Ecological Aquaculture

ECOLOGICAL AQUACULTURE The Evolution of the Blue Revolution

ECOLOGICAL AQUACULTURE The Evolution of the Blue Revolution

ECOLOGICAL AQUACULTURE The Evolution of the Blue Revolution Edited by Barry A. Costa-Pierce

Rhode Island Sea Grant College Program Graduate School of Oceanography Department of Fisheries, Animal and Veterinary Science University of Rhode Island

# 2002 Blackwell Science Ltd, a Blackwell Publishing Company Editorial Offices: Osney Mead, Oxford OX2 0EL, UK Tel: +44 (0)1865 206206 Blackwell Science, Inc., 350 Main Street, Malden, MA 02148-5018, USA Tel: +1 781 388 8250 Iowa State Press, a Blackwell Publishing Company, 2121 State Avenue, Ames, Iowa 50014-8300, USA Tel: +1 515 292 0140 Blackwell Science Asia Pty, 54 University Street, Carlton, Victoria 3053, Australia Tel: +61 (0)3 9347 0300 Blackwell Wissenschafts Verlag, KurfuÈrstendamm 57, 10707 Berlin, Germany Tel: +49 (0)30 32 79 060 The right of the Author to be identified as the Author of this Work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.

First published 2002 by Blackwell Science Ltd Library of Congress Cataloging-in-Publication Data Costa-Pierce, Barry A. Ecological aquaculture: the evolution of the blue revolution/Barry A. Costa-Pierce. p. cm Includes bibliographical references (p.). ISBN 0-632-04961-8 (alk. paper) 1. AquacultureÐEnvironmental aspects. 2. AquacultureÐEconomic aspects. I. Title. SH135 .C67 2002 338'71Ðdc21 2001043507 ISBN 0-632-04961-8 A Catalogue record for this title is available from the British Library Set in 10/13pt Times by DP Photosetting, Aylesbury, Bucks Printed and bound in Great Britain by MPG Books Ltd, Bodmin, Cornwall For further information on Blackwell Science, visit our website: www.blackwell-science.com

I dedicate this book to the memory of my beloved father, Edward A. Pierce Sr., who passed away while this book was being completed; to my mother, Thelma Pierce; and to my United Nations family ± Lily Mae Ho-Pierce, Lia Kaiulani Ho Costa-Pierce, Sierra Shaiming Ho Costa-Pierce ± and to the Elders of the Wampanoag Tribe of Massachusetts, USA.

Contents

List of Contributors Foreword Preface Acknowledgments Part 1

The Background of Ecological Aquaculture

1

The History of Aquaculture in Traditional Societies Malcolm C.M. Beveridge and David C. Little

2

The Ahupua'a Aquaculture Ecosystems in Hawaii Barry A. Costa-Pierce

Part 2 3

4

The Methods of Ecological Aquaculture

ix x xii xv 1 3 30

45

Development and Application of Genetic Tags for Ecological Aquaculture Theresa M. Bert, Michael D. Tringali and Seifu Seyoum

47

Aquaculture Escapement, Implications and Mitigation: The Salmonid Case Study C.J. Bridger and Amber F. Garber

77

5

Farming Systems Research and Extension Methods for the Development of Sustainable Aquaculture Ecosystems Barry A. Costa-Pierce

103

6

A Market-driven, Social Ecological Approach to Planning for Sustainable Aquaculture: A Case Study of Tilapia in Fiji Barry A. Costa-Pierce

125

Part 3 7

The Context of Ecological Aquaculture

Village-based Aquaculture Ecosystems as a Model for Sustainable Aquaculture Development in Sub-Saharan Africa Randall E. Brummett and Barry A. Costa-Pierce

143 145

viii

Contents

8

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems William J. FitzGerald, Jr

161

9

An Integrated Fish and Field Crop System for Arid Areas James E. Rakocy

263

10

Sustainability of Cage Aquaculture Ecosystems for Large-Scale Resettlement from Hydropower Dams: An Indonesian Case Study Barry A. Costa-Pierce

11

The Role of Aquaculture in the Restoration of Coastal Fisheries Mark A. Drawbridge

Part 4 12

Conclusion

Ecology as the Paradigm for the Future of Aquaculture Barry A. Costa-Pierce

Index

286 314

337 339 373

List of Contributors

Theresa M. Bert Florida Marine Research Institute,100 Eighth Avenue South, St. Petersburg, Fl 33701, USA Malcolm C.M. Beveridge FK9 4LA, Scotland

Institute of Aquaculture, University of Stirling, Stirling

Christopher J. Bridger College of Marine Sciences, University of Southern Mississippi, 703 East Beach Drive, Ocean Springs, MS 39564, USA Randall E. Brummett International Center for Living Aquatic Resources Management (ICLARM), BP 2008 (Messa), YaoundeÂ, Cameroon Barry A. Costa-Pierce Rhode Island Sea Grant College Program, Department of Fisheries, Animal and Veterinary Science, University of Rhode Island, Narragansett, RI 02882, USA Mark A. Drawbridge Hubbs-Sea World Research Institute, 2595 Ingraham Street, San Diego, CA 92109, USA William J. FitzGerald, Jr

PO Box 6997, Tamuning, Guam 96931

Amber F. Garber Department of Zoology, North Carolina State University, Raleigh, NC 27695-7617, USA David C. Little Scotland

Institute of Aquaculture, University of Stirling, Stirling FK9 4LA,

James E. Rakocy University of the Virgin Islands, Agricultural Experiment Station, Kingshill, VI 00850, USA Seifu Seyoum Florida Marine Research Institute, 100 Eighth Avenue South, St. Petersburg, Fl 33701, USA Michael D. Tringali Florida Marine Research Institute, 100 Eighth Avenue South, St. Petersburg, Fl 33701, USA

Foreword

During the early years of the modern aquaculture era ± which I think of as starting in the 1960s ± practitioners of the art undoubtedly considered themselves to be strong environmentalists. Working with nature to produce food, bait, and ornamental species for the enjoyment of humans was the goal, but the practice involved establishing and maintaining the best possible conditions for production of the target species. Our understanding of pond dynamics, physiological requirements of culture species, and aquatic animal nutrition and disease were just a few among the many topics addressed by researchers. A tremendous amount of information on the relationships between aquatic organisms and their environment, e.g. ecology, was generated and continues to be generated by aquacultural researchers. Yet, by the mid-1980s, at least some forms of aquaculture were being branded as detrimental to the environment. It took a bit of time for the aquaculture community to become sufficiently introspective to recognize that, indeed, some of their practices had negatively impacted the environment. For at least the past decade, a considerable amount of time, money, and effort have been expended by the aquacultural community in addressing both real and perceived problems. The mantra of the aquacultural community became focused on sustainability. Many publications have appeared over that period which address the criticisms and relate how aquaculture has responded to them. Ecological Aquaculture takes a somewhat different approach in that it provides interesting insight into what we would now consider to be primitive culture systems and then ties those activities into modern aquaculture approaches. The book is edited and authored to a considerable degree by Barry Costa-Pierce who has been one of the leaders in the discussion of responsible and sustainable aquaculture development. A prolific author of often thought-provoking articles, he has assembled some of the other leading thinkers in the field to provide a pot-pourri of information spanning the spectrum from artisanal to high technology approaches to producing aquatic animals with an eye on maintaining balanced ecosystems. The approaches advocated in this volume represent the future of aquaculture around the world. Criticisms will continue to be lodged by opponents, but the fact is that if the demand for seafood is to be met in the future, a major source will have to be from aquaculture as capture fisheries are currently being exploited at or beyond maximum sustainable yields. Aquaculture often cannot be practiced without some

Foreword

xi

environmental impact, but that impact can be reduced, hopefully to insignificance, if the proper approaches are adopted. This book helps define those approaches. It should be required reading for anyone interested in producing aquatic organisms in an environmentally responsible manner; which means that it should be read by everyone involved in aquaculture. While the minds of the opposition may not be changed by volumes such as this, those members of society who are interested in the facts concerning how environmentally responsible aquaculture has been practiced in the past, how it is currently being practiced, and where it might be going in the future will find this book to be an excellent primer on the topic. Robert R. Stickney Director Texas Sea Grant College Program Texas A&M University College Station, Texas

Preface

`It is the arrogance of the rich to teach the virtue of poverty to the poor.' Dr P.M.S. Blackett, the late Nobel Prize-winning physicist `Don't underestimate the power of a small group to change the World. In fact, that's the only ones who ever have.' Dr Margaret Mead Ecological aquaculture is an integral part of our common planetary wisdom and cultural heritage, and is an essential part of our future evolution as a sophisticated species living in peace with the Earth's complex ecosystems. Traditional aquaculture systems are closely integrated with the management of indigenous land and water food production systems, and rely on intact natural ecosystems. Aquaculture ecosystems evolved as sophisticated forms of agriculture in areas of Asia, the Pacific and Europe where human populations overshot the carrying capacities of the traditional agro- and aquatic ecosystems to support these societies. Aquaculture evolved to take pressure off natural and cultivated land and water ecosystems and ecosystem services. In these `human ecosystems', aquaculture evolved as a part of ± not the dominating feature of ± the wonderful variety of aquatic life, and the evolutionary diversity of cultures that decided to undertake such an extraordinarily specialized art. `Blue revolutions' ± as the natural evolution of societies ± have been happening for over 2000 years. In this regard, the concept of `ecological aquaculture' is nothing new ± especially in Asia. In the West, however, the evolution of the `blue revolution', and avoiding the social and environmental blunders of the `green' one, are more recent. Development of an aquaculture ecosystems pedagogy and systems `mentality' began only in the late 1970s and 1980s (MacKay, 1983). The purpose of this book is to stimulate discussion among aquaculture's modern scientific, education, extension communities ± and the larger aquatic resource management community ± about the principles, practices, and policies needed to develop ecologically and socially sustainable aquaculture. It is one of the first primers on ideas of how we are to become twenty-first century stewards of the Earth's cultivated aquatic ecosystems using the singular unifying rubrics of ecological aquaculture and aquaculture ecosystems. However, serious students of aquaculture will recognize that many of the ideas conveyed in this book are nothing new, but have

Preface

xiii

been repackaged in a way to stimulate discussion on how the latest science and outreach advances can assist the evolution of aquaculture into the modern era. The authors of the chapters in this book demonstrate how aquaculture can be a valuable player in the evolution of planning for sustainable aquatic resource management in a more crowded, protein-hungry world. To meet the protein needs for 8 billion people, and at the same time protect the oceans and freshwater ecosystems of the Earth from the uncontrolled exploitation of hunting (capture fisheries), aquaculture must expand dramatically in the twenty-first century. To accomplish expansion, however, aquaculture must be planned as part of, not separate from, a comprehensive management strategy for the restoration of fisheries ecosystems. Fisheries legislation throughout the world, such as the US Magnuson-Stevenson Sustainable Fisheries Act, falls far short of this vision, completely ignoring aquaculture in the planning for sustainable capture fisheries. To accomplish major worldwide expansion, and to ensure its social acceptance, aquaculture needs family and community roots in addition to corporate ones. Thousands of new, ecologically and economically sustainable family farms and progressive start-up companies need to be developed in the twenty-first century. These farms will need to practice `input management', recycle water, nutrients and materials, and produce healthy, uncontaminated products without discharges. Farmers incorporating aquaculture into family farms and new companies will internalize, not externalize, plans for more efficient resource recycling and enhance natural ecosystem services as part of their business plans and economic projections ± not neglect their social and environmental responsibilities, as suggested in a classic volume by Ken Boulding some 40 years ago (Boulding, 1962). Aquaculture developments need greater planning in the larger regional and community contexts. In short, aquaculture must become less short-term and less production oriented, and become more ecologically, community, and culturally based. Such approaches make good business sense! If aquaculture doesn't evolve with an environmentally friendly/socially responsible pedagogy in nations where it is new, but evolves principally as a `corporate' undertaking, aquaculture will never bring its full potential benefits. Environmental regulations, management difficulties, and resource and social conflicts coming in the crowded twenty-first century will halt its progress. And it is certain that the public ± worldwide ± will not accept any new forms of food production that exploit people, cause environmental harm, or produce new sources of aquatic pollution. In short, the `blue revolution' will quickly go bust unless it `greens up'. I conclude this book with these concerns about the role of aquaculture in the twenty-first century, contending that aquaculture needs to return to its historical roots ± its ecosystem and community-based roots ± and that ecological aquaculture needs to become the basic level of analysis for development planning for aquaculture worldwide. We contend that in this century, ecological aquaculture needs to emerge as the dominant method not only for smallholder farms, but also for commercial aquaculture. In addition, ecological aquaculture scholarship needs to emerge as a `new' field of applied environmental scholarship throughout the world at the major land and sea universities to assist aquaculture's rapid transition to social and

xiv

Preface

environmental sustainability, and to integrate aquaculture into mainstream planning for sustainable fisheries and coastal zone management. By adopting ecological principles as the basis of development planning aquaculture will play an important role in creating new social constructs that tie together thousands of new, knowledge-based family farmers and companies, producing huge benefits to society. Management of these `ecotones' between society, natural ecosystems, and sustainable environmental development is the key to the future of sustainable, ecological aquaculture.

References Boulding, K. (1962) The Reconstruction of Economics. Science Editions, New York. MacKay, K. (1983) Ecological aquaculture, new approaches to aquaculture in North America. Journal of the World Mariculture Society, 14, 704±713.

Barry A. Costa-Pierce University of Rhode Island Narragansett, RI

Acknowledgments

Ideas for this book were planted in the late 1970s while I studied with Murray Bookchin (Bookchin, 1985) and Jim Nolfi at the Institute for Social Ecology, Goddard College in Plainfield, Vermont. Murray and Jim were early pioneers in the study of the social ecology of food systems, alternative energy and social strategies. In addition, I am forever grateful to my teachers, who have helped me (and many others) to formulate the ideas and create the inspiration needed to produce this book, especially: Joseph Kiefer, Roger Pullin, Ron Zweig, Daniel Pauly, John Lyle, Ian Smith, John Bardach, Rich Merrill, Ken MacKay, Ed Laws, Lee Swenson, E.F. Schumacher, Amory Lovins, Eugene Odum, David Brower, Page Nelson, Margaret Mead, Bill McLarney, and John Todd. Special thanks go to all the pioneers at the New Alchemy Institute and the Farallones Institute of the 1970s and 1980s. This book would not be possible without the stimulation from a host of aquatic friends who've been `swimming in these same blue waters' ± special people who have served as examples to me and countless others by dedicating their lives towards giving the next generation a rehabilitated blue-green planet ± Pete Bryant, Bob Stickney, Otto Soemarwoto, Orten Msiska, John Wagner, Randy Brummett, Dave Penn, Arlo Fast, Sutandar Zainal, Reg Noble, Pam Sager, Pepen Effendi, Gelar Wiraatmadja, Jean Davidson, Reg Noble, Joseph Ofori, Jay Maclean, Jim McVey, Gail Work, John Lyle, Glenn Jones, John Munro, Sarvahara Judd, Fredson Chikafumbwa, Chris Bridger, Earl Barnhart, Hilde Maingay Barnhart, Russell Cuhel, Dave Karl, Catalino de la Cruz, Joseph Weinstock, Dan Chodorkoff, Charles Woodard, Calley O'Neill, Dick Jacobs, Daniel Jamu, Don Heacock, Spencer Malecha, Rick Weisburd, Clive Lightfoot, Mark Prein, Jim Rakocy, Les Behrends, Bill Engler, and Anne van Dam. Special thanks to Kay Bruening for all her professional assistance in preparing the book for publication; and to the staff of the Mississippi-Alabama Sea Grant Consortium.

Reference Bookchin, M. (1985) The Ecology of Freedom. Cheshire Books, Palo Alto, CA.

Part 1

The Background of Ecological Aquaculture

Chapter 1

The History of Aquaculture in Traditional Societies Malcolm C.M. Beveridge and David C. Little Institute of Aquaculture, University of Stirling The origins of aquaculture The origins of aquaculture are lost in history and little evidence remains to direct even a serious investigator of the subject. There are, after all, no aquaculturespecific artifacts to guide archaeologists. There is often little to distinguish abandoned ponds, even supposing we were able to find them, from dams of various types or from systems for producing inundated arable crops such as rice. Surface water would be stored to support communities or households in many cultures, and the predominant use for domestic/agricultural purposes may disguise a secondary role of holding or growing fish. Where rainfall is seasonal the focus of communities is around surface water bodies that may have originated as little more than natural depressions but became modified by the dependence these communities had on them. The presence of fish bones and shells in refuse heaps at early human settlements, or representations of fish on cooking pot shards, indicate only that the occupants ate these foods, not how they obtained them. The tools used in aquaculture are common to farmers and fisher folk, and remains of net-like materials or hooks tell us only about how fish met their fate, not whether they were caught from a river or from a fishpond. With few exceptions there is also no genetic record of domestication to draw upon, an important distinction between fish and livestock being that fish didn't need to be domesticated in the way livestock did in order to rear them in captivity. In this chapter, we examine what is known about the origins and development of aquaculture among traditional societies: who practiced it, and why, what distinguishes it from industrial and post-industrial aquaculture and its legacy. We briefly consider agriculture and current theories concerning its origins and development. We draw upon examples from traditional societies in four continents ± Africa, the Americas (with the exception of Hawaii, which is considered in detail in the following chapter), Asia and Europe ± and from various periods in history. Unfortunately, our review is not entirely balanced as there remain enormous gaps in knowledge. We begin with a consideration of what aquaculture is and how it differs from hunting.

