; Wunsch and Webb, 1979: Eriksen, 1982a, 1982b, 1985, 1991; Kaneko et al., 1986; Brink, 1989; Genin et al., 1989]. Such phenomena can lead to the generation of periodic, small-scale, fast, short-duration bottom currents. The depth of the seamount summit below the ocean surface is one of the most important physical factors in determining the abundance and diversity of benthic communities on seamounts and has been used to classify them (e.g. ICES, 2006]. Seamounts with a depth of less than 250 m reach into the euphotic zone, where enough light penetrates to allow photosynthesis, and _ therefore communities that include algae can develop. Seamounts with a summit depth down to 1 000 m are likely to interact with layers of zooplankton that undergo a daily vertical migration in the water column [Wilson and Boehlert, 2004). These migrating plankton form a relatively thin layer of organisms detectable by echo sounders [deep scattering layer, or DSL}. Several observations indicate that the topography of seamounts can trap descending layers of zooplankton, which provide a source of food for seamount- associated species (Rogers, 1994; Seki and Somerton, 1994; Haury et al., 2000). Whether or not this takes place depends on the depth of the seamount summit in relation to the vertical depth range over which the plankton migrate. It also depends on the intensity of horizontal currents that advect the DSL over the seamount at night. Studies of the fish populations of the Great Meteor Seamount have shown that they prey on the DSL and are concentrated around the margins of the summit to maximize chances of encountering zooplankton [Fock et al., 2002). Such mechanisms may also be important in the nutrition of abundant benthic communities on seamounts. For example, over the Nasca and Sala Y Gomez Seamounts in the southeastern Pacific, the lower depth of distribution of the lobster Projasus bahamondei, a dominant megabenthic predator, coincided with the deepest depth of migration of the DSL (Parin et al. 1997]. Other mechanisms of concentration of food may also operate around seamounts associated with eddies or up- or down-welling currents and the relative movement behaviour of zooplankton (Genin, 2004). It is important to note that currently there is little understanding of the ecological links between the pelagic ecosystem, especially of larger predators such as fish, and communities of benthic organisms living on seamounts. Thus it is unknown how the removal of large quantities of fish biomass, by fisheries, from the vicinity of seamounts would affect the benthic community (Commonwealth of Australia, 2002; Lack et al., 2003). The distribution of sediments and benthic communities on seamounts Is a function of the current velocity near the seabed. Such currents may displace material off the 22 seamount and resuspend organic material. Many seamounts also have distinct ‘moats’ around the base where currents scour out sediments lying around the seamount (e.g. Anton Dohrn Seamount, northeastern Atlantic). Some seamounts, known as guyots, are flat-topped and often covered in sediment as a result of wave-erosion when they were exposed as islands. However, seamounts are notable for the occurrence of hard substrata and complex small- scale topography, which show a marked contrast to the surrounding deep seabed - which tends to comprise fine sediments (hard substrata can occur elsewhere on banks and the slopes of continental shelves). The occurrence of terraces, canyons, pinnacles, crevices, craters, rocks and cobbles can exert a strong influence on the distribution of animals and plants on seamounts [reviewed in Rogers, 1994]. Topographic relief controls local current flow regimes, and filter-feeding organisms such as corals are frequently concentrated in areas of strongest currents near ridges and pinnacles (Genin et al., 1986). The following chapter will examine in greater detail the biological communities that seamounts can support, and ask how well their diversity can be assessed on a global scale. Deep-sea corals and seamount biodiversity 3. Deep-sea corals and Seamount biodiversity THE DIVERSITY OF LIFE ON SEAMOUNTS he occurrence of hard substrata on seamounts means | seamount communities can be dominated by sessile organisms that are permanently attached to the seabed — not possible on the soft sediments of most of the surrounding deep-sea floor On seamounts with very shallow summits that penetrate the euphotic zone, such as the Vema Seamount in the southeastern Atlantic Ocean or the Gorringe Bank in the northeastern Atlantic Ocean, plant life can occur with kelp and encrusting calcareous algae dominating hard substrates (Simpson and Heydorn, 1965; Oceana 2006). The deepest records of living marine plants are of encrusting coralline algae from seamounts in the Caribbean living at 268 m depth [Littler et al., 1985). In the tropics, reef-forming corals such as Acropora spp., Pocillopora spp., Porites spp. and Montastrea spp. can occur on shallow seamounts which are often drowned atolls, such as the Raita Bank on the Hawaiian Ridge. Other animal groups that occur commonly on hard substrata on shallow seamounts include sponges, hydroids, azooxanthellate corals, molluscs, echinoderms and ascidians (sea squirts] (Simpson and Heydorn, 1965; Oceana, 2006). On seamounts with deeper summits, the dominant megafauna [i.e., generally those animals that can be easily seen in photographs or video) are the attached, sessile organisms that feed on particles of food suspended in the water. The predominant suspension feeders are from the phylum Cnidaria and include sea anemones, sea pens, hydroids, stony corals, gorgonian corals and black corals (reviewed in Rogers, 1994; see also Koslow and Gowlett- Holmes, 1998; Koslow et al., 2001; Rowden et al., 2002). Other common suspension feeders include barnacles, bryozoans, polychaete worms, molluscs, sponges, ascid- lans, basket stars, brittle stars and crinoids. There is also an associated mobile benthic fauna that includes echinoderms (starfish, sea urchins and sea cucumbers) and crustaceans such as crabs and lobsters, some of which have commercial value (reviewed in Rogers, 1994]. Deep-sea or cold-water corals (Box 1] are a group of organisms that have drawn a great deal of public attention recently. Whilst their existence has been known since the 18th century, it was only with the advent of modern technologies — which allowed fisheries, oil exploration and scientific observations to penetrate into deeper areas — that the scale and abundance of cold-water coral ecosystems Deep Atlantic Stepping Stones ScienceTeam/IFE/URI/NOAA Box 1: What is a coral? Corals are found within the phylum Cnidaria (coming from the Greek word cnidos, which means stinging nettle]. Four main classes of Cnidaria are known: the Anthozoa [which contains the true corals, anemones and sea pens); Hydrozoa (the most diverse class, comprising hydroids, siphono- phores and many medusae]; Cubozoa [the box jellies); and Scyphozoa (true jellyfish). Corals can exist as individuals or in colonies, and stony corals may secrete external skeletons made of aragonite, a form of calcium carbonate. Corals can be found in the photic zone of the ocean, where sunlight penetrates (with symbiotic photosynthetic zooxanthellae, a type of alga), as well as in the deep sea — the so-called ‘cold-water corals’. Cold-water coral ecosystems are populated by members from two classes of the Cnidaria. The main corals that will be discussed in this report are: scleractinians (stony corals], octocorals (which include the gorgonians), antipatharians [black corals) and zooanthideans (anemone-like hexa- corals), which are all found within the Anthozoa, and the stylasterids (hydrocorals), which are found within the Hydrozoa. 23 Seamounts, deep-sea corals andfisheries Chrysogorgia sp., Davidson Seamount. (NOAA/MBARI) were revealed. Deep-sea coral reefs are common features of continental shelves, slopes, banks, ridges and seamounts (Rogers, 1999; Friewald et al., 2004; Roberts et al., 2006). Today, as knowledge of their biology and ecology expands, it is becoming clear that deep-sea corals are particularly vulnerable to physical disturbance such as bottom trawling (Koslow et al., 2001; Clark and O'Driscoll, 2003; Freiwald et al., 2004; Rogers, 2004). Furthermore, because deep-sea corals have slow growth rates and poor post-disturbance recovery potential (Roberts et al., 2006), major research efforts on their conservation are emerging globally le.g. Weaver et al., 2004]. However, in addition to the direct effects of disturbance on deep-sea corals, it is becoming increasingly evident that they are an integral component of the overall species assemblage, and that the disturbance of deep-sea coral will have an equally destructive impact on the wider biological community. Whilst hard substrata are more common on seamounts than elsewhere in the deep sea, sediments are common towards the base of seamounts or on terraces or summits of flat-topped seamounts {so-called guyots]. These sediments originate from different sources, and their distribution and particle size depend on the local current regime and biological activity. Sites characterized by low exposure to currents exhibit fine, poorly sorted sediments, whilst those that are exposed to stronger currents tend to be coarser and may also be associated with bedforms such as ripples or sand waves [Levin and Thomas, 1989]. There are only a few studies on the biology of seamount sediments, but it is known that they host a wide diversity of organisms that 24 Holothurian, cerianthid anemone and Hymenaster koehleri, Davidson Seamount, 2 854 m. (NOAA/MBARI] may burrow into sediments, or live amongst the sediment’s particles or on its surface. The animals found in the sediment, known as the infauna, are classed according to size. The macrofauna [animals typically 500-250 pm in size} are dominated by polychaetes in the few studies on seamount infauna. These include many families common in other deep-sea habitats such as Paraonidae, Cirratulidae, Sabellidae, Syllidae and Ampharetidae (Levin and Thomas, 1989). Other common groups include crustaceans, molluscs, ribbon worms, peanut worms and oligochaetes. The smaller animals that live amongst the sand grains, known as the meiofauna (250-48 um in size) include nematode worms, tiny crustaceans and some more unusual groups of marine invertebrates such as loriciferans and kinorhynchs. Observations indicate that there can be an inverse relationship between diversity of the infaunal community and current strength. This is because vigorous currents lead to more coarse sediments, with a lower content of bacteria and organic food particles and higher incidence of abrasion resulting from turbation {Levin and Thomas, 1989}. The summit of Great Meteor, in the Northeast Atlantic, is covered in coarse, calcareous sediments that are home to a highly unusual community of tiny meiofaunal animals. These include new species of Loricifera (Gad, 2004a) epsilonematid nematode worms (Gad 2004b) and harpacticoid copepods (George and Schminke, 2002). The species, genera and families are not typical for deep-sea sediments and are more characteristic of littoral or shallow subtidal sediments. Larger animals living on the surface of sediments include sea pens, sponges, stalked-barnacles, gorgonians, cerianthid sea anemones, crinoids, brittle stars, sea urchins and sea cucumbers. Xenophyophores, giant single-celled organisms that agglutinate different types of particles [e.g. foraminiferan shells, sand, volcanic glass) to create elaborate dwellings of a variety of shapes, are particularly common on seamount sediments (Rogers, 1994]. Many of these organisms are suspension feeders and tend to favour areas exposed to strong currents. THE RELATIONSHIP BETWEEN CORALS AND OTHER LIFE The most spectacular benthic communities on seamounts are those associated with biological habitats or bioherms, such as cold-water coral reefs (Koslow et al., 2001). It has been suggested that cold-water coral reefs are ‘the most three-dimensionally complex habitats in the deep ocean’ (Roberts et al., 2006). As a result, there may be an associated, complex community of organisms that is dynamically linked to either the habitat structure provided by coral, or the living coral (Koslow et al., 2001; Freiwald et al., 2002). As such, cold-water coral reefs can play a similar ecological role to that of shallow-water coral reef systems (Rogers, 1999). The diversity of animals associated with cold-water coral reefs is extremely high and comparable to, or higher than, their tropical shallow-water counterparts (Rogers, 1999; Buhl-Mortensen and Mortensen, 2005). For example, greater than 1 300 species have been reported to date as being closely associated with cold-water coral reefs in the northeastern Atlantic Ocean (Roberts et al., 2006). A varying proportion of associated species may be new to science, depending on geographic area investigated [e.g. Richer de Deep-sea corals and seamount biodiversity Forges et al., 2000). The reasons for this highly diverse association are not fully understood. However, the added habitat complexity to the environment is thought to offer refugia for numerous invertebrates and fish within the living and dead coral reef framework, coral rubble and sediments, while at the same time providing hard substrates for colonization by other sessile or encrusting organisms such anemones, bryozoans and other corals (Freiwald et al., 2002). In this sense, some cold-water corals may be regarded as ecosystem engineers — that is, they create, modify and maintain habitat for other organisms (Jones et al, 1994]. Many fish species, including several of com- mercial significance, show spatial co-occurrence with deep- water corals (Auster et al., 2005: Stone, 2006), and fish catches have been found to be higher in and around deep- water coral reefs (Husebg et al., 2002). The reefs formed by some stony corals (scleractinians) are not the only three-dimensional structures built by corals. Large branching and treelike corals such as antipatharians (black corals} and octocorals [including the gorgonians) can also provide an extension of the benthic habitat through forming so-called coral beds or gardens (Stone, 2006). The branches of these corals are raised off the seabed into the overlying water [emergent epifaunal, providing rigid platforms for other sedentary and sessile species, thereby allowing them better access to food brought by prevailing currents (Stone, 2006). Such non-reef forming corals, along with other organisms such as sponges, therefore have an important role in providing habitat for other species. In the Aleutian Islands, 97 per cent of juvenile rockfish and 96 per cent of juvenile golden king crabs have Lepidion sp., swimming amongst coral framework, Hatton Bank. (DT! SEA Programme, S/o Bhavani Narayanaswamy) 25 Seamounts, deep-sea corals and fisheries been observed as associated with emergent epifauna such as octocorals and sponges (Stone, 2006). Such observations do not necessarily indicate dependence by fish on emergent epifauna. Recent studies in the Hawaiian archipelago on associations between black corals (Antipathes spp.) and fish in shallow water have indicated that many fish may routinely pass through the branches of coral colonies, treating it as general habitat. A few species regularly used the coral for protection from perceived threats, and only one species of fish was restricted to the branches of coral trees (Boland and Parrish, 2005). The fish communities of deeper slopes in Hawaii also use octocorals and zoanthids as shelter interchangeably with non-biotic habitat (Parrish and Baco, in press]. In some cases observations suggest that fish and corals occur together because they may have similar habitat requirements on seamounts and banks (Mundy and Parrish, 2004; Parrish and Baco, in press]. In a similar way, despite concentrations of orange roughy on the Tasmanian Seamounts, juveniles or young fish of this species have not been found associated with the corals; and though the adults occur in the same physical environment as the epibenthic fauna, no interaction has been observed between them (Smith, 2001). The association of fish and corals may attract large predators. For example, the endangered Hawaiian monk seal (Monachus schauinslandi] forages preferentially for fish amongst beds of deep-sea octocorals and antipatharians (Parrish et al., 2002) In addition to the general coexistence of coral and non- coral species, some animals have formed strong relationships with their coral hosts. A recent review of direct dependencies on cold-water corals globally has shown that of the 983 coral associated species studied, 114 were characterized as mutually dependent, of which 36 were exclusively dependent to cnidarians (Buhl- Mortensen and Mortensen, 2004). Such commensal relationships may come in a variety of forms: some animals are obligate inhabitants on or within the coral skeleton, such as the polychaete Gorgoniapolynoe caeciliae on the gorgonian Candidella imbricata {Eckelbarger et al., 2005); the amphipod Pleusymtes comitari associated with the gorgonian Acanthogorgia sp. (Myers and Hall-Spencer, 2004]; and the polychaete Eunice norvegicus associations with the scleractinians Lophelia pertusa and Madrepora oculata (Rogers, 1999; Mortensen, 2001; Roberts, 2005). E. norvegicus lives in tubes that become calcified by Lophelia pertusa or Madrepora oculata as they grow, conferring protection to the worm which also acts as a kleptoparasite on the corals (Mortensen, 2001; Roberts, 2005]. The worm tubes aggre- gate coral colonies, strengthening the coral framework, and the worms defend the coral vigorously from predators. Numerous species of ophiuroid brittle stars are obligate inhabitants of tree-forming corals such as the anti- patharia (Stewart, 1998 and references therein; Buhl- Mortensen and Mortensen, 2004}, which in exchange ‘clean’ corals of the build-up of detrital material that could clog their polyps. Such coral associates may be regarded as important structural species in that they may be important for the viability of the key structural species in reef and coral garden habitats (ICES, 2006). Other types of relationship also exist, for example epitoniid gastropods are specifically adapted to feed on coral polyps (B Marshall, personal communication, Museum of New Zealand Te Papa Tongarewa, Wellington, New Zealand). Lophelia pertusa framework with rich associated invertebrate fauna, Hatton Bank. (DTI SEA Programme, ©/o Bhavani Narayanaswamy] 26 Pycnogonids found on slope and base of Davidson Seamount (1 570 m); also note the chiton. (NOAA/MBARI| ATTEMPTING TO DETERMINE GLOBAL SEAMOUNT FAUNAL DIVERSITY It is important to understand the relationships between coral colonies and the fauna that Is likely to be dependent on them for food and habitat. In this sense, better understanding of the entire deep-sea coral community on seamounts will lead to a more comprehensive view of the potential impact on them from human activities such as bottom trawling In order to examine the global benthic invertebrate community composition on seamounts where corals have also been found, a freely available online resource of seamount related biological data, SeamountsOnline (Stocks, 2006) was used. SeamountsOnline is currently the largest database of its kind, spanning a wide taxonomic and geographic range of published accounts of animal and plant species occurrences on seamounts, as well as unpublished data provided voluntarily by seamount researchers. Data from SeamountsOnline used here are the most up to date at the time of the analysis (last accessed July 13, 2006) Our analysis was constrained to those seamounts for which coral has been sampled. For this, we consider in total the members of the Antipatharia, Octocorallia, Scleractinia, Stylasterida and Zoanthidea to be the coral community with potential for providing substrate or unique habitat. At the time of this analysis, the database held approximately 15 841 observations of 3 701 species from 287 seamounts around the globe. We use the term observation’ to mean a record of the occurrence of a species on a seamount Both corals and other members of the benthic community have been sampled on 47 seamounts [including seamounts <1000 mj). However, it should be noted that there can be a sampling bias towards the communities that Deep-sea corals and seamount biodiversity Fragment of live stony coral Lophelia pertusa with polychaete worm Eunice norvegicus. (Paul Tyler, School of Ocean & Earth Science, University of Southampton) were targeted [e.g. whilst hard substrates have generally been sampled, some studies have targeted soft substrates], and that the majority of seamounts have been under- sampled, so that the number of species should be considered an underestimate. Among the 47 seamounts, 322 coral species and 1 158 non-coral species are recorded from 5 541 observations. However, it must be noted that co- occurrence of corals and other benthic species does not necessarily indicate an association. That is, coral and non- coral species examined here were not necessarily collected simultaneously, nor were they necessarily collected from the same area of the seamount. In addition, some scleractinian coral species that form frameworks may have an exceptional influence on non-coral species diversity, as the reefs they form have a high associated biodiversity [e.g. Rogers, 1999; Freiwald et al., 2004; Roberts et al., 2006) Therefore, the main assumption for this analysis is that coral and non-coral species collected on the same seamount are possibly associated, in the sense that any impact on the seamount would potentially affect both coral and non-coral communities. GLOBAL DISTRIBUTION OF SAMPLING ON SEAMOUNTS The geographic distribution of 47 seamounts examined here, and the number of observations from which data were generated, is shown in Table 3.1. The map [Figure 3.1] shows that generally the North Atlantic and Southwest Pacific are the two main centres, with the highest numbers of observations of both corals and non-coral seamount species. Most seamounts examined fall within national EEZs, although exceptions include seamounts In the eastern Atlantic (Josephine, Great Meteor, Plato, Hyeres, Cruiser 27 Seamounts, deep-sea corals and fisheries Seamount Name (Number refers to Figure 3.1) Ocean Area Table 3.1: Ocean area and FAO area of seamounts used for the biodiversity analysis, together with the predicted number of large seamounts per FAO area (* indicates seamounts <1 000 m elevation). FAO Area Estimated total no. of large seamounts per FAO area = . Galicia Bank . Joao de Castro Bank . Josephine Seamount . Le Danois Bank* . Ormonde Seamount Atlantic, Northeast 27 325 . Atlantis Seamount . Cruiser Tablemount 9. Great Meteor Tablemount 10. Hyeres Seamount 11. Plato Seamount 12. Seine Seamount 2 3 4 5. Lousy Bank 6 7 8 Atlantic, Eastern Central 34 536 13. Andy's Seamount* 14. Dory Hill* 15. Hill 38* 16. Macca’s Seamount* 17. Main Pedra Seamount* 18. Sister | Seamount* Indian, Eastern 57 588 19. Kinmei and Koko Seamounts Pacific, Northwest 61 1 350 20. Dickens Seamount 21. Giacomini Seamount 22. Pratt Seamount 23. Welker Seamount Pacific, Northeast 67 265 24. Antigonia* 25. Argo Seamount 26. Jumeau East Seamount 27. Jumeau West Seamount* 28. Kaimon Maru Seamount 29. Nova Bank 30. Titov Seamount* Pacific, Western Central 71 2 670 31. Bank 8 32. Bonanza Seamount* 33. Brooks Banks 34. Cross Seamount 35. Fieberling Tablemount 36. Horizon Tablemount 37. Ladd Seamount 38. Loihi Seamount 39. Middle Bank 40. Raita Bank 41. Salmon Bank 42. Twin Banks 43. Volcano 6 Pacific, Eastern Central 77 2 735 44. Britannia Guyot 45. Gascoyne Tablemount 46. Gifford Tablemount 47. Taupo Seamount Pacific, Southwest 81 996 28 Deep-sea corals and seamount biodiversity f y Bel ie 8 % 1315 Figure 3.1: Locations of seamounts from where coral and non-coral species data were compiled for the biodiversity analysis. The numbers refer to the seamounts Listed in Table 3.1. and Atlantis Seamounts], the eastern Pacific (Fieberling and Volcano 6 Seamounts], the western Pacific (Kinmei and Koko