4

Ecological Aquaculture

Aquaculture: some definitions It is important at the outset to be able to distinguish aquaculture from fisheries and agriculture. It may be differentiated from fishing because, as in agriculture, some measure of care or cultivation is involved. Reay's (1979) definition of aquaculture as `Man's attempt, through inputs of labour and energy, to improve the yield of useful aquatic organisms by deliberate manipulation of their rates of growth, mortality and reproduction' appeals from a biological perspective. However, it omits the other key component that distinguishes it from hunting: the concept of ownership or the extension of access and exploitation rights. The recently revised definition of aquaculture used by the UN Food and Agricultural Organization states that aquaculture is: `. . . the farming of aquatic organisms including crocodiles, amphibians, finfish, molluscs, crustaceans and plants, where farming refers to their rearing to their juvenile and/or adult phase under captive conditions. Aquaculture also encompasses individual, corporate or state ownership of the organism being reared and harvested . . .' (Rana, 1998). In this definition, both husbandry and ownership are seen as intrinsic. However, many traditional forms of aquaculture are based on the exploitation of multipurpose water bodies in which the organisms themselves are `common property', i.e. `owned' neither by an individual nor by some corporate body or the state. For present purposes we assume the key criteria distinguishing farming from hunting are that: there is some form of intervention(s) to increase yields; and l there is either ownership of stock or there are controls on access to and benefits accruing from the interventions (for parallels in agriculture see Bromley, 1992). l

Another key point in the FAO definition: end purpose is not at issue and fish owned and reared other than for food are regarded as the products of aquaculture. Nevertheless, as is discussed further below, differentiating between hunting and farming in the aquatic environment remains fraught with difficulties, in large measure because comparatively little effort has been expended on documenting and analyzing the range of methods used in exploitation of aquatic environments. A further set of definitions is necessary in order that we can compare aquaculture in traditional societies with contemporary practices from an ecological standpoint: these relate to resource use, or the differences between `intensive', `semi-intensive' and `extensive' aquaculture. According to Coche (1982), in extensive aquaculture the aquatic animals must rely solely on available natural food, such as plankton, detritus and seston. Semi-intensive aquaculture involves either fertilization to enhance the level of natural food in the systems and/or the use of supplementary feed. Such feeds are often low-protein (generally <20% DM), usually compounded from locally available plants or agricultural by-products, and complement the intake of natural

The History of Aquaculture in Traditional Societies

5

food which is higher in protein (Hepher, 1988). In intensive aquaculture, animals are almost exclusively reliant on an external supply of high-protein food (generally >20%), usually based on fishmeal and fish oil. These definitions broadly relate to use of environmental resources ± so-called `goods' ± although they ignore other resources such as land and water and seed. The intensity of production methods also has implications for use of environmental `services'; the more external food that is supplied per tonne production, the greater the wastes and the greater the demands on the environment to disperse and assimilate these wastes. However, the terminology is insufficiently well defined to be any more than a general guide. The term `semiintensive aquaculture', in particular, covers a huge diversity of aquaculture practices and ranges from minimal inputs to fairly substantial inputs of feed.

The origins of agriculture `People did not invent agriculture and shout for joy. They drifted or were forced into it, protesting all the way.' Tudge (1998) Neanderthals, Bandits and Farmers

Introduction The long accepted view of how and why agriculture began and spread has recently undergone some revision (Harris, 1996; Diamond, 1997; Tudge, 1998). While the established view remains that it started some 10 000 to 12 000 years ago, there is a growing recognition that different peoples adopted food production at different times. While some cultures, such as the Chinese, developed agriculture independently, others learned from neighbors or colonizers while a few, such as the Aboriginal Australians, appear never to have acquired agriculture at all (Flannery, 1994). The advantages of farming are readily apparent. Much plant and animal biomass is difficult or dangerous to harvest, labor intensive to prepare or of poor nutritional value. By farming, it is possible to select and grow crops and animals that give high nutritional returns per unit expenditure thereby increasing food supplies. Agriculture also generates food surpluses and food storage, prerequisites to the development of settled, politically centralized, socially stratified, economically complex and technologically innovative societies (Diamond, 1997). However, Tudge (1998) has recently argued that this view still underplays the importance of farming throughout much of our two-million-year history, especially from the late Paleolithic ± some 40 000 years ago ± onwards. He believes that a variety of `proto'-farming activities, a term coined to describe an ad hoc collection of activities that coaxed more food out of the environment, such as crop protection and game management, were part of the repertoire of responses to times when demands for wild foods outstripped supplies. Tudge also contends that when food supplies improved through upturns in abundance of game or more clement weather, or death or emigration of people, they returned to what they enjoyed best: hunting. His contention is that farming was hard work and to be avoided unless absolutely necessary.

6

Ecological Aquaculture

Although he makes little reference to farming prior to 12 000 years ago, Harris (1996) too believes in the concept of `proto-agriculture' and has elaborated an evolutionary classification of systems of plant and animal exploitation; a simplified, unified history of agriculture (Fig. 1.1a). There is much food for thought in these ideas and we believe that many of the concepts currently being considered by agricultural historians, especially with regard to terrestrial animal production, provide insights into aquaculture and into how it evolved and was practiced among traditional societies.

Fig. 1.1a An evolutionary classification of systems of animal exploitation (modified from Harris, 1996). Long seasonal migrations, examples including permanent winter settlements and summer migrations.

Aquaculture and `proto-aquaculture' Aquaculture too began in different societies, both agriculture- and fishing-based, and followed a pattern of development in many respects similar to that of agriculture. There is good evidence in aquaculture for Tudge's theory of people opting in and out of plant and animal cultivation according to their needs, although as we hope to show these needs were not always related to food. There is also evidence of `protoaquaculture', defined here as activities designed to extract more food from aquatic environments, such as: transplantation of fertilized eggs; entrapment of fish in areas where they could thrive and be harvested as required; l environmental enhancements, such as development of spawning areas, enhancement of food, exclusion of competitors or predators, etc.; l holding of fish and shellfish in systems ± ponds, cages, pens ± until they had increased in biomass or until their value had improved. l l

Each activity might on its own be considered as no more than stock enhancement and thus within the definition of what might be considered as managed fisheries. However, we apply the critical concept of control of access and benefits to draw a line between where managed fisheries end and proto-aquaculture begins. If the effect of such interventions increased supplies sufficiently to satisfy needs and to result in an equitable distribution of benefits, then it may have evolved no further. We propose that the relatively small degree of control over the life cycle of the animal and the low impact of the intervention on fish or shellfish production is used to distinguish proto-

The History of Aquaculture in Traditional Societies

7

aquaculture from aquaculture. The definition of proto-aquaculture is compatible with those of Coche (1982) regarding extensive, semi-intensive and intensive aquaculture and is further characterized by low consumption of energy (see above). The proposed definitions do not neatly distinguish fishing from fish farming, but perhaps this is only to be expected when dealing with something like the transition from hunting to farming. It also ignores some of the more contentious issues, such as ranching, a term used to describe the release of juveniles into the wild, only to be recaptured later as adults. However, it is useful in helping explain how aquaculture might have first developed. It is very much a working hypothesis and others with more detailed and accurate information, more insight and more time for reflection and debate will undoubtedly construct a better framework.

Proto-aquaculture and the origins and pattern of development of aquaculture Why might people in traditional societies have begun farming fish and shellfish? The answer, according to the agricultural historians, would be because it was necessary (Boserup, 1965). It is clear that aquaculture began in various parts of the world and at various points along the aquatic food supply line, between water and plate. The farming of fish and shellfish is by definition an activity of settled societies, originating among both fishing and wetland farming cultures as well as at points of trade. While we may surmise that conditions similar to those favouring the development of agriculture would have usually been necessary, i.e. that foraging and hunting (fishing) were insufficient to satisfy demands for fish, provision of food was not always the most important driver for the development of aquaculture. Stewart (1994) and others believe the importance of fish in early hunter±gatherer societies to have been underestimated. Rudimentary proto-aquaculture techniques would probably have evolved among such societies, although evidence is scant. Native North American peoples living on the Pacific seaboard are believed to have transplanted the eggs of spawning salmon in an attempt to improve fish survival and returns. Many proto-aquaculture activities relied on some sort of holding facility. The simplest to construct would have been earth ponds. In some parts of the world these would have been little more than mud walls constructed to temporarily hold water and fish following the seasonal flooding of a river. Such systems are still in use in some parts of the world today. The whedos or fish holes of Benin are one such example (see Welcomme, 1972, for details). The practice of communal construction of weirs on small rivers and streams in Asia to store water outside of the monsoons principally to ensure adequate irrigation for wet rice cultivation during lulls in the rains and allow early seed bed preparation is also common. Attempts to increase fish yields would have been a logical next step, by affording protection from predators and, perhaps, by feeding fish with household scraps or farm wastes (see Fig. 1.1b). Among fishing-based societies, a number of scenarios in which proto-fish or protoshellfish farming arose are readily envisaged: the short-term storage of catches until there were sufficient fish or shellfish to make the journey to market worthwhile; the transport of live fish to market; the holding of catches until prices improved. These

8

Ecological Aquaculture

Fig. 1.1b A scenario of how aquaculture may have evolved.

strategies are still seen among fisher folk today: modified traps, netted-off shallow areas of lakes, cages of the sort still seen in parts of Indonesia, traditional floating cages used in the Great Lake area of Cambodia (see Beveridge, 1996). If the theories of the agricultural historians hold true for aquaculture, then we can also expect to see aquaculture wax and wane as the result of changes in supply relative to demand for fish and shellfish.

Africa The earliest evidence of fish culture of sorts purportedly comes from ancient Egypt where fish often had a sacred as well as prosaic role in society. They were strongly associated with the cyclical life-giving forces of the Nile and the New Kingdom Egyptian view of the world, tilapia in particular being strongly linked to the goddess Hathor and the concept of rebirth (Desroches-Noblecourt, 1954). In his account of tilapia in ancient Egypt, Chimits (1957) reproduces a 4000-year-old bas relief figure from the tomb of Thebaine showing what appears to be an artificial, drainable pond being fished by a nobleman (Fig. 1.2). Many New Kingdom tomb scenes also show tomb owners sitting on chairs, fishing tilapias from their ponds, their wives standing behind and assisting with the catches. Although rod and line fishing is believed to have been common among all classes in Egypt at that time, the fishing activities of the

Fig. 1.2 Bas relief from the tomb of Thebaine, showing an Egyptian nobleman catching tilapia from an artificial pond (re-drawn from Chimits, 1957).

The History of Aquaculture in Traditional Societies

9

nobility were limited to their ponds. Their interest in fishing stemmed from religious rituals associated with death and rebirth and not with pleasure or sustenance (Desroches-Noblecourt, 1954; Brewer & Friedman, 1989). This was aquaculture at its most simple, more proto-aquaculture than aquaculture, involving little in the way of inputs or husbandry or pond management. Tilapia would have been transferred from nearby rivers to the ponds where they readily would have bred. While some food may have been provided, it is unlikely to have been important since production of fish for food was not the objective. It is believed that the practice persisted into the New Kingdom until the importance of rebirth in the world view of Egyptians waned. Although Brewer & Friedman (1989) detail many peculiar beliefs and taboos among the priesthood associated with fish, early travelers to Egypt confirm that fish was of tremendous importance in the Egyptian diet. The Roman traveler Diodorus Scullus is quoted as saying that `. . . the Nile contains every variety of fish and in numbers beyond belief: for it supplies the native not only with fish freshly caught but also yields an unfailing multitude for salting.' Herodotus, too, who traveled here some 2500 years ago, reported that `. . . all Egyptians in the Nile Delta possess a net with which, during the day, they fish. . . .' Given the fertility of the river, the abundance of fish and the skills of the fishermen, it is not surprising that these proto-aquaculture activities developed no further and indeed waned with the change in religious significance of fish. Its revival would have to wait several millennia until the early years of the present century.

Asia China and freshwater fish farming Although possibly pre-dated by events in Africa, Asia ± and China in particular ± is widely regarded as the cradle of aquaculture. The chronology is complicated, but is summarized in Tables 1.1 and 1.2. Many factors are thought to have constrained livestock development and predisposed China to develop aquaculture earlier than elsewhere in Asia. The wet rice agro-environment evolved relatively late in China's history, population pressure stimulating colonization of low-lying deltas. Such conditions would have both inhibited any development of mixed farming based on ruminant livestock and crops and supported fish culture and production of livestock such as pigs and poultry that thrived on rice by-products and water-based scavenging respectively. The process of agricultural evolution in southern China, from cropdominated to mixed farming (Grigg, 1974; Little & Edwards, 1997) was, therefore, molded by the limits and potentials of a flood-prone environment. Any diversification from a rice monoculture required a process of `ditching and diking' that produced deeper areas suitable for fish and higher dike areas for horticulture. Although areas of wet rice production were relatively sparsely populated until the Han dynasty (Grigg, 1974), the adoption of increasingly intensive, and irrigated, production in suitable areas prompted a rapid increase in population and demand for aquatic products that would also have been an important factor in stimulating aquaculture. With the development of ditch-dike systems, other crops such as beans, green vegetables and

10

Ecological Aquaculture

Table 1.1 Chronology of significant aquaculture developments in China, based on written records. Information derived from Lin (1991), Li (1994) and Yang (1994) Date (BP)*

Period

Event

2300

Zhou Dynasty

Publication of monograph by Fan Li, detailing design and layout of ponds, propagation, fry and fingerling production

2200±2100

Han Dynasty

Publication of You Hou Bin, detailing integration of fish with aquatic plant and vegetable production Development of cage, cove and pen culture

1975±1780 1380±1100

Culture of fish in rice paddies Tang Dynasty

1100

Development of polyculture Publication of Lin Biao Lu Yi by Liu Xun, detailing theory of mutualism in rice±fish culture Integration of fish and fruit production

1040±780

Song Dynasty

Increasing collection and distribution of wild fish fry for pond rearing

632±350

Ming Dynasty

The Complete Agricultural Art written by Xu Guangqi and Treatise on Fish Culture by Huang Shenchen detail extensive to intensive fish farming methods in Jiangxi Province, including rotation of fish with aquatic plant production; integration of fish with livestock and effects of manuring on pond fish production

500 360±90

Development of integrated mulberry±dike±fishpond production system Qing Dynasty

40

Detailed written accounts of fry production, sorting and transportation Widespread success in artificial propagation of carps

* BP = Before present.

Table 1.2 Schema showing the evolution of farming systems for livestock and fish. It illustrates the probable evolution of farming systems from traditional, crop-dominated systems (Settled Agricultural Phase I) through mixed farming in which the importance of livestock was enhanced through their integration with crops (Settled Agricultural Phase II), to industrial agriculture characterized by monoculture (Settled Agricultural Phase III). (From Little & Edwards, 1997.) Category System definition

Evolutionary stage/trend Traditional crop dominated

Mixed farming

Agro-industrial

Intensity level

Extensive

Semi-intensive

Intensive

Descriptor

Settled Agricultural Phase I

Settled Agricultural Phase II

Settled Agricultural Phase III

Livelihood

Part of a complex of activities

Specialized activity

Knowledge

Indigenous

Scientific

Resource use

Land, water*

Cash, fossil fuel energy intensive

Market

Rural, subsistence, local

Urban, cash, export

Cultured species

Polyculture

Monoculture

* Based on pond fish culture.

The History of Aquaculture in Traditional Societies

11

tobacco could also be grown alongside (Bray, 1984). As today, fish would have been a common component of seasonally inundating rice paddies. Given the lack of animals and the prevalence of wetlands it is thus not surprising that fish was a prominent component of the diet in many areas. The development of irrigation in China and elsewhere was driven by the fact that rice grows best when provided with water of the right quality and in the right quantity at the right time. While there is no neat chronology of evolution of irrigation systems in China, it is clear from the widespread existence of clay models of irrigated agriculture systems recovered from graves throughout southern China that by the Han Dynasty (2300±1700 BP) ponds were being widely employed for water storage (Bray, 1984; Li, 1994). To some this indicates the earliest that aquaculture might have been developed. In a single grave over 18 varieties of aquatic plants and animal, that are still used by the Chinese today, were found within an intact rice-field model. These included lotus flowers, seeds and leaves, water chestnuts, soft-shelled turtles (Trionyx sinensis), grass carp (Ctenopharyngodon idella) and goldfish Carassius auratus (Guo, 1985, in Li, 1992). These areas of southern China had high densities of people culturally dependent on aquatic foods. As population densities increased, demand for fish and other aquatic foods would have increased and the practice of holding and growing fish would become increasingly attractive compared with reliance on increasingly exploited and inconsistent wild stocks. In the floodplains of China and elsewhere in Asia, soil is excavated to construct elevated, better-drained areas for establishing homesteads and raising crops. Although the resultant excavations may be referred to as fishponds by aquaculturists, farmers refer to them simply as ponds, an indication of their multipurpose nature. Others, however, including the Fisheries Society of China, refer to the short treatise published by the statesman Fan Li some 2500 years ago (2500 BP). It describes common carp (Cyprinus carpio) farming in sufficient detail to provide incontrovertible evidence that fish culture had developed well beyond a proto-aquaculture activity and that aquaculture was well established by this time (Li, 1994). The monograph details the design and layout of fishponds, carp breeding, and fry and fingerling rearing techniques. Fan Li's account is of `semi-intensive' monoculture of carp, although there remains some debate as to the species (see Balon, 1995). The integration of fish ± presumably carp ± culture with that of aquatic plants and vegetables is apparent from written records dating from 2200 to 2100 BP while written records of rice-fish culture date from the period 1975±1780 BP (Yang, 1994). Despite the long history of freshwater fish farming in China, there are few documentary accounts and details are fragmentary at best (see Table 1.1). According to Li (1994) fish culture expanded from rice paddies and ponds to lakes ± this implies the use of cages, pens and/or enclosures ± during the Han Dynasty (206 BC±AD 220). Provided herbivorous species were used, culture could have relied on semi-intensive or even extensive methods. If, however, omnivores were farmed, then there would have been a greater reliance on supplementary feed. There is strong evidence that small dams, constructed by farmers primarily for water storage purposes, were also used to produce lotuses, water chestnuts, fish and turtles (Bray, 1984). A milestone for aquaculture in China seems to have been reached at the beginning of the Tang