Seamount] and the South Pacific (Gifford Seamount). Seamounts where comparative data exist are restricted to eight FAO areas (Table 3.1]. Within FAO areas, the number of seamounts sampled is limited to only a small fraction of the total estimated number of large seamounts. Some of the seamounts were subject to recent and/or continued studies, which produced useful species inventories. In the North Atlantic Ocean, the Atlantis, Cruiser, Great Meteor, Hyeres and Josephine seamounts have been particularly well studied (Figure 3.1]. Numerous species of sessile (e.g. brachiopods, bryozoans, fan worms, sponges, barnacles, tunicates) and mobile [e.g. crinoid feather stars] suspension feeders have been observed. However, the occurrence of species that typically live within soft sediments [e.g. Echinocardium heart urchins, cuspidarid bivalves and numerous polychaete families) Suggests that soft sediment habitats also exist on these seamounts. Seamounts in the Southwest Pacific have received much recent attention, and represent the most comprehensively studied region in terms of their benthic communities. The Antigonia, Jumeau East, Jumeau West, Kaimon Maru and Nova seamounts have useful species inventories, where, in addition to those components found in North Atlantic seamounts, species of ascidians, hydroids and anemones were commonly sampled. The coral communities of the Central Pacific and Northeast Pacific seamounts have been generally poorly studied. This may reflect an historical and present day scientific interest restricted to seamount fisheries of these regions. In the Central Pacific, however, the Cross and Horizon Seamounts have been well studied, and show similar community components to those described above at a similar taxonomic level. Finally, common to most seamount species inventories are numerous observations of mobile epifauna such as decapods, gastropods, nudibranchs, pycnogonids and Sea urchin on sediments on Rockall Bank. (DTI SEA Programme, ¢/o Bhavani Narayanaswamy) 29 Seamounts, deep-sea corals and fisheries Table 3.2: Summary of sampling effort on seamounts Number of Number of Mean number of coral Mean number of Observations seamounts species recorded per seamount non-coral species recorded per seamount <10 7 2 3 11-25 5 5 10 26-50 12 11 9 51-100 11 8 32 101-200 3 26 64 201-500 6 17 119 501-1 000 3 21 172 Source: SeamountsOnline echinoderms. These groups are typically categorized as detritivores or predators, suggesting higher levels of trophic complexity within seamount communities. A PRELIMINARY ASSESSMENT OF GLOBAL SEAMOUNT FAUNAL DIVERSITY A summary of the sampling effort on large seamounts is given in Table 3.2, expressed in terms of the total number observations per seamount, mean number of coral species and mean number of non-coral species observed per seamount. The majority of seamounts have had a total of 26- 100 reported observations, with means of up to 11 species of coral and 32 non-coral species observed on each seamount. The numbers of non-coral species observed increased as total number of observations increased, with the highest mean number of non-coral species observed being 172 per seamount. The mean number of coral species observed varied from two species per seamount to 26 species per seamount. Although these data could be interpreted as suggesting a link between numbers of coral species and numbers of non-coral species in the community overall, the data are highly dependent on the numbers of observations made. This reflects the inadequate sampling of the fauna on all of the seamounts studied. As a result of the limited sampling on seamounts where both corals and non-coral species have been observed, any conclusion on the relationship between coral and non-coral diversity or further analysis and interpretation of these data would be inappropriate at this time. HOW TO ALTERNATIVELY ASSESS SEAMOUNT FAUNAL DIVERSITY The comparative global analysis of the few well-sampled seamount assemblages indicates that a complex community of invertebrates may exist on those seamounts that harbour corals. However, the most evident finding is that there are Significant geographic gaps in the distribution of studied seamounts. This is highlighted by the limited number of FAO 30 areas represented by studied seamounts, and the limited number of seamounts studied in areas beyond national jurisdiction in general. Examination of coral communities is limited primarily to a few seamounts in the North Atlantic and Southwest Pacific Oceans, representing only a fraction of the total number of seamounts where biological collections have been made worldwide. Furthermore, taxonomic gaps in species inventories are likely where sampling or research aims have targeted specific components of the community, such as in the central, western and eastern North Pacific. Nonetheless, the examination of SeamountsOnline data has been useful for identifying these taxonomic and geographic gaps in the global picture of seamount biodiversity. lt has been widely suggested that negative impacts on seamount coral assemblages are likely to have significant impacts on a wider benthic community le.g. Fossa et al., 2002; Koslow et al., 2001; Lack et al., 2003), and may have possible cascading effects on the benthic and pelagic community as a whole, although these are poorly understood (Commonwealth of Australia, 2002; Lack et al., 2003). Currently there is insufficient global data to assess directly the potential vulnerability of seamount communities. Assessing the potential impacts of disturbance by bottom trawling on the seamount coral community using available cold-water coral data as a proxy for the whole seamount benthic community is a prudent alternative. The first step in this approach is taken in the following chapter of this report. Distribution of corals on seamounts PADisimiptiinenot conals on seamounts Graneledone boreopacifica and Trissopathes sp., Davidson Seamount, 1 973 m depth. (NOAA/MBARI) THE NEED TO ASSESS THE DISTRIBUTION OF CORALS Analyses of the diversity of seamount communities have generally aimed at assessing the overall diversity of seamount communities and levels of potential endemism (e.g. Richer de Forges et al., 2000). However, such studies have revealed little about how species within specific groups are distributed on seamounts at regional and global scales. Such information is critical in understanding what environmental factors influence species diversity on seamounts. It is also important in predicting the impacts of human activities on seamount communities in the absence of detailed data. Data on the occurrence of species on seamounts Is sparse and scattered over a variety of sources. For some groups of animals, most notably those comprising large, conspicuous organisms, there are a substantial number of observations. Fortunately, data for corals was sufficient for a detailed analysis of the distribution of corals on seamounts that had several principle aims: [i]. to identify global hotspots in seamount coral diversity; {ii]. to compare the distribution of different coral groups; and (iii).to understand the limitations of available data for corals in terms of geographic coverage (Rogers et al., in press}. Antipatharian coral, Munidopsis sp. and Paramola sp., Hatton Bank. (DT! SEA Programme, S/o Bhavani Narayanaswamy) THE TASK OF COMPILING USEFUL DATA A database was generated for records of all known occurrences of corals on seamounts, including some shallower features of <1 000 m elevation and some banks associated with the continental margin [n = 3 235; Rogers et al., in press]. The coral database consisted of records of the presence of a coral species at a locality and could not be used to infer species absence. This included records of Scleractinia (stony corals}; Octocorallia [including gorgonians}; Antipatharia (black corals]; Stylasterida (stylasterids/hydrocorals) and Zoanthidea (zoanthids). These records included all species of corals including those that are reef-forming, contribute to reef formation, or occur as isolated colonies. Corals were chosen as they are the most commonly recorded group of benthic animals recorded from seamounts (Stocks, 2004) and are also often associated with a diversity of other species (Rogers, 1999; Freiwald et al., 2004; Buhl-Mortensen and Mortensen, 2005; Roberts et al., 2006). As such, corals may be representative of the biological diversity of the hard- substrata benthic communities on seamounts in general {see arguments in previous chapter). These records were extracted from the primary scientific literature, from museum databases, from online data 31 Seamounts, deep-sea corals and fisheries % ‘ge eas of o s - aa °F v* o « fe a ard ks 4 = « *,. r * e é é * s ee Fig. 4.1 Global distribution of seamounts with records of corals (Scleractinia, Octocorallia, Antipatharia, Stylast- erida and Zoanthidea). Source: Rogers et al. [in press] sources [Seamounts Online; Biogeoinformatics of Hexa- corals}, from reports and from the records held by scientists Records were included in the database if corals were identified to species level or occasionally to genus if this represented a single unidentified species within the genus on a seamount. Information recorded for each record, if available, included the species name; ocean region; seamount; location, which was the exact latitude and longitude of the specimen collection if available or that of the seamount given in IOC-IHO GEBCO database; depth or Bathypathes sp., Davidson Seamount, 2 467 m. (NOAA/MBARI} depth range from which the specimen was collected; whether the specimen was alive, dead or if this information was unknown, the origin of the record; and any other pertinent notes. GLOBAL DISTRIBUTION OF RECORDS FOR SEAMOUNT CORALS Analyses of the corals on the seamount database demonstrated that sampling of seamounts has not taken place across the world’s oceans evenly (Rogers et al., in press]. Examination of a map of all coral (Scleractinia, Octocorallia, Antipatharia, Stylasterida and Zoanthidea) records shows that for some regions very few seamount samples have been taken, including the entire Indian Ocean Munidopsis sp., orange hydroid and amphipods on drifting kelp, Davidson Seamount, 1 400 m (NOAA/MBARI] 32 ___ Distribution of corals on seamounts Table 4.1: Numbers of records, species, genera and families recorded for all coral groups from seamounts Total Number Number Number Group of records of species of genera of families Scleractinia 1783 249 85 20 Octocorallia 957 161 68 21 Stylasterida 372 68 18 2 Antipatharia 157 34 22 6 Zoanthidea 28 14 6 3 Source: Rogers et al. (in press) and other regions, such as the South Atlantic, central southern Pacific and much of the Southern Ocean (Figure 4.1). It is also apparent that some areas have been well sampled, such as around New Zealand, Hawaii, off western North America and in the Northeast and Northwest Atlantic. In total, fewer than 300 seamounts have been sampled for corals, representing 2.1 per cent of the identified number of large seamounts in the oceans globally [or 0.03 per cent when assuming there are 100 000 seamounts with elevation greater than 1 000 m). PATTERNS OF CORAL DIVERSITY One of the most notable results of analyses of the database was the finding that most coral species found on seamounts are restricted to a single ocean and most of these to a single region within an ocean (Rogers et al., in press}. Only a relatively small number of species have wide geographic distributions, and very few have near-cosmopolitan distributions. Often the taxonomy and systematic status of such globally distributed species is not entirely resolved, and it is possible that some of these species represent clusters of morphologically similar sibling or cryptic species {see Le Goff-Vitry et al., 2004). Many of the widely distributed species are the primary framework building corals of cold-water reefs (e.g. Lophelia pertusa, Solenosmilia variabilis and Madrepora oculata). It is not known to what extent the limited sampling of seamounts influenced this result, and certainly some coral species have a wider geographic distribution than is apparent from the occurrences recorded on seamounts (Rogers et al., in press]. A global analysis of the species richness of corals on seamounts on a 10° by 10° latitudinal and longitudinal grid was also carried out (Rogers et al., in press]. This analysis showed that several geographic areas appeared to be hotspots of coral diversity. However, an analysis of the relationship between the numbers of coral samples for each grid box indicated that species richness was strongly dependent on sampling effort (Rogers et al., in press). Species richness of corals was also analysed by latitude (Rogers et al., in press] because there has been a suggestion that biological diversity in the oceans peaks at mid-latitudes (Worm et al., 2003). This suggestion seemed to be confirmed by the coral diversity on seamounts, which also peaked at mid-latitudes. However, this proved to be an artefact, caused by an equatorial gap in the sampling of searnount fauna (Rogers et al., in press). Despite the limitations of the coral on seamounts dataset, some broad patterns in distribution were detected. Table 4.2: Numbers of species of the two main coral groups that occur on seamounts in different ocean regions Coral group Ocean Region Scleractinia Octocorallia Northeast Atlantic Northwest Atlantic Southeast Atlantic Southwest Atlantic Northeast Pacific Northwest Pacific Southeast Pacific Southwest Pacific Southern Ocean Source: Rogers et al. [in press) 33 Seamounts, deep-sea corals and fisheries Figure 4.2: Known locations of scleractinian corals on seamounts. Source: Rogers et al. (in press] Key (above and below) Scleractinian corals are the most diverse and commonly observed group, with 249 species having been recorded. This Summit depth (m) Summit depth (m) is followed by the octocorals, the stylasterids, the 0-500 | 2 500-3 000 antipatharians and the zoanthids in order of diversity and H 500-1 000 | 3 000-3 500 number of records (Table 4.1]. Fewer than 1 500 species of 1 000-1 500 B 3 500-4 000 scleractinian corals have been described, and seamounts a 1 500-2 000 ee 4 000-4 500 therefore potentially host a substantial fraction of the er 2 000-2 500 | 4 500-5 000 global scleractinian fauna, and a very large fraction of Fig. 4.3: Known locations of octocorals on seamounts. Source: Rogers et al. [in press] 34 Biers Distribution of corals on seamounts Figure 4.4: Known locations of antipatharian corals on seamounts. Source: Rogers et al. [in press) Key Summit depth (m) Summit depth (m) head 0-500 MB .2.500-3 000 a 500-1 000 BBs: 000-3 500 © 1: 000-1 500 HB s3:500-4 000 MiSs 1 500-2000 ME 4000-4 500 WH. 2000-2 500 BB 4500-5 000 azooxanthellate coral species living in deeper waters (Rogers et al., in press). Comparison of the relative diversity of the coral groups in different regions of the oceans revealed significant differences (Table 4.2). In most parts of the world, Scleractinia were the most diverse group, followed by the Octocorallia. However, in the northeastern Pacific, this trend was reversed. Here, octocorals are markedly more diverse than scleractinians (Rogers et al., in press}. The north- eastern Pacific is characterized by a shallow aragonite saturation horizon, which may explain the lower relative diversity of stony corals in this region (Guinotte et al., 2006). Scleractinia need to accumulate large quantities of aragonite to build the coral skeleton. Undersaturation of aragonite makes this process more difficult and may result in the dissolution of dead coral skeletons, potentially preventing the occurrence of cold-water coral reefs. Given the present evidence of acidification of the oceans, this has significant implications for the global distribution of cold- water corals and coral reefs (Orr et al., 2005; Royal Society, 2005; see Chapter 5]. It is also notable that the seamounts of the northeastern Pacific are very isolated, and differences in dispersal capacity between the two coral groups may also influence their distribution. The feeding ecology of scleractinians and octocorals is also different, and this may also result in contrasting environmental preferences of the two coral groups. THE RELATIVE OCCURRENCE AND DEPTH DISTRIBUTION OF THE MAIN CORAL GROUPS Analysis of the depth distribution of the four main different coral groups, the Scleractinia, Octocorallia, Stylasterida, Antipatharia and Zoanthidea, found that the different coral groups occurred at different depths (Figures 4.2-4.4). Most scleractinian and stylasterid species occur in the upper 1 000- 1 500 m (Rogers et al., in press]. Octocorals can be found in greater depths, with most species occurring in the upper 2000 m. Antipatharians also occurred in the upper 1 000 m, although a higher proportion of species occurs deeper than scleractinians or stylasterids. A variance analysis, using a Generalised Linear Model {[GLM], of the whole dataset showed that the depth distributions were different between the four coral groups. Analysis in pairs showed the depth distributions of scleractinian and stylasterids to be similar and different from both octocorals and antipatharians (Rogers et al., in press). Sampling effort to date limits our understanding of coral distribution below 2 500 m. GETTING A BETTER UNDERSTANDING OF CORAL DISTRIBUTION ON SEAMOUNTS The relative occurrence and distribution of corals on 35 Seamounts, deep-sea corals and fisheries seamounts demonstrate that the depth of the seamount summit will have a significant influence on the composition of the coral communities present. This is likely to apply also to other groups of sessile organisms (Rogers et al., in press). The greatest diversity of corals observed on seamounts occurs in the upper 1 000 m of the oceans, and the depth ranges with the highest coral diversity overlap with those where most deep-sea fishing currently takes place (250- 1 500 m; Koslow et al., 2000; ICES, 2005). Given that depth is one of the major factors influencing physical classification of seamounts (Rowden et al., 2005; ICES, 2006), this will be a significant factor in predicting the diversity of coral communities on unsampled seamounts. However, it should be noted that even for mean depths, the results for the GLM indicated that taxonomic groups of coral have only a relatively small influence on depth distribution [it explains about 10 to 13 per cent of the variation), and that many other factors — such as the physical environment of a seamount - will also determine species composition and distribution (Rowden et al., 2005). Overall, the analyses revealed new patterns in the regional and vertical distribution of coral species. The reasons for differences in the depth and regional distribution of the different coral groups are most likely related to the nature of substrates available for attachment; the quantity, quality and abundance of food at different depths [see Chapters 2 and 3]; but also to the aragonite saturation horizon, temperature and the amounts of different essential elements and nutrients (Bonilla and Pifdn, 2002). The dataset also revealed significant areas of weakness in our knowledge of seamount coral diversity, especially in the lack of sampling of seamounts in equatorial latitudes. Thus, in order to make a reasonable assessment of the vulnerability of corals and, by proxy, non-coral communities on seamounts to bottom trawling, it is currently necessary to use models to predict the global occurrence of suitable coral habitat. 36 Stony corals on seamounts 9. Predicting global distribution of stony corals on seamounts Paragorgia arborea, Davidson Seamount, 1 779 m. (NOAA/MBARI) KNOWN CORAL DISTRIBUTION The previous chapter demonstrates that our knowledge of the distribution of corals on seamounts is limited. Most records come from heavily sampled regions such as the Northeast Atlantic and around New Zealand, a pattern that is unlikely to represent the true distribution of these corals. There are very few data from seamounts in some regions, such as the south-central Pacific and the Indian Ocean, and the vast majority of large seamounts have not been sampled at all. In order to improve our knowledge of where and why deep-sea corals are found on seamounts, further sampling and research has to be conducted, but this is time- consuming and expensive. A short-term alternative, although not replacing the need for further sampling, is to use a modelling approach. A common problem in biology is attempting to predict in which areas an organism is likely to be found, given a limited set of observations of its distribution. Understanding the factors, such as climate and food availability, that drive its distribution (Gaston, 2003) can help. Models (Box 2] can be used to predict the distribution of a species from observed occurrences and absences of individuals and their relationship to measurable environmental parameters [Guisan and Zimmermann, 2000; Guisan and Thuiller, 2005). In this chapter we construct a habitat suitability model to gain insight into the global distribution of deep-sea corals on seamounts. Some scleractinian corals form complex structures and frameworks such as reefs that provide habitat for other deep-sea species (Rogers, 1999; Freiwald et al., 2004). Better knowledge of the distribution of such species supplies a useful proxy for the biodiversity of benthic communities of seamounts (see Chapter 3) Other groups of coral, such as octocorals, for example, can also form important habitats such as coral gardens [e.g Stone, 2006; see Chapter 3). These corals may have very different distributions from that of stony corals, which would also be useful to appreciate in the context of determining the vulnerability of seamounts communities to bottom trawling Unfortunately, the available data for octocorals are currently too limited to enable appropriate modelling Box 2: What is a model? A model, in this context, is a simplified, abstracted representation of a real-world system. Models are typically constructed using mathematical equations or statistical functions that are programmed into a computer. For example, existing data [such as known seamount coral distributions) are fed into the model, and the output (such as predicted habitat suitability maps for seamount corals) is used to aid in the understanding of patterns and processes and to make predictions. Models are often compared and tested against one another. There is a trade-off between simplicity and complexity. A simple model that captures the essential features of the system in question is often preferable to a more complex model where more assumptions have to be made because there is normally not enough known about parts of the ecosystem. It is important to remember that a model can never be perfect or ‘right’. It is a simplified representation of reality. A good outcome would be for the model to capture large-scale features of the system in question. It is also important to calibrate a model against known data and knowledge, and to statistically assess its accuracy. Only when the uncertainty in a model can be quantified is it of significant use. 37 Seamounts, deep-sea corals andfisheries —_ Table 5.1: Environmental parameters used to predict habitat suitability [GLODAP = Global Ocean Data Analysis Project; SODA = Simple Ocean Data Assimilation 1.4.2; VGPM = Vertically Generalized Productivity Model; WOA = World Ocean Atlas 2001] Parameter Units Source Reference Temperature °C WOA Conkright et al., 2002 Salinity Pss WOA Conkright et al., 2002 Depth m WOA Conkright et al., 2002 Surface chlorophyll yg tl WOA Conkright et al., 2002 Dissolved oxygen mlt! WOA Conkright et al., 2002 Per cent oxygen saturation % WOA Conkright et al., 2002 Overlying water productivity mg C m2 yr! VGPM Behrenfeld and Falkowski, 1997 Export primary productivity gC m2 yr-] VGPM Behrenfeld and Falkowski, 1997 Regional current velocity cm s7! SODA Carton et al., 2000 Total alkalinity umol kg-! GLODAP Key et al., 2004 Total dissolved inorganic carbon umol kg-! GLODAP Key et al., 2004 Aragonite saturation state umol kg-! Derived from Key et al., 2004; USING HABITAT SUITABILITY MODELLING TO PREDICT STONY CORAL DISTRIBUTION Statistical techniques for the modelling of habitat suitability have been used since the 1970s, and since then have branched into a variety of different approaches (Guisan and Thuiller, 2005). There is no single model that is ‘best’ in all situations; typically a model is selected because it is thought to be the most appropriate for the type of data, or several competing models are tested against one another. Crinoid (Florometra serratissima) and brisingid seastar on black coral, Davidson Seamount 1 950 m. (NOAA/MBARI) 38 GLODAP data Orr et al., 2005; Zeebe and Wolf-Gladrow, 2001 The modelling technique used in this analysis is ‘environmental niche factor analysis (ENFA], developed by Hirzel et al. (2002). ENFA compares the observed distribution of a species, or group of species, to the background distribution of environmental factors (temperature and salinity, for example). In this way, it assesses how different the environmental niche a tax- onomic group occupies Is relative to the mean background environment (its ‘marginality’], and how narrow this niche is [its Specialization’). The model also reveals factors that can be important in determining the distribution of the studied organisms. ENFA can then use this information to predict habitat suitability for unsampled regions. ENFA is ideal when there is reliable presence data, but no reliable absence data (Hirzel et al., 2001; Brotons et al., 2004), as is the case for coral data from seamounts. We know where scleractinians have been found, but even for those seamounts that have been sampled, we cannot infer true absence, since coral species may be living on an unsampled region of the same seamount or coral material has not been identified and sorted from samples. ENFA has been previously used in the marine environment to model coral distributions on the Canadian Atlantic continental shelf (Leverette and Metaxas, 2005). Further details of the model are given in Appendix II. STONY CORAL DISTRIBUTION AND ENVIRONMENTAL DATA The location of records of scleractinian corals on seamounts came from the database generated for the analysis of coral distribution (see Chapter 4). These data were then combined with physical, biological and chemical oceanographic data from a variety of sources, as outlined in Table 5.1 (full details in Appendix |]. Data on large seamount locations were obtained from the data used for Chapter 2 (Kitchingman and Lai, 2004). The coral data and the seamount locations do not completely match, since some of the coral records come from small seamounts Thus we cannot model habitat suitability for stony corals on seamounts directly. Instead, we use the coral data to model habitat suitability in various regions and depth zones of the global oceans, initially ignoring the locations of large seamounts. The habitat suitability maps can then be used in two ways: [i] to examine the habitat suitability for as yet unknown seamounts and other sea floor features within a particular region of the marine environment; and [ii] fitted to the summits of known/predicted seamounts. Habitat suitability for scleractinians on seamounts may be very different depending upon whether the corals are sited on the seamount summit or slope, as these are at different depths and potentially in different oceanographic regimes. Caution should therefore be used when interpreting habitat suitability fitted to seamount summits. The ENFA model assumes that the data span the environmental range of actual scleractinian occurrence, |.e that Scleractinia do not reside outside the environmental extremes that have been sampled; otherwise, the model will not predict areas of suitable habitat beyond these extremes This appears to be a reasonable assumption in this instance. Nonetheless, we limited the model to 2 500 m in depth, as below 2 500 m data are more limited by sampling, and there is a marked change in the species composition of the scleractinians (Rogers et al., in press} PREDICTED HABITAT SUITABILITY FOR STONY CORALS The predicted habitat suitability for scleractinians found on seamounts is shown in Figures 5.1 to 5.6 in 250-500 m bands from 0 m to 2 500 m depth. The following assessment of these maps refers to the main FAO fishing areas (FAO, 2005) and to the areas beyond national jurisdiction given in the Reference Maps 1 and 2 on the back cover. In near-surface waters, suitable seamount habitat ies in the southern North Atlantic {mostly FAO area 31), the South Atlantic (FAO area 41), much of the Pacific especially FAO areas 77, 81 and 87], and the southern ndian Ocean (FAO areas 51 and 57). The Southern Ocean and northern North Atlantic are, however, unsuitable. Habitat suitability patterns change substantially below this depth. In depths from 250 m to 750 m, a narrow band around 30°N + 10° and a broader band of suitable habitat occur around 40°S + 20° [areas 81 and 87 in the South Pacific, 41 and 47 in the South Atlantic, and 51 and 57 in the Indian Ocean). Below 750 m, the North Pacific and Stony corals on seamounts northern Indian Ocean become particularly unsuitable. The circum-global band of suitable habitat at around 40°S narrows with depth (to + 10°), breaking up into smaller suitable habitat areas around the southeast coast of South America and the tip of South Africa. Suitable habitat remains in parts of the Atlantic to 2 500 m depth [especially the North and tropical West Atlantic, most consistently in FAO areas 31 and 34, with FAO areas 21 and 27 becoming more prominent with depth]. The global extent of habitat suitability for stony corals on seamounts was predicted to be at its maximum between 250 m and 750 m [Figure 5.2] The majority of the suitable habitat for stony corals occurs in areas beyond national jurisdiction. However, suitable habitats are also predicted in deeper waters under national jurisdiction, especially in the EEZs of countries [i] between 20°S and 60°S off South Africa, South America and the Australian/New Zealand region, [ii] off northwest Africa, and [iii) around 30°N in the Caribbean. The results of combining the predicted habitat suitability with the summit depth and location of large seamounts are shown in Figure 5.7. The majority of the large seamounts that could provide suitable habitat on their summits are located in the Atlantic Ocean (all Atlantic FAO areas - 21, 27, 31, 34, 41 and 47]. The rest are mostly clustered in a band between 20°S and 60°S. A few seamounts elsewhere, such as in the South Pacific, have summits in the high suitability depth range between 0m and 250 m. In the Atlantic, a large proportion of suitable seamount summit habitat is beyond national jurisdiction, whereas in the Pacific it is mostly within EEZs. In the southern Indian Ocean, suitable habitat appears both within and outside of EEZs. When analysing the habitat suitability on the basis of summit depth, it should be Primnoid coral with shrimp, Davidson Seamount, 1570 m depth. (NOAA/MBARI) 39 Seamounts, deep-sea corals and fisheries Table 5.2: Variance explained by the first eight ecological factors in the ENFA model. Factor one explains the z| marginality, the remainder the specialization. The cumulative explained specialization of the first eight factors is 88.6 per cent. Factor 1(Marginality) 2 3 4 5 6 il 8 Explained 0.12 0.19 0.17 0.12 0.10 0.08 0.06 0.05 specialization Alkalinity -0.30 0.04 0.22 0.07 0.13 0.23 0.23 0.35 Aragonite 0.34 0.07 0.06 0.83 0.12 0.00 0.14 0.12 saturation state Surface chlorophyll 0.25 0.02 0.13 0.05 0.01 0.00 0.03 0.30 Depth -0.21 0.15 0.32 0.18 0.05 0.16 0.07 0.07 Dissolved 02 0.22 0.66 0.11 0.08 0.41 0.48 0.68 0.35 Per cent 02 0.27 0.61 0.22 0.41 0.74 0.70 0.46 0.40 saturation Primary productivity 0.46 0.00 0.00 0.06 0.01 0.04 0.08 0.49 Export productivity 0.31 0.02 0.00 0.01 0.01 0.00 0.04 0.09 Salinity 0.24 0.05 0.03 0.13 0.04 0.03 0.16= 0.00 Total CO2/DIC -0.29 0.29 0.57 0.12 0.51 0.48 0.04 0.49 Temperature 0.35 0.29 0.49 0.11 0.01 0.09 0.46 0.40 Regional current -0.03 0.03 0.01 0.07 0.03 0.03 0.03 0.01 velocity i remembered that suitable habitat for stony corals might also occur on the slopes of seamounts, i.e. at depths greater than the summit. WHAT ENVIRONMENTAL FACTORS ARE IMPORTANT IN DETERMINING STONY CORAL DISTRIBUTION? Table 5.2 shows the importance of each environmental parameter in constraining the distribution of scleractinians on seamounts. The first column (Factor 1] is the marginality of the group, and the remaining factors its specialization (see Appendix II). The predicted value of the marginality is 0.918, which indicates that optimal habitat for scleractinians iS quite different from the background mean values. This is true for almost all environmental variables except regional current velocity. The remaining factors (2-8) show the parameters that are important in driving the observed distribution. The specialization value is 1.369, indicating that stony corals are highly specialized and occupy a relatively narrow environmental niche. The environmental parameters that are most important in determining suitable habitat for seamount stony corals are dissolved oxygen and per cent oxygen saturation, total dissolved inorganic carbon and the aragonite saturation state. The comparison of Figures 5.8 with Figures 5.2 and 5.3 shows that high levels of aragonite saturation and dissolved oxygen correspond with suitable habitat for scleractinians. Similarly, high values of per cent oxygen saturation and low 40 values of total dissolved inorganic carbon also correspond with suitable habitat. Interestingly, neither surface chlorophyll nor regional current velocities apparently are important in determining global scleractinian distributions on seamounts, although these may have an effect at a smaller spatial scale, such as an individual seamount. That is at a scale not captured by the size of the grid used within the model. Temporal variability in the environmental factors is not captured by the model, and daily, seasonal and annual changes may all play a role in driving stony coral distributions. It is also worth noting that other important factors may not have been included in the model. For example, biological factors, such as competitive exclusion, are typically not captured by habitat suitability models. Those factors that are included in the model may not actually be responsible for driving the distribution of Scleractinia, but simply correlated with unknown factors. This could mean that even if a region is predicted as being highly suitable for scleractinians, it does not mean that these corals will actually be found there. The result that dissolved oxygen availability is a major factor affecting habitat suitability for stony corals, and thus influences their distribution, is significant in a global oceanographic context. It means that oxygen minimum zones [Helly and Levin, 2004), which can be extensive in some parts of the worlds oceans, would not be very suitable habitat for these corals. —S—s_. —— Stony corals on seamounts Figure 5.1: Predicted habitat suitability for seamount stony corals from 0-250 m depth. Key Habitat suitability Habitat suitability % % i: 0-10 Bi 50-60 &: 10-20 ww 60-70 & 20-30 eal 70-80 al 30-40 eB 80-90 & 40-50 HS sé"0-i100 Aragonite is a form of calcium carbonate that scleractinians use to form their hard skeletons. It has been speculated that stony corals will have their distribution limited by the level of aragonite saturation (Orr et al., 2005; Guinotte et al., 2006). In the oceans, carbon dioxide and carbonate ions react with each other to form bicarbonate (Royal Society, 2005). Increased amounts of aqueous COz (e.g. from anthropogenic sources) cause a decrease in the availability of carbonate ions, which corals and other organisms use to build calcareous skeletons [Royal Society, 2005). Simultaneously, this decreases the pH of the ocean [makes it more acidic]. Thus not only are there fewer resources available with which to produce coral skeletons, but they are also dissolved more quickly by the higher acidity (Orr et al., 2005). Thus, we would expect high levels of aragonite saturation to be suitable habitat, and this is indeed the case. Total dissolved inorganic carbon, however, is inversely correlated with aragonite High percentage values indicate more suitable habitat. saturation, so low levels provide suitable stony coral habitat. The model output must be examined in an appropriate context, and the habitat suitability maps in this chapter can be considered as testable hypotheses. If additional sampling were to be carried out and found scleractinians on seamounts outside their current environmental envelope, then this would change the model predictions. The model may perform better in some regions than in others. Furthermore the distribution of deep-sea stony corals in non-seamount regions may be different from that on seamounts. Previous chapters have built a picture of where seamounts are located, what lives on them - in particular corals - and where corals may be found beyond areas that have been sampled. The next step in the sequence is to look at what fish species occur on seamounts, and where fisheries for them have, or may, occur. 41 Seamounts, deep-sea corals and fisheries Figure 5.2: Predicted habitat suitability for seamount stony corals from 250-750 m depth. Key (above and below) Habitat suitability Habitat suitability High percentage values % % indicate more suitable ie 0-10 al 50-60 habitat. bay 10-20 | 60-70 Wr 20-30 al 70-80 Fe 30-40 & 80-90 el 40-50 HM ss0-100 42 Stony corals on seamounts Figure 5.4: Predicted habitat suitability for seamount stony corals from 1 250-1 750 m depth. Key (above and below) Habitat suitability Habitat suitability High percentage values % % indicate more suitable 0-10 50-60 habitat. eek 10-20 60-70 Be 20-30 70-80 a 30-40 80-90 i 40-50 90-100 43 Seamounts, deep-sea corals and fisheries Figure 5.6: Predicted habitat suitability for seamount stony corals from 2 250-2 500 m depth. Key (above and below) Habitat suitability Habitat suitability High percentage values % % indicate more suitable F sa 0-10 50-60 habitat. fe 10-20 60-70 fel 20-30 70-80 eal 30-40 80-90 @ 40-50 90-100 44 I Oe Stony corals on seamounts Figure 5.8: Aragonite saturation state (left panels) and dissolved oxygen (right panels). Top panels are at a depth of 500 m, lower panels at 1 000 m. Key (left upper and lower) Key (right upper and lower) Aragonite (umol kg”) Dissolved oxygen (ml lt") ex! 50-25 mM 50-75 bait 0-05 M 35-40 a -25-0 Mas-100 eh 0.5-1.0 Mm 40-45 Si 0-25 Mi io0-125 hat 10-15 M 45-50 eal 25-50 MB ins-150 bal 15-20 Mm 50-55 (ed 20-25 MM 55-60 Bal 25-3.0 | es a 3.0-3.5 Mm 65-70 45 Seamounts, deep-sea corals and fisheries 6 SesnncUA ish ane ishenes FISH BIODIVERSITY Seamounts support a large number and wide diversity of fish species. Wilson and Kaufman [1987] were the first to review seamount biota worldwide and reported about 450 fishes collected from more than 60 seamounts. Rogers (1994) provided a list of 77 commercial species fished on seamounts. Since then, more detailed studies of certain seamounts and seamount chains have provided more comprehensive species lists. Froese and Sampang (2004) compiled a list of 535 fish species, which was augmented by Morato and Pauly (2004) to a total of 798 species. Most of these fish species are not exclusive to seamounts and occur widely on the continental shelf and slope habitat (Morato and Clark, in press]. Fish communities around and on seamounts are therefore complex, being composed of pelagic species living in the surface water layers, mesopelagic species such as myctophids occurring in deeper water, and species living close to or on the seabed of the seamount itself [sometimes termed the seamount community; Commonwealth of Australia, 2002). It is known that different elements of these communities may share common prey species, although the trophic relationships between different groups of fish around seamounts are not well understood at present (e.g. Parin and Prut’ko, 1985; Commonwealth of Australia, 2002). Seamounts can be an important habitat for commer- Bathysaurus mollis, Davidson Seamount, 2 375 m depth; ambush predator. (NOAA/MBARI) 46 cially valuable species that may form dense aggregations for spawning or feeding (Clark, 2001; Roberts, 2002; ICES, 2005}, and on which a number of large-scale fisheries have developed. Because many fisheries on seamounts target aggregations, catches can be relatively clean li.e. they are composed of one or a few species]. However, in other cases, the by-catch of seamount fisheries include a variety of other species that are often discarded (Roberts, 2002). Levels of by-catch can be such that non-target species of seamount can become depleted. For example, records over 10 years of the orange roughy fishery on the Chatham Rise showed that 13 out of 17 by-catch species recorded lower biomass in 1994 compared to 1984 (Clark et al., 2000). Depletion of sharks and rays either by targeted fishing or as by-catch from deep-sea fisheries is a major cause for concern {Lack et al., 2003; Royal Commission on Environmental Pollution, 2005; see also UN General Assembly, 2004b, Paragraphs 47 and 48). DEEP-WATER FISHERIES DATA Information on distribution and depth ranges of commercial fish species were obtained from global databases available on the Internet, namely FishBase [www.fishbase.org) and Ocean Biogeographic Information System (OBIS; www. iobis.org]. Both sources have a number of distributional maps available, based on point locality information. Fish Base also offers facilities to map a projected distribution; however, location data for areas outside of national jurisdiction are often missing, and the maps overestimate the potential distribution for some species. Both types of distributional data were examined, and a subjective assessment was applied based on expertise and experience of one of the contributors to this report (M Clark} to define the likely distribution of the fish species in areas outside of national jurisdiction. The only international source of global fisheries catch data is that compiled by the FAO. While FAO statistics do not make a distinction between EEZs and areas beyond national jurisdiction, and reporting areas are very large, data available from the FAO provides a useful input for assessing the deep-water catch by species in areas outside of national jurisdiction in various parts of the world where seamounts are important fishing grounds. Clark et al. [in press} have _ Seamount fish and fisheries Chimaerid, probably Chimaera monstrosa, Hatton Bank, Northeast Atlantic. By-catch of this group of fish species is a major concern related to deep-sea fisheries. (DTi SEA Programme, ©/o Bhavani Narayanaswamy) used FAO data, together with fisheries statistics and data of Soviet, Russian and Ukrainian scientific research and exploratory cruises, and published reports for some seamount fisheries conducted by Japan, New Zealand, Australia, Spain, other EU countries and Namibia. Personal contacts and data extracted by Clark et al. [in press) were used in some cases to provide ‘guesstimates of likely species composition and catch for some seamount regions. The report by Gianni (2004) on high seas [areas outside of national jurisdiction) fishing in general was examined for some areas where much of the high seas catch was thought to be from seamounts. The catch figures given in this chapter are known to be incomplete. Some countries’ data were not available, there is known to have been misreporting or non-reporting of catches from areas outside of national jurisdiction in the past (e.g. Lack et al., 2003). In addition, many catch statistics (e.g. FAO records; catches from ICES sub-areas] are on a scale that does not allow the approximate location to be determined, let alone assign the catch to a particular seamount. The effect of this variable quality is that some fishing may have occurred outside areas of national jurisdiction, or that effort and catch levels in some areas could be much higher. However, the compilation is the most comprehensive attempted to date for seamount fisheries, and is believed to give a reasonable indication of the general distribution of seamount catch over the last four decades. GLOBAL DISTRIBUTION OF DEEP-WATER FISHES Current deep-water trawl fisheries occur in areas beyond national jurisdiction for a number of species. These include alfonsino (Beryx splendens]; black cardinalfish {Epigonus telescopus}; orange roughy (Hoplostethus atlanticus); boarfish (Pseudopentaceros richardsoni}; macrourid rattails (primarily roundnose grenadier Coryphaenaides rupestris); oreos (several species of the family Oreosomatidae, including smooth oreo (Pseudocyttus maculatus}, black oreo (Allocyttus niger], warty oreo {Allocyttus verrucosus) and spiky oreo (Neocyttus rhomboidalis). Many of these fisheries use bottom-trawl gear. Other fisheries occur over seamounts, such as those for pelagic species [mainly tunas) and target species for smaller-scale line fisheries (e.g. black scabbardfish Aphanopus carbo) (FAO 2004). The depth distribution of these fish species is given in Table 6.1. Many species cover a very wide depth range, which can vary with the life history stage of the species (e.g. juveniles are often found in shallower depths than adults). Typically, the depth range in which fishing takes place is smaller than the actual range of the species, as fishers target depths where the adult fish often aggregate for spawning or feeding. The depth distribution of most species differs in various parts of the world, as water masses vary. For example, orange roughy typically occurs on seamounts at depths of 800-1000 m in the Southwest Pacific and southern Indian Ocean, 500-800 m in the South Atlantic, and at greater than 1 000 m in the North Atlantic. 