12

Ecological Aquaculture

Dynasty in AD 618 with the culture of combinations (polyculture) of carps (Ling, 1977; Li, 1994). However, while we might expect that this would have been promoted as a means of increasing yields from semi-intensively managed systems, historians claim polyculture to have been less due to an appreciation of ecology and the synergies of growing together species with complementary feeding habits than to the fact that the word for common carp in Chinese ± `Li' ± sounded the same as the emperor's surname. As a result, the catching, selling and eating of this species was banned for the next 300 years or so. Availability of suitable wild seed was critical to China's aquaculture development, and the same was true for Indian carp polyculture in the subcontinent and culture of Vietnamese silver carp (Hypophthalmichthys harmandi ), mud carp (Cirrhinus molitorella) and common carp (Cyprinus carpio) in the Red River Delta, Vietnam (Chevey & Lemasson, 1937). There is also evidence that there were traditional methods of producing seed that pre-date the now widespread use of hypophysation in hatcheries. Dry `bundhs', seasonal ponds that fill quickly at the time of the first rains, were used to stimulate spawning of Indian major carps in West Bengal over a hundred years ago (Sharma & Rana, 1986). Other key events in Chinese aquaculture include the gradual integration of fish ponds with various crop and livestock production systems (see Table 1.1), leading to what is widely regarded as the most complex integrated aquaculture system of all, the fishpond±dyke±mulberry system of Zhujiang, southern China (Ruddle & Zhong, 1988). The fishpond±dyke±mulberry system was strongly output orientated to meet both local demand for a variety of products ± live fish, fruit, etc. ± and distant markets for products such as silk. Until recent decades the yields of Chinese carp polycultures are likely to have remained low (<2 metric tons per hectare per year), however, and mainly dependent on feeding low nutrient, volunteer fodder to the grass carp. This situation, as indicated by the oft-quoted Chinese proverb `One grass carp feeds three other fish', reflects the paucity of, and competition for, nutrient inputs on traditional farms in southern China. The macrophagous grass carp, whose manure enhanced the productivity of food organisms for other fish species in the pond, filled a `niche' usually filled by ruminants, for such low value feed (Little & Edwards, 1997, 2000). Collectivization of agriculture in the 1950s saw great changes in the types of nutrient inputs, labour and markets available and a resultant increase in complexity of Chinese integrated aquaculture. The influence of China on aquaculture in the region has undoubtedly been great, but poorly studied. Chinese emigrants may have taken common carp Cyprinus carpio and techniques of fish farming with them first to Korea and then to Japan (Drews, 1951; Ling, 1977), from where McLarney (1984) believes the practice gradually spread to other parts of Southeast Asia. The culture of the exotic common carp with indigenous anabantids from swamps, such as giant gourami (Osphronemus gouramy), kissing gourami (Helostoma temmincki) and snakeskin gourami (Trichogaster pectoralis), and the riverine carp, silver barb (Barbodes gonionotus) and nilem (Osteochilus hasselti) in Indonesia, is believed to stem from their introduction by Chinese immigrants over two centuries ago (Edwards et al., 1997). Until recently, ethnic Chinese immigrants growing fish around urban centres imported Chinese carp in well-boats and, latterly, by air. Nevertheless, as we say above, the creation of exca-

The History of Aquaculture in Traditional Societies

13

vated areas in floodplains, and the presence of high densities of people used to exploiting aquatic foodstuffs are all the conditions necessary to initiate proto-aquaculture/aquaculture. There is evidence that away from the major floodplains wild stocks of aquatic animals are more prone to over-exploitation and that the stocking of juvenile and breeding wild fish was more likely to be a response to local fish shortages in these circumstances. The stocking and management of local common carp in ricefields in mountainous northern Laos and Vietnam (Chevey & Lemasson, 1937, Little & Pham, 1999) is an example. The accumulation of land and other resources by elites may also be a prerequisite to the development of aquaculture in agricultural societies. The construction of large ponds would require mobilization and organization of labor beyond the resources of poorer households. This perhaps explains the domination of large Hindu landowners in traditional aquaculture in Bengal (Lewis et al., 1996). There are almost certainly rich histories of aquaculture development in India, Indonesia and elsewhere. Unfortunately, these are much more poorly documented than in China (see Table 1.3 for summary chronology for Southeast Asia).

Mariculture Documentary evidence suggests that mariculture in Asia is less than 1000 years old. The culture of milkfish (Chanos chanos) in brackish water coastal ponds is believed to have developed in Java, Indonesia, some time between AD 1200 and 1400 under the influence of the Hindu rulers (Schuster, 1952). Convicts were apparently sent to the coastal fringes to work in the salt marshes and to guard coastal fires. They were forbidden to practice agriculture or trade and their conditions were so harsh that they are believed to have survived largely through damming of creeks and trapping of fish and crustaceans. Schuster (1952) believes coastal pond fish farming evolved among these `untouchable' fish-eating peoples from the seasonal flooding of salt ponds during the wet season. Mariculture developed independently in Japan during the Tokugawa era (1603± 1868) (Ling, 1977; McLarney, 1984). Some traditional forms of coastal mariculture, such as the use of barachois in Mauritius, are still practiced to a limited extent. Here, shallow lagoons, isolated from the sea by stone walls fitted with screens, are stocked with mullet (Mugil cephalus) or rabbit fish (Siganus sp.) fingerlings. Some oyster farming may also be practiced.

Europe Introduction Historical evidence suggests that mariculture in the Mediterranean pre-dates that in Asia by a millennium or so. The Etruscans practiced active management of coastal lagoons along the Adriatic and Tyrrhenian coasts in the fourth and fifth centuries BC. These proto-aquaculture activities evolved into what is generally known today as `vallicoltura', involving the installation of permanent or semi-permanent embankments to enclose the lagoons. Sluices were incorporated within the structures in order

14

Ecological Aquaculture

Table 1.3 Traditional and modified artisanal inland aquaculture systems in Southeast Asia. Historical is >100 years; recent is <100 years old. Improved implies benefits from scientifically derived knowledge. (Modified from Edwards et al., 1997.) Traditional Country/area

Culture facility

Historical

Brunei Cambodia Indonesia

Ponds Cages and pens Rice±fish Cages Ponds

+ + + +

Laos PDR Mountains Lowland Malaysia Myanmar Papua NG Philippines Singapore Thailand Vietnam Mountains Red River Delta Mekong Delta

Rice±fish Rice±fish Ponds Rice±fish Ponds Ponds Ponds Rice±fish Cages and pens Ponds Ponds Rice±fish Cages and pens Ponds Rice±fish Cages Ponds Ponds Rice±fish Cages Ponds Rice±fish Cages

+

+

+ + + + + +

Recent

Improved

+

+ + + +

+ + + + + + + +

+ + + + + + + + +

+

+ + +

+ +

+ + + + + + +

+ + + +

to trap fish inside the shallow, productive lagoon environment where they could be harvested when desired. This form of mariculture is still practiced to a limited extent around the Northern Adriatic, Tyrrhenian Sea, Sicily and Sardinia (Kirk, 1987; Ardizzone et al., 1988). Shellfish farming, involving the culture and relaying of oysters, was well established in the Adriatic 2100 to 2200 years before present (Balon, 1967b, in Balon, 1995), but probably pre-dates this by many hundreds of years. The Romans are documented as being among the first to build coastal aquaculture ponds, most likely before the end of the second century BC (2200 BP). In his Natural History (Vol. IX) written some hundred years later, Pliny the Elder notes that a certain Lucinus Murena had excavated substantial areas of fish ponds (piscinae) at Grotta Ferraia, close to the summer residence of Cicero. This idea was apparently adopted by other members of the nobility, partly for the purpose of holding live food fish, but also as a demonstration of wealth and status. Pliny the Elder also recounts that:

The History of Aquaculture in Traditional Societies

15

`. . . at Baculo in Baiae District, the pleader Hortensius had a fishpond containing a lamprey that he fell so deeply in love with that he is believed to have wept when it expired. At the same country house Drusus' wife Antonia adorned her favourite lamprey with earrings, and its reputation made some people extremely eager to visit Baculo.' Aquaculture perhaps, but scarcely for food! However, while these saltae or maritimae were built by the rich nobility primarily for spectacle, a practice that was widespread and persisted for several centuries, common people also built ponds (dulces) for food production for income generation (Kirk, 1987; Zeepvat, 1988). Both freshwater and saltwater ponds were built, the former being stocked with a variety of coarse fish and salmonids, the latter with eel, mullets, turbot and sea bass. According to Zeepvat, classical Roman literature gives the impression that the keeping of fish in artificial ponds was commonplace throughout the Mediterranean provinces of the Empire.

Continental Europe The common carp (Cyprinus carpio) was undoubtedly key to the development of aquaculture in continental Europe over much of the past two millennia. During the first and second centuries AD, Roman attempts to culture local fishes in piscinae on a significant scale to increase supplies of fresh fish failed and as a result the common carp from the Danube were imported (Balon, 1995). Gradually the practice of common carp culture developed by the Romans spread westwards, although it did not reach England until the late fourteenth century (Currie, 1991). The monasteries, the repository of knowledge throughout the Middle Ages, were undoubtedly important in the development of fish ± especially carp ± culture among the farming communities of central and Western Europe. There are several reasons for this. First, fish played an important symbolic role in early Christian motifs, fish representing Christian souls, and ponds of fish are common in mosaics decorating early Byzantine churches (Drewer, 1981). Second, a number of the Christian monastic orders, such as the Cistercians, were skilled agriculturists. The Cistercians also led in the development of fulling mills in which the processing of large quantities of wool took place. The requirements for waterwheels to drive the heavy timber and stone hammers that pounded the wool to clean and thicken it, in the mid-fourteenth century led to the construction of dams that were usually stocked with fish (Binnie, 1987). According to Binnie (1987), the Cistercians and other religious orders influenced the populace at large to adopt the practice of eating fish on Fridays and during Lent. The growing importance of fish and the preference for fresh rather than salted led to not only the royal and ecclesiastical establishments but also wealthy landowners to construct `stew' ponds. No single authoritative history of fish culture in Europe has yet been produced, the principal reason being the lack of study. Hoffman (1996), however, has marshaled current knowledge and understanding to give an excellent preliminary synthesis of pond fish culture. Hoffman associates the development of pond fish culture with

16

Ecological Aquaculture

environmental degradation and population growth that led to growing demand for freshwater fish. By the reign of Charlemagne (AD 768±814) carp culture was sufficiently important for the emperor to have issued edicts to tenant farmers regarding the maintenance and management of ponds, protection from poaching, the regulation of fishing and the sale of fish (Balon, 1995). By medieval times, fish had become an important staple in the diet (Hoffman, 1996). Initially, fishponds seem to have been built in Europe to ensure supplies of fish for the owner(s) and his/their dependents, especially during Lent when eating of meat was prohibited. Small fish were returned to the ponds to ensure continuity of production. Any surpluses tended to be used as gifts to cement social relations or for sale, although as Hoffman (1995) states, medieval fishponds were not managed to cater for what were undoubtedly considerable markets for fish for stocking and for consumption. A number of the medieval power dams, such as the Cistercian-owned Lucelle dam in northwestern Switzerland and the Bruna dam, Sienna, Italy, were also concurrently used for fish culture (Schnitter, 1994). The latter was owned by the State, which had a certain responsibility to ensure there was sufficient food for its populace and woe betide any that failed in this (Adams, 1984; Hoffmann, 1996). Economic growth during the fourteenth to sixteenth centuries promoted widespread development of fishponds throughout much of continental Europe (Adams, 1984; Schnitter, 1994) although for reasons given below not in England. Carp farming was very extensive and often practiced in modified floodplains difficult to exploit profitably except for pasture. In Hungary, for example, ponds were often located in areas of poor, unproductive soils that were likely to flood. The ponds ensured a continuous supply of food, particularly during times of war, feeding both rich and poor. In the Czech Republic, in the late fourteenth century there were 75 000 ha of ponds although production was very low, yielding 2250 mt of fish (Adamek & Kouril, 2000). Bohemia appears to have had the greatest concentration of ponds. During the fifteenth and sixteenth centuries, some 700 ponds, totaling an area of 1800 km2, were built (Votruba, 1987). Ponds also flourished in the district of La Dombes, between Lyon and Bourg, France, for a period of several hundred years from the early Middle Ages, eventually covering an area of some 180 km2 (Sahrhage & Lundbeck, 1992). From the late sixteenth century onwards, pond fish farming in Germany and France fell into decline and did not begin to recover until the development of trout culture in the mid-nineteenth century. The causes were various. In Bohemia, they have been ascribed to over-production coupled with the pauperization of farmers that culminated in the Peasants' Wars of 1524 (Sahrhage & Lundbeck, 1992), while in the Czech Republic frequent wars and the development of agriculture were blamed (Adamek & Kouril, 2000). In France, the secularization of the church by Napoleon in 1803 is accorded some of the blame (Sahrhage & Lundbeck, 1992). The main reasons, however, were economic. Just as it had once proved highly profitable to convert farmland into fishponds, so the fortunes of fish culture turned full circle: increased profitability of other crops, coupled with State incentives to convert ponds into farmland, and a contracting market finished carp farming in many areas. The decline in the consumption of fish, a commodity always more expensive than meat, had

The History of Aquaculture in Traditional Societies

17

begun during the seventeenth century, first among the nobility and later among the middle classes.

England As in continental Europe, the Romans must be credited with the introduction of fish culture to England. At the time Zeepvat (1988) wrote his review, some dozen ponds, ranging in size from 30 m2 to 1770 m2, had been uncovered at Roman villas in southern England and there are undoubtedly more to be discovered. The earliest of the ponds dates from AD 65, the latest, at Bancroft, Buckinghamshire, from the first half of the fourth century AD. Many of the ponds were primarily ornamental in nature, while others were principally designed for bathing. However, many served a secondary purpose of production and storage of fish for the table and a few even are surmised to be primarily for food production: Zeepvat (1988) presents data from an excavation at Shakenoak, Oxfordshire, dating from the second century BC. Here a complex of ponds was uncovered in the 1970s, all of which seemed to have been purpose-built for fish production. One pond is surmised to have been a breeding pond, another a production pond, while a third is believed to have been used as a holding tank for fish awaiting shipment to market. The villas and their ponds were abandoned when the Romans left Britain in the fifth century AD. According to many historians, the greatest consumption of fish in medieval England was by the religious communities such as the Cistercians, in whose houses `meat and lard were never eaten . . . save for those seriously ill and by the hired workmen. The diet was therefore confined to bread, vegetables and dishes mainly from fish, flour, eggs and cheese, with honey as an exception. Throughout Lent and Advent and on certain vigils, eggs and cheese were banned, as were all kinds of imported spices' (Knowles, 1950). However, recent research by Currie (1989, 1991) has cast doubts on the traditional view that the monastic orders were primarily responsible for the development and dissemination of pond fish culture in England. There is much evidence to show that the ascendant Norman aristocracy built the first large-scale fish pond developments, primarily in order to cement their new-found status as landowners. There are records of fishponds from several areas of Norman England, dating back to the end of the eleventh century (Roberts, 1966; Stearne, 1971; McDonnell, 1981; Roberts, 1985; Binnie, 1987; Currie, 1991). In 1069 William the Conqueror, the French nobleman who had subjugated England only a few years before, ordered the River Foss at York to be dammed just above its confluence with the Ouse in order to protect his newly built castle, leading, incidentally, to the creation of a fishpond. Many more fishponds were built in England over the next two and a half centuries. Indeed, Roberts (1966) claims that `. . . fishponds were normal appurtenances to manors over much of lowland England during the middle ages . . .' and that `. . . few monastic establishments were without a supply of freshwater fish, frequently kept in ponds . . .' while `. . . fish production [was] on some scale from royal ponds during the thirteenth century'. By the late Middle Ages, there were at least 135 pond systems in Oxfordshire alone (Chambers & Gray, 1988). Other accounts highlight the role of royalty in pond fish production. Henry III

18

Ecological Aquaculture

(1216±72), for example, made numerous well documented gifts of live fish to his nobles, promoting the development of fishponds on the estates of the wealthy (McDonnell, 1981; Binnie, 1987). The ponds owned by the Cistercians, Augustinians and Gilbertines and other monastic orders were either granted by wealthy secular patrons in association with gifts of land or were built somewhat later during the medieval period and on a smaller scale. As mentioned above, fish were important in the early Christian church in both symbolic and practical terms. More than 100 fasting days per year were introduced, during which only cold-blooded animals such as fish and crustaceans could be eaten and even more elaborate rules regarding consumption of fish were followed by monastic orders (see Knowles, 1950). However, the idea that monastic fishponds provided much-needed fresh fish for the monastic orders during fasting days is rejected by Dyer (1988), Currie (1989) and others. They assert that cheap salted sea fish was eaten during periods of penance and cite contemporary records stating that fresh pond fish were to be kept for special occasions. The populace at large was rarely able to eat meat or fresh fish. Dyer (in Currie, 1991) states that fishponds were seen as bastions of privilege and often a focus of peasant discontent, such as during the Peasants' Revolt of 1381, in which the `stews of Southwark' were wrecked. The royal fishponds fell into decline during the period 1300±1480 due to the decline in the number of royal houses, the costs of pond maintenance and the relatively poor economic returns and because of the rising numbers of retail fishmongers (Stearne, 1988). With the dissolution of the monasteries in the sixteenth century, many monastic fishponds were abandoned, never to be productive again. In the latter part of the medieval period, a few wealthy freemen began to build ponds (Roberts, 1966; McDonnell, 1981; Currie, 1991). There is evidence that wealthy peasants owned ponds in the Forest of Arden, Warwickshire, for example, and that part of the produce was sold. Currie (1991) also cites evidence of fishmongers in mid-fourteenth century London feeding and keeping fish in ponds for sale. However, the practice could never be described as widespread. In post-medieval times, the keeping of common carp in ponds became popular for a while among land-owning nobility, the fish being used for both domestic consumption and sale (Currie, 1991). However, Taverner (1600), in his discourse on fish culture, mentions that fishpond keeping was no longer as popular as it once had been. In leasing estates to tenants, fishponds were often neglected. There was a decline in eating freshwater fish such as carp, bream and perch during the seventeenth and eighteenth centuries. According to Chambers & Gray (1988), the 1869 edition of the redoubtable Mrs Beeton's Book of Household Management states that freshwater fish are seldom purchased.