47 Seamounts, deep-sea corals and fisheries _ Table 6.1: Depth distribution of commercial fish species on seamounts Species Code Scientific name Main depth Total depth (common name) range (m) * range (m) * Alfonsino BYX Beryx splendens 300-600 25-1 300 Cardinalfish EPT Epigonus telescopus 500-800 75-1 200 Rubyfish RBY Plagiogenion rubiginosum 250-450 50-600 Blue ling LIN Molva dypterygia 250-500 150-1 000 Black scabbardfish SCB Aphanopus carbo 600-800 200-1 700 Sablefish SAB Anoplopoma fimbria 500-1 000 300-2 700 Pink maomao MAO Caprodon spp. 300-450 To 500 Southern boarfish LBO Pseudopentaceros richardson 600-900 To 1 000 Pelagic armourhead ARM Pseudopentaceros wheeleri 250-600 To 800 Orange roughy ORH Hoplostethus atlanticus 600-1 200 180-1 800 Oreos OEO (BOE, SSO) Pseudocyttus maculatus, 600-1 200 400-1 500 Allocyttus niger Bluenose BNS Hyperoglyphe antarctica 300-700 40-1 500 Redfish RED Sebastes spp. (S. marinus, 400-800 100-1 000 S. mentella, S. proriger) Roundnose grenadier RNG Coryphaenoides rupestris 800-1 000 180-2 200 Toothfish PTO Dissostichus spp. 900-1 500 50-3 850 Notothenid cods NOT Notothenia spp. 200-600 100-900 * Main depth range refers to the commercial fishing depths; total depth range refers to the full known depth range of adult fish (from FishBase). The geographical distribution of the main commercial A number of southern hemisphere species are found in the fish species in the world’s oceans is summarized in Table North Atlantic, but do not extend into the North Pacific (e.g. 6.2. Many have a widespread occurrence, especially through orange roughy, oreos]. Some species are more localized to the Atlantic Ocean, Indian Ocean and South Pacific Ocean. the North Atlantic (e.g. roundnose grenadier, blue ling, and Table 6.2: Geographical distribution of commercial fish species (+ indicates occurrence in that ocean) Species North South North South Indian Southern (common name) Atlantic Atlantic Pacific Pacific Ocean Ocean Alfonsino + + + + + Cardinalfish + + + + Rubyfish + + + Blue ling + Black scabbardfish + + Sablefish Pink maomao Southern boarfish Pelagic armourhead Orange roughy + Oreos Bluenose Redfish + + Roundnose grenadier Toothfish Notothenid cods + |e lt |+ f+ + [+ |+ |+ [+ + [+ [+ |+ [+ |+ + ft f+ [+ [+ 48 - Seamount fish and fisheries Figure 6.1: Distribution of (clockwise from top left) orange roughy, roundnose grenadier, Patagonian toothfish and alfonsino. Source: OBIS and FishBase databases redfishes Sebastes mentella and S. marinus}, and sablefish (Anoplopoma fimbria) occur only in the North Pacific. Hence, depending on the target fish species, certain geographic areas, including parts of areas beyond national jurisdiction, are more likely to be searched by fishing vessels than others. Distributions of four of the most important (for either their abundance or commercial value] seamount fish species are shown in Figures 6.1 and 6.2. The first shows recorded location data taken from OBIS [which is linked to FishBase], whereas the second shows the distributions modelled and generated with the Aquamap function within FishBase. Location data for areas outside of national jurisdiction are poor, because research vessels work mainly in national waters. The modelled distribution of some of the species is uncertain. The overall distribution and relative densities predicted are based on limited distributional and environmental data. In some cases they are known to be too extensive. However, they do serve as an approximate guide to the likely distribution when viewed together with the actual location data. Orange roughy is widely distributed throughout the North and South Atlantic Oceans (FAO areas 27, 47), the mid-southern Indian Ocean [FAO areas 51, 57) and the South Pacific (FAO areas 81, 87). The species does not extend into the North Pacific, and is unlikely to occur in the northern parts of the Indian Ocean [although the modelling does suggest the latter; cf. Figure 6.2). It is frequently associated with seamounts for spawning or feeding, although it is also widespread over the general continental slope. Alfonsino has a global distribution, being found in all the major oceans. It is a shallower species than orange roughy, occurring mainly at depths of 400 m to 600 m. It is associated with seamount and bank habitat. Roundnose grenadier is restricted to the North Atlantic (FAO areas 21, 27). It occurs on both sides of the North Atlantic, as well as on the Mid-Atlantic Ridge, where aggre- gations occur over peaks of the ridge. Patagonian toothfish (Dissostichus eleginoides) - and in some areas Antarctic toothfish (Dissostichus mawsoni) - have a restricted southern distribution (FAO areas 48, 58, 88). Having a very wide depth range, the species is sometimes associated with seamounts, but also general slope and large bank features (Rogers et al., 2006). GLOBAL DISTRIBUTION OF MAJOR SEAMOUNT TRAWL FISHERIES The intensive search for fisheries resources on seamounts around the world’s oceans was initiated by the former Soviet Union, and soon after by Japan, in the late 1960s and 1970s (Rogers, 1994). Seamounts with concentrations of fish and invertebrates were found initially in the Pacific Ocean but later in other parts of the Atlantic and Indian Oceans, and offshore seamounts became established as important habitat for global fisheries (Figure 6.3). In subsequent decades other countries such as Korea, and later China, Cuba, Australia and New Zealand, and countries in the European Union and southern Africa, also developed fisheries on seamounts. Table 6.3 shows that in total, the 49 Seamounts, deep-sea corals and fisheries Figure 6.2: Predicted distribution of (clockwise from top left) orange roughy, roundnose grenadier, Patagonian toothfish and alfonsino. High probability of occurrence values indicate more suitable habitat. Key (all above) Habitat suitability Habitat suitability fal 0-0.1 & 05-0.6 iba 0.1-0.2 ia 0.6-0.7 oe) 0.2-03 3 0.7-0.8 a 0.3-0.4 B 0.8-0.9 a 0.4-0.5 8 0.9-1.0 international catch of demersal fishes on seamounts by distant-water fishing/trawling fleets is estimated to be about 2 million tonnes of fish since the 1960s (derived from data in Clark et al., in press). The largest seamount trawl fisheries have occurred in the Pacific Ocean. In the 1960s to 1980s large-scale fisheries for pelagic armourhead and alfonsino occurred on the Hawaiian and Emperor seamount chains in the North Pacific [FAO area 77} (Figure 6.3). In total about 800 000 tonnes of pelagic armourhead were taken, and about 80 000 tonnes of alfonsino. In the southwestern Pacific (FAO areas 81, eastern part of 57], fisheries for orange roughy, oreos and alfonsino have been large, and continue to be locally important. Orange roughy has also been the target of fisheries on seamounts on the Reykjanes Mid-Atlantic Ridge in the North Atlantic, off the west coast of southern Africa, and in the southwestern Indian Ocean. Roundnose grenadier was an important fishery for the Soviet Union in the North Atlantic (FAO area 27], where catches have been over 200 000 tonnes. Smaller fisheries for alfonsino, mackerel and cardinalfish have occurred on various 50 seamounts in the mid-Atlantic and off the coast of North Africa. In the Southern Ocean, fisheries for toothfish, notothenioids and icefish can occur on seamounts as well as slope and bank areas. Most of these seamounts are fished with bottom trawl, but several are also subject to mid-water trawl and long-line fisheries. In most cases It has not been possible to distinguish between bottom trawl and mid-water trawl. Many of these fisheries are historical. Most of these fisheries have not been sustainably managed, with many examples of boom and bust’ fisheries, which developed and declined rapidly, sometimes within a few years or a decade (e.g. Uchida and Tagami, 1984; Koslow et al., 2000; Clark, 2001; Lack et al., 2003). A prime example of this, in areas beyond national jurisdiction, is the recent fishery in the Southwest Indian Ocean, which collapsed after only four years in the late 1990s (FAO, 2002; Lack et al., 2003). Recovery of shallow-water fish stocks that have been collapsed or severely depleted have rarely taken place after 15 years (Royal Commission on Environmental Pollution, 2004). The life history characteristics of many deep-water fish species are more conservative than shallow water species (e.g. slow growth rate, low rates of natural mortality in adult fish, late age of sexual maturity, sporadic repro- duction, high longevity; Rogers, 1994; Koslow et al., 2000; Lack et al., 2003). This makes the rebuilding and recolon- ization of previously fished seamounts extremely slow, and many have shown no signs of recovery to date (Tracey and Horn, 1999; Cailliet et al., 2001; Lack et al., 2003). and main gear types used in the seamount fisheries Table 6.3: Total estimated historic catch of main commercial fish species from seamounts, major fishing periods, Seamount fish and fisheries Species Total historical catch (t) Main fishery years Gear type Alfonsino 166 950 1978-present Bottom and mid-water trawl, some long-line Cardinalfish 52 100 1978-present Bottom (and mid-water trawl) Rubyfish 1.500 1995-present Bottom and mid-water trawl Blue ling 10 000 1979-1980 Bottom trawl Black scabbard fish 75 000 1973-2002 Bottom and mid-water trawl Sablefish 1 400 1995-present (Bottom trawl), line Pink maomao 2 000 1972-1976 Bottom and mid-water trawl Southern boarfish 9 600 1982-present Bottom trawl Pelagic armourhead 800 000 1968-1982 Bottom and mid-water trawl Orange roughy 419 100 1978-present Bottom trawl Oreos 145 150 1970-present Bottom trawl Bluenose 2500 1990-present Bottom and mid-water trawl Redfish 54 450 1996-present Bottom and mid-water trawl Roundnose grenadier 217 000 1974-present Bottom and mid-water trawl Toothfish 12 250 1990-present Bottom trawl, long-line Notothenid cods 36 250 1974-1991 Bottom trawl 148 200 Mackerel species 1970-1995 (Bottom) and mid-water trawl Total Source: Clark et al. [in press] 2 153 470 AREAS OF EXPLORATORY FISHING IN AREAS BEYOND NATIONAL JURISDICTION Offshore seamount fisheries in international waters generally require large freezer trawlers. Such fleets need to target aggregations of high-value species in order to operate economically. For this reason we have presented distribution maps of orange roughy, toothfish and alfonsino, which are all relatively valuable species. Roundnose grenadier is of lesser value, but can occur in large quantities, and the North Atlantic region, where this species is most commonly found, is readily accessible to trawlers compared with the southern hemisphere oceans. Over the last decade, exploratory fishing for deep-water species in many areas beyond national jurisdiction has focused on alfonsino and orange roughy on seamounts. Toothfish have also been targeted, although this species occurs in areas under the management of the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR), and illegal, unreported and unregulated (IUU) fishing in waters of the Southern Ocean is the focus of major international preventative measures. Hence, we do not cover this species here. The two fisheries for alfonsino and orange roughy are, to an extent, discrete in that they operate at different depths on seamounts. Alfonsino fisheries: approximately 250-750 m. Commercially valuable by-catch species include black cardinalfish, southern boarfish, bluenose. Orange roughy fisheries: approximately 750-1250 m. Commercially valuable by-catch species include various oreos (black, smooth and sometimes spiky). This depth difference, although not clear-cut, can help when trying to evaluate seamounts that could be of commercial interest. Seamounts with a summit shallower than the species distribution may still have that species present down its slopes, |.e. at greater depth than the summit. Hence seamounts with summits shallower than 750 m can have orange roughy at 750 m and deeper down their flanks. However, although caution needs to be exercised, summit depth is a useful parameter to examine against the dis- tribution of seamounts in areas beyond national jurisdiction. The distribution of large ones with summit depths in the two depth ranges are shown in Figures 6.4 and 6.5. At alfonsino depths (250-750 ml], there are seamount chains in the central and eastern Pacific that are beyond areas of national jurisdiction, near the Challenger Fracture Zone and along the Sala y Gomez Ridge respectively {FAO area 87). Further areas with fishable seamounts are at the 51 Seamounts, deep-sea corals and fisheries hi } oom Figure 6.3: Distribution (top panel) and relative size (bottom panel) of major historical seamount fisheries. Circle size in the bottom panel is proportional to the total catch for that one-degree grid square, maximum is 85 000 tonnes. 50 001-100 000 @ 10 001-50 000 @ ae A | @ 1001-10 000 BYX See Table 6.1 for codes to the fish species. 5 ® 11-100 101-1 000 ORH OEO B’ ORH OEO BYX CDL 52 Seamount fish and fisheries Ye e - - e * - . - e “=° os D 4 3, “% ae a %., = % , se ° « C] es fea * aA * w 0 Ve £,.*% ay o . Nal Ly “4 s $ x2 ve % s° : z* "\ eT 5 A *? ye “ 5S a te . ee ap Ae yep es Sat Td 3 os ar * le ae ' pe oro . ‘ i i” fe | * Porm aif “Sa Ap e Ph > ° =: *e e bd e e ® ad e ‘ s e e ee we 3 nee - eye % e val Figure 6.4: Location of predicted large seamounts with summit depths between 250 m and 750 m, the main depth range for alfonsino fisheries. southwestern end of the Walvis Ridge and in the Gulf of Guinea [FAO area 47) in the South Atlantic; in the Indian Ocean along the Southwest Indian Ocean Ridge (FAO area 51) and near the Ninety East Ridge [FAO area 57); along the Emperor Seamount chain (FAO area 77) in the North Pacific; and south of the Azores in the North Atlantic (FAO areas 27, 34]. Most of these areas are thought to have been explored, or commercially exploited, already. At orange roughy depths (750-1200 mJ, there are seamounts in the South Pacific Ocean, along the Louisville Ridge (FAO area 81), and further east near the Challenger Fracture Zone and Sala y Gomez Ridge (FAO area 87). The Walvis Ridge, Atlantic-Indian Ridge, and southern end of the Mid-Atlantic Ridge all FAO area 47) also have seamounts at appropriate depths. In the Indian Ocean, areas of the South- west Indian Ridge, Ninety East Ridge, and Broken Ridge (FAO areas 51, 57) are also at orange roughy depths, but towards the northern limit of the species distribution. In the North Atlantic, there are features along the Mid-Atlantic Ridge from about 30°N northwards. Seamounts further south into the northern South Atlantic are getting outside the geographical distribution of orange roughy. It is difficult to determine which areas outside national jurisdiction have been extensively explored. The data sources used by Clark et al. [in press) are known to be incomplete, and FAO catch reporting is on a spatial scale that does not allow individual seamounts, clusters or chains to be identified. Clark et al. {in press} have determined that some of the areas of potential seamount fisheries have been searched in the late 1980s to 1990s and early 2000s. Where large-scale fisheries have not developed, it may be a sign that commercial concentrations of target species are not there. Alternatively, rough patches of sea floor are common on seamounts, and bottom trawling may have been unsuccessful due to gear damage or the bottom being too rough to even attempt trawling. Modern deep-water trawis have large bobbin or rock-hopper ground gear, and together with advances in navigational and electronic fishing aids since the 1980s, these have made trawling on rough seamounts much more feasible than 20 years ago (Roberts, 2002; Lack et al., 2003). Small seamounts and trawlable paths can routinely be located and fished. Nevertheless, Spectrunculus grandis, Davidson Seamount, 2 677 m depth, 60 cm long. (NOAA/MBARI) 53 Seamounts, deep-sea corals and fisheries Fig 6.5: Location of predicted large seamounts with summit depths between 750 m and 1 200 m, the main depth range for orange roughy fisheries. there are still some limitations on fishers’ ability to bottom trawl on seamounts. When clusters of seamounts occur, fish may not be distributed evenly between them, or may only be evident at certain times of the day or year, and so intensive trawling may be required to locate commercial quantities. Operating costs of large offshore vessels are relatively high, and if there are no signs of fish, the vessel may move on rather than continue to explore a small area. | Therefore, even where fishing has occurred, there may be | potential for small stocks of deep-water species to exist, and to support future exploratory fishing operations. ‘ The depth and geographical distribution of the alfonsino and orange roughy trawl fisheries overlap with the predicted distribution of large seamounts and deep-sea coral distribution. The next chapter will discuss the results of the previous chapters, and bring various sources of information together to evaluate the vulnerability of seamount benthic communities to deep-water fishing activities. 54 The vulnerability of stony corals on seamounts 7. Assessing the vulnerability of stony corals on seamounts RATIONALE Corals are a prominent component of the seamount fauna, which can be highly diverse and abundant, and may be associated with many species new to science. Deep-sea corals can form complex biological structures on the seabed and thus provide crucial habitat for a diversity of associated invertebrates and fish. Up to 100 000 large seamounts may exist in the world’s oceans, but the fauna of only a small fraction has been documented Commercial fishing has targeted numerous fish species on seamounts, and there is mounting concern over the damage that deep-sea trawling can cause to the benthic communities that live on them. The biology and life histories of deep-sea corals make them highly vulnerable to bottom trawling. Their destruction can potentially have knock-on effects for seamount ecosystems Many seamounts are located in areas beyond national jurisdiction, and are increasingly targeted by commercial fishing activities taking place on the high seas. In the light of concern about the impacts and ecological ramifications of fishing on seamount habitats and the biological commun- ities in these areas, countries and some stakeholders called on intergovernmental bodies to discuss and develop appro- priate multilateral action on a regional and/or global scale. The United Nations General Assembly (UNGA} has repeatedly addressed the issue (UN General Assembly, 2003, 2004a, 2004b, 2005a, 2005b, 2006). In the resolutions on oceans and the law of the sea’ and sustainable fisheries, the UNGA has called upon States and international organizations to urgently take action to address destructive practices that have adverse impacts on marine biodiversity and vulnerable ecosystems, and to consider the interim prohibition of such destructive fishing practices. Common to all of these calls was [i] the need to take action ona scientific basis, and (ii) the specific mentioning of seamounts and cold-water corals as examples of vulnerable marine biodiversity and ecosystems Protection of marine biodiversity in coastal areas within EEZs and particularly on the high seas has been weak [e.g. Royal Commission on Environmental Pollution, 2004), and only 0.5 per cent of the world’s marine environment is currently protected (Kimball, 2005]. However, there are general obligations in the 1982 United Nations Convention on the Law of the Sea (UNCLOS) to protect and preserve the marine environment and to conserve and manage high seas Atrawled seamount off Tasmania. (T Koslow, CSIRO Marine and Atmospheric Research] living resources (Kimball, 2005]. These obligations apply both within and beyond waters of national jurisdiction. The enforcement of international legal regimes on vessels is the responsibility of flag states. Obligations under UNCLOS are also implemented through regional agreements and, in the case of fisheries, through Regional Fisheries Management Organizations (RFMOs). UNCLOS (Kimball, 2005), regional agreements (e.g. OSPAR; see Johnston, 2004) and RFMOs have all emphasized the requirement to base conservation measures on the best scientific information available. This may be justified because of the risk of displacing harmful activities, such as deep-sea trawling, to as yet unexplored but potentially more sensitive habitats if decisions are made without sufficient scientific information (ICES, 2006) However, lack of scientific data should not be used as an excuse for inactivity and should also be balanced by the app- lication of the precautionary principal through ecosystem- based management practices (Vierros et al., 2006; WWF, 2006) The ecological importance of corals on seamounts has been clearly demonstrated through a growing body of 55 Seamounts, deep-sea corals and fisheries Figure 7.1: Main areas under risk from alfonsino seamount fisheries (250-750 m depth horizon). Above: Predicted habitat suitability for stony corals in 250-750 m depth. High percentage values indicate more suitable habitat. Upper, opposite page: Predicted seamount summit depths 250-750 m depth. Lower, opposite page: Seamounts with known historical alfonsino group catches. scientific evidence. Scientific investigations have also identified that these organisms and their associated bio- logical communities are highly vulnerable to fishing. To evaluate the vulnerability of seamounts to putative impacts by trawling, the distribution of coral habitat needs to be compared with that of seamount fisheries worldwide. However, corals have only been sampled from a small fraction of seamounts worldwide, whilst because of the rapid expansion of deep-sea fisheries, a global perspective on seamount conservation is required. Scientific surveys of seamount communities are extremely expensive and time- consuming and are unlikely in the short to medium term (tens of years] to identify the majority of seamount habitats that require protection from harmful activities. In the present report, a new approach to identifying the occurrence of marine habitats that are sensitive to particular activities — in this case fishing, primarily by deep-sea bottom trawling - has been adopted by scientists within the CenSeam programme. This approach was to use modelling - based on existing observations of the occurrence of stony corals — to predict where seamounts with favourable environmental conditions for the development of diverse coral communities 56 Key (above) Habitat suitability Habitat suitability % % ‘has 0-10 Mm 50-60 ne 10-20 0-70 Pal 20-30 Pa 70-80 | 30-40 & 80-90 al 40-50 HM ss0-100 are likely to occur. Combining this information with the known geographical occurrence of commercially valuable seamount fish species identifies which seamounts are in urgent need of measures to protect biodiversity. A note of caution here is that other types of corals, particularly octocorals, and other organisms, such as sponges, form diverse biological communities and have markedly different distributions from that of stony corals. Thus, whilst large areas of the North Pacific may be relatively unsuitable for stony corals, the area is suitable for octocorals, which form coral gardens with a high diversity of associated species. Octocoral gardens are as vulnerable to fishing activities as cold-water coral reefs formed by stony corals. OVERLAP BETWEEN STONY CORALS AND FISHERIES A key finding from the qualitative comparisons of the predicted global distribution of stony coral habitat on seamounts with the distribution of seamount fisheries is the considerable spatial overlap between the likely distribution of stony corals and past, current and potential future seamount fisheries. The predicted distribution of seamount habitat suitable eo oe : ee 2 3 ss ee “a : ¢* ; for stony corals (scleractinians) is extensive on a global scale. The majority of this suitable habitat is located in areas beyond national jurisdiction, mainly at depths between 250 m and 750 m. High levels of oxygen saturation and aragonite (a form of calcium carbonate used by corals to form hard skeletons) are among the most important environmental factors in determining habitat suitability for stony corals. Predicted habitat suitability indicates that seamounts provide coral habitat mainly in a band across all oceans between 20°S and 60°S, and in other areas of the Atlantic The vulnerability of stony corals on seamounts 5 “a a Py a, * aS 8 “ts “4 % t s b: af . Se Te Le fe ¥ st - ee . “ oo pe x * . ar ‘ eo? e e 2 ots, 4° s e ee be ad Ocean. In the Atlantic, a large proportion of suitable sea- mount coral habitat lies beyond areas of national juris- diction, whereas in the Pacific it lies mostly within national EEZs. In the southern Indian Ocean, suitable coral habitat on seamounts appears both within and outside of areas of national jurisdiction. Examinations of seamount fisheries information revealed that the main deep-sea fish species of commercial value have a widespread distribution, and for at least parts of their life history can be found associated with seamounts. The two fish species of highest commercial value that are 57 Seamounts, deep-sea corals and fisheries targeted on seamounts in areas beyond national jurisdiction are alfonsino and orange roughy. Fisheries for these two species are, to an extent, discrete in that they operate at different depths: the alfonsino fishery operates primarily between 250 m and 750 m, whilst the fishery for orange roughy occurs largely at water depths of 750-1 200 m. Throughout the world’s oceans, there are numerous large seamounts that a) have summits within the depth range of the fish and fisheries; b] are located outside of areas of national jurisdiction; and c] lie within the known or predicted distribution of alfonsino and orange roughy. Most of the areas where these seamounts occur are thought to have already been explored or commercially exploited, but, especially at orange roughy depths, there are seamounts in some areas that appear to be within the distributional and depth range of the species that may not yet have been the subject of extensive fishing. VULNERABILITY OF CORALS ON SEAMOUNTS TO BOTTOM TRAWLING Many long-lived epibenthic animals such as corals have an important structural role within sea floor communities, Figure 7.2: Main areas under risk from orange roughy seamount fisheries (750-1 250 m depth horizon). Below: Predicted habitat suitability for stony corals in 750-1 250 m depth. High percentage values indicate more Suitable habitat. Upper, opposite page: Predicted seamount summit depths between 750-1 250 m depth. Lower, opposite page: Seamounts with known historical orange roughy group catches. providing essential habitat for a large number of species (Rogers, 1999; Freiwald et al., 2004; Roberts et al., 2006). Consequently, the loss of such key structural species lowers survivorship and recolonization of the associated fauna, and has spawned analogies with forest clear-felling on land (e.g. Watling, 2005). Such comparisons stem principally from destructive fishing practices that are mostly in the form of bottom-contact trawling. A considerable body of evidence on the ecological impacts of trawling is available for shallow waters [e.g. Watling and Norse, 1998; Hall, 1999; Kaiser and de Groot, 2000), but scientific information on the effects of fishing on deep-sea seamount ecosystems is much more limited. The scientific literature of the effects of fishing on seamount habitat is summarized by Clark and Koslow [in press]. Their key findings include: 1. The impacts of trawling on seamounts have been studied most intensively within the EEZs of Australia and New Zealand [e.g. Koslow et al., 2001; Clark and O'Driscoll, 2003). 2. On seamounts off Tasmania (Australia), the fished seamounts had typically fewer species (reduced by about Key (below) Habitat suitability Habitat suitability % % 0-10 50-60 10-20 60-70 20-30 70-80 30-40 80-90 40-50 90-100 58 half] and had lower biomass of benthic invertebrates (by about seven times) (Koslow and Gowlett-Holmes, 1998; Koslow et al., 2001). 3. On New Zealand seamounts, the composition of larger benthic invertebrates was different on ‘fished’ seamounts, which had a smaller amount of coral habitat formed by live Solenosmilia variabilis and Madrepora oculata than on unfished’ seamounts. In addition, trawl marks were observed over six times more frequently on seabed images from ‘fished’ seamounts (Clark and O'Driscoll, 2003, Rowden et al., 2004) The vulnerability of stony corals on seamounts The intensity of trawling on seamounts can be very high. For example, Soviet fishing effort for pelagic armourhead on relatively few seamounts in the Southern Emperor and Northern Hawaiian Ridge system was around 18 000 trawler days during the period from 1969 to 1975 (Borets, 1975). Koslow et al. (2001) and Clark and O'Driscoll (2003) have reported that between several hundred and several thousand trawls have been carried out on small seamount features in the orange roughy fisheries around Australia and New Zealand Similarly, O'Driscoll and Clark (2005) documented that a ted Seon, ain . os et hd *\e ? Da ie %, - a % ais a ¥ ¢ ‘e Sore F e e < . oe “t *30 he Same o* 7 | ge s & : a bo =. %° ? ee* 4 > : ¥ ea - (cae Se ¢ oem “PY o, Ba er he RE ye Se eee » Hy ba “4 oo - gor ; ps .* ‘ 2g pe tape 76 gees ee * ° Lar f a 4 *. os Ff " : fe ; o? if 59 Seamounts, deep-sea corals and fisheries _ the total length of bottom tows per square kilometre of seamount area off New Zealand averages 130 km of trawled sea floor. Such intense fishing means that the same area of the sea floor can be repeatedly trawled, causing long-term damage to the coral communities and preventing any recovery or recolonization. The impact of trawling on sea floor biota can differ depending on the gear type used. Information about the potential impact of trawling practices for alfonsino, where mid-water trawls are often used on seamounts, is currently lacking. Mid-water trawls may have only a small impact if they are deployed well above the sea floor. However, in many cases the gear is most effective when fished very close to, or even lightly touching, the bottom. Thus, it is likely that the effects of the alfonsino fisheries on the benthic fauna would be similar to that of the orange roughy fisheries. WHERE ARE THE MAIN AREAS OF RISK AND CONCERN? The spatial extent of the likely vulnerability of seamount biodiversity on seamounts in areas beyond national jurisdiction can be gauged by combining the three sets of information (Figures 7.1 and 7.2) produced in this study: 1. the predicted global distribution of suitable habitat for stony (scleractinian) corals; 2. the location of predicted large seamounts with summits in depth ranges of the fishery for alfonsino (250 m- 750 m) and orange roughy (750 m-1 250 m); and 3. the distribution of the fishing activity on seamounts for these two species. The spatial overlaps highlight a broad band of the southern Atlantic, Pacific and Indian Oceans between about 30°S and 50°S where there are numerous seamounts at fishable depths, and high habitat suitability for corals at depths between 250 m and 750 m, and again (but somewhat narrower) between 750 m and 1 250 m depth. There are also some areas of overlap in the North Atlantic Ocean. This spatial concordance of fishable seamounts within the depth band of orange roughy suggests there could be further commercial exploration for orange roughy fisheries on seamounts in the central-eastern southern Indian Ocean (as evidenced by the Southwest Indian Ocean fisheries rush between 1998 and 2003], the southern portions of the Mid- Atlantic Ridge in the South Atlantic, and some regions of the southern-central Pacific Ocean. Importantly, since these areas also contain habitat suitable for stony coral, impacts on deep-water corals - and seamount ecosystems in general - are likely to arise in such a scenario. It is uncertain whether fisheries exploration will expand further. Often, fish aggregations are very localized, and given the large number of seamounts and smaller features in the oceans, they may be difficult to locate. Hence, there may be further fisheries 60 potential, but if stocks are small and localized, they may not currently be economic. Thus, this study has for the first time revealed the global scale of the likely vulnerability of stony {scleractinian) corals on seamounts - including habitat-forming species, and by proxy a diverse assemblage of other species - to the impacts of trawling on seamounts in areas beyond national jurisdiction. This report provides some of the best scientific evidence to date to support the need for management practices on the high seas to protect seamounts vulnerable to the adverse effects of deep-water fishing. 8. A Way Forward HOW CAN THE IMPACT OF FISHING ON SEAMOUNTS BE MANAGED IN AREAS BEYOND NATIONAL JURISDICTION? The Report of the Secretary-General on Oceans and the Law of the Sea (2003), Paragraph 183, states "..fisheries governance has focused its attention on reducing fishing efforts, improving compliance with and enforcement of conservation and management measures established by regional fisheries bodies.... The international community has yet to devote sufficient attention to the protection of vulnerable marine ecosystems from the adverse impacts of fishing and non- fishing activities, an important step towards fisheries conservation within an ecosystem-based management of capture fisheries.” Examples of vulnerable marine ecosystems in this document include seamounts (Report of the Secretary- General, 2003, Paragraph 180). In 2005 the Secretary- General published a further report detailing deep-sea ecosystems, threats to the marine environment and the legal framework associated with protecting the marine environment both within and beyond waters of national jurisdiction (Report of the Secretary General, 2005) This report has reviewed scientific evidence that where seamounts, deep-sea corals and fisheries come together, there is a need for management. It has also demonstrated that deep-sea corals, and by proxy benthic communities, on as yet unexplored/unfished seamounts in areas beyond national jurisdiction are at risk from the potential expansion of alfonsino and orange roughy fisheries. Consequently, it is sensible for appropriate management strategies to be in place prior to these fisheries being established, so as to prevent the adverse effects of fishing on these seamount ecosystems. Management initiatives for seamount fisheries within national EEZs have increased in recent years. Several countries, such as New Zealand and Australia, have closed seamounts to fisheries, established habitat exclusion areas and stipulated method restrictions, depth limits, individual seamount catch quotas and by-catch quotas [e.g., Smith, 2001; Commonwealth of Australia, 2002; Gianni, 2004; Gjerde, 2006; Brodie and Clark, 2004; Melo and Menezes, 2003). In comparison, fisheries beyond areas of national jurisdiction have often been entirely unregulated (FAO, 2004; A way forward Benthodytes sp. (sea cucumber), Davidson Seamount, 2 789 m. (NOAA/MBARI) Gianni, 2004; Gjerde, 2006). There are 12 Regional Fisheries Management Organizations [RFMOs] with responsibility to agree on binding measures that cover areas beyond national jurisdiction (Kimball, 2005}, including some of the geographical areas identified in this report that might see further expansion of exploratory fishing for alfonsino and orange roughy on seamounts. However, it should be noted that only the five RFMOs for the Southern Ocean (CCAMLR], Northwest Atlantic {NAFO), Northeast Atlantic (NEAFC], Southeast Atlantic (SEAFO) and the Mediterranean [(GFCM!) currently have the legal competence to manage most or all fisheries resources within their areas of application, including the management of deep-sea stocks beyond national jurisdiction (Kimball, 2005). The other RFMOs have competence only with respect to particular target species like tuna or salmon (Kimball, 2005). SEAFO covers parts of the eastern South Atlantic where exploratory fishing has occurred in recent decades, and where further trawling could occur. However, the western side of the South Atlantic is not similarly covered by an international management organization. There have been recent efforts to improve cooperative management of fisheries in the Indian Ocean, although there are no areas covered by an RMFO. In addition, efforts are underway, for example in the South Pacific, to establish a new regional fisheries convention and body that would fill a large gap in global fisheries management. 61 Seamounts, deep-sea corals and fisheries Farrea sp., a sponge that blankets large areas at or near crests on Davidson Seamount (1 400 m); associated with crabs, basket stars, seastars and brittle stars. (NOAA/MBARI) In light of the recent international dialogues concerning the conservation and sustainable manage- ment and use of biodiversity in areas beyond national jurisdiction held within and outside the United Nations system (Report of the Secretary-General, 2003 and 2005; CBD, 2004; Kimball, 2005), various fisheries bodies (e.g. NEAFC, NAFO, SEAFO) are more actively updating their mandates and including benthic protection measures as part of their fisheries management portfolio. Very recent initiatives include the formation of a Southwest Indian Ocean Fisheries Commission. There have also been recent proposals by industry to designate large voluntary Benthic Protection Areas [BPAs]. These are areas that are closed to bottom trawling primarily to protect the benthic fauna but also to preserve areas of outstanding scientific interest and potentially to act as a refuge for commercial fish species. In general, they have been proposed to give a wide representative coverage of geological structures, sediment overlays, bottom types and benthic habitat types. The New Zealand deep-water fishing industry has proposed BPAs mainly inside the New Zealand EEZ but some of which also encompass areas outside of the national EEZ. The Southern Indian Ocean Deepwater Fisheries Operators Association (SIODFOA) has also proposed a number of BPAs in the southern Indian Ocean. It appears that a growing legislation and policy framework, including an expanding RFMO network, particularly in the southern hemisphere, could enable the adequate protection of and management of the risks to vulnerable seamount ecosystems and resources identified 62 in this report. In order to be successful, a number of challenges will have to be overcome, including: 1. Establishing adequate data reporting requirements for commercial fishing fleets. Some unregulated and unreported fishing activities take place, even in areas where there are well-defined fishery codes of practice and allowable catch limits [e.g. Patagonian toothfish fishery]. Some countries require vessels registered to them to report detailed catch and effort data, but many do not. Therefore it is difficult at times to know where certain landings have been taken. 2. Ensuring compliance with measures, especially in areas that are far offshore and where vessels are difficult to detect. Compliance monitoring is also an acute problem in southern hemisphere high seas areas, where there are no quotas for offshore fisheries 3. Facilitating RFMOs, where necessary, to undertake ecosystem-based management of fisheries on the high seas. 4. Establishing, where appropriate, dialogue to ensure free exchange of information between RFMOs, governments, conservation bodies, the fishing industry and scientists working on benthic ecosystems. The experiences gained by countries in the protection of seamount environments in their EEZs and in the management of their national deep-water fisheries can provide useful case examples for the approach to be taken under RFMOs. Other regional bodies, such as Regional Sea Conventions and Action Plans, might be able to provide lessons learned from regional cooperation to conserve, protect and use coastal marine ecosystems and resources sustainably, including the implementation of an ecosystem approach in oceans management and the establishment of marine protected areas [MPAs] (Johnston and Santillo, 2004). Regional Sea Conventions and Action Plans also provide a framework for raising awareness of coral habitats in deep water areas under national jurisdiction, and coordinating and supporting the efforts of individual countries to conserve and manage these ecosystems and resources sustainably (e.g. ICES, 2005, 2006). In calling for urgent action to address the impact of destructive fishing practices on vulnerable marine ecosystems, Paragraph 66 of UN General Assembly Resolution 59/25 (UN General Assembly, 2005b) places a strong emphasis on the need to consider the question of bottom-trawl fishing on seamounts and other vulnerable marine ecosystems on a scientific and precautionary basis, consistent with international law. In this regard, it is important to recognize the role of science and the extent that scientific information, or lack thereof, is a prerequisite for management action The UN Fish Stocks Agreement (FSA) Articles 5 and 6 - ‘General principles’ and the Application of the precautionary approach (Kimball, 2005) also establish clear obligations for fisheries conservation and the protection of marine biodiversity and the marine environment from destructive fishing practices. The Articles also establish that the use of science is essential to meeting these objectives and obligations. Article 5{k) calls on States to promote and conduct scientific research in support of fishery conservation and management, and Article 6.3{a] requires States to improve decision making by obtaining and sharing the best scientific information available and implementing improved tech- niques for dealing with risk and uncertainty. Article 5{d} calls on States to assess the impacts of fishing on target stocks and species belonging to the same ecosystem, or those associated with or dependent upon the target stocks. And Article 6.3(d) calls for the development of data collection and research programmes to assess the impact of fishing on non-target and associated or dependent species and their environment, and for adopting plans necessary to ensure the conservation of such species and to protect habitats of special concern (Kimball, 2005) At the same time, the FSA recognizes that scientific understanding may not be complete or comprehensive, and in such circumstances, caution must be exercised. Articles 6.2 and 6.3{c] require taking into account uncertainties relating to the impact of fishing activities on non-target and associated or dependent species - that States be more cautious when information is uncertain, unreliable or inadequate. The absence of adequate scientific information shall not be used as a reason for postponing or failing to take conservation and management measures. A precautionary approach, consistent with the general principles for fisheries conservation contained in the FSA, as well as the UN FAO Code of Conduct for Responsible Fisheries and the principles and obligations for biodiversity conservation in the Convention on Biological Diversity (Kimball, 2005), would require the exercise of considerable caution in relation to permitting or regulating bottom-trawl fishing on the high seas on seamounts. This is because of the widespread distribution of stony corals and associated assemblages on seamounts in many high seas regions, and the likelihood that seamounts at fishable depths may also contain other species vulnerable to deep-sea bottom trawling even in the absence of stony corals. In this regard, a prudent approach to the management of bottom-trawl fisheries on seamounts on the high seas would be to first ascertain whether vulnerable species and ecosystems are A way forward associated with a particular area of seamounts of potential interest for fishing, and only then permitting well-regulated fishing activity provided that no vulnerable ecosystems would be adversely impacted FURTHER AND IMPROVED SEAMOUNT RESEARCH The conclusions of this report apply only to the association of stony corals with large seamounts. In order to consider other taxonomic groups on a wider range of seamounts, further sampling and research is required Development, implementation and review of effective management measures rely on sound scientific data and assessments. As already acknowledged in Principle 15 of the Rio Declaration on Environment and Development (Agenda 21], gaps in information and knowledge often cause a lack of full scientific certainty, and a precautionary approach has to be applied to protect the environment from threats of serious or irreversible damage and to prevent environmental degradation. UN General Resolutions 59/24 (Paragraph 81] and 60/30 (Paragraph 85] [UN General Assembly, 2005a, 2006) call for scientific research to: ../Mprove understanding and knowledge of the deep sea, including, in particular, the extent and vulnerability of deep-sea biodiversity and ecosystems..." The preparation of this report has identified a number of shortcomings and gaps in the data and in our knowledge of seamounts, deep-sea corals and fisheries. These gaps need to be addressed and closed in order to answer questions from policy makers, managers and scientists - answers that at present cannot be provided at the required level of certainty. Anthomastus sp. (mushroom soft coral), Davidson Seamount, 1 580 m. (NOAA/MBARI) 63 Seamounts, deep-sea corals and fisheries These include: 1. 64 Obtain better seamount location information: The two most recent seamount position datasets, based on satellite altimetry measures, both contain location information for about 15 000 predicted large seamounts. This number is thought to be an underestimate, with extrapolative techniques predicting the global seamount number to be 100 000. Fisheries often operate on much smaller seamounts, but such seamounts cannot be identified by large-scale remote sensing methods. However, it will be possible, with more extensive satellite measurements of the Earth's ocean surface with improved altimetry technology [to reduce loss of signal by wave ‘noise’] and closer spacing of satellite tracks, to greatly improve location data for large seamounts. Address geographic data gaps: Fewer than 300 seamounts have been biologically surveyed worldwide, which represents a very small {less than 2 per cent] fraction of existing seamounts in the world’s oceans. Only 80 of these seamounts have had at least a moderate level of sampling, and far fewer have received sampling sufficient to characterize the biological communities present. Thus, the fauna on the vast majority of seamounts remains unknown. Past surveys have tended to concentrate on a few geographic areas (e.g. North Atlantic, Southwest Pacific], while few data exist for seamounts in other regions such as the Indian Ocean and the Southern Ocean. Although seamounts are particularly common in the tropics, existing data come mostly from temperate regions at higher latitudes, and therefore the biological communities of tropical seamounts remain poorly documented for large parts of the oceans. Most biological surveys on seamounts have been relatively shallow (e.g. mostly less than 1 500 ml, and thus the great majority of deeper seamounts remains largely unexplored. Field programmes are required to address these deficiencies. Inclusion of other deep-sea habitats: To assess to what degree seamounts present ‘unique’ ecosystems, comparative data are required from other deep-sea environments such as the abyssal plains surrounding seamounts, and direct comparisons with slope environments - particularly island slopes and continental margins. Thus, field programmes should target both seamounts and such comparative environments whenever possible. Assessment of the spatial scale of variability: The distribution of deep-sea corals and other benthic invertebrate fauna on seamounts is likely to be patchy at a range of spatial scales — for example, on a seamount, and within and between seamounts on different clusters and chains. Very few individual seamounts have been comprehensively surveyed to determine the variability of faunal assemblages within a single seamount, where, for example, small-scale differences may occur between hard and soft substrates. It is important to understand the spatial scales at which variation in fauna community composition occurs, in order to develop management strategies that ensure the effective protection of this level of biodiversity and associated ecosystem function. Availability of data: For many seamount studies, only summary data are publicly available. Analysis of species distribution patterns and studies on assemblage composition across different seamounts and regions does, however, require access to species catch data for individual stations and/or samples {i.e. non-aggregated data]. In addition, many seamount studies are contained in the ‘grey literature’ and not always readily accessible. Increased accessibility of full (non-aggregated) datasets from seamount expeditions [after an appropriate time to publish] through searchable, integrated databases like SeamountsOnline and the Ocean Biogeography Information System (OBIS) is required. Collection methods: While different gear types are required to sample different types of faunal assembl- ages (e.g. otter trawls for fish, benthic sleds and dredges for macro-invertebrates], past studies have also used different gear types for the same faunal group. Since different collecting gears have different performances, often compounded by differences in deployment techniques and operations, direct comparisons of data may be confounded to some (unknown) degree. A minimum set of standardized seamount sampling protocols should be adopted as widely as possible by seamount sampling programmes. Taxonomic resolution: Different taxonomists (scientists who classify living things) or different groups of taxonomists often work on collections from different seamount studies. In fact, much of past and current seamount research relies fundamentally on the availability of specialized taxonomic expertise, a critical resource that continues to decline globally. Datasets may need careful taxonomic intercalibration before regional and global analysis can be undertaken with confidence. Similarly, for some faunal groups, few taxonomic specialists are available, often limiting the scope of analysis. More funding for existing taxonomic experts and training of new taxonomists — particularly . Recovery from trawling impacts for faunal groups that are currently poorly analysed globally — is required. This provision should also enable