The Americas In pre-Hispanic Mexico the chinampas that developed and flourished in lakes Xochimilco and Texcoco in the Valley of Mexico under the Aztecs, represent only one of several integrated wetland agriculture/aquaculture systems (Coe, 1964). The term `chinampa' derives from the Nahuatl language and means `net of branches'. The

The History of Aquaculture in Traditional Societies

19

chinampas were created by cutting and piling turves on to the lake bottom around lake margins to form a patchwork of peninsulas and islands, each typically 100 m long by 5±10 m wide, interspersed by a grid-like pattern of canals. The key feature of the chinampas is that lake water could infiltrate the agricultural beds, maintaining a supply of water to the roots of the plants throughout the year. A wide range of crops was grown, including maize and tomatoes. Organically rich mud recovered from the canals would be piled on to the chinampas, thus maintaining fertility (Armillas, 1971). Although there have been many changes and much has been lost, the systems still exist and remain `the biologically richest agro-ecosystem . . . in which most of the flora is managed and used' (JimeÂnez-Osornio & Gomez-Pompa, 1991). Fish play no part in the present-day system, and although it is believed that they once did it is difficult to say how significant or to what extent fish production was deliberately enhanced. It is thought that fish were purposely trapped in the canals where they could readily be harvested as required, and that crop and vegetable by-products may have been used to enhance fish production. There were attempts to experiment with a more highly integrated form of chinampa-based aquaculture in Tabasco, Mexico, and in the Akanyaru marshlands of Rwanda in the 1980s, with mixed success (Micha & Chavez, 1997).

Fish farming methods Introduction Almost without exception, pre-twentieth century fish farming was of herbivorous/ omnivorous species such as carps and roach, milkfish, mullets and tilapias. Culture of piscivorous pike, as was practiced in Medieval European fishponds, was done in polyculture systems where the pike undoubtedly fed on smaller stocked carps and other coarse fishes. The monoculture of carnivorous salmonids was initiated in the nineteenth century as a response to the need to meet the growing demand of recreational fishing and to environmental degradation on the north Pacific coast of the Americas, and was limited to the production of juvenile stages for stocking. The culture of aquatic carnivores for the table had to wait until after the Second World War. Aquaculture for religious or for social status reasons aside, proto-aquaculture and aquaculture were about increasing fish and shellfish production. The fundamental principles involved can be derived from the figure of Pitcher & Hart (1982) for the factors that determine exploitable stock biomass (Fig. 1.3). To increase the harvestable yields from a stock, it is necessary to increase stock biomass (recruitment and growth) while minimizing losses through disease or predation. Stocking, increases in food supply, either through fertilization or provision of additional food, concern for animal welfare and exclusion of predators may all be employed in fish farming, although profitability is usually the ultimate criterion. Here we examine what was known about aquaculture methods among practitioners in traditional societies.

20

Ecological Aquaculture

Growth rate

Recruitment rate

Increase

Farmed stock biomass

Decrease

Harvesting mortality rate

Natural mortality rate

Fig. 1.3 Categories of factors governing exploitable farmed stock biomass (developed from Pitcher & Hart, 1982).

Broodstock management and spawning Early fish farming relied either on trapping or transplantation of wild fry (carp, milkfish, mullet, etc.) or on breeding and production of fish such as tilapia, roach, etc. within the production system. Accounts of catching and transportation of fry for stocking of fishponds are apparent from two written works dating from the early thirteenth century (Song Dynasty) in China (Li, 1994). How well did early fish farmers understand the life cycle and principles of farming fish? It is clear from the treatise on fish culture written by Fan Li some 2500 BP that the principles of carp breeding were well appreciated by that time in China (Li, 1994). McDonnell (1981) believes that in England the basic concepts behind holding fish in separate enclosures for store (`stews') and for breeding would have been familiar to medieval Cistercian sheep farmers, implying that in many of the larger fishpond systems there were separate breeding/fry rearing ponds and production ponds. Hoffman (1995) also cites evidence of separate carp fry rearing ponds or ditches associated with the ducal fishpond complex at LaperrieÁre-sur-SaoÃne, Burgundy, in the mid-fourteenth century. Most of the fish cultured in ponds in England at this time ± bream (Abramis brama), dace (Leuciscus leuciscus), perch (Perca fluviatilis), roach (Rutilus rutilus), tench (Tinca tinca) and pike (Esox lucius) ± (McDonnell, 1981) would have readily acclimatized to the muddy, stagnant pond environment, and, provided there were a few aquatic plants present, such fish would have happily bred. During harvest of medieval European fishponds, smaller fish were usually retained for restocking purposes (see Hoffman, 1995, for example). However, there is also extensive documentary evidence showing considerable and well organized movements of fish for stocking purposes (Roberts, 1966; McDonnell, 1981). Over a 20-year period, for example, parcels of fish, typically numbering 45 bream, 30 pike and larger numbers of smaller fish, were sent as gifts from the Royal Foss pond in York to stock the ponds of the Archbishop of York, the Fountains and Byland Abbeys in Yorkshire and, further afield, ponds belonging to the Earls of Cornwall and Lincoln (McDonnell, 1981; Binnie, 1987). The first carps kept in ponds in Roman and medieval continental Europe would have been wild Danube forms. Little is known about how these fish were propagated

The History of Aquaculture in Traditional Societies

21

or whether any selection took place. Although the first documentary accounts of common carp breeding in ponds did not appear until the thirteenth century, Balon (1995) asserts that breeding in captivity had occurred as early as in Roman times in Europe, albeit that spawning would probably have been a rather hit-or-miss affair. By the mid±late sixteenth century articles and even scientific investigations of carp production were being published. It is apparent from around this time that domesticated strains that differed in colour, shape and scale pattern from wild forms, were being produced. Balon (1995) believes domestication of the goldfish (Carassius auratus) in China preceded that of the common carp Cyprinus carpio in Europe, probably some decades earlier in the sixteenth century. Balon cites Chen (1952) in asserting that domestication in China resulted from the Buddhist belief that freeing a captive animal was an act of self-purification, and that the freeing of a rare, naturally occurring mutant form was a greater deed. `Ponds of mercy' were established at temples and the rich began to culture fish in jade jars, leading to inbreeding of aberrant forms ± trailing fins, bulging eyes and scale and pigment: a perverse form of selective breeding, in other words. The culture of koi carp in Japan some 200 years later may or may not have arisen from this practice (see Balon, 1995, for review). Today, the release of captive aquatic animals, especially fish and turtles, is still an everyday practice in many parts of Southeast Asia as is the husbanding of wild stocks in and around temple grounds. Protection and feeding fish in rivers and canals proximal to Buddhist temples and within temple ponds to make merit is widespread in Thailand, Laos and Cambodia. Such refuges may be an important part of conservation of wild stocks in areas with heavy exploitation and extensive water control.

Pond construction and management Much effort and expenditure often went into the construction and maintenance of ponds, as is particularly evident from European records. Medieval European fishponds varied in size from a few hundred square metres to more than 80 ha, and depths could be as much as 6±8 m. Most fishponds were formed by the damming of a stream in a valley, the dam wall being constructed from earth and stone and strengthened with wood. Early ponds tended to exploit the contours of the land; ponds built from the sixteenth century onwards were more likely to be linear in conformation, thereby facilitating netting (Chambers & Gray, 1988). Often the ponds were multipurpose; however, purpose-built fishponds were also constructed in close association with a millpond or moat surrounding a manor house or large farmhouse. Few of the ponds built in England during the early medieval period could be drained without creating a breach in the wall, and a simple overflow was usually all that was present to ensure that ponds did not burst their banks. Many of the ponds built in France, however, had deep drainage ditches fitted with wooden sluices that could be lifted, thereby allowing complete drainage of the ponds (Hoffman, 1995). Once built, ponds had to be maintained. Binnie (1987) asserts that silting was not a problem in medieval English fishponds and that pond catchments deliberately may not have been ploughed (i.e. used for crop culture) but instead were used for animal grazing. However, there are many contemporary written accounts of periodic

22

Ecological Aquaculture

de-silting. A detailed account at one such medieval pond is given by Roberts (1985). The Alresford fishpond, built around 1200, was an enormous structure (a magnum vivarium), covering some 80 ha. In the winter of 1252±53, after fish had first been transferred to a nearby stew pond, a breach was deliberately made in the 70-m long, 6-m high weir in order to allow the pond to drain. The following winter 122 labourers using 24 carts were employed to clean out the silt, the breach was mended and the pond was allowed to refill and was then restocked with 1072 roach, 603 bream, 229 perch and 115 pike. The factors that most often precipitated pond draining were deteriorations in stock `through disease or fouling of ponds' (Binnie, 1987). Although breaching of pond walls and de-silting were expensive, especially with such large ponds, the removal of silt was seen as essential in maintaining pond volume, and the practice of allowing pond soils to dry out seems to have long been recognized as important in maintaining productivity (Binnie, 1987). It is known that during dry fallowing periods, mineralization of sediment-bound nutrients occurs, resulting in renewed productivity once ponds are refilled. Such recycling of nutrients is further enhanced if grazing animals are allowed within these temporal pastures. There was also a tradition of inter-cropping in some parts of Europe. Taverner (1600) advised crops should be grown and cattle herded in fishponds during alternate dry years. According to Hoffman's account of the management of fishponds at LaperrieÁre-surSaoÃne, Burgundy, during the middle of the fourteenth century (Hoffman, 1995) draining and harvesting of ponds were often finished in time for planting of summer crops. During 1339±40 and 1350±51, extensive and costly repairs necessitated that the dry pond beds be farmed for a year, yielding crops of barley, oats and hemp. In central Europe the practice of rotations of 3 years of crops being followed by three fish crops was widely practiced (Wunder, 1949, in Hickling, 1962).

Feeding practices Technical literature related to feeding of fish is scant. Rumsey (1994) quotes mention by John Halver and Peter Milne and others of illustrations dating from more than three millennia of people feeding fish in impoundments. However, as still seen in some of the more extensive forms of aquaculture today, it is likely that the majority of the diet of fish grown in ponds and rice paddies would have been natural, principally plankton and benthic invertebrates and detritus. Natural productivity alone would have been able to sustain production of several hundred kilograms per hectare from monoculture. Production from polyculture, where synergistic feeding relationships are exploited, is likely to have been somewhat higher, although probably no more than 600±700 kg/ha/year, depending on the nature of the water body, proximity to human habitation and how nutrients drained into water bodies. Manuring has been widely used for centuries in Asia. In England too, Chambers & Gray (1988) cite evidence of human effluent being deliberately channeled in monastery fishponds during the later medieval period. Feeding was certainly known to the Chinese and to the Romans. However, Hickling (1971) and Currie (1991) believe that it would have been beneath the upper classes that owned fishponds in eleventh to thirteenth century England to try to increase revenues through feeding. But this was

The History of Aquaculture in Traditional Societies

23

certainly not the case with the aspiring lower orders who owned or leased fishponds for commerce and who did feed their fish. In many systems fish would also have been fed, most likely on wastes from the table and agricultural by-products and surpluses. Such practices would have been carried out on an experimental basis initially, but would have become accepted practice as yields were observed to increase. The integration of fish culture with agriculture in the pond±dyke system in southern China would have encouraged the use of such surpluses and wastes as fish food. Supplementary feeding would almost certainly have been necessary for fish reared in cages, there being insufficient plankton or seston available to sustain growth under most circumstances (see Beveridge, 1996, for review). With the development of salmonid farming in the nineteenth century, the use of animal protein such as oysters and horseflesh in diets became necessary (Maitland, 1887). Rumsey (1994) quotes anecdotal mention of fish feeds in books by Garlick, Norris and Stone published in North America during the latter half of the nineteenth century, although he says little detail is provided. Fish diet development started in earnest in the 1920s (Ellis, 1994; Rumsey, 1994). In Germany Schaeperclaus (1933) published a monograph on pond fish culture in which he discusses the inclusion of marine fish, meat and offal from warm-blooded animals, fish and shrimp flours and waste meals from oil extraction of soya, sunflower, etc., in trout diets. He also recommends the inclusion of 25% of beech or poplar sawdust as binder!

Discussion Despite the lack of serious archaeological or historical research there is sufficient evidence to conclude that the origins of aquaculture date back several ± possibly as many as five ± millennia. Early aquaculture ± proto-aquaculture as we have termed it here ± would probably have involved some simple form of intervention, such as transplantation of fertilized eggs or spawning fish or the enhancement of spawning grounds to increase production. Such actions would require that a claim to ownership or the control of access to and exploitation rights of stocks were established. A parallel development may have been the use of feed or shelter, probably first to attract fish in natural ecosystems and make them easier to exploit, before these aspects were incorporated into more formal culture systems. These forms of proto-aquaculture have parallels with the development of animal husbandry (see Fig. 1.1 and Table 1.2). The concept would have been readily understood, and perhaps informed from experience with livestock, and if inappropriate food were given it would have acted as a fertilizer, stimulating the productivity of the food web. As with agriculture, aquaculture would have to wait until the twentieth century for the arrival of inorganic fertilizers, but scientific inquiry into their use, and widespread adoption by fish farmers lags behind their use in terrestrial systems by several decades. Evidence suggests that captive breeding of fish and the management of ponds were later developments, as indicated in the aquaculture development scenario illustrated in Fig. 1.1b. Without control over propagation, selection for improved growth and disease resistance, let alone shape, colour, flesh quality or domestication, such a significant

24

Ecological Aquaculture

component of agricultural husbandry could not occur. In fact, the opposite may have happened. Although in certain contexts there would have been a large degree of passive selection for response to cultured environments, the practice of harvesting larger fish, leaving the smaller, slower-growing fish to serve as broodstock was, and still is, commonplace (see Silliman, 1975, for discussion). Removing the largest individual appears to be a common practice in the history of aquaculture the world over, perhaps illustrating residual close links to `hunting' and the role of raising fish in building social capital. That the fish thrived and grew often owed more to chance than to a profound understanding of how to satisfy the animals' needs; that had to wait until the development of the biological sciences in the nineteenth century. Proto-aquaculture activities have been recorded in many parts of the ancient world, including Egypt, China and Mexico. Why did aquaculture develop in some places rather than others, and what governed the methods of culture and the pattern of development? Fish has almost always been universally important in the Asian diet. Although extensive, China's coastline was small relative to the inland land mass. With its wetland-based agriculture and, from the Han Dynasty onwards, its high densities of aquatic food-eating people it had strong incentives to develop inland aquaculture and it is here that such advanced aquaculture methods as controlled spawning of fish, polyculture and integrated aquaculture first developed. By contrast, until recently in Japan and other countries with extensive coastlines, marine fish and shellfish resources were sufficient to supply their populations' needs. It should also be reiterated that religion, politics and power and social issues were sometimes more important in aquaculture development than increasing the supply of food or economic viability. In central and western Europe, for example, fish farming was primarily developed in the first instance by the monastic orders in order to ensure supplies of fish for days when eating of meat was forbidden. Subsequent pond aquaculture developments in continental Europe occurred in areas where soils were too poor to sustain agriculture (Hickling, 1962). In medieval England pond fish culture was also used by the arriviste Norman rulers to help consolidate political power. There are many parallels between the development of agriculture and aquaculture, with people opting in and out as the need for fish/food changed. However, unlike agriculture which has been the most important way of obtaining food on land for at least 10 000 years, aquaculture has until recently contributed little in real terms to world fish or shellfish production. Instead of evolving towards cultivation, hunter± gatherer methods of procuring food from the aquatic environment developed along a different path: by improvements in tracking methods and by increases in killing power. There are several reasons why the exploitation of food in terrestrial and aquatic environments did not develop in the same way. First, food in inland waters and in the sea has, until recently, been abundant. Changes in fishing grounds, increases in fishing pressure and development of fisheries technology were sufficient to meet demands and there was therefore little incentive to learn to farm. Moreover, the aquatic environment was hostile and something to be feared. It must have seemed impossible that a structure that could hold fish securely and withstand the forces of the tides and currents, waves and storms, could be built in the sea. There were other

The History of Aquaculture in Traditional Societies

25

technical problems, too, to overcome. While the breeding and husbandry of animals and the harvesting and planting of seeds was readily achieved on land, it has proved difficult to breed many aquatic species and to hatch the eggs and successfully rear the offspring. These problems in part stemmed from the fact that people were dealing with organisms that were very different from themselves and with an environment about which they were largely ignorant. It was not until the rise of the biological sciences in the nineteenth century that the mysteries surrounding the physiology and reproduction of aquatic animals and the role the environment played in controlling these processes began to be understood. These constraints were greater for coastal than inland environments, which may explain the earlier development of the latter. The environments for fish culture to develop were to some extent a consequence of the overall development of intensified floodplain agriculture in Asia. Like agriculture and fishing, aquaculture was shaped by changes in society ± wars and colonialism, politics and power, the waxing and waning of religions with their associated cults and schisms, the rise of science and industrialization (Tables 1.2, 1.3). However, it is not solely concerned with the production of food but has also played a role in a number of religions and in consolidating social position. The production of fish for religious purposes in ancient Egypt, Christian Europe and Buddhist Japan has contributed much to the development of aquaculture as a whole, especially with regard to selective breeding. Certain ironies are also apparent. Unlike hunting whose image is associated with nobility, fishing is often associated with poor people of low social standing (Sahrhage & Lundbeck, 1992). The construction of fishponds and the keeping of fish as in the Roman Empire and medieval England have long been associated with social position and would have required significant resources unavailable to ordinary people. Where demands for fish for religious or social purposes declined or the economics of production changed, as in ancient Egypt and in eighteenth century France, fish farming withered and died. Despite its ancient roots, however, fish and shellfish farming should be regarded as very much a post-Second World War phenomenon. Although locally important, in global terms farmed aquatic production prior to 1950 was insignificant. With the possible exception of China, few of today's important industries, particularly where mariculture is concerned, owe much to ancient traditions or technologies or have a history extending back more than 30 or 40 years. Even in China until recently farmed production was but a fraction of that from capture fisheries. Data assembled by Li (1994) and the FAO (Lu, 1992; Immink, 1997) show that the spectacular ± approximately 50-fold ± growth in cultured freshwater fish production has occurred since the founding of the People's Republic in 1949 as a result of political, economic, technical and demographic changes. Some 82 000 man-made water bodies have been constructed for hydropower, flood control and irrigation purposes, increasing the inland water surface area by 2.05 million ha (Lu, 1992). In the late 1950s the techniques for spawning Chinese carps were developed, freeing Chinese farmers from their reliance on wild-caught fry, while the development of the mixed economy some 20 years later has given further incentives to increase productivity. Elsewhere in Asia aquaculture was rare until recent decades. Huet's (1972) generalization that `In the

26

Ecological Aquaculture

Far East where all farmers are fish farmers and vice versa' was based on observations of a few clusters of activity around Bangkok and in West Java (Edwards et al., 1997). Edwards et al. estimate that probably far fewer than 10% of small-scale farms in the region culture fish and of those that do, most produce fewer fish than their resource base allows. And what is the legacy of traditional aquaculture? Much present day mariculture, particularly the intensive salmon and shrimp farming industries that have grown so rapidly during the past two decades, has a basis in industry. It has been technologyled and is heavily dependent upon the environment not only for the supply of essential resources such as food, but also to disperse and assimilate waste food, faeces, excreta and chemical residues (Beveridge et al., 1997). In parts of China traditional aquaculture practices are being abandoned in favour of intensification. Land prices are too high, water is in short supply, cheap feeds are becoming ever more easily available and there is a strong demand for fish from the increasingly affluent and urbanized population. Throughout Asia, the stigma associated with use of night soil in ponds has led to a decline in its use, compounded by increased mixing of human and industrial waste in urban sewage systems (P. Edwards, pers. comm.). However, for large areas of rural Asia, and elsewhere in the developing world, resource-intensive aquaculture is not, and is unlikely to be, a viable option. Trends to separate aquaculture from broader farming systems are largely commercially driven but are in conflict with longer term, and rational, practice. Just as industrial livestock production is now being challenged in Western Europe on public health, animal welfare and, most immediately, environmental grounds, we expect a similar fate for unsustainable aquaculture development.

Acknowledgments Colleagues James Muir, David Penman and Krishen Rana kindly helped with information and references. Last, but not least, we thank Dr Barry Costa-Pierce for his knowledge and enthusiasm on the subject of the ecology of aquaculture systems.

References Adamek, Z. & Kouril, J. (2000) A long aquaculture tradition in the Czech Republic. Aquaculture Europe, 25 (1), 20±23. Adams, N. (1984) Architecture for fish: the Sienese Dam on the Bruna River ± structures and designs, 1468±ca. 1530. Technology and Culture, 768±797. Ardizzone, G.D., Cataudella, S. & Rossi, R. (1988) Management of Coastal Lagoon Fisheries and Aquaculture in Italy. FAO Fisheries Technical Paper, 293. FAO, Rome. Armillas, P. (1971) Gardens on swamps. Science, 174, 653±661. Balon, E.K. (1995) Origin and domestication of the wild carp Cyprinus carpio: from Roman gourmets to the swimming flowers. Aquaculture, 129, 3±49. Beveridge, M.C.M. (1996) Cage Aquaculture, 2nd edn. Blackwell Science, Oxford.

The History of Aquaculture in Traditional Societies

27

Beveridge, M.C.M., Phillips, M.J. & Macintosh, D.C. (1997) Aquaculture and the environment: the supply and demand for environmental goods and services by Asian aquaculture and the implications for sustainability. Aquaculture Research, 28, 101±111. Binnie, G.M. (1987) Early Dam Builders in Britain. Thomas Telford, London. Boserup, E. (1965) The Conditions of Agricultural Growth. Allen & Unwin, London. Bray, F. (1984) Biology and Biological Technology. Part II: Agriculture. Science and Civilisation in China. Vol. VI (ed. J. Needham). Cambridge University Press, Cambridge. Brewer, D.J. & Friedman, R.F. (1989) Fish and Fishing in Ancient Egypt. Aris & Phillips, Warminster, UK. Bromley, D.W. (1992) Making the Commons Work. Theory, Practice and Policy. ICS Press, San Francisco. Chambers, R.A. & Gray, M. (1988) The excavation of fishponds. In: Medieval Fish, Fisheries and Fishponds in England (ed. M. Aston), pp. 113±128. British Archaeological Reports, British Series, 182 (i). Chevey, P. & Lemasson, J. (1937) Contribution aÁ l'EÂtude des Poissons des Eaux Douces Tonkinoises. Institut OceÂanographic de I'Indochine. 33e Note. Gouvernement GeÂneÂral de L'Indochine, Hanoi. Chimits, P. (1957) Tilapia in ancient Egypt. FAO Fisheries Bulletin, 10, 211±215. Coche, A.G. (1982) Cage culture of Tilapias. In: Biology and Culture of Tilapias (eds R.S.V. Pullin & R.H. Lowe-McConnell), pp. 205±246. ICLARM, Manila, Philippines. Coe, M.D. (1964) The chinampas of Mexico. Scientific American, 211, 90±98. Currie, C.K. (1989) The function of fishponds in the monastic economy. In: The Archaeology of Rural Monasteries (eds R. Gilchrist & H. Mytum), pp. 147±172. British Archaeological Reports, British Series, 203. Currie, C.K. (1991) The early history of the carp and its economic significance in England. Agricultural History Reviews, 11, 97±107. Desroches-Noblecourt, C. (1954) Poissons, tabous et transformations du mort. Nouvelles considerations sur les pelerinages aux villes saintes. Kemi, 13, 33±42. Diamond, J. (1997) Guns, Germs and Steel. A Short History of Everybody for the Last 13,000 Tears. Chatto & Windus, London. Drewer, L. (1981) Fishermen and fish pond: from the sea of sin to the living waters. Art Bulletin, 63, 533±547. Drews, R. (1951) The Cultivation of Food Fish in China and Japan: A Study Disclosing Contrasting National Patterns for Rearing Fish Consistent with the Differing Cultural Histories of China and Japan. PhD thesis, University of Michigan, Ann Arbor. Dyer, C. (1988) The consumption of fish in medieval England. In: Medieval Fish, Fisheries and Fishponds in England (ed. M. Aston), pp. 27±38. British Archaeological Reports, British Series, 182 (i). Edwards, P., Little, D.C. & Yakupitiyage, A. (1997) A comparison of traditional and modified inland artisanal aquaculture systems. Aquaculture Research, 28, 777±788. Ellis, J.N. (1994) In search of history: early days and trout feeds. Aquaculture Magazine, July/ August, 50±51, 54, 56±59. Flannery, T.F. (1994) The Future Eaters. Reed Books, Australia. Grigg, D.B. (1974) The Agricultural Systems of the World. Cambridge University Press, Cambridge.

28

Ecological Aquaculture

Harris, D.R. (Ed.) (1996) The Origins and Spread of Agriculture and Pastoralism in Eurasia. Smithsonian Institution Press, Washington. Hepher, B. (1988) Nutrition of Pond Fishes. Cambridge University Press, Cambridge. Hickling, C.F. (1962) Fish Culture. Faber & Faber, London. Hickling, C.F. (1971) Prior More's fishponds. Medieval Archaeology, XV, 118±123. Hoffman, R.C. (1995) `Carpes pour le Duc . . . :' the operation of fishponds at LaperrieÁre-surSaoÃne, Burgundy, 1332±1352. Archaeofauna, 4, 33±45. Hoffman, R.C. (1996) Economic development and aquatic ecosystems in Medieval Europe. American Historical Reviews, 101, 630±669. Huet, M. (1972) Textbook of Fish Culture. Fishing News Books, Farnham, Surrey. Immink, A.J. (1997) World Aquaculture Production 1950±1983. Report for the FAO, Rome. Institute of Aquaculture, University of Stirling. JimeÂnez-Osornio, J.J. & Gomez-Pompa, A. (1991) Human role in shaping of the flora in a wetland community, the chinampa. Landscape and Urban Ecology, 20, 47±51. Kirk, R. (1987) A History of Marine Fish Culture in Europe and North America. Fishing News Books, Oxford. Knowles, D. (1950) The Monastic Order in England. Cambridge University Press, Cambridge. Lewis, D.J., Wood, G.D. & Gregory, R. (1996) Trading the Silver Seed. Intermediate Technology Development Group, London. Li, K.M. (1992) Rice±fish farming systems in China: past, present and future. In: Rice±Fish Research and Development in Asia (eds C. R. de la Cruz, C. Lightfoot, B.A. Costa-Pierce, V.R. Carangal & M.P. Bimbao), pp. 17±26. ICLARM Conference Proceedings 24, ICLARM, Manila. Li, S.F. (1994) Introduction: freshwater fish culture. In: Freshwater Fish Culture in China: Principles and Practice (eds S. Li & J. Mathias), pp. 1±25. Elsevier, Amsterdam. Lin, Z. (ed.) (1991) Pond Fisheries in China. Pergamon, London. Ling, S.W. (1977) Aquaculture in Southeast Asia. University of Washington Press, Seattle. Little, D.C. & Edwards, P. (1997) Contrasting strategies for inland fish and livestock production in Asia. In: Recent Advances in Animal Nutrition in Australia (eds J.L. Corbett, M. Choct, J.V. Nolan & J.B. Rowe), pp. 77±85. University of New England, Armadale, Australia. Little, D.C. & Edwards, P. (2002) Integrated Livestock ± Fish Farming Systems: the Asian Experience and its Relevance for Other Regions. FAO Fisheries Technical Paper. FAO, Rome (in press). Little, D.C. & Pham, T.A (1999) Common carp indigenous ricefield-based culture in Northern Vietnam. AARM Newsletter, 4 (3), 4±5. Lu, X. (1992) Fishery Management Approaches in Small Reservoirs in China. FAO Fisheries Circular 854. FAO, Rome. Maitland, J.R.G. (1887) The History of Howietoun. J.R. Guy, Stirling. McDonnell, J. (1981) Inland fisheries in medieval Yorkshire 1066±1300. Borthwick Papers, 60. University of York. McLarney, W. (1984) The Freshwater Aquaculture Book. Hartley & Marks, Point Roberts, Washington. Micha, J.-C. & Chavez, M. (1997). Development of agro-piscicultural ecosystems in tropical marshlands. In: Integrated Fish Farming (eds J.A. Mathias, A.T. Charles & H. Baotong), pp. 347±358. CRC Press, Boca Raton.

The History of Aquaculture in Traditional Societies

29

Pitcher, P.T.J. & Hart, P.J.B. (1982) Fisheries Ecology. Croom Helm, London. Rana, K.J. (1998) Recent developments in aquaculture statistics. Fishery and aquaculture statistics in Asia. In: Proceedings of the FAO/SEAFDEC Regional Workshop on Fishery Statistics. 19±21 August 1997, 11, 242±254. FAO, Rome. Reay, P.J. (1979) Aquaculture. Arnold, London. Roberts, B.K. (1966) Medieval fishponds. Amateur Historian, VII, 119±126. Roberts, E. (1985) Alresford Pond, a medieval canal reservoir: a tradition assessed. Proceedings of the Hampshire Field Club and Archaeological Society, 41, 127±138. Ruddle, K. & Zhong, G. (1988) Integrated Agriculture±Aquaculture in South China. The Dike± Pond System of the Zhujiang Delta. Cambridge University Press, Cambridge. Rumsey, G.L. (1994) History of early diet development in fish culture, 1000 BC to AD 1955. Progressive Fish-Culturist, 56, 1±6. Sahrhage, D. & Lundbeck, J. (1992) A History of Fishing. Springer-Verlag, Berlin. Schaeperclaus, W. (1933) Lehrbuch der Teichwirtschaft. Paul Parey, Berlin. Schnitter, N.J. (1994) A History of Dams. A.A. Balkema, Rotterdam. Schuster, W.H. (1952) Fish Culture in Brackish-Water Ponds of Java. Indo-Pacific Fisheries Council Special Publication No. 1. FAO, Rome. Sharma, K.P. & Rana, R. (1986). Seasonal ponds ± a potential source of fish seed in Rajasthan, India. In: The First Asian Fisheries Forum (eds J.L. Maclean, L.B. Diizon & L.V. Hosillos), pp. 87±88. Asian Fisheries Society, Manila, Philippines. Silliman, R.P. (1975) Selective and unselective exploitation of experimental populations of T. mossambicus. Fisheries Bulletin, 73, 495±507. Stearne, J.M. (1971) The medieval fishponds of Northamptonshire. Northants Past and Present, IV (5), 299±310. Stearne, J.M. (1988) The royal fishponds of medieval England. In: Medieval Fish, Fisheries and Fishponds in England (ed. M. Aston), pp. 39±68. British Archaeological Reports, British Series, 182 (i). Stewart, K.M. (1994) Early hominid utilization of fish resources and implications for seasonality and behaviour. Journal of Human Evolution, 27, 229±245. Taverner, J. (1600) Certaine Experiments Concerning Fish and Fruite. William Ponsonby, London. Tudge, C. (1998) Neanderthals, Bandits and Farmers. How Agriculture Really Began. Weidenfield & Nicolson, London. Votruba, L. (1987) Historissche Talsperren und Wasserseicher in der CSSR. Historische Talsperren, Band 1, pp. 401±406. K. Witter, Stuttgart. Welcomme, R.L. (1972) An evaluation of the acadja method of fishing as practiced in the coastal lagoons of Dahomey (West Africa). Journal of Fish Biology, 4, 39±55. Yang, H. (1994) Integrated fish farming. In: Freshwater Fish Culture in China: Principles and Practice (eds S. Li & J. Mathias), pp. 219±270. Elsevier, Amsterdam. Zeepvat, R.J. (1988) Fishponds in Roman Britain. In: Medieval Fish, Fisheries and Fishponds in England (ed. M. Aston), pp. 17±26. British Archaeological Reports, British Series, 182 (i).

Chapter 2

The Ahupua'a Aquaculture Ecosystems in Hawaii Barry A. Costa-Pierce Rhode Island Sea Grant College Program `The whole distance to the village of Whyeete is taken up with innumerable artificial fishponds extending a mile inland from shore, in these the fish taken by nets in the sea are put, and though most of the ponds are fresh water, yet the fish seem to thrive and fatten. . . . The ponds are several hundred in number and are the resort of ducks and other waterfowl' T. Bloxam, British naturalist on H.M.S. Blonde describing Waikiki in 1825 (Handy & Handy, 1972).

Introduction Beveridge & Little (Chapter 1, this volume) describe what is known about the origins of aquaculture in traditional societies in China, Egypt, Europe and the Americas. Most of these examples are inland, freshwater developments associated with rivers or other water courses indigenous to large continental land masses. The ancient mariculture systems of Hawaii are unique in that they connect an isolated island society with sophisticated ocean harvesting and integrated sea farming activities to an entire watershed management/food production system (the ahupua'a). Ancient Hawaiian mariculture systems are remarkable in terms of their diversity, distinctive management, and sheer extent of development, especially given the small size of Hawaii. Although the Hawaiian systems are relatively recent (only 1500±1800 years old) by Chinese and Egyptian standards, the evolution of ocean fishing to ranching and onwards to true ocean farming systems (mariculture) is notable. Evolution of such sophisticated farming systems may be a natural evolutionary part of societies whose population densities exceed the carrying capacity of natural ecosystems to support them. As a result, rapid evolution of new, innovative farming systems ± such as ecological aquaculture and mariculture systems ± occurred in Hawaii.

The social ecology of ancient Hawaii `The shores of Hawaii are by no means so well stocked with fish as those of the Society Islands . . . The industry of the Hawaiians in a great degree makes up for

The Ahupua'a Aquaculture Ecosystems in Hawaii

31

the deficiency in fish, for they have numerous small lakes and ponds, frequently artificial, wherein they breed fish of various kinds and in tolerable abundance' (Ellis, 1826) The ancient Hawaiian fishponds were part of a large, integrated, and complex Hawaiian subsistence and barter economy that included agriculture, aquaculture, and animal rearing. The political aspects of this sociocultural system contributed greatly to the development of the expansive aquaculture±agriculture network. Hierarchical political control and redistribution of food was essential to the smooth functioning of the ancient integrated farming systems, because construction and management of the huge fishpond complexes required sizeable labor forces. Massive ponds such as the Kaloko pond in Kona, Hawaii, have a 229 m long wall about 2 m high that is 11 m thick at the base. This wall contains an estimated 150 000 m3 of rock and fill (Apple & Kikuchi, 1975). The Kuapa pond at Maunalua, Oahu, was reportedly built over several years by thousands of people who formed long human chains to transport rocks from the Ko'olau Mountains. Efforts of this magnitude obviously required tremendous social organization. Ancient Hawaii had highly stratified chiefdoms with a well defined class structure separating chiefs, advisors, stewards, and commoners. This organization was similar to that of the chiefdoms found in Tonga, Samoa and the Society Islands (Sahlins, 1958). Prior to 1848, all Hawaiian land ± its natural resources, fishponds, communal and spiritual centers ± were owned by the kings (ali'i). The kings would contract the bulk of the land and fishponds to subchiefs (konahiki), but keep sacred resources such as fishponds under their direct control. Couriers would transport from royal fishponds to the court plump fish in water-filled gourds or by hand (Rice, 1923). Subchiefs were granted large, wedge-shaped areas (ahupua'a) of the Hawaiian islands that encompassed the watersheds of entire valleys, stretching from the mountains to the sea (Lind, 1938). Ahupua'a were generally not physically demarcated in Hawaii. No evidence of erect stones marking individual land holdings, such as in Tahiti, have been found (Handy & Handy, 1972). It appears that the ahupua'a were mainly political subdivisions granted by the kings to the subchiefs to assure subsistence supplies of food, firewood, timber, thatch, and ornamentation. Handy & Handy (1972) have described a share-cropping arrangement between tenant families and the subchiefs that was `comprehensive and reciprocal in its benefits.' Within an ahupua'a, sections of land ('ili) were granted to individual extended families ('ohana) for cultivation. These land divisions within the ahupua'a carried individual titles. `It was said that in every community there were individuals who were well versed in the local lore of land boundaries, rights, and history' (Handy & Handy, 1972). All harvests from the fishponds were distributed in a politically institutionalized manner by the subchiefs to extended family groups and pond workers living in the ahupua'a. Kikuchi (1976) suggested that the fishponds were symbols of the chiefly right to conspicuous consumption and the exclusive ownership of the land and its resources, and that the fishponds were the subject of frequent inter- and intratribal conflicts.