the research community to analyse specimens collected across multiple seamounts in multiple programmes Increase genetic studies: One of the critical questions for seamount conservation is whether they support isolated populations and, if so, on what scale that isolation occurs. Genetic studies can inform, for example, whether a single seamount is an appropriate scale for protection, or whether multiple seamounts ina chain have connected populations and should be protected Assessment of trawling impacts: Better studies on the impacts of trawling are needed. Studies to date on seamounts and in the deep sea have been limited. More and improved studies would improve our understanding of the extent to which the large fauna associated with corals and other structure-forming organisms are impacted. Studies should also investigate the nature of impact from different gear types, so that fishing gear can be optimized to reduce damage to the benthic fauna, while still catching fish effectively, Bottom fishing undoubtedly has severe impacts on seamount biota, particularly corals. The physical destruction caused by bottom-contact fishing gear is clearly visible on the seabed, and the removal of corals has significant consequences for the biodiversity and biomass of the associated fauna. It is, however, not known how long these communities take to recover from fishing impacts and what the trajectory of any such recovery may be Based on the slow growth and longevity of deep- sea corals, recovery of corals is predicted to be extremely slow, but is essentially unknown for field situations. However, such information on the time and nature of recovery [if any) is essential for ecosystem- based fisheries management on seamounts, and for evaluating the efficiency of MPAs on seamounts. Thus it is essential that the time frames and nature of recovery be documented. . Functional understanding: Our understanding of seamount biota has improved over the last few decades, but many of these advances have been made in documenting structural properties of seamount communities {e.g. species composition, distribution, growth rates, etc.). By contrast, much less is known about the processes operating in seamount ecosystems and how functional aspects of seamount assemblages A way forward Neolithodes, Davidson Seamount, 1 319 m. |NOAA/MBARI) may be altered by human activities. Therefore, future research should include aspects of community and ecosystem processes such as > food-web architecture on and above seamounts; 9 linkages of the bottom fauna with water-column and geological processes; > mechanisms and rates of recruitment (addition of organisms through reproduction or immigration] to seamount communities [e.g. larval dispersion, retention, oceanographic drivers of recruitment variability, etc.); >) processes promoting increased primary and secondary production on seamount and coupling to sea floor communities; > trophic (food-chain) links between seamount- associated fish and prey populations; and 9 the relative role of corals and other structure- forming fauna in promoting biodiversity and providing essential habitat for fish. 12. Fisheries information: At present, data collection from fishing vessels operating in areas beyond national jurisdiction is largely ad hoc, and FAO records also appear incomplete for many offshore fisheries. It is important for effective management of such fisheries to obtain accurate information on what is being caught, how much, and where. With seamount fisheries, this requires location data on a small-scale [individual tow data, recorded to at least a 1 minute of a degree accuracy], so that fishing on individual seamounts can be identified Without a concerted effort by a number of organizations, institutions, consortia and individuals to attend to the identified gaps in data and understanding, the ability of any body to effectively and responsibly manage and mitigate the impact of fishing on seamount ecosystems will be severely constrained. Considering what this report has revealed about the vulnerability of seamount biota — particularly deep-sea corals - to fishing, now is the time for this collaborative effort to begin in earnest 65 Seamounts, deep-sea corals and fisheries CCAMLR CoML DAWG DSL EEZ ENFA GLODAP ETOPO2 ERS] FAO GEBCO GFCM GLM GLODAP ICES |IHO OC NAFO NEAFC OBIS RFMO SAUP SEAFO SODA SWIOFC NOAA UN UNEP UNGA VGPM WOA WOCE 66 Acronyms Commission for the Conservation of Antarctic Marine Living Resources Census of Marine Life Data Analysis Working Group Deep Scattering Layer Exclusive Economic Zone Environmental Niche Factor Analysis Global Ocean Data Analysis Project Used to describe a 2-minute global bathymetry grid generated from a combination of sources including satellite altimetry observation and shipboard echo-sounding measurements European Remote-Sensing Satellite-1 Food and Agriculture Organization of the United Nations General Bathymetric Chart of the Oceans General Fisheries Commission for the Mediterranean Generalised Linear Model Global Ocean Data Analysis Project International Council for the Exploration of the Sea International Hydrographic Organization International Oceanographic Commission Northwest Atlantic Fisheries Organization Northeast Atlantic Fisheries Commission Ocean Biogeographic Information System Regional Fisheries Management Organizations Sea Around Us Project Southeast Atlantic Fisheries Organization Simple Ocean Data Assimilation Southwest Indian Ocean Fisheries Commission National Oceanic and Atmospheric Administration (USA United Nations United Nations Environment Programme United Nations General Assembly Vertically Generalized Production Model World Ocean Atlas World Ocean Circulation Experiment Glossary Algae: a group of plants [i.e. capable of photosynthesis] that occur in aquatic habitats, or in moist environments on land. Anthozoa: A class of animals within the Cnidaria that contains the corals and anemones. Antipatharia: An order within the Anthozoa (sub-class Hexacorallia), the so-called black corals. Aragonite: A form of calcium carbonate used by scleractinian corals to build their skeletons. Ascidians: a class of animals (Ascidiacea], the sea squirts. Azooxanthellate: without Zooxanthellae. Beam trawl: A trawl in which the horizontal opening is maintained by a wood or metal beam. Benthic: Related to the sea floor, includes fauna and flora that live on or in the seabed. Biodiversity: {1} The number and variety of organisms found within a specified geographic region; (2) The variability among living organisms including within and between species and within and between ecosystems. Biota: The plant and animal life of a region. Bottom trawling: Method of trawling where the net remains in contact with the sea floor - can comprise multiple nets i.e. twin-rigged trawls. Chlorophylls: A group of green pigments found in photosynthetic organisms including phyto- plankton that absorb energy from sunlight. Cnidaria: Phylum of more-or-less radially symmetrical invertebrate animals that lack a true body cavity, possess tentacles studded with nema- tocysts [stinging structures), and include the hydroids, Jjellyfishes, sea anemones and corals. Synonomous with the Coelenterates. Coelenterates: See Cnidaria. Corals: A group of benthic anthozoans that can exist as individuals or in colonies and may secrete calcium carbonate external skeletons. Corals can be found in the photic zone (with symbiotic zooxanthellae) as well as in the deep sea, the so called cold-water corals. Crinoid: Marine animals that make up the class Crinoidea (phylum Echinodermata). Also known as sea lilies’ or ‘feather-stars’. Deep scattering layer: A relatively thin layer of organisms, composed of migrating plankton forms, which can be detected by echo sounders. Detritivores: Scavengers that feed on dead plants and animals or their waste. Dissolved inorganic carbon (DIC): All inorganic carbon dissolved in a volume of water at a given Glossary temperature and pressure. Diversity: (1] The number of taxa in a group or place (species richness) (2) a parameter used to describe richness and evenness within a collection of species. Echinoderms: A phylum of marine animals found at all depths (from the Greek for spiny skin) Exclusive economic zone (EEZ): 1) A zone under national jurisdiction [up to 200-nautical miles wide) declared in line with the provisions of the 1982 United Nations Convention of the Law of the Sea, within which the coastal State has the right to explore and exploit, and the responsibility to conserve and manage, the living and non-living resources; 2) The area adjacent to a coastal state which encompasses all waters between: [a] the seaward boundary of that state, (b) a line on which each point is 200 nautical miles (370.40 km) from the baseline from which the territorial sea of the coastal state is measured (except when other international boundaries need to be accommodated], and [c) the maritime boundaries agreed between that state and the neighbouring states. Endemic: A taxa that is restricted in its distribution, only found in a specific area/region. Environmental Niche Factor Analysis (ENFA): A habitat suitability modelling technique. Epipelagic: The part of the oceanic zone into which enough sunlight enters for photosynthesis to take place. See also euphotic/photic. Epibenthic: Living on the bottom or sea floor Euphotic: The part of the oceanic zone into which enough sunlight enters for photosynthesis to take place. See also epipelagic/photic. Fauna: Animals, especially those of a particular region, considered as a group. GLM: Generalised Linear Model. A statistical linear model that can relate one dependent factor to one or more independent factors. Gorgonacea: An order within the Anthozoa characterized by having a flexible, often branching skeleton of horny material. Guyot: Flat topped seamount which is often covered in sediments from when they were exposed islands. Habitat: The area or environment where an organism or ecological community normally lives or occurs. Hexacorals: A subclass of the Anthozoans. Includes the Antipatharia and Scleractinia. High seas: denotes [in municipal and international law) all 67 Seamounts, deep-sea corals and fisheries of that continuous body of salt water in the world that is navigable in its character and that lies outside of the territorial waters and maritime belts of the various countries [also called open seas). Hydrozoa (hydroids): A class within the phylum Cnidaria. Marginality: An ENFA term indicating how different the optimal habitat for a taxonomic group is from the mean environment. Mid-water trawling: Method of trawling where the net is towed through mid-water i.e. above, and not in contact with the sea floor. Modelling: Representing a system through mathematical ; or statistical equations. Niche: The role an organism fills in an ecosystem. Octocorals: A sub-class of corals within the Anthozoa which are characterized by having eight tentacles on each polyp. Otter trawl: A trawl in which the horizontal opening is maintained by a pair of trawl doors [or otter boards). Pelagic: Of relating to or living in the open sea, away from the sea bottom. Photic: A zone in the water column that is penetrated by sufficient sunlight for primary productivity/ production. Photosynthesis: The process by which carbohydrates are synthesized from carbon dioxide and water using light as an energy source. Most forms of photosynthesis release oxygen as a byproduct. Plankton: Minute pelagic organisms that float or drift in great numbers in fresh or salt water, especially at or near the surface, and serve as food for fish and other larger organisms. Polyp: A single individual of a colony or a solitary attached cnidarian. Primary productivity/production: The rate of carbon fixation by phytoplankton [marine photosynthetic organisms). Seamount: An elevation of the seabed with a summit of limited extent that does not reach the surface. They can have a variety of shapes but are generally conical with a circular, elliptical or elongate base, and do not breach the surface. 68 There is no unified consensus of what does or does not constitute a seamount. Some definitions are based on elevation e.g. must be greater than 1 000 m whilst others will class a seamount as a topographic feature that rises more than 50 m above the sea floor. Scleractinia: An order within the Anthozoa [sub-class Hexacorallia), the so called stony corals. Specialization: An ENFA term indicating how stringent are the environmental requirements of a taxonomic group (how narrow a niche it occupies). Sponge: A phylum (Porifera] of sessile (attached) forms that are spongy or stony to the touch. No obvious animal features and often mistaken for a plant. Stylasteridae: A family of corals within the class hydrozoa. Taxonomy: The science of classifying living things e.g. Phylum, Class, Order, Family, Genus, Species. Taylor column: Models predict that the steady flow of a uniform water column past a seamount results in a stationary vortex over the seamount, a so- called a Taylor column. However, stratification of water layers above a seamount may reduce the column to a cap - a Taylor cap. Trawl: Trawls are nets consisting of a cone-shaped body closed by a bag or cod end and extended at the opening by wings. They are actively pulled through the water and kept open in the vertical plane by various methods e.g. floats, and on the horizontal plane e.g. by trawl doors. They can be towed by 1 or 2 boats and according to type, are used on the bottom (demersal) or mid-water (pelagic). Trophic: Of, or involving, the feeding habits or food relationship of different organisms in a food chain. Zooxanthellae: Algae that live symbiotically within the cells of other organisms e.g. corals in the photic zone. Zooanthid: An order of anemone like hexacorals which have a colonial lifestyle. Zooplankton: General term for the animal component of the plankton. in aquatic habitats, or in moist environments on land. References Anderson OF, Clark MR (2003) Analysis of bycatch in the fishery for orange roughy Hoplostethus atlanticus, on the South Tasman Rise. Marine and Freshwater Research, 54 (3), pp. 635-652. Anderson OF, Clark MR (2003) Analysis of bycatch in the fishery for orange roughy Hoplostethus atlanticus, on the South Tasman Rise. Marine and Freshwater Research, 54 (3), pp. 635-652. Andrews AH, Cordes EE, Mahoney MM, Munk K, Coale KH, Cailliet GM, Heifetz J (2002). Age, growth and radiometric age validation of a deep-sea, habitat-forming gorgonian (Primnoa resedae- formis) from the Gulf of Alaska. Hydrobiologia, 471, pp. 101-110. Auster PJ, Moore J, Heinonen KB, Watling L (2005). A habitat classification scheme for seamount landscapes: assessing the functional role of deep-water corals as fish habitat. In: Freiwald A, Roberts JM leds) Cold-Water Corals and Eco- systems. Springer-Verlag, Berlin, Heidelberg, pp. 761-769. Behrenfeld MJ, Falkowski P (1997). Photosynthetic rates derived from _ satellite-based chlorophyll concentrations. Limnology and oceanography, 24, pp. 1-20. Bell TH (1975). Topographically generated internal waves in the open ocean. Journal of Geophysical Research, 80, pp. 320-327. Boland RC, Parrish FA (2005). A description of fish assemblages in the black coral beds of Lahaina, Maui, Hawai'i. Pacific Science, 59, pp. 411-420. Bonilla HR, Pihén GC (2002). Influence of temperature and nutrients on species richness ofdeep water corals from the western coast of the Americas. Hydrobiologia, 471, pp. 35-41. Borets LA (1975). Some results of studies on the biology of the pelagic armourhead ([Pentaceros richardsoni Smith). Investigations of the biology of fishes and fishery oceanography, TINRO, Vladivostok, pp. 82-90 (in Russian). Boyce MS, Vernier PR, Nielsen SE, Schmiegelow FKA (2002). Evaluating resource selection functions. Ecological Modelling, 157, pp. 281-300. Brink KH (1989). The effect of stratification on seamount- trapped waves. Deep-Sea Research, 36, pp. 825- 844. Brodie S, Clark MR (2004). The New Zealand seamount management strategy - steps towards con- serving offshore marine habitat. In: Aquatic Protected Areas: what works best and how do References we know? Proceedings of the World Congress on Aquatic Protected Areas {eds JP Beumer, A Grant, DC Smith], Australian Society for Fish Biology, pp. 664-673, Cairns 2002, Australia. Brotons L, Thuiller W, Araujo MB, Hirzel AH (2004). Presence-absence presence-only modelling methods for predicting bird habitat suitability. Ecography, 27, pp. 437-448. Buhl-Mortensen L, Mortensen PB (2004). Symbiosis in Deep-Water Corals. Symbiosis, 37, pp. 33-61. Buhl-Mortensen L, Mortensen PB (2005). Distribution and diversity of species associated with deep- sea gorgonian corals off Atlantic Canada. In: Cold-water Corals and Ecosystems leds A Freiwald, JM Roberts], Springer Publishing House, pp. 849 - 879, Heidelberg, Germany. Cailliet GM, Andrews AH, Burton E.J, Watters DL, Kline DE, Ferry-Graham LA (2001). Age determination and validation studies of marine fishes: do deep- dwellers live longer? Experimental Gerontology, 36, pp. 739-764. Carton JA, Chepurin G, Cao X, Giese B (2000). A simple ocean data assimilation analysis of the global upper ocean 1950-1995. Part 1: Methodology. Journal of Physical Oceanography, 30, pp. 294- 309. CBD (2004). Convention on Biological Diversity. Decisions adopted by the conference of the parties to the Convention on Biological Diversity at its seventh meeting. Decision VII/5 Marine and coastal biological diversity. Paragraphs 57-62, pp. 140- 141. Conservation and sustainable use of biological diversity in marine areas beyond the limits of national jurisdiction. Available at: www. biodiv.org/decisions/default.aspx?m=COP.0 7&id=7742&lg=0 Chapman DC (1989). Enhanced subinertial diurnal tides over isolated topographic features. Deep-Sea Research, 36, pp. 815-824. Cheney RE, Richardson PL, Nagasaka K (1980). Tracking a Kuroship cold ring with a free-drifting surface bouy. Deep-Sea Research, 27, pp. 641-654. Clark MR (1999). Fisheries for orange roughy (Hoplostethus atlanticus) on seamounts in New Zealand. Oceanologica Acta, 22, pp. 593-602. Clark MR (2001). Are deepwater fisheries sustainable? - the example of orange roughy (Hoplostethus atlanticus) in New Zealand. Fisheries Research, 51, pp. 123-135. Clark MR, Anderson OF, Francis RICC, Tracey DM (2000). versus 69 Seamounts, deep-sea corals and fisheries The effects of commercial exploitation on orange roughy [(Hoplostethus atlanticus) from the continental slope of the Chatham Rise, New Zealand, from 1979 to 1997. Fisheries Research, 45, pp. 217-238. Clark MR, Koslow JA [in press). Impacts of fishing on seamounts. In: Seamounts: Ecology Fisheries and Conservation {eds TJ Pitcher, PJB Hart, T Morato, R Santos, M Clark], Blackwell Fisheries and Aquatic Resources Series, Blackwell Scientific. Clark MR, O'Driscoll R (2003). Deepwater fisheries and aspects of their impact on seamount habitat in New Zealand. Journal of Northwest Atlantic Fishery Science, 31, pp. 441 - 458 Clark MR, O'Shea S, Tracey D, Glasby B (1999) New Zealand region seamounts. Aspects of their biology, ecology and fisheries. Report prepared for the Department of Conservation, Wellington, New Zealand, August 1999. 107 pp. Clark, MR, Vinnichenko VI, Gordon JDM, Kukharev NN, Kakora, AF (in press). Large scale distant water trawl fisheries on seamounts. Chapter 17. In: Seamounts: Ecology Fisheries and Conservation, (eds TJ. Pitcher, PJB Hart, T Morato, R Santos, M Clark], Blackwell Fisheries and Aquatic resources Series, Blackwell Scientific. Commonwealth of Australia (2002) Tasmanian Seamounts Marine Reserve Masnagement Plan. Environment Australia, Canberra, Australia. ISBN 0642 547 742. 54 pp. Conkright ME, Locarnini RA, Garcia HE, O’Brien TD, Boyer TP, Stephens C (2002). World Ocean Atlas 2001: Objective analyses, data statistics, and figures. CD-ROM documentation. Rep. IR-17. National Oceanography Data Centre, 17, Silver Spring, Md. Dower J, Freeland H, Juniper K (1992). A strong biological response to oceanic flow past Cobb seamount. Deep-Sea Research, 39, pp. 1139-1145. Druffel ERM, Griffin S, Witter A, Nelson E, Southon J, Kashgarian M, Vogel J (1995) Gerardia: bristlecone pine of the deep-sea? Geochimica et Cosmochimica Acta, 59, pp. 5031-5036. Eckelbarger KJ, Watling L, Fournier H (2005). Reproductive biology of the deep-sea polychaete Gorgoniapolynoe caeciliae ([Polynoidae], a commensal species associated with octocorals. Journal of the Marine Biological Association of the United Kingdom, 85, pp. 1425 - 1433. Epp D, Smoot NC (1989). Distribution of seamounts in the North Atlantic. Nature, 337, pp. 254-257. Eriksen CC (1982a). Observations of internal wave 70 reflection off sloping bottoms. Journal of Geophysical Research, 87, pp. 525-538. Eriksen CC (1982b). An upper ocean moored current and density profiler applied to winter conditions near Bermuda. Journal of Geophysical Research, 87, pp. 7879-7902. Eriksen CC (1985). Implications of ocean bottom reflection for internal wave spectra and mixing. Journal of Physical Oceanography, 15, pp. 1145-1156. Eriksen CC (1991). Observations of amplified flows atop a large seamount. Journal of Geophysical Research, 96, pp. 19,227-15,236. FAO (2002). Report of the second Ad-Hoc meeting on management of deepwater fisheries resources of the Southern Indian Ocean. FAO Fish Report, No. 677, 106 pp FAO (2004). The State of World Fisheries and Aquaculture. Food and Agricultural Organization of the United Nations, Rome, Italy. 153 pp. FAO (2005). Review of the state of world marine fishery resources. FAO Fisheries Technical Paper, 457, 235 pp. Fock H, Uiblein F, Koster F, Westernhagen H (2002). Biodiversity and species-environment relation- ships of the demersal fish assemblage at the Great Meteor Seamount (subtropical NE Atlantic], sampled by different trawls. Marine Biology, 141, pp. 185-199. Fossa J, Mortensen P, Furevik D (2002). The deep-water coral Lophelia pertusa in Norwegian waters: distribution and fishery impacts. Hydrobiologia, 471, pp. 1-12. Freiwald A, Huhnerbach V, Lindberg B, Wilson JB, Campbell J (2002). The Sula Reef Complex, Norwegian shelf. Facies, 47, pp. 179 - 200. Freiwald A, Fossa JH, Grehan A, Koslow JA, Roberts JM (2004). Cold-water coral reefs: out of sight-no longer out of mind. UNEP-WCMC.Cambridge, UK. 84 pp. Froese R, Sampang A (2004). Taxonomy and biology of seamount fishes. In: Seamounts: Biodiversity and Fisheries (eds T Morato, D Pauly}. Fisheries Centre Research Report, 12, pp. 25-31. Gad G (2004a). The Loricifera fauna of the plateau of the Great Meteor Seamount. Archive of Fishery and Marine Research, 51 (1-3), pp. 9-29. Gad G (2004b). Diversity and assumed origin of the Epsilonematidae (Nematoda) of the plateau of the Great Meteor Seamount. Archive of Fishery and Marine Research, 51 (1-3), pp. 30-42. Gad G, Schminke HK (2004). How important are seamounts for the dispersal of interstitial meiofauna? Archive of Fishery and Marine Research, 51 (1-3), pp. 43-54. Gaston KJ (2003). The structure and dynamics of geographic ranges. Oxford University Press, Oxford, UK. pp. 276. George KH, Schminke HK (2002). Harpacticoida (Crustacea, Copepoda) of the Great Meteor Seamount, with first conclusions as to the origin of the plateau fauna. Marine Biology, 144, pp. 887-895. Genin A (2004). Bio-physical coupling in the formation of zooplankton and fish aggregations over abrupt topographies. Journal of Marine Systems, 50, pp. 3-20. Genin A, Dayton PK, Lonsdale PF, Spiess FN (1986). Corals on seamount peaks provide evidence of current acceleration over deep-sea topography. Nature, 322, pp. 59-61. Genin A, Noble M, Lonsdale PF (1989). Tidal currents and anticyclonic motions on two North Pacific seamounts. Deep-Sea Research, 36, pp. 1803- 1815. Gianni M (2004). High Seas bottom trawl fisheries and their impacts on the biodiversity of vulnerable deep-sea ecosystems: options for international action. UCN, Gland, Switzerland. 83 pp. Gjerde KM (2006). Ecosystems and biodiversity in deep waters and high seas. UNEP Regional Seas Report and Studies, No. 178. 