32

Ecological Aquaculture

Kamakau (1976) argues, however, that the presence of the fishponds did not indicate any contempt on the part of the subchiefs for the local populace. He stated, `How could they have worked together in unity and made these walls if they had been frequently at war and in opposition against one another? If they did not eat the fruit of their efforts?' Indeed, a native Hawaiian, David Malo, wrote of a Big Island chief who was killed because of his cruel efforts to exploit his people when he ` made the people of Ka'u sweat and groan . . . [with] the building of heavy stone walls about several fishponds' (Malo, 1951). Contact with Europeans, which began in 1778, had dramatic effects on all levels of Hawaiian society. It destroyed the ancient religion and the chief's supernatural right to control all the land, its resources, and its people. The economy changed from the traditiona1 barter system to a monetary economy. Contact with foreigners brought new diseases, which led to the massive depopulation of Hawaii. The Hawaiian land decision of 1848 decision (known as the Great Mahele) allowed the purchase of crown lands by Hawaiian commoners and by foreigners. In many areas the largest purchase of these lands was by foreigners. Some of these purchased thousands of acres for $0.25±$0.50 per acre (Kelly, 1980). The Great Mahele was a pivotal point not only in Hawaiian history but also in the history of the integrated mariculture farming ecosystems. Decline of the fishpond complexes and Hawaiian integrated aquaculture ecosystems was rapid after the Great Mahele. Once the harvests from the lands and fishponds became economic entities with prices, their distribution tended no longer to follow either an institutionalized pattern of sharing (Handy & Handy, 1972) or exploitation of the commoners by the chiefs (Kikuchi, 1973), as before the Great Mahele. With the general demise of native Hawaiian society, the majority of Hawaiian integrated farming systems fell into disuse and disrepair. When Captain James Cook reached Hawaii in 1778, at least 360 fishponds existed. They produced an estimated 900 metric tons (mt) of fish/year (Costa-Pierce, 1987). Farber (1997) speculated there were as many as 488 ponds. According to the State of Hawaii, only 28 ponds were suitable for fish culture in 1977 (Madden & Paulsen, 1977). By 1985 only seven ponds were in commercial or subsistence use, producing an estimated 15 000±20 000 kg/year. However, accelerated community development efforts in the 1980±90s, especially on Molokai, which included studies, conferences, and formation of a fishpond task force, identified 31 ponds on Molokai alone that the Hawaiian community wanted to restore (Wyban, 1992; Farber, 1997).

Hawaiian integrated aquaculture ecosystems Four basic types of fishponds and one fish `trap' were known in ancient Hawaii and were integrated to various degrees with the staple carbohydrate crop of the Hawaiians, wetland taro (Colocasia esculenta). Ponds were fed with cut grass, mussels, clams, seaweeds, and taro leaf from adjacent agricultural or natural ecosystems (Wilder, 1923; Titcomb, 1952). In contrast to modern integrated aquaculture systems in Asia, Hawaiian fishponds did not receive fertilization from animal or human

The Ahupua'a Aquaculture Ecosystems in Hawaii

33

wastes or kitchen refuse. Hawaiian chiefs prohibited such waste introductions (Kikuchi, 1976). The diversity, extent, and number of fishponds in Hawaii before contact with Europeans is impressive. The various fishponds spanned the natural salinity range of water. The four types of fishponds (Fig. 2.1) developed within the ahupua'a were: freshwater taro fishponds (loko i'a kalo) l other freshwater ponds (loko wai) l brackish water ponds (loko pu'uone) l seawater ponds (loko kuapa) l

Fig. 2.1 Four types of Hawaiian integrated aquaculture ecosystems developed in historical times: (a) lo'i were for the paddy culture of taro (Colocasia esculenta); and loko i'a kalo were taro patches modified to include aquaculture. These upland ponds are depicted in a valley with elevation contours indicated; (b) loko wai were artificial (and modified natural) freshwater lakes excavated for aquaculture; (c) loko pu'uone were brackishwater lakes separated from the sea by a pu'uone (a spit of land reinforced by mud, silt, and refuse) and connected to the sea by a ditch that had grates to trap and hold large fish; (d) loko kuapa were ponds built along the ocean shore usually on top of a reef flat with volcanic rock and/or coral rock to form a wall (kuapa). Controlled harvests were accomplished using a canal, net, and grate system. Modified from Kikuchi (1976).

34

Ecological Aquaculture

An additional type of pond (a fish trap) was known as loko `umeiki (Summers, 1964).

Freshwater taro fishpond ecosystems The taro fishponds (loko i'a kalo) were developed in the uplands to cultivate taro and simultaneously grow a limited range of euryhaline and freshwater fish, such as mullet (Mugil cephalus; ama 'ama), silver perch (Kuhlia sandwicensis; aholehole), and Hawaiian gobies (Eleotris sandwicensis and E. fusca; 'o'opu). Freshwater prawn (Macrobrachium sp.; opae) and green algae (Spirogyra sp. and Cladophora sp.; limu kalawai) were also grown. These integrated freshwater fishpond ecosystems probably arose originally from shallow ponds (lo'i) created by the diversion of stream runoff for the irrigation of wetland taro agriculture. Over time the Hawaiians added aquaculture to the design of these ponds. In addition, surplus fish from abundant sea harvests of milkfish (Chanos chanos), mullet and silver perch were put in shallow freshwater taro ponds located close to the sea. Fish also directly entered the taro patch-fishponds by migrating from the sea up the newly created artificial estuary. It is likely that the originators of the stocking practice observed that fish put temporarily in these freshwater ponds near the ocean not only survived the harsh transition in salinity from salt to fresh but also grew well. They also probably noticed that their taro grew more luxuriantly and had fewer pests, owing to the continual grazing and pruning activities of herbivorous and benthic-feeding fish such as milkfish and mullet. Taro was planted in mounds to leave channels for swimming fish to feed on the insects and ripe leaf stems of the taro (Kamakau, 1976).

Other freshwater pond ecosystems The second type of freshwater ponds, loko wai, were inland ponds or lakes typically excavated by hand from a natural depression, lake, or pool and supplied with water diverted by ditches from streams, rivers, or by natural groundwater springs or aquifers. Native species of freshwater prawn and Hawaiian gobies (Eleotris sandwicensis, E. fusca, and Gnatholepis anjernesis) and migrants from the sea that move into freshwater (mullet, milkfish, and silver perch) were stocked, grown, and harvested from these ponds. Milkfish were particularly abundant in these ponds, having been procured in shallow shoreline areas and carried long distances in large gourds filled with water (Beckley, 1883). These ponds were harvested by woven reed nets (hala) placed across a channel to capture the fish during their seaward spawning migrations, oftentimes during full moons in the spring.

Brackishwater pond ecosystems Brackishwater ponds, loko pu'uone, were coastal ponds excavated by hand from a natural body of water stranded by eustatic sea level changes (Kikuchi, 1976), or

The Ahupua'a Aquaculture Ecosystems in Hawaii

35

formed by piling mud, sand, and coral to form earth embankments parallel to the coast (Fig. 2.1). A sand bar, coastal reef, or two adjacent edges of land masses isolated these ponds from the open sea. These loko pu'uone were connected to the ocean by a canal constructed so that seawater would enter the fishpond on a rising tide. Loko pu'uone usually had some freshwater inputs, either from springs, streams entering the pond, submarine groundwater discharges, or water percolating from adjacent aquifers. The combination of fresh- and saltwater produced a brackishwater environment that was very productive and very diverse in fish species that could acclimate to both fresh- and saltwater. Two types of loko pu'uone have been described, a commoner's pu'uone and a king's pu'uone (Handy & Handy, 1972), classified by their ownership and the effort and elaboration used in their construction.

Nearshore mariculture ecosystems The fourth type of fishpond, the seawater ponds, loko kuapa, were the ultimate aquaculture achievement of the native Hawaiians and a valuable contribution to native engineering and the evolution of subsistence food production. Mariculture, or the farming of euryhaline and marine aquatic animals in seawater, appears to have reached a sophisticated level in ancient Hawaii. Summers (1964) states that loko kuapa are found nowhere else in Polynesia. The main isolating feature of these ponds was a seawall (kuapa) constructed of coral or lava rock. Kikuchi (1973) noted that the lengths of 90 fishponds studied ranged from 46 to 1920 m, with the greatest frequency of lengths between 366 and 610 m, containing an average of 955 m3 of rocks and fill. Some of the stones used in the walls have been estimated to weigh as much as half a ton. On the island of Molokai, which has a protected, regular, shoal, southern coastline, more loko kuapa were constructed per area of land than anywhere else in Hawaii (Fig. 2.2). Large numbers of these ponds were also developed in the Kaneohe Bay, Waikiki,

Fig. 2.2 A map of the Hawaiian island of Molokai with its long, shoal, southern coastline. Darkened areas indicate the areas of some 28 marine fishponds (loko kuapa), Two brackish water ponds (loko pu'uone) are indicated by letters. Numbers refer to the location of fish traps (loko 'umeiki). Modified from Cobb (1902).

36

Ecological Aquaculture

and Pearl Harbor areas of Oahu (Fig. 2.3). In some of the Molokai ponds coralline algae, which secrete a natural cement, were used to strengthen the walls. Women and children gathered coralline algae from the sea for this purpose (Summers, 1964). Ponds on Molokai were built on a reef flat, with the walls extending in a semicircle from the shoreline. The ponds thereby contained all of the marine aquatic biota of the original reef environment. At least 22 species of marine life f1ourished in these ponds (Fig. 2.4).

Fig. 2.3 A map of the Pearl Harbor and Kalihi Basin areas of the island of Oahu, Hawaii. Darkened areas depict the locations of more than 30 loko wai, loko pu'uone and loko kuapa. Modified from McAllister (1933).

Loko kuapa had an additional feature that can only be described as an ancient ecological engineering marvel. Canals (auwai) were constructed into the walls of the ponds for the stocking, harvesting, and cleaning of the seawater ponds with minimal human effort. The canals connected the ponds directly to the sea and had, in the middle, a single, immovable grate (makaha) made of dense native woods (Fig. 2.5). These grates were constructed by vertically lashing solid timbers ('ohi'ia or lama) to two or three cross beams with ferns, so that the individual timbers were separated by approximately 0.5±2.0 cm wide spaces. As a result, only water and very small fish could pass freely in and out of the pond. The pond was therefore automatically stocked by normal recruitment of juvenile fish from the sea. The grates were fixed in the canal, and large fish trying to migrate to the sea to spawn were harvested by setting nets on the pond side of the grate, or by hand capture (Kamakau, 1976). Harvesting was attuned to the behavior of the fish. Loko kuapa were used to culture mainly two species of fish (milkfish and mullet). Both are sea spawners (catadromous). During spring moons in Hawaii, they return from their freshwater and brackishwater habitats to spawn in coastal seawater. Salmon, being anadromous fish, have an exactly opposite life cycle. During the migration periods the keepers of the fishpond (kia'i) would joyously watch hundreds of fish swim into

The Ahupua'a Aquaculture Ecosystems in Hawaii

Fig. 2.4 The Hawaiian integrated aquaculture ecosystem spanned the entire normal salinity range of water, and comprised a continuum from agriculture to aquaculture. An impressive number of species were harvested from seawater fishponds and traps; the ponds enclosed a reef-flat environment and all of the reef-flat species. Modified from Kikuchi (1976).

37

38

Ecological Aquaculture

Fig. 2.5 Details of the sluice grate (makaha) that was permanently fixed in a canal (auwai o ka makaha) that connected the marine fishponds to the open sea. Nets were set on the pond side of the canal to capture fish gathering in the canal attempting to migrate to the sea. A single, immovable grate was used in the ancient design but was modified in recent times to have two grates on the ocean and pond sides of the canal. Modified from Apple & Kikuchi (1975).

the canal in a futile attempt to reach the sea. Nets set on the pond side of the makaha close off the migratory route. Later in Hawaiian history, the canals were modified to have one or two vertically movable makaha substituting for the set net and immovable makaha used earlier. With this modification, as the fish entered the canal and tried to migrate to the sea, the seaside makaha was lowered (or was permanently fixed) and the pond side makaha was raised and fish crowded into the harvest canal. The pond side makaha was then lowered, trapping the fish, which were simply netted out or hand harvested from the canal. The process was then repeated. Thus through the use of keen observational skills and knowledge of fish behavior, a method was devised of allowing the fish to harvest themselves!

The Ahupua'a Aquaculture Ecosystems in Hawaii

39

`When the keeper of the pond wished to remove some fish, he would go to the makaha (grate) while the tide was coming . . . the keeper would dip his foot into the water at the makaha . . . and if the sea pressed in like a stream and felt warm, then he knew that the sluice would be full of fish. The fish would scent the fresh sea and long for it! I have seen them become like wild things. At a sluice where the fish had been treated like pet pigs, they would crowd to the makaha where the keepers felt of them with their hands and took whatever of them they wanted.' S.M. Kamakau, 9 December 1869. Translated from a Hawaiian Newspaper, Ke Au 'Oko'a (Kamakau, 1976) Over time the loko kuapa would become filled with sediments, either washed in during heavy rains or accumulated from the settling of particles in the water. In some of the larger ponds on Molokai that tended to become silted, the grates and canals were operated to clean the ponds, in a clever example of practical pond maintenance. In these cases, more than one canal was constructed in an orderly pattern in the pond walls, with grates set across from each other into the direction of the prevailing longshore current. On a rising tide the grate on the upstream end of the longshore current was opened. This washed the sediment accumulated at this upstream grate downstream toward the middle of the fishpond. On the next ebb tide this upstream grate was closed, and the downstream grate on the opposite side of the pond opened. The ebb tide therefore tended to pull the accumulated sediment from the middle of the pond toward the downstream grate. By a regular program of following the tidal cycle and opening and closing the proper grates the ponds could be effectively cleaned of sediment. In addition, if a pond was silted up after a particularly heavy rain, weighted bamboo rakes (kope 'ohe) were towed behind outrigger canoes to facilitate movement of the accumulated sediment out of the fishponds. Cordover (1970) discovered another type of seawater fishpond on Molokai with no grates. These ponds were stocked with fingerling mullet (Mugil cephalus) only once, and it was reported by Hawaiians that mullet spawned and grew there successfully. Hamre (1945) reported more of these types of ponds on Molokai prior to the 1946 tsunami. Although modern scientists have had great difficulty in spawning mullet in captivity (this has improved only recently), Phelps (1937) was assured by Hawaiian elders that mullet had indeed spawned regularly in these ponds. He states, `The Hawaiian knowledge of the natural history of fishes, in the old days, should not be underestimated' (Phelps, 1937).

Nearshore fish traps The last type of fishpond used by the Hawaiians, loko 'umeiki, was actually a trap rather than a pond (Fig. 2.5). Hawaiian fish traps are very similar to those in much of Oceania. Like loko kuapa, these traps were shoreline ponds with low, semicircular pond walls. However, unlike the loko kuapa, pond walls were partially or wholly submerged at high tide and contained numerous openings, or lanes, leading into or

40

Ecological Aquaculture

out of the trap. Most known loko 'umeiki were located on the island of Molokai, possibly owing to the favorable orientation of the island with regard to longshore currents. However, it is claimed that Pearl Harbor, on Oahu, had three or four of these types of traps and that one fish trap may have existed on land. These lanes connecting the traps with the ocean were used to catch fish migrating down the coastline, which were attracted to the surge of water at the lane entrances (Fig. 2.6). Fishermen simply set nets facing the sea across the opening of the lane to capture fish swimming into the trap on an incoming tide. When the tide reversed, fishermen faced their nets toward the traps, capturing fish as they swam out toward the sea. It was reported that the right to fish during different portions of the tidal cycle was divided among family groups.

Fig. 2.6 Plan of a fish trap (loko'umeiki) called Papa'ili'ili on the island of Molokai wedged between two marine fishponds (loko kuapa) (called Kaina'ohe and Keawanui). Details of three pond outlet canals (A,B,C) and one pond inlet canal (D) are shown. Note the enlarged wall sections on canals B, C, and D accommodating fishers. These areas indicate where nets were set to capture fish on rising (D) and falling (A,B,C) tides. Modified from Stokes (1909).