58 pp. Guinotte JM, Orr J, Cairns S, Freiwald A, Morgan L, George R (2006). Will human-induced changes in seawater chemistry alter the distribution of deep-sea scleractinian corals? Frontiers in ecology and the environment, 1, pp. 141-146. Guisan A, Zimmermann NE (2000). Predictive habitat distribution models in ecology. Ecological Modeling, 135, pp. 147-186. Guisan A, Thuiller W (2005). Predicting species distribution: offering more than simple habitat models. Ecology Letters, 8, pp. 993-1009. Hall SJ (1999). The effects of fishing on marine ecosystems and communities. Blackwell Scientific, Oxford, U.K. 274 pp. Hall-Spencer J, Allain V, Fossa JH (2002). Trawling damage to Northeast Atlantic ancient coral reefs. Proceedings of the Royal Society of London, Series B: Biological Sciences, 269, pp. 507-511. Haury L, Fey C, Newland C, Genin A (2000). Zooplankton distribution around four eastern North Pacific seamounts. Progress in Oceanography, 45, pp. 69- 105. References Helly JJ, Levin LA (2004). Global distribution of naturally occurring marine hypoxia on continental margins. Deep-Sea Research /, 51, pp. 1159- 1168. Hirzel AH, Arlettaz R (2003). Modeling habitat suitability for complex species distributions by environmental-distance geometric mean Environmental Management, 32, pp. 614-623. Hirzel AH, Hausser J, Chessel D, Perrin N (2002). Ecological-niche factor analysis: how to compute habitat-suitability maps without absence data? Ecology, 83, pp. 2027-2036. Hirzel AH, Helfer V, Metral F (2001). Assessing habitat- suitability models with a virtual species. Ecological Modelling, 145, pp. 111-121. Husebg A, Nottestad L, Fossa JH, Furevik D, Jorgensen S (2002). Distribution and abundance of fish in deep-sea coral habitats. Hydrobiologia, 471, pp. 91-99. ICES (2005). Report of the Working Group on Deep-water Ecology {WGDEC}, 8-11 March, 2005, ICES Headquarters, Copenhagen. ICES CM 2005/ACE:02. 76pp. ICES (2006). Report of the Working Group on Deep-water Ecology [(WGDEC), 4-7 December 2005, Miami, USA. ICES CM 2006/ACE:04. 79pp. Johnston PA, Santillo D (2004). Conservation of seamount ecosystems: application of a Marine Protected Areas concept. Archive of Fishery and Marine Research, 51, pp. 305-319. Jones CG, Lawton JH, Shachek M [1994]. Organisms as ecosystem engineers. Oikos, 69, pp. 373-386. Kaiser MJ, de Groot SJ (eds) (2000). Effects of fishing on non-target species and habitats. Blackwell Scientific, Oxford, U.K. 399 pp. Kaneko A, Honji H, Kawatate K, Mizuna S, Masuda A, Miita T (1986). A note on internal wavetrains and the associated undulation of the sea surface observed upstream of seamounts. Journal of the Oceanographic Society, Japan, 42, pp. 75-82. Key RM, Kozyr A, Sabine CL, Lee K, Wanninkhof R, Bullister J, Feely RA, Millero F, Mordy C, Peng T-H (2004). A global ocean carbon climatology: results from Global Data Analysis Project (GLODAP). Global Biogeochemical Cycles, 18, doi:10.1029/2004GB002247. Kimball LA (2005). The International Legal Regime of the High Seas and the Seabed Beyond the Limits of National Jurisdiction and Options for the Establishment of Marine Protected Areas (MPAs) in Areas Beyond the Limits of National Jurisdiction. Secretariat of the Convention on 71 Seamounts, deep-sea corals and fisheries Biological Diversity, Montreal, Canada, CBD Technical Series No. 19. 64pp. Kitchingman A, Lai S (2004). Inferences on potential seamount locations from mid-resolution bathymetric data. In: Seamounts: Biodiversity and Fisheries {eds T Morato, D Pauly} UBC Fisheries Centre, 78, pp. 261, Vancouver, B.C. Kitchingman A, Lai S, Morato T, Pauly D (in press). Seamount abundance and _ locations. In: Seamounts: Ecology Fisheries and Conservation (eds TJ Pitcher, PJB Hart, T Morato, R Santos, M Clark], Blackwell Fisheries and Aquatic Resources Series, Blackwell Scientific. Koslow JA, Boehlert GW, Gordon JD, Haedrich RL, Lorance P, Parin N (2000). Continental slope and deep-sea fisheries: implications for a fragile ecosystem. ICES Journal of Marine Science, 57, pp. 548-557. Koslow JA, Gowlett-Holmes K (1998). The seamount fauna off southern Tasmania: benthic com- munities, their conservation and impacts of trawling: Final Report to Environment Australia and the Fisheries Research and Development Corporation. Rep. FRDC Project 95/058, CSIRO, Hobart, Tasmania, Australia, pp. 104. Koslow JA, Gowlett-Holmes K, Lowry JK, O’HaraT, Poore GCB, Williams A (2001). Seamount benthic macrofauna off southern Tasmania: community structure and impacts of trawling. Marine Ecology Progress Series, 213, pp. 111-125. Lack M, Short K, Willock A (2003). Managing risk and uncertainty in deep-sea fisheries: lessons from Orange Roughy. TRAFFIC Oceania and WWF Endangered Seas Programme, 73pp. Le Goff-Vitry MC, Pybus OG, Rogers AD (2004). Genetic structure of the deep sea coral Lophelia pertusa in the northeast Atlantic revealed by micro- satellites and internal transcribed spacer techniques. Molecular Ecology, 13, pp. 537 - 549. Leverette TL, Metaxas A (2005). Predicting habitat for two species of deep-water coral on the Canadian Atlantic continental shelf and slope. In: Cold- water corals and ecosystems (eds A Freiwald, JM Roberts) Springer Publishing House, Heidelberg, Germany. Levin LA, Thomas CL (1989). The influence of hydrodynamic regime on infaunal assemblages inhabiting carbonate sediments on central Pacific seamounts. Deep-Sea Research, 36 (12), pp. 1897-1915. Littler MM, Littler DS, Blair SM, Norris JN (1985). Deepest known plant life discovered on an uncharted seamount. Science, 227, pp. 57-59. 72 MCBI (2003 et seq.) Scientists Statement on Protecting the World's Deep-Sea Coral and Sponge Ecosystems. Marine Conservation Biology Institute and Oceana. Available at www.mcbi.org/ what/what_pdfs/dsc_signatures.pdf. 37pp (1 452 signatures). Meincke J (1971). Observation of an anticyclonic vortex trapped above a seamount. Journal of Geo- physical Research, 76, pp. 7432-7440. Melo 0, Menezes G (2002). Exploratory fishing of the orange roughy (Hoplostethus atlanticus) in some seamounts of the Azores archipelago. [CES C.M. 2002/M, 26, p 11. Morato T, Clark MR [in press). Seamount fishes: ecology and life histories. In: Seamounts: Ecology Fisheries and Conservation (eds TJ Pitcher, PJB Hart, T Morato, R Santos, M Clark], Blackwell Fisheries and Aquatic resources Series, Blackwell Scientific. Morato T, Pauly D (eds) (2004). Seamounts: Biodiversity and Fisheries. Vancouver: Fisheries Centre Research Report 12 Mortensen PB (2001). Aquarium observations on the deep-water coral Lophelia pertusa (L., 1758) (Scleractinia) and selected associated invertebrates. Ophelia, 54, pp. 83-104. Mundy BC, Parrish FA (2004). New records of the fish genus Grammatonotus (Teleostei: Perciformes: Percoidei: Callanthiidae) from the Central Pacific, including a spectacular species in the north- western Hawaiian Islands. Pacific Science, 58, pp. 403-417. Myers AA, Hall-Spencer J (2004). A new species of amphipod crustacean, Pleusymtes comitari sp. nov., associated with gorgonians on deep-water coral reefs off Ireland. Journal of the Marine Biological Association of the United Kingdom, 84, pp. 1029-1032. Noble M, Cacchione DA, Schwab WC (1988). Observations of strong mid-Pacific internal tides above Horizon guyot. Journal of Physical Oceano- graphy, 18, pp. 1300-1306. Noble M, Mullineaux LS (1989). Internal tidal currents over the summit of Cross seamount. Deep-Sea Research, 36, pp. 1791-1802. Oceana (2006) The seamounts of the Gorringe Bank. Available at www.oceana.org. 71pp. O'Driscoll RL, Clark MR (2005). Quantifying the relative intensity of fishing on New Zealand seamounts. New Zealand Journal of Marine and Freshwater Research, 39, pp. 839-850. Orr JC, Fabry VJ, Aumont O, Bopp L, Doney SC, Feely RA, Gnanadesikan A, Gruber N, Ishida A, Joos F, Key RM, Lindsay K, Maier-Reimer E, Matear R, Monfray P, Mouchet A, Najjar RG, Plattner GK, Rodgers KB, Sabine CL, Sarmiento JL, Schlitzer R, Slater RD, Totterdell IJ, Weirig MF, Yamanaka Y, Yool A (2005). Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature, 437, pp. 681-686. Pankhurst NW (1988). Spawning dynamics of orange roughy, Hoplostethus atlanticus, in mid-slope waters of New Zealand. Environmental biology of fishes, 21, pp. 101-116. Parin NV, Mironov AN, Nesis KN (1997]. Biology of the Nazca and Sala y Gomez submarine ridges, an outpost of the Indo-West Pacific fauna in the eastern pacific ocean: compostion and distribution of the fauna, its communities and history. Advances in Marine Biology, 32, pp. 145- 242. Parin NV, Prut’ko VG (1985). The thalassial mesobenthopelagic icthyocoene above the equator seamount in the western tropical Indian Ocean. Oceanology, 25 (6), 781-783. Parrish FA, Abernathy K, Marshall GJ, Buhleier BM (2002). Hawaiian monk seals (Monachus schauninslandi) foraging in deep-water coral beds. Marine Mammal Science, 18 (1), 244-258. Parrish FA, Baco AR Chapter 8: State of the U.S. Deep Coral Ecosystems in the Hawaiian Archipelago and the United States Pacific Islands Region. NOAA Technical Memorandum NMFS-OPR-29, In press. Report of the Secretary General (2003) United Nations General Assembly. Oceans and the Law of the Sea. Report of the Secretary-General 58/65 Available at: daccessdds.un.org/doc/UNDOC/GEN/N03/266/68/ PDF/N0326668.pdf?OpenElement 80pp. Report of the Secretary General (2005) United Nations General Assembly. Oceans and the Law of the Sea. Report of the Secretary-General 60/63 Addendum 1 Available at: daccessdds.un.org/ doc/UNDOC/GEN/N05/425/11/PDF/N0542511.pd f?O0penElement 87pp. Richer de Forges B, Koslow JA, Poore GCB (2000). Diversity and endemism of the benthic seamount fauna in the southwest Pacific. Nature, 405, pp. 944-947. Risk MJ, Heikoop JM, Snow MG, Beukens R (2002) Lifespans and growth patterns of two deep-sea corals: Primnoa resedaeformis and Desmo- phyllum cristagalli. Hydrobiologia, 471, pp. 125- 131. References Roberts CM (2002). Deep impact: The rising toll of fishing in the deep sea. Trends in Ecology and Evolution, 17 (5), pp. 242-245 Roberts JM (2005). Reef-aggregating behaviour by symbiotic eunicid polychaetes from cold-water corals: Do worms assemble reefs? Journal of the Marine Biological Association of the U.K., 85, pp. 813-819. Roberts JM, Harvey SM, Lamont, PA, Gage JD, Humphery JD (2000). Seabed photography, environmental assessment and evidence for deep-water trawling on the continental margin west of the Hebrides. Hydrobiologia, 441, pp. 173-183. Roberts J, Wheeler AJ, Freiwald A (2006). Reefs of the Deep: The Biology and Geology of Cold-Water Coral Ecosystems. Science, 312, pp. 543-547. Roden GI (1991). Mesoscale flow and thermohaline structure around Fieberling seamount. Journal of Geophysical Research, 96, pp. 16,653-16,672. Roden Gl, Taft BA (1985). Effect of the Emperor Seamounts on the mesoscale thermohaline structure during the summer of 1982. Journal of Geophysical Research, 90, pp. 839-855. Roden Gl, Taft BA, Ebbesmeyer CC (1982). Oceanographic aspects of the Emperor Seamounts region. Journal of Geophysical Research, 87, pp. 9537- 9552. Rogers AD (1994). The biology of seamounts. Advances in Marine Biology, 30, pp. 305-350. Rogers AD (1999). The biology of Lophelia pertusa (Linnaeus 1758) and other deep-water reef- forming corals and impacts from human activities. International Review of Hydrobiology, 84, pp. 315 - 406 Rogers AD (2004). The biology, ecology and vulnerability of seamount communities. |UCN, Gland, Switzerland. Available at: www.iucn.org/themes/ marine/pubs/pubs.htm 12 pp. Rogers AD, Baco A, Griffiths H, Hall-Spencer JM [in press). Corals on seamounts. In: Seamounts: Ecology Fisheries and Conservation {eds TJ Pitcher, PJB Hart, T Morato, R Santos, M Clark], Blackwell Fisheries and Aquatic Resources Series, Blackwell Scientific. Rogers AD, Morley S, Fitzcharles E, Jarvis K, Belchier M (2006). Genetic structure of Patagonian toothfish (Dissostichus eleginoides} populations on the Patagonian Shelf and Atlantic and western Indian Ocean Sectors of the Southern Ocean. Marine Biology, 149 (4), pp. 915-924. Rowden AA, Clark MR, O'Shea S (2004). The influence of deepwater coral habitat and fishing on benthic 73 Seamounts, deep-sea corals and fisheries faunal assemblages of seamounts on the Chatham Rise, New Zealand. ICES CM2004/ AA:09. Rowden AA, Clark MR, O'Shea S, McKnight D (2002). Benthic biodiversity of seamounts on the northwest Chatham Rise. New Zealand Marine Biodiversity Biosecurity, Report No. 2. 21 pp. Rowden AA, Clark MR, Wright IC (2005). Physical characterisation and a biologically focused classification of ‘seamounts’ in the New Zealand region. New Zealand Journal of Marine and Freshwater Research, 39, pp. 1039-1059. Royal Commission on Environmental Pollution (2004). Turning the Tide - Addressing the Impact of Fisheries on the Marine Environment. Royal Commission on Environmental Pollution, Westminster, London, U.K. ISBN 0 10 1639228. 480pp. Royal Society (2005) Ocean acidification due to increasing atmospheric carbon dioxide. The Royal Society, London, U.K. Policy Document 12/05, available at www.royalsoc.ac.uk ISBN 0 85403 617 2. 60pp. Sagalevitch AM, Torohov PV, Matweenkov VV, Galkin SV, Moskalev LI (1992). Hydrothermal activity on the underwater volcano Peepa (Bering Sea]. Izvestiya RAN. Seroes Biology, 9, pp. 104-114. (In Russian]. Seki MP, Somerton DA (1994). Feeding ecology and daily ration of the pelagic armourhead Pseudo- pentaceros wheeleri at Southeast Hancock Seamount. Environmental Biology of Fishes, 39, pp. 73-84. Sherwood OA, Scott DB, Risk MJ (2006) Late Holocene radiocarbon and aspartic acid racemization dating of deep-sea octocorals. Geochimica et Cosmochimica Acta, 70, pp. 2806-2814. Simpson ESW, Heydorn AEF (1965). Vema Seamount. Nature, 207, pp. 249-251. Smith DK, Cann, JR (1990). Hundreds of small volcanoes on the median valley floor of the Mid-Atlantic ridge at 24-30° N. Nature, 348, pp. 152-155. Smith DK, Jordan TH (1988). Seamount Statistics in the Pacific Ocean. Journal of Geophysical Research, 93, pp. 2899-2918. Smith PJ (2001) Managing biodiversity: Invertertebrate bycatch in seamount fisheries in the New Zealand Exclusive Economic Zone. World Fisheries Trust, IRRC/CRDI & UNEP. 30pp. Sokal RR, Rohlf FJ (1995). Biometry: the principles and practice of statistics in biological research. WH Freeman, New York. Stewart B (1998). Can a snake star earn its keep? 74 Feeding and cleaning behaviour in Astrobrachion constrictum (Farquhar) (Echinodermata: Ophiu- roidea}, a euryalid brittle-star living in association with the black coral, Antipathes fiordensis (Grange, 1990). Journal of Experimental Marine Biology and Ecology, 221, pp. 173-189. Stocks KI (2004). Seamount invertebrates: composition and vulnerability to fishing. In: Seamounts: Biodiversity and Fisheries [eds T Morato, D Pauly], UBC Fisheries Centre, 78, pp. 17-24, Vancouver, B.C. Stocks KI (2006). SeamountsOnline: an online information system for seamount biology. World Wide Web electronic publication. seamounts.sdsc.edu. Last accessed 13 July 2006. Stocks KI, Boehlert GW, Dower JF (2004). Towards an international field programme on seamounts within the Census of Marine Life. Archive of Fishery and Marine Research, 51, pp. 320-327. Stone GS, Madin, LP, Stocks, K, Hovermale, G, Hoagland, P, Schumacher, M, Etnoyer, P,Sotka, C, Tausig, H. (2004). Seamount biodiversity, exploitation and conservation. In: Defying Ocean's End (eds LK Glover, SA Earle}, pp. 45-70, Island Press, Washington. Stone RP (2006). Coral habitat in the Aleutian Islands of Alaska: depth distribution, fine-scale species associations, and fisheries interactions. Coral reefs, 25, pp. 229-238. Tracey DM, Horn PL (1999). Background and review of ageing of orange roughy (Hoplostethus atlant- icus] from New Zealand and elsewhere. New Zealand Journal of Marine and Freshwater Research, 33, pp. 67-86. Tsukamoto K (2006). Oceanic biology: Spawning of eels near a seamount. Nature, 439, p. 929. Uchida RN, Tagami DT (1984). Groundfish fisheries and research in the vicinity of seamounts in the North Pacific Ocean. Marine Fisheries Review, 46, pp. 1-17. UN General Assembly (2003) Resolution adopted by the General Assembly. 57/141. Oceans and the law of the sea. Available at: daccessdds.un.org/doc/ UNDOC/GEN/NO02/547/54/PDF/N0254754.pdf?0 penElement 13pp. UN General Assembly (2004a) Resolution adopted by the General Assembly. 58/240. Oceans and the law of the sea. Available at: daccessdds.un.org/doc/ UNDOC/GEN/N03/508/92/PDF/N0350892.pdf?0 penElement 15pp. UN General Assembly (2004b) Resolution adopted by the General Assembly 58/14. Sustainable fisheries, including through the 1995 Agreement for the Implementation of the Provisions of the United Nations Convention on the Law of the Sea of 10 December 1982 relating to the Conservation and Management of Straddling Fish Stocks and Highly Migratory Fish Stocks, and related instruments. Available at: daccessdds.un.org/ doc/UNDOC/GEN/N03/453/75/PDF/N0345375.pd f?0penElement 12pp. UN General Assembly (2005a) Resolution adopted by the General Assembly. 59/24. Oceans and the law of the sea. Available at: daccessdds.un.org/doc/ UNDOC/GEN/N04/477/64/PDF/N0447764.pdf?0 penElement 18pp. UN General Assembly (2005b) Resolution adopted by the General Assembly 59/25. Sustainable fisheries, including through the 1995 Agreement for the Implementation of the Provisions of the United Nations Convention on the Law of the Sea of 10 December 1982 relating to the Conservation and Management of Straddling Fish Stocks and Highly Migratory Fish Stocks, and related instruments. Available at: daccessdds.un.org /doc/UNDOC/GEN/N04/477/70/PDF/N0447770.p df?O0penElement 16pp. UN General Assembly (2006) Resolution adopted by the General Assembly. 60/30. Oceans and the law of the sea. Available at: daccessdds.un.org/doc/ UNDOC/GEN/N05/489/34/PDF/N0548934. pdf?0 penElement 19pp. US Department of Defense (1989). Gazetteer of undersea features. CD-ROM, US Department of Defense, Defense Mapping Agency, USA Vastano AC, Hagen DE, McNally GJ (1985). Lagrangian observations of the surface circulation at the Emperor seamount chain. Journal of Geo- physical Research, 90, pp. 3325-3331. Vastano AC, Warren BA (1976). Perturbations of the Gulf Stream by Atlantis Il seamount. Deep-Sea Research, 23, pp. 681-694. Vierros M, Douvere F, Arico S (2006) /mplementing the Ecosystem Approach in Open Ocean and Deep- Sea Environments. United Nations University- Institute of Advanced Studies Report. UNU-IAS, Yokohama, Japan, 40pp. References Watling L (2005). The global destruction of bottom habitats by mobile fishing gear. In: Marine conservation biology: the science of maintaining the sea’s biodiversity (eds EA Norse, LB Crowder], Island Press, Washington. pp. 198-210. Watling L, Norse EA (1998). Disturbance of the seabed by mobile fishing gear: a comparison to forest clearcutting. Conservation biology, 12, pp 1180-1197. Weaver PPE, Billett DSM, Boetius A, Danovaro R, Freiwald A, Sibuet M (2004). Hotspot ecosystem research on Europes deep-ocean margins Oceanography 17 (4), pp. 132-143. Wessel P (1997). Sizes and ages of seamounts using remote sensing: Implications for intraplate volcanism. Science, 277, pp. 802-805. Wessel P (2001). Global distribution of seamounts inferred from gridded Geosat/ERS-1 altimetry. Journal of Geophysical Research. B. Solid Earth, 106, pp. 19,431-19,441. Wilson CD, Boehlert GW (2004). Interaction of ocean currents and resident micronekton at a seamount in the central North Pacific. Journal of Marine Systems, 50, pp. 39-60. Wilson RR, Kaufman RS (1987). Seamount biota and biogeography. In: [eds BH Keating, P Fryer, R Batiza, G Boehlert) Seamounts, Islands and Atolls, Geophysical Monograph, 43, pp. 355-377. Worm B, Lotze HK, Myers RA (2003). Predator diversity hotspots in the blue ocean. Proceedings of the National Academy of Sciences, USA, 100, pp. 9884-9888. Wunsch C, Webb S (1979). The climatology of deep ocean internal waves. Journal of Physical Ocean- ography, 9, pp. 225-243. WWF (2006) Policy proposals and operational guidance for ecosystem-based management of marine capture fisheries. WWF International, Gland, Switzerland, 80pp. Available at: www.panda.org. Zeebe RE, Wolf-Gladrow DA (2001). C02 in seawater: equilibrium, kinetics, isotopes. Elsevier Oceano- graphy Series, 65, pp. 346, New York. 75 Seamounts, deep-sea corals and fisheries Selection of institutions and researchers working on seamount and cold-water coral ecology Country/Institution Contact AUSTRALIA Commonwealth Scientific and Industrial Research Organization (CSIRO) Marine and Atmospheric Research University of the Sunshine Coast University of Tasmania CANADA Dalhousie University University of British Columbia University of Victoria GERMANY Alfred Wegener Institute for Polar and Marine Research Coral Reef Ecology (CORE) Working Group, GeoBio-Centre, Tony Koslow Tony.Koslow(@csiro.au Alan Williams alan.williams(@csiro.au Thomas Schlacher tschlach{dusc.edu.au Karen Miller karen.miller(dutas.edu.au Derek Tittensor derekt(dmathstat.dal.ca Adrian Kitchingman a.kitchingman(dfisheries.ubc.ca Telmo Morato t.morato(dfisheries.ubc.ca Daniel Pauly d.pauly(dfisheries.ubc.ca Tony Pitcher t.pitcher(dfisheries.ubc.ca John Dower dower(duvic.ca Verena Tunnicliffe verenat(duvic.ca Heino O Fock hfock(dawi-bremerhaven.de Christian Wild c.wild{dlrz.uni-muenchen.de Ludwig Maximilians University Munich Mainz Academy of Sciences / The Institute for Polar Ecology, University of Kiel Max Planck Institute for Marine Microbiology, Bremen 76 Dieter Piepenburg dpiepenburg(ipoe.uni-kiel.de Antje Boetius aboetius(dmpi-bremen.de Country/Institution Contact University of Erlangen André Freiwald andre.freiwald(dpal.uni-erlangen.de Universitat Hamburg Bernd Christiansen bchristiansen(uni-hamburg.de INDIA National Institute of Oceanography Baban Ingole baban(ddarya.nio.org IRELAND National University of Ireland, Anthony J Grehan anthony.grehan(dnuigalway.ie Martin White Martin.White(dnuigalway.ie Galway Andrew Wheeler a.wheeler(ducc.ie University College Cork JAPAN Extremobiosphere Research Center Shinji Tsuchida Japan Agency for Marine-Earth Science and Technology (JAMSTEC) tsuchidas(djamstec.go.jp NETHERLANDS Royal Netherlands Institute for Sea Research (NIOZ] Gerard CA Duineveld duin{dnioz.nl Matt Gianni matthewgiannildnetscape.net Fisheries, Oceans, Marine Biodiversity NEW CALEDONIA Institut de Recherche pour Bertrand de Forges le Développement (IRD) bertrand.richer-de- forgesfdnoumea.ird.nc NEW ZEALAND National Institute of Water and Atmospheric Research (NIWA) Malcolm Clark m.clark(aniwa.co.nz Mireille Consalvey m.consalvey(dniwa.co.nz Ashley Rowden a.rowden(Gniwa.co.nz Dianne Tracey d.tracey(aniwa.co.nz Keith Probert keith.probert(dstonebow.otago.ac.nz University of Otago Country/Institution NORWAY Contact Institutions and researchers Country/Institution Institute of Marine Research (IMR) | Lene Buhl-Mortensen PORTUGAL University of the Azores RUSSIA P.P. Shirshov Institute of Oceanography UNITED KINGDOM Joint Nature Conservation Committee (JNCC) Scottish Association for Marine Science (SAMS) United Nations Environment Programme World Conservation Monitoring Centre {UNEP-WCMC] University of Plymouth Lene.Mortensen(difm.uib.no Jan Helga Fossa jhfdimrno Pal Mortensen Paal.mortensen(dimr.no Gui Menezes gui(dnotes.horta.uac.pt Ricardo Santos ricardo(dnotes.horta.uac.pt Andrey Gebruk agebruk(docean.ru Tina Molotsodova tina(dsio.rssi.ru Charlotte Johnston charlotte.johnston(@jncc.gov.uk Mark Tasker mark.tasker(djncc.gov.uk Bhavani Narayanaswamy Bhavani.Narayanaswamy(dsams.ac.uk J Murray Roberts Murray.Roberts(dsams.ac.uk Stefan Hain Stefan.Hain(@unep-wemc.org Jason Hall-Spencer jason.hall-spencer(aplymouth.ac.uk Zoological Society of London, Institute of Zoology Kerry Howell kerry.howell(@ plymouth.ac.uk Alex Rogers Alex.Rogers(dioz.ac.uk UNITED STATES OF AMERICA Natural History Museum of Los Angeles, County in California National Undersea Research Center (NURP]} National Oceanic & Atmospheric Administration (NOAA) University of California San Diego Florida Atlantic University Smithsonian Institution Paul Wessel University of Maine University of Kansas Woods Hole Oceanographic Institution [(WHOI) Contact Peter Etnoyer peter(daquanautix.com Peter Auster peter.auster(duconn.edu Bob Embley embley(@pmel.noaa.gov