The Ahupua'a Aquaculture Ecosystems in Hawaii

41

`Such was the case of Mikiawa Pond at Ka'amola, Molokai. When the tide was coming in, the people of Keawanui could use the lanes. When the sea ebbed, the fish belong to Ka'amola.' Timoteo Keaweiwi, 1853, Foreign Native Testimony Book 16, State of Hawaii Archives, Honolulu, Hawaii (Summers, 1964)

The context of the Hawaiian innovations in the evolution of mariculture ecosystems It is evident that the ancient Hawaiians supported a high population density by managing an ecologically complex integrated farming system that connected agricultural watershed ecosystems to nearshore mariculture/fisheries ecosystems ± the ahupua'a aquaculture ecosystem (Fig. 2.7). These historical developments are

Fig. 2.7 An idealized Hawaiian ahupua'a aquaculture ecosystem showing topographic placement of freshwater, brackishwater, and nearshore integrated aquaculture ecosystems on the landscape. Stars indicate placement of settlements. Hawaiian aquaculture ecosystems were diverse, unique, and well adapted to the wide range of natural environments and social structures present. Typical valley ecosystems of this type would be approximately 10 km from mountains to ocean and 10±20 km along the shoreline.

42

Ecological Aquaculture

remarkably similar in principle to integrated farming systems developed in ancient China and Egypt. Hawaiian society, like other ancient civilizations, was subject to droughts, climatic disruptions, natural disasters, and famines; it may have developed these integrated farming systems in response. The limited archeological and aquaculture research, as well as exploration in the Pacific Basin, allows no conclusions to be drawn either regarding the uniqueness of the Hawaiian integrated farming systems among the Pacific islands, as some have suggested (Summers, 1964; Kikuchi, 1973), or their possible relationships to Chinese or other Asian systems. The Hawaiians appear to be one of the originators of mariculture; there is no evidence of another ancient culture using oceanic resources in this manner (Costa-Pierce, 1987). Most of the previous work on early Hawaiian aquaculture focused on the marine fishponds. These studies concluded that the ponds were technologically primitive, ecologically inefficient, and unproductive in biomass per unit area when compared with Asian practices (Hiatt, 1947a, b; Kikuchi, 1973, 1976). But these earlier interpretations may be inappropriate in light of the total farming ecosystem, which spanned an extensive salinity range of water, encompassed entire valleys, and integrated watersheds and nearshore marine ecosystems in unique and possibly unprecedented ways.

References Apple, R.A. & Kikuchi, W.K. (1975) Ancient Hawaii Shore Zone Fishponds: An Evaluation of Survivors for Historical Preservation. National Park Service, Honolulu. Beckley, E.M. (1883) Hawaiian Fisheries and Methods of Fishing. Advertiser Stream Print, Honolulu. Cobb, J.N. (1902) Commercial fisheries of the Hawaiian Islands. US Fish Commission, 12, 383± 499. Cordover, J. (1970) Molokai fishponds: manmade ecosystems. In: Molokai Studies: Preliminary Research in Human Ecology (ed. H.T. Lewis), pp. 27±44. University of Hawaii Press, Honolulu. Costa-Pierce, B. (1987) Aquaculture in ancient Hawaii. BioScience, 37, 320±331. Ellis, W. (1826) Journal of William Ellis. Notes from a Tour of Hawaii (Owhyhee) 1823. London. Farber, J. (1997) Ancient Hawaiian Fishponds. Neptune House Publications, Encinitas, CA. Hamre, C.J. (1945) A Survey of Nine Commercial Fishponds. Cooperative Fisheries Research Unit, University of Hawaii, Honolulu. Handy, E.S.C. & Handy, E.G. (1972) Native Planters in Old Hawaii. Bishop Museum Press, Honolulu. Hiatt, R.W. (1947a) Food-chains and the food cycle in Hawaiian fish ponds ± Part I. The food and feeding habits of mullet (Mugil cephalus), milkfish (Chanos chanos), and the ten-pounder (Elops machnata). Transactions of the American Fisheries Society, 74, 250±261. Hiatt, R.W. (1947b) Food-chains and the food cycle in Hawaiian fish ponds ± Part II. Biotic interactions. Transactions of the American Fisheries Society, 74, 262±280.

The Ahupua'a Aquaculture Ecosystems in Hawaii

43

Kamakau, S.M. (1976) The Works of the People of Old. Bishop Museum Press, Honolulu. Kelly, M. (1980) Majestic Ka'u: Mo'olelo of Nine Ahupua'a. Report 80-2. Bishop Museum Press, Honolulu. Kikuchi, W.K. (1973) Hawaiian aquacultural systems. Ph.D. Dissertation, University of Arizona, Tucson. Kikuchi, W.K. (1976) Prehistoric Hawaiian fishponds. Science,193, 295±299. Lind, A.W. (1938) An Island Community, Ecological Succession in Hawaii. University of Chicago Press, Chicago. Madden, D. & Paulsen, C.L. (1977) The Potential for Mullet and Milkfish Culture in Hawaiian Fishponds. State of Hawaii Department of Planning and Economic Development, Honolulu. Malo, D. (1951) Hawaiian Antiquities (Mo'olelo Hawaii). Bishop Museum Press, Honolulu. McAllister, J. (1933) Archaeology of Oahu. Bishop Museum Bulletin 104, Honolulu. Phelps, S. (1937) A Regional Study of Molokai. Bishop Museum Press, Honolulu. Rice, W.H. (1923) Hawaiian Legends. Bishop Museum Press, Honolulu. Sahlins, M.D. (1958) Social Stratification in Polynesia. University of Washington Press, Seattle. Stokes, J. (1909) Molokai Survey Field Notebook. Bishop Museum Press, Honolulu. Summers, C.C. (1964) Hawaiian Fishponds. Bishop Museum Press, Honolulu. Titcomb, M. (1952) Native Use of Fish in Hawaii. Polynesian Society Press, Wellington, New Zealand. Wilder, G.P. (1923) The Taro, or Kalo, in Hawaii. University of Hawaii Extension Service, Honolulu. Wyban, C. (1992) Tide and Current: Fishponds of Hawai'i. University of Hawai'i Press, Honolulu.

Part 2

The Methods of Ecological Aquaculture

Chapter 3

Development and Application of Genetic Tags for Ecological Aquaculture Theresa M. Bert, Michael D. Tringali and Seifu Seyoum Florida Marine Research Institute What is a genetic tag? The term `genetic tag' can refer to any DNA-based marker that can be used to identify individuals (Palsboll, 1999), parentage and family relationships (Garcia De Leon et al., 1995, 1998), members of populations (Moritz, 1994; Triantafyllidis et al., 1999), or even members of species (e.g. Bilodeau et al., 1999; MaKinster et al., 1999; Frischer et al., 2000). Two common uses of genetic tags are to identify specific wild individuals that may be present within groups of individuals (e.g. Paetkau & Strobeck, 1994; Richard et al., 1996) and to track released or escaped offspring of parents bred in captivity (e.g. Garcia De Leon et al., 1998; Hedrick et al., 2000). Ideally, the perfect genetic tag at the most precise level would allow discrimination of each individual in a group from every other individual. The recent, relatively inexpensive availability of DNA sequencing (the determination of the sequence of the nucleotides that comprise DNA) and the development of numerous ways to `look at' DNA have essentially opened the door for the discovery of all types of genetic tags. In many cases, genetic tags with the appropriate level of specificity can be developed in only a few weeks or months. Of course, the more that is known about the DNA of a species or group of species, the easier it usually is to develop a genetic tag, particularly a highly specific one. The use of genetic tags has enhanced our ability to understand contemporary patterns of genetic divergence within and among populations (e.g. Stepien, 1995, 1999); levels of gene flow (Taggart & Ferguson, 1986); many biological and life history characteristics of animals and plants valuable for both theoretical biology and conservation biology (e.g. Packer et al., 1991; Coltman et al., 1998); and the genetic effects of various types of captive breeding program, including aquaculture (e.g. ProdoÈhl et al., 1995), and accidental or purposeful release of captively bred individuals into wild environments (e.g. Mathias et al., 1985; Doyle et al., 1995; Tessier et al., 1997). Particularly the specific identification of individuals via genetic tags has enabled the estimation of parameters that are relevant for studying important ecological and conservation issues (Palsboll, 1999). Thus, the search for and development of genetic tags, particularly for threatened or managed species, is a field of intense activity and paramount importance. Genetic tags used for studies of organismal biology, population biology, ecology,

48

Ecological Aquaculture

evolutionary biology, conservation biology, or applied biology (including aquaculture) can be classified into two general categories: those that are produced by artificial (anthropogenic) selection and those that occur naturally. Selected genetic tags are those genetic alleles or traits that are rare in natural populations but specifically enhanced in cultured populations through a process of selective breeding (see Gharett & Seeb, 1990). Individuals that have a particular rare genotype at a specific gene locus (physical location within the genome) are sought after, brought into captivity, and bred. Subsequent generations of offspring then carry the selected rare allele in high frequencies (up to 100%, if each parent is bred to be homozygous for [i.e. to have identical forms of] the allele). These genetic tags allow cultured organisms to be identified if they are released or escape into wild environments containing a conspecific population in which the targeted allele is very low or absent (e.g. King et al., 1993). Naturally occurring genetic tags are usually composed of one or more segments of highly variable or hypervariable DNA. The advantage of finding this type of genetic tag is that no directed selection is involved in the development or use of the tag. Indeed, most types of DNA that are investigated for their potential as useful genetic tags are thought to be essentially free from selection. A naturally occurring genetic tag allows an individual to be identified in samples collected from a population because the individual's genotype is naturally unique. If individual identity is the objective of the search for a genetic tag, the tag must be so hypervariable that, essentially, every individual has a unique genotype. The discriminatory power of this tag is such that the probability of discerning each individual from another approaches 100%. A genetic tag of this specificity is usually termed a DNA fingerprint (Wright, 1993; O'Reilly & Wright, 1995) and might involve numerous loci.

Types of genetic tag Molecular identification of individuals or groups of individuals can be accomplished by directly looking at DNA sequence or some property of DNA (e.g. length, fragment pattern, mobility) or by looking at DNA products (e.g. the electromobility of alleles at allozyme loci, protein sequences). The choice of DNA tool to use depends on the ability to perform the laboratory analyses needed to find and resolve the DNA or its products and the specific level of discrimination required for the project at hand. For animal research, DNA markers can be developed from two basic types of DNA: nuclear (chromosomal) and mitochondrial (mtDNA). In general, any DNA segment that can be analyzed with reliable repeatability by the technique chosen by the researcher, for which the data can be clearly interpreted, and that possesses a high to hyper degree of variation is a suitable candidate for exploration as a genetic tag. For certain situations, a single segment of DNA serves adequately as the tag; in other cases, the level of discrimination needed may require the tag to be composed of several DNA segments and multiple types of DNA markers. Typical types of nuclear DNA explored for use in developing genetic tags include highly repetitive DNA such as minisatellite and microsatellite DNA, and

Development and Application of Genetic Tags

49

DNA with only moderate to low numbers of repeats, such as introns, genes of the major histocompatibility complex (MHC), and anonymous single-copy, nuclear DNA (ascnDNA). Using certain techniques, even the entire nuclear genome can be used to generate a genetic tag. Minisatellite and microsatellite DNA are collectively known as VNTRs (DNA with variable numbers of tandem nucleotide repeats); microsatellites are also termed SSRs (simple sequence repeats), or STRs (simple tandem repeats). A review of VNTRmarker development and applications can be found in O'Connell & Dillon (1997). Briefly, mini- and microsatellites are regions of DNA composed of sequential repeats of short nucleotide sequences (e.g. AGCCTAGTAGCCTAGTAGCCTAGT . . . , or ACACAC . . .). The typical number of nucleotides in a single minisatellite repeat unit ranges from 10 to 100 base pairs (bp); the typical number in a single microsatellite repeat unit ranges between 2 and 5 bp (Dowling et al., 1996). Allelic polymorphism in mini- and microsatellite DNA is the product of variability in the number of repeated units of these short sequences and is manifested as DNA fragments of different lengths. Levels of polymorphism and numbers of alleles at VNTR loci are generally high within species or populations and may vary greatly among taxa. Minisatellites have an advantage of showing the highest degree of Mendelian polymorphism of all known loci. Thus, they are particularly useful for establishing pedigrees (Burke et al., 1996). Microsatellite loci are essentially replacing allozyme loci as the molecular tool of choice for many types of population genetics studies, including those for detection of interspecific hybridization, within- and between-population genetic diversity, relationships among individuals (including parentage), reproductive success (Bruford et al., 1996), and stock identification (Perez-Enriquez & Nobuhiko, 1999). Often, microsatellite regions are located in regions of DNA called introns. Introns are segments of genes that undergo DNA transcription but not translation. They are distributed throughout the nuclear genome and may be duplicated multiple times in different regions of the genome; each duplicated segment more or less undergoes an independent evolutionary pathway. Introns themselves can be highly variable and can be used as molecular markers. They are highly variable principally because they are non-coding regions of DNA; however, they are likely to be (and some evidence shows that they are) very much affected by selective forces because they are tightly linked to their associated exons (transcribed and translated segments of DNA). In some cases, balancing selection, which can facilitate the maintenance of polymorphisms in loci, has been proposed to account for the high level of genetic diversity in introns. For these reasons, and because a high degree of specificity is needed to identify individuals, several intron loci (hopefully from different linkage groups ± e.g. different chromosomes) used congruently, are usually needed for a valid study. MHC genes belong to a large `family' of functional genes that play critical roles in the immune response of vertebrates to foreign pathogens (Edwards & Potts, 1996). Certain classes of MHC genes have been recognized as being the most polymorphic of all known functional genes (see Hedrick, 1994); some loci contain as many as 100 alleles. There is an ongoing debate about the mechanisms of selection thought to maintain this incredible amount of diversity (e.g. Potts & Wakeland, 1990; Watkins et

50

Ecological Aquaculture

al., 1992). Regardless of the debate, these genes have proven to be useful in genetic surveys of vertebrate populations and in fingerprinting applications (see Edwards & Potts, 1996). For non-insect invertebrates, different hypervariable supergene families are known to occur, but these are not well characterized (Marchalonis & Schluter, 1990). AscnDNA is DNA that is arbitrarily chosen from anywhere in the genome and that has an unknown function. These fragments may be segments of coding regions or of non-coding regions of DNA. Thus, some fragments may be under selection whereas others may not. Thus, and for other reasons, the levels of variability in nucleotide sequence of this DNA vary greatly. Advantages to using ascnDNA fragments are that they are randomly chosen from throughout the entire nuclear genome and, overall, the fragments being assayed probably are not under direct selection pressure. AscnDNA fragment data has been used to determine levels of genetic diversity, various types of population genetic structure, and family relationships (see references in Karl, 1996). As genetic tags, ascnDNA fragments will probably be best utilized in combination with other types of DNA-based information. Mitochondrial DNA, which is extranuclear DNA that resides in the mitochondria of organisms, is composed almost exclusively of transfer RNA genes, ribosomal RNA genes, and genes that code for proteins. These genes are arranged in tandem in the circular mtDNA molecule. Unlike nuclear DNA, which undergoes genetic recombination during replication and is biparentally inherited, mtDNA does not undergo recombination and is effectively inherited without change from the mother to offspring (but see Zouros et al., 1994). Because of this unique inheritance pattern, the mtDNA molecule is frequently considered to be a single, linked genetic unit. Nevertheless, different mtDNA genes (e.g. those for cytochrome b and 16S rRNA) vary considerably in their rates of nucleotide substitution (reviewed in Palumbi, 1996, and Kocher & Stepien, 1996). On the average, the mtDNA genome is more variable than the majority of nuclear genes because of its mode of inheritance and because the rate of mutational error during the process of mtDNA replication is generally higher than that during nuclear DNA replication. Some mtDNA genes that have relatively high substitution rates can be used together in various ways to construct genetic tags (see Methods, below). One, usually hypervariable, non-coding region of mtDNA, the control region (called the D-loop in vertebrates and the A+T-rich region in noninsect invertebrates; Taberlet, 1996), is frequently used as a genetic tag (e.g. Stabile et al., 1996; Bentzen et al., 1997; Seyoum et al., 2000). Its overall substitution rate is among the highest known because, in addition to the factors cited above, it does not contain genes that are involved in the coding process. The ability to easily use the mtDNA control region as a genetic tag varies greatly among taxa because of differences among taxa in accessibility of the region and in its degree of variability. In vertebrates, the gene sequence in mtDNA is relatively constant among taxonomic groups (e.g. fish, mammals). Segments of the control region or the entire region can be accessed using standard, published primers (short segments of DNA that, after the mtDNA has been denatured, attach to segments of conserved mtDNA genes that flank the mtDNA region of interest [e.g. the control region] and, through the polymerase chain reaction [PCR], facilitate the synthesis of many copies