Paul Brewin pebrewin(dsdsc.edu Lisa Levin llevinfducsd.edu Karen Stocks kstocks{dsdsc.edu Jon Moore jmoore(dfau.edu Stephen Cairns Cairnss{dsi.edu University of Hawaii pwessella@hawaii.edu Les Watling watling{dmaine.edu Daphne G Fautin fautin(aku.edu Amy Baco-Taylor abacofdwhol.edu Tim Shank tshank(a@whoi.edu 77 Seamounts, deep-sea corals and fisheries Selection of coral and seamount resources censeam.niwa.co.nz CenSeam [a global census of marine life on seamounts] is a Census of Marine Life Field Programme aiming to provide the framework needed to prioritize, integrate, expand and facilitate seamount research efforts. www.coml.org The Census of Marine Life [CoML] is a network of researchers in more than 70 nations engaged in a 10-year initiative to assess and explain the diversity, distribution, and abundance of marine life in the oceans — past, present and future. www.fao.org/DOCREP/003/X2465E/x2465e0h.htm FAO FISHERIES TECHNICAL PAPER 382 ‘Guidelines for the Routine Collection of Capture Fishery Data’ www.fishbase.org/search.php FishBase is a relational database with information to cater to different professionals such as research scientists, fisheries managers, zoologists and many more. FishBase on the web contains practically all fish species known to science. [eds R Froese, D Pauly; version 16 February 2004). bure.unep-wcmc.org/marine/coldcoral Global cold-water coral database and GIS, an interactive mapping tool developed by UNEP which provides easy access to a wealth of information on cold-water coral ecosystems, drawing on the data and collective expertise of scientists, national agencies and regional organizations from around the world. cdiac.ornl.gov/oceans/glodap/Glodap_home.htm The GLobal Ocean Data Analysis Project (GLODAP) is a cooperative effort to coordinate global synthesis projects funded through the National Oceanic and Atmospheric Administration (NOAA), the U.S. Department of Energy (DOE), and the National Science Foundation (NSF) as part of the Joint Global Ocean Flux Study - Synthesis and Modeling Project (JGOFS-SMP). www.ngdc.noaa.gov/mgg/gebco General Bathymetric Chart of the Oceans {GEBCO) aims to provide the most authoritative, publicly-available bathymetry datasets for the world’s oceans. GEBCO operates under the auspices of the International Hydrographic Organization (IHO) and the United Nations’ (UNESCO} Intergovernmental Oceanographic Commission (IOC). www.eu-hermes.net Hotspot Ecosystems Research on the Margins of European Seas [HERMES], a multidisciplinary deep-sea research project 78 with 50 partners under the EC Framework Six Programme. HERMES work packages include, inter alia, cold-water coral reefs and carbonate mounds. www.kgs.ku.edu/Hexacoral/ Biogeoinformatics of Hexacorals is intended to: (1) provide a public information resource of data, interpretation and methods related to the taxonomy, biogeography and_ habitat characteristics or environmental correlates of the Hexacorallia and allied taxa (2) connect and integrate the activities of the individual and institutional partners [3] keep a wide range of project information updated and available to all interested parties and (4) provide a directory and communication links to participants and related projects. www.lophelia.org/index.htm Lophelia.org is dedicated to the cold-water coral Lophelia pertusa and is an information resource on the cold-water coral ecosystems of the deep ocean. Wwww.mar-eco.no MAR-ECO (patterns and processes of the ecosystems of the northern mid-Atlantic] is Census of Marine Life Field Programme. MAR-ECO is an international exploratory study of the animals inhabiting the northern mid-Atlantic. Scientists from 16 nations around the northern Atlantic Ocean are participating in research of the waters around the mid-Atlantic Ridge from Iceland to the Azores oceanexplorer.noaa.gov/ NOAA Ocean Explorer is an educational Internet offering for all who wish to learn about, discover, and virtually explore the ocean realm. It provides public access to current information on a series of NOAA scientific and educational explorations and activities in the marine environment with links to numerous cold-water coral expeditions. www1.uni-hamburg.de/OASIS OASIS [Oceanic seamounts: an integrated study) is a European Commission supported project aiming to describe the functioning characteristics of seamount ecosystems. marine.rutgers.edu/opp IMCS Ocean Primary Productivity Team's (OPPT) home page aims to provide: (1) Access to datasets of primary productivity measurements based on 14C uptake and stimulated fluorescence techniques, with the hope that these data will be used for productivity model development and testing; (2) Computer source code, input data fields and ocean productivity estimates for the Vertically Generalized Production Model (VGPM] developed by the OPPT, and; (3) Information on activities of the NASA-sponsored Ocean Primary Productivity Working Group [OPPWG], which has been conducting round-robin algorithm testing exercises since 1994 to compare, in an investigator-independent manner, the performance of various productivity models with the intent of establishing a NASA resident ‘consensus’ algorithm for the routine generation of ocean productivity maps. www.seaaroundus.org The Sea Around Us Project (SAUP] is devoted to studying the impact of fisheries on the world’s marine ecosystems. To achieve this, project staff have used a Geographic Information System (GIS) to map global fisheries catches from 1950 to the present, under explicit consideration of coral reefs, seamounts, estuaries and other critical habitats of fish, marine invertebrates, marine mammals and other components of marine biodiversity. The data presented, which are all freely available, are meant to support studies of global fisheries trends and the development of sustainable, ecosystem-based fisheries policies. seamounts.sdsc.edu SeamountsOnline is a freely-available online resource of seamount related data. It is a NSF-funded project designed to gather information on species found in seamount habitats, and to provide a freely-available online resource for accessing and downloading these data. It is designed to facilitate research into seamount ecology, and to act as a resource for managers. earthref.org The Seamount Catalog [search under databases for the Seamount Catalog) is a digital archive for bathymetric seamount maps that can be viewed and downloaded in various formats. This catalog also contains morphological data and sample information. Related grid and multibeam data files, as well as user-contributed files, can be downloaded as well. www.nodc.noaa.gov/OC5/W0A01/pr_woa01.html The World Ocean Atlas 2001 [WOA01) contains ASCII data of statistics and objectively analysed fields for one-degree and five-degree squares generated from World Ocean Database 2001 observed and standard level flagged data. The ocean variables included in the atlas are: in situ temperature, salinity, dissolved oxygen, apparent oxygen utilization, per cent oxygen saturation, dissolved inorganic nutrients (phosphate, nitrate and silicate), chlorophyll at standard depth levels, and plankton biomass sampled from 0-200m. Resources Appendix | Physical data All physical data were compiled onto a one-degree resolution global grid, centred on the midpoint of each degree cell. Physical data were gridded at 0, 500, 1000, 1500, 2000 and 2500 m depth. These resolutions were chosen to fit with data availability [{WOA and GLODAP data are available at this grid resolution}. Physical data and primary productivity model output were all long-term annual means. Composite annual data were derived from cruises and sampling covering a variety of time periods; where possible, data were selected from the 1990s World Ocean Atlas 2001 data (Conkright et al., 2002) were composite annual objectively analysed means. GLODAP gridded data {Key et al., 2004) were mostly derived from 1990s WOCE (World Ocean Circulation Experiment] cruises. VGPM model outputs [Behrenfeld and Falkowski, 1997) were depth- integrated surface values corrected for cloudiness, derived from data collected between 1977 and 1982. SODA modelled current velocities (Carton et al., 2000) were the grand mean of the annual means for the period 1990-1999, using the 1.4.2 version of the model; the velocity layer nearest to each depth grid layer was used. The aragonite saturation state was calculated using GLODAP data and following the A[CO32-]a method of Orr et al. (2005), with constants as described in Orr et al. (2005) and equations following Zeebe and Wolf-Gladrow (2001). Positive [CO32-]a indicates supersaturation; negative undersaturation. Depth is included as a parameter not because it is important per se, but because it may correlate with unmeasured factors such as pressure. 79 Seamounts, deep-sea corals and fisheries Appendix II The habitat suitability model ENFA is a predictive habitat suitability modelling technique designed to work with presence-only data (Hirzel et al., 2002). We bin scleractinian seamount data records to the one-degree global grid and assign them to the closest depth layer. We used only coral records above 2 500 m depth. Physical data were normalized using the Box-Cox transformation (Sokal and Rohlf, 1995). A mismatch occurs between some coral locations and predicted seamount locations in that some corals are found on seamounts that are not detected by the bathymetric analysis (Kitchingman and Lai, 2004). To resolve this, we model habitat suitability for the whole ocean, but restrict coral presences to seamounts. We used the geometric mean algorithm in ENFA (Hirzel and Arlettaz, 2003). ENFA outputs species marginality {absolute difference between the global mean and the species mean in the multidimensional environmental space) and specialization {ratio of variance between the global distribution and species distribution). All environmental variables are converted into uncorrelated factors in a manner similar to principal com- ponent analysis. Habitat suitability maps were constructed following Hirzel et al. (2002) using the isopleth method. Eight factors were used to construct habitat suitability maps, following a broken stick distribution (Hirzel et al., 2002). Assessing model performance presents a different challenge for presence-only models than for presence-absence models (Boyce et al., 2002). In this case, validation for habitat suitability maps was carried out using a cross-validation technique (Boyce et al., 2002). Data were partitioned into four bins followed by a 10-fold cross validation. For each validation subset, area- adjusted frequency was compared with that of a randomly distributed species using Spearman's rank correlation to assess the monotonicity of the curve [Table A1). This coefficient varies between -1 and 1; a value near 1 indicates area-adjusted frequency model predictions monotonically increasing with increasing habitat suitability and deviating from a random curve, Suggesting good model performance. 80 Table A1: Cross-validation results; Spearman's rank coefficient Replicate Rs 0.8 0.8 1 1 1 0.8 1 0.8 0.8 0.8 Mean 0.88 0.10 © NO (oR [|e |= so = i=) wn Sy Key assumptions of ENFA are that data are multinormal, that species occurrence data span the complete environmental range, and that the species is at equilibrium. Hirzel et al. (2002) suggest that ENFA is robust to deviations from normality, and the method has also been shown to be robust to quality and quantity of data (Hirzel et al., 2001). Spatial autocorrelation was not directly accounted for but is unlikely to be a major issue with this data (Leverette and Metaxas, 2005]. REFERENCE MAPS Map 2. FAO Major marine fishing areas Area Name - = Area ——— Name ee 18 Arctic Sea 57 Indian Ocean, Eastern 21 Atlantic, Northwest 58 Indian Ocean, Antarctic and Southern 27 Atlantic, Northeast 61 Pacific, Northwest 31 Atlantic, Western Central 67 Pacific, Northeast 34 Atlantic, Eastern Central 71 Pacific, Western Central 37 Mediterranean and Black Sea 77 Pacific, Eastern Central 41 Atlantic, Southwest 81 Pacific, Southwest 47 Atlantic, Southeast 87 Pacific, Southeast 48 Atlantic, Antarctic 88 Pacific, Antarctic 51 Indian Ocean, Western Map 3. Regional sea conventions and action plans 4 Convention on the Protection of the Marine Environment of the Baltic Sea Area (HELCOM) 2 Bucharest Convention and Black Sea Environment Programme 3 Cartagena Convention for the Wider Caribbean Region, Caribbean Environment Programme (CEP) and Action Plan® 4 East Asian Seas Action Plan (COBSEA]@ 5 Nairobi Convention and East African Action Plan@ 6 Barcelona Convention and Mediterranean Action Plan (MAP]@ 7 Antigua Convention and North-East Pacific Action Plan 8 North West Pacific Action Plan (NOWPAP]@ 9 Jeddah Convention and Red Sea and Gulf of Aden Action Plan (PERSGA]> 10 Kuwait Convention and ROPME Sea Area Action Plan® 11 Noumea (or SPREP] Convention and Pacific Action Plan® 12 South Asian Seas Action Plan (SAS) and South Asian Seas Cooperative Environment Programme (SACEP)> 13 Lima Convention and South-East Pacific Action Plan (CPPS}> 14 Abidjan Convention and West and Central Africa Action Plan@ 15 Regional Programme of Action for the Protection of the Arctic Marine Environment from Land-based Activities (PAME}H,¢ 16 Convention on the Conservation of Antarctic Marine Living Resources (CCAMLRJH.¢ 17 Framework Convention for the Protection of the Marine Environment of the Caspian Sea {Teheran Convention) and Caspian Sea Strategic Action Programme’ 18 OSPAR Convention for the Protection of the Marine Environment of the North-East Atlantic (OSPAR)H,¢ H: with a high sea mandate / competence. In general, UNEP administered Conventions and Action Plans apply only to the national waters of member states, incl. EEZs, where appropriate. a: UNEP administered b: Non-UNEP administered c: Independent Programme Map 4. Regional marine fisheries bodies that can directly establish management measures The map shows only the areas of competence of those Regional Marine Fisheries Bodies that can directly establish management measures. In addition to those listed and displayed, the International Whaling Commission [IWC] is a global bodies without a defined area of competence. 1 Convention on the Conservation of Antarctic Marine Living Resources (CCAMLR]> 2 Convention on the Conservation and Management of the Pollock Resources in the Central Bering Sea (CCBSP] 3 Commission for the Conservation of Southern Bluefin Tuna (CCSBT) 4 General Fisheries Commission for the Mediterranean (GFCM)@»9 5 Inter-American Tropical Tuna Commission [IATTC) 6 International Commission for the Conservation of Atlantic Tunas (ICCAT) 7 Indian Ocean Tuna Commission (IOTC)@ 8 International Pacific Halibut Commission (IPHC) 9 Northwest Atlantic Fisheries Organization (NAFO}D 10 North Atlantic Salmon Conservation Organization (NASCO} il North East Atlantic Fisheries Commission (NEAFC)® 12 North Pacific Anadromous Fish Commission [NPFAC} 13 Pacific Salmon Commission (PSC) 14 South East Atlantic Fisheries Organization (SEAFO)b 15 South Indian Ocean Fisheries Agreement (SIOFA)© 16 South Pacific Regional Fisheries Management Organisation (SPRFMO) 17 Western and Central Pacific Fisheries Commission [WCPFC) a: FAO administered b: legal competence to manage most or all fisheries within their areas of application, including management of deep sea stocks beyond national jurisdiction c: under negotiation Seamounts, deep-sea corals and fisheries Appendix II The habitat suitability model ENFA is a predictive habitat suitability modelling technique designed to work with presence-only data (Hirzel et al., 2002). We bin scleractinian seamount data records to the one-degree global grid and assign them to the closest depth layer. We used only coral records above 2 500 m depth. Physical data were normalized using the Box-Cox transformation (Sokal and Rohlf, 1995). A mismatch occurs between some coral locations and predicted seamount locations in that some corals are found on seamounts that are not detected by the bathymetric analysis (Kitchingman and Lai, 2004). To resolve this, we model habitat suitability for the whole ocean, but restrict coral presences to seamounts. We used the geometric mean algorithm in ENFA (Hirzel and Arlettaz, 2003). ENFA outputs species marginality [absolute difference between the global mean and the species mean in the multidimensional environmental space) and specialization {ratio of variance between the global distribution and species distribution). All environmental variables are converted into uncorrelated factors in a manner similar to principal com- ponent analysis. Habitat suitability maps were constructed following Hirzel et al. (2002) using the isopleth method. Eight factors were used to construct habitat suitability maps, following a broken stick distribution (Hirzel et al., 2002). Assessing model performance presents a different challenge for presence-only models than for presence-absence models (Boyce et al., 2002). In this case, validation for habitat suitability maps was carried out using a cross-validation technique [Boyce et al, 2002). Data were partitioned into four bins followed by a 10-fold cross validation. For each validation subset, area- adjusted frequency was compared with that of a randomly distributed species using Spearman's rank correlation to assess the monotonicity of the curve [Table A1). This coefficient varies between -1 and 1; a value near 1 indicates area-adjusted frequency model predictions monotonically increasing with increasing habitat suitability and deviating from a random curve, Suggesting good model performance. 80 Table A1: Cross-validation results; Spearman's rank coefficient Replicate Rs 1 0.8 2 0.8 3 1 4 1 5 1 6 0.8 7 1 8 0.8 9 0.8 10 0.8 Mean 0.88 s. D. 0.10 Key assumptions of ENFA are that data are multinormal, that species occurrence data span the complete environmental range, and that the species is at equilibrium. Hirzel et al. (2002) suggest that ENFA is robust to deviations from normality, and the method has also been shown to be robust to quality and quantity of data {Hirzel et al., 2001). Spatial autocorrelation was not directly accounted for but is unlikely to be a major issue with this data (Leverette and Metaxas, 2005). REFERENCE MAPS Map 1. Exclusive economic zones Prepared using the Global Maritime Boundaries Database [February 2006 edition, © General Dynamics Advanced Information Systems, 1998-2006) EEZs and fishing zones in the Mediterranean not displayed. Map 2. FAO Major marine fishing areas Source and further information; http://www lao. org/figis/serviet/static?dom=root&xml=geagraphy/fao_fishing_areaxml Seamounts, deep-sea corals and fisheries Appendix II The habitat suitability model ENFA is a predictive habitat suitability modelling technique designed to work with presence-only data (Hirzel et al., 2002). We bin scleractinian seamount data records to the one-degree global grid and assign them to the closest depth layer. We used only coral records above 2 500 m depth. Physical data were normalized using the Box-Cox transformation (Sokal and Rohlf, 1995). A mismatch occurs between some coral locations and predicted seamount locations in that some corals are found on seamounts that are not detected by the bathymetric analysis (Kitchingman and Lai, 2004). To resolve this, we model habitat suitability for the whole ocean, but restrict coral presences to seamounts. We used the geometric mean algorithm in ENFA (Hirzel and Arlettaz, 2003). ENFA outputs species marginality [absolute difference between the global mean and the species mean in the multidimensional environmental space) and specialization {ratio of variance between the global distribution and species distribution). All environmental variables are converted into uncorrelated factors in a manner similar to principal com- ponent analysis. Habitat suitability maps were constructed following Hirzel et al. (2002) using the isopleth method. Eight factors were used to construct habitat suitability maps, following a broken stick distribution (Hirzel et al., 2002). Assessing model performance presents a different challenge for presence-only models than for presence-absence models (Boyce et al., 2002). In this case, validation for habitat suitability maps was carried out using a cross-validation technique [Boyce et al., 2002]. Data were partitioned into four bins followed by a 10-fold cross validation. For each validation subset, area- adjusted frequency was compared with that of a randomly distributed species using Spearman's rank correlation to assess the monotonicity of the curve (Table A1). This coefficient varies between -1 and 1; a value near 1 indicates area-adjusted frequency model predictions monotonically increasing with increasing habitat suitability and deviating from a random curve, Suggesting good model performance. 80 Table A1: Cross-validation results; Spearman's rank coefficient Replicate Rs 1 0.8 2 0.8 3 1 4 1 5 1 6 0.8 7 1 8 0.8 9 0.8 10 0.8 Mean 0.88 Ss. D. 0.10 Key assumptions of ENFA are that data are multinormal, that species occurrence data span the complete environmental range, and that the species is at equilibrium. Hirzel et al. (2002) suggest that ENFA is robust to deviations from normality, and the method has also been shown to be robust to quality and quantity of data (Hirzel et al., 2001). Spatial autocorrelation was not directly accounted for but is unlikely to be a major issue with this data (Leverette and Metaxas, 2005). REFERENCE MAPS Map 3. Regional sea conventions and action plans Source and further information: http://www.unep.org/regionalseas/ MAP 4. Regional marine fisheries bodies that can directly establish management measures Source and further information: FAO, 1999-2006, Regional Fishery Bodies - Map of competence area, http://www. fao.org/fi/body/rfb/index.htm ‘=. ae Seamounts, deep-sea corals and fisheries An ubiquitous ocean floor feature, a key marine ecosystem and an important human activity: together these have created one of the most critical ocean issues. Seamounts, deep-sea corals and fisheries reveals the global scale of the vulnerability of habitat-forming stony corals on seamounts - and that of associated marine biodiversity and assemblages - to the impacts of trawling, | especially in areas beyond national jurisdiction. It provides some of the best { scientific evidence to date to support the call for concerted and urgent action on j the high seas to protect seamount communities and their associated resources | from the adverse effects of deep-water fishing. 3 Seamounts, deep-sea corals and fisheries describes the results of data | analyses that were used to understand the global distribution of deep-sea corals { on seamounts, to model the distribution of suitable habitat for stony corals, and | to appreciate the extent of trawl fisheries on seamounts in areas beyond national jurisdiction. These results were combined, along with knowledge of the effects of trawling on corals and other seamount species, to identify the main areas at risk from the impact of current and future trawling on the high seas. In particular, seamount ecosystems in the Indian, North and South Atlantic, and South Pacific 4 Oceans are threatened by the expansion of alfonsino (250-750 metres) and orange roughy (750-1 200 metres) fisheries. ~ ~ Seamounts, deep-sea corals and fisheries aims to raise the awareness | of managers, decision makers and stakeholders about the distribution of deep- sea corals on seamounts and their vulnerability to trawling. It provides facts and information to support and guide the international processes within and outside the United Nations system to find solutions for the conservation, protection and sustainable management of seamount ecosystems -— before it is too late. UNEP Regional Seas Report and Studies No 183 www.unep.org UNEP-WCMC Biodiversity Series No 25 ace ae Tel: +254 (0) 20 7621234 Fax: +254 (0) 20 7623927 Email: uneppub@unep.org Website: www.unep.org UNEP World Conservation Monitoring Centre 219 Huntingdon Road, Cambridge ISBN 978-92-807-2778-4 CB3 ODL, United Kingdom Tel: +44 (0) 1223 277314 A Fax: +44 (0) 1223 277136 Email: info@unep-wemc.org Website: www.unep-wcmc.org