Development and Application of Genetic Tags

51

of that region). For example, the D-loop of many fish species can be accessed using primers published in the now classic paper by Kocher et al. (1989). However, in noninsect invertebrates that are commonly cultured (e.g. shrimp, bivalve mollusks), the A+T-rich region is not always in a fixed position relative to flanking conserved genes, and in some cases (e.g. certain bivalve species; Gjetvaj et al., 1992), it cannot be located by standard techniques. Thus, compared with vertebrates, relatively little is known about the levels of variability of this region in non-insect invertebrates. Even within taxonomic groups, levels of genetic variation in the control region differ greatly. For example, in some fish, segments of the D-loop are extremely hypervariable (see, for example, Seyoum et al., 2000), and relatively short segments (e.g. 450±600 bp) can serve alone as genetic tags for some needs in genetic monitoring of aquaculture activities. In contrast, in other fish species, the level of variability is comparatively low, possibly due to genetic bottlenecking or founder events in the recent past (see, for example, Wilson et al., 1997). Despite these shortcomings, the mtDNA control region is probably the segment of DNA most frequently used for differentiating intraspecific interbreeding populations (e.g. Taberlet, 1996). It also has great utility for examining maternal contribution to reproductive output, and, either alone or together with other types of DNA, for assessing effective population size (Tringali & Bert, in prep.). Nuclear DNA products such as proteins can be used as genetic tags, but because of the redundancy of the genetic code (see Hartl & Clark, 1997), the variation associated with these molecules is inherently less than that of the DNA that codes for these gene products. If protein sequences or allozyme alleles are used as genetic tags, batteries of these need to be used, and probably used in conjunction with other types of DNA markers. These types of markers are typically used to discriminate groups or organisms from each other (e.g. Taggart & Ferguson, 1984; Crozier & Moffett, 1995; Bert et al., 1996). Allozyme electrophoresis is the oldest widely used technique for assaying genetic variation at the molecular level. The widespread use of allozyme electrophoresis for theoretical and applied population-level studies began with the publication of Lewontin & Hubby's (1966) paper on genetic variation in Drosophila pseudoobscura. Allozyme electrophoresis has been used for numerous studies of phylogenetics, population genetic structure, interspecific hybridization, species identification, genetic stock identification, and studies that require genetic tags (e.g. Taggart & Ferguson, 1984, 1986; Nyman & Ring, 1989). For example, rare allozyme alleles have been used as selected genetic tags a number of times for stock enhancement (e.g. Murphy et al., 1983; Taggart & Ferguson, 1984, 1986; Knut et al., 1994). However, because in allozyme electrophoresis gene products, proteins, are assayed, the process is a very conservative estimate of genetic variation. Thus, the utility of allozyme electrophoresis as a natural genetic tag for identifying individuals or specific family lineages is limited to that of either a preliminary screening test or a secondary technique, to be used in conjunction with other, more sophisticated methods of analysis. Morphological tagging is the oldest method of demarcating cultured organisms. Morphological tags often represent phenotypic expressions of genetic traits. They have been used to distinguish hatchery-bred fish (e.g. cultured red sea bream (Pagrus

52

Ecological Aquaculture

major) in Japan; T.M. Bert, personal observation) and mollusks (e.g. color morphs of bay scallops (Argopecten irradians), Adamkewicz & Castagna, 1988; notata form of hard clams (Mercenaria mercenaria), Humphrey & Walker, 1982; Castagna & Manzi, 1989) from wild conspecific individuals. The advantage of morphological markers is that, if they are reliable (i.e. optimally controlled by a single locus and homozygous in the cultured population) and obvious, they allow the easy and unambiguous distinction of the marked individuals from all others. However, the potential drawbacks associated with using morphological tags are numerous. As with molecular tags, genetic selection for a phenotypic marker is unavoidable (in fact, it's the objective), and because several generations of offspring must usually be bred to render the marker homozygous in the cultured population, changes in genetic diversity are inevitable. The genes coding for the marker may be closely linked to genes at flanking or nearby polymorphic loci; the allele frequencies of those closely linked genes may be altered also because their alleles may be `carried' along with those of the marker genes during DNA recombination. Thus, selection on genes of unknown (and perhaps critical) function may occur. In addition, researchers cannot track offspring of cultured animals if the inheritance pattern of the phenotypic trait is not known. Specifically, hybrids are not identifiable if the alleles for the locus controlling the trait are not co-dominant and hybrids are not expressed as intermediates. Because the genetic basis of morphological tags is generally poorly understood and because the use of morphological tags almost invariably involves selection, they will not be further addressed here.

Methodologies for obtaining and utilizing genetic tags Thanks in large part to the Human Genome Project, a plethora of molecular techniques to identify DNA at various levels of precision have been developed in recent years. Many techniques used to analyze human DNA have also been used to search for and develop genetic tags. The following descriptions of the techniques and procedures are greatly simplified; for a more detailed explanation, the reader should consult the cited references. The most direct method for analyzing DNA is to determine its nucleotide sequence. DNA sequencing can be used to analyze any DNA segment, including the various types of highly variable DNA of interest in the context presented here (e.g. introns, ascnDNA, mtDNA control region). DNA sequencing offers the researcher an opportunity to directly observe individual nucleotide bases along a targeted region of DNA. The first step in this process is to find and isolate the segment of DNA that can serve as a candidate for a genetic tag. This step entails searching for hypervariable segments of either nuclear DNA or mtDNA, which is usually accomplished by cloning or by using primers. Cloning is a process in which a desired segment of DNA is isolated and inserted into the genome of a host organism (bacterium). Because bacteria multiply rapidly, the inserted segment of DNA is reproduced (cloned) in large quantities for genetic assay. A review of the various cloning techniques and applications may be found in Sambrook et al. (1989). Primers usually range from

Development and Application of Genetic Tags

53

about 20 to 35 nucleotides long. Pairs of primers are designed to be complementary to the two opposing strands of conserved (non-variable) regions of DNA that flank the region of interest (in this case, a hypervariable region). Many primers exist for all types of hypervariable DNA, both nuclear and mitochondrial, and their nucleotide sequences can be easily found in the scientific literature or via Genbank (http:// www.ncbi.nlm.nih.gov/irx/cgi-bin/genbank). Primers are often used for isolation of the target DNA, a process that involves the polymerase chain reaction (PCR; Saiki et al., 1989). PCR is a chemical process in which large quantities of identical copies of specific segments of DNA can be synthesized from limited and even somewhat degraded amounts of DNA. A review of the techniques, common protocols and primers used, and general applications may be found in Palumbi et al. (1991). The PCR product, which is composed of millions of copies of the desired segment of DNA along with the original (and much smaller) number of copies of all other DNA in the organism, is readily utilizable in direct DNA sequencing. Using cloned DNA or PCR products, DNA sequencing can be accomplished manually and visualized on acrylamide gels (Sanger et al., 1977; Craxton, 1991), or by using an automated DNA sequencing system. Often, preliminary sequencing is first performed on a few individuals (e.g. 10±20). This effort represents a developmental step in which `nested' primers are developed. These primers are obtained using DNA sequence segments that are interior to the segments to which the original primers attached. These specific primers are more custom-designed for the target species and are employed in subsequent, large-scale genetic assays (see, for example, Seyoum et al., 2000). Also, by first sequencing a small number of individuals, a researcher usually can form a preliminary opinion on the amount of nucleotide sequence variation that is contained in the specific DNA segment of interest and thereby evaluate the potential usefulness of that segment as a genetic tag. For surveys involving genetic tags, sequencing usually is the method of analysis used for hypervariable ascnDNA, introns, MHG genes, and the mtDNA control region. Because the sample sizes needed to ascertain the uniqueness of a genetic tag and to survey populations in which tagged individuals constitute some (possibly small) fraction of the population can be substantial (e.g. many hundreds of individuals), it may be beneficial to utilize one or more `short-cut' techniques that take advantage of such variation as the tertiary configuration of DNA or subtle differences in DNA charge that are associated with conformation of the DNA molecule and nucleotide base pairing. Several of these types of techniques are available; most are well summarized in Girman (1996). These techniques screen the DNA for sequence differences without the necessity of sequencing each individual. With these techniques, large numbers of individuals can be preliminarily screened and then only individuals of interest or uncertain genotypes sequenced. Among these techniques are singlestranded conformation polymorphism (SSCP) analysis, denaturing gradient gel electrophoresis (DGGE), and allele-specific oligonucleotide (ASO) hybridization. In aquaculture, these techniques are useful for determining and quantifying the parentage of hatchery broods. In SSCP analysis, short DNA fragments (100±300 bp provides the best resolution) are amplified by PCR and radiologically or chemically labeled for later visualization.

54

Ecological Aquaculture

The amplified DNA is then denatured (put through a process to separate the two sides of the DNA nucleotide ladder). Under specific conditions, the single-stranded DNA will then fold on to itself and assume a tertiary configuration that is characteristic of the particular base-pair sequence in the strand. Then the DNA is run through a polyacrylamide gel under a very specific and precise set of conditions. These conditions must be determined for each project and must be strictly adhered to in all SSCP procedures performed on all individuals that are analyzed to insure that each different single-stranded DNA nucleotide sequence always remains in its characteristic sequence-specific folded position. Ideally, the differences in folded positions among DNA nucleotide sequences cause each different sequence to migrate a unique distance through the gel. Thus, individuals with unique nucleotide sequences can be identified. The DNA of these individuals can be sequenced to confirm their uniqueness. The DGGE technique is highly sensitive to mutations in DNA sequence. In DGGE analysis, double-stranded, PCR-amplified DNA from different individuals is run through an acrylamide gel that has a gradient of denaturing chemicals (typically urea and formamide) within it. As the DNA runs through the gel matrix, it reaches an area in the gel where it begins to denature. This markedly slows the migration of the DNA through the gel. Ideally, DNA with different nucleotide sequences runs different distances through the gel. Single-nucleotide substitutions in the DNA of different individuals can be distinguished by these differences in migration distance from the original position of the DNA in the gel. To ascertain the nucleotide sequence of the target DNA segment for any individual, the DNA can be excised and extracted from the gel and then sequenced using standard techniques. A modification of the DGGE technique is temperature gradient gel electrophoresis (TGGE). In this technique, temperature, rather than chemicals, is used to generate the gradient in the gel. ASO hybridization can be used to determine the genotypes of many individuals quickly and inexpensively. The nucleotide sequence of a target DNA segment must be known for some reference individuals (e.g. broodstock individuals). In addition, each individual must possess at least one mutation in this segment of DNA that is unique compared to the other reference individuals. For each individual, an allele-specific probe (short segment of DNA with a nucleotide sequence that is complementary to a segment of the target DNA that contains the unique mutation) is created (this can be custom-ordered from a company specializing in such products). The target DNA of many individuals is annealed to filter paper. Under a specific temperature regime, the DNA is denatured and a probe for one specific mutation is applied; prior to applying the probe to the DNA affixed to the filter paper, the probe is labeled with a chemical that, later on in the process, will allow the probe to be visualized. If the conditions are correct, the probe will attach at the appropriate location on the filter paper only at sites where the DNA of individuals with precisely the complementary nucleotide sequence to the probe have been placed. Then, through a series of chemical reactions, the reconstituted probe is visualized. Because the probe is situated on the filter paper only at sites where the DNA of individuals that have the probe-specific mutation has been placed, those individuals can be identified. Kits are available for labeling the probe and detecting the labeled probe. The invention and affordability of DNA sequencing machines has facilitated the

Development and Application of Genetic Tags

55

generation of a veritable explosion of DNA sequence data for all types of research. For organismal biology, DNA sequencing data is fundamental to studies of phylogenetics, population genetics, conservation biology, and increasingly, fisheries biology (for stock identification) and aquaculture (for pedigree analysis). Review and applications can be found in Hillis et al. (1996), Kocher & Stepien (1996), and Smith & Wayne (1996). Examples abound in all journal series that publish articles on evolutionary biology, ecology, genetics, fisheries, and many other topics that deal with organismal biology. Microsatellite DNA loci are usually developed from genomic `libraries' of DNA fragments that are produced by treating total, purified DNA with restriction enzymes (enzymes that recognize specific DNA nucleotide sequence sites 4±6 nucleotide bases long, attach to the DNA at those locations, and sever the DNA at or near those locations). Libraries may be enriched for microsatellite-containing DNA fragments by a variety of means (see, for example, Bruford et al., 1996). The library is then checked for DNA fragments that contain microsatellite DNA regions by cloning. Next, bacterial clones having microsatellite DNAs are identified by hybridization (denaturing the target DNA and reconstituting with radio- or chemically labeled, synthesized, simple-sequence [e.g. CACACA . . . , ATTATT . . .] DNA) and sequencing. Specific primers are then developed for the cloned DNA fragments that contain microsatellites and are used in PCR reactions performed on the total DNA extracted from tissues or DNA-containing fluids obtained from individuals. The genotypes of the individuals are visualized by running the PCR products through an acrylamide gel electrophoresis system and staining the gel with an appropriate dye or stain or by using an automated DNA analyzer and an appropriate microsatellite DNA protocol based on fluorescent dye chemistry. Assays for microsatellite loci may be streamlined by `multiplexing' (see O'Reilly et al., 1996). In this procedure, two or more microsatellite loci can be co-amplified in a single PCR reaction and assayed during a single automated sequencing reaction. Like most allozyme loci (see below) and single-locus minisatellite loci, microsatellite loci usually have co-dominant alleles, which allow homozygous and heterozygous genotypes to be distinguished in individuals. These loci can usually be analyzed in accordance with the assumptions associated with Mendelian inheritance (e.g. Hardy±Weinberg genotype equilibrium assumptions; see Hartl & Clark, 1997). Minisatellites are generally assayed using restriction enzyme digestions of total DNA. The resulting mixtures of DNA fragments are separated according to their sizes (lengths) via gel electrophoresis. The minisatellite loci are detected by using one or more minisatellite-specific DNA probes (for detailed methodologies, see Jeffreys et al., 1985a, b, and Burke et al., 1996, and references therein). The raw data for individuals are often in the form of complex patterns (restriction fragment profiles) in which multiple loci are present. Thus, individual genotypes for particular loci are indistinguishable. However, because the minisatellite profiles are often individualspecific, they are highly useful for DNA fingerprinting and for determining parentage in individuals. Allele frequencies cannot be calculated and standard analyses of population genetic parameters (e.g. estimation of gene flow and effective population size) cannot be performed without making extensive simplifying assumptions (Burke

56

Ecological Aquaculture

et al., 1996), but measures of genetic similarity between individuals and within and between populations can be calculated based on `band-sharing' coefficients (Gilbert et al., 1990, 1991). In some cases, the analytical disadvantages of standard minisatellite assays have been overcome by the development of single-locus minisatellite markers (e.g. Bentzen et al., 1993). This usually requires cloning (Bentzen & Wright, 1993) during the initial period of marker development, which can be a time-consuming and expensive process. In general, VNTR markers have proven to be highly successful for understanding the genetic population structure within species (Bentzen et al., 1996; O'Connell et al., 1998; Ruzzante et al., 1998), gene flow (Ruzzante et al., 1996), the genetic basis of phenotypic traits (Garcia De Leon et al., 1998), and parentage determinations (O'Reilly et al., 1998; Colbourne et al., 1999). Most importantly for our purpose, they have proven their utility as genetic tags for use in ecology (Woods et al., 1999), conservation biology (Lehman et al., 1991; Wayne & Jenks, 1991; Avise, 1994), and applied biology (e.g. see review by O'Connell & Dillon, 1997), including aquaculture (Harris et al., 1991; Harris & Wright, 1995; McConnell et al., 1995; Morris et al., 1996). PCR primers for single-locus VNTR markers of many fish and shellfish species are available from the scientific literature (e.g. Bentzen et al., 1994; Takagi et al., 1996; O'Connell et al., 1998; Tassanakajon et al., 1998; Turner et al., 1998). As with any nuclear DNA marker, if a number of independent loci can be resolved for the species or population of interest, the probability that the genetic analyses are not strongly biased by undetected selection at any single locus is reduced. The advantages of VNTR markers are that they are highly polymorphic and generally provide information on both paternal and maternal contributions to offspring. When VNTR characters are co-dominant, they are amenable to many types of population genetic analyses (e.g. standard analyses of genetic variability; associations among populations, McConnell et al., 1997; gene flow, O'Connell et al., 1998; migration rate; and effective population size, Herbinger et al., 1997. In addition, co-dominant VNTR markers have been useful in statistical `assignment tests' in which individuals are assigned to their source populations on the basis of their genotypes (Paetkau et al., 1997). Restriction fragment length polymorphism (RFLP) analysis is a technique that randomly surveys genetic variation in a targeted region of DNA (e.g. ascnDNA or a segment of mtDNA); or, in the case of mtDNA, the whole molecule may be assayed. Purified mtDNA, nuclear DNA, DNA PCR products, or total genomic DNA can be surveyed in this way. A number of restriction enzymes are used separately to cleave the DNA at their respective recognition sites, yielding for each restriction enzyme a reproducible array of variously sized DNA fragments for each individual assayed. The fragments are then separated according to size by gel electrophoresis. For each individual, the fragment pattern (composite of all of the variously sized DNA fragments) produced by each restriction enzyme is given a numerical or letter label. For each restriction enzyme, all individuals that produced the same fragment pattern are given the same label. The composite of the labels for all restriction enzymes is used to characterize the RFLP profile of each individual. Typically, only the restriction enzymes that produce fragment patterns that vary among individuals in the sample are included in the RFLP profile (see Sambrook et al., 1989, for methodological

Development and Application of Genetic Tags

57

details). Dozens of restriction enzymes are available for testing their ability to produce fragment patterns on the target species. Although RFLP analysis offers less resolving power than DNA sequencing, it remains a cost-effective alternative when large portions of a genome or large numbers of individuals need to be screened. Through RFLP analysis, nucleotide substitutions, large additions or deletions to the genome, or gene sequence rearrangements can be detected. RFLP analyses are typically performed on ascnDNA fragments and mtDNA (either segments composed of specific genes or the control region, or the entire molecule). The data generated are frequencies of specific RFLP patterns. By using multiple, hypervariable ascnDNA loci in RFLP analysis, individualspecific genetic tags may be obtained, but this may require considerable searching. The initial search for these fragments is performed by digesting total nuclear DNA with a restriction enzyme. Then a lengthy procedure (described in detail in Karl, 1996, and references therein) is performed, in which a recombinant nuclear DNA library is generated and screened for potential single-copy sequences. Sequence-specific primers are then designed for those sequences that are targeted for surveys of degree of polymorphism. These primers are used in PCRs; the RFLP analysis is applied to the PCR products and a number of restriction enzymes are used. A battery of those RFLPs that prove to be highly polymorphic may be used as a genetic tag. However, even when 4-bp restriction enzymes are used (these produce the most complex RFLPs), RFLP patterns usually exhibit insufficient polymorphism for use as genetic tags, at least when a tag that can discriminate among individuals is needed. In this case, nuclear DNA markers can be combined with mtDNA RFLP data to increase resolution. For aquatic organisms, RFLP studies have been used to resolve species relationships (phylogenetics; e.g. Ohland et al., 1995); identify species (e.g. Seyoum, 1990; Seyoum & Kornfield, 1992; Tringali et al., 1999a,b); determine genetic population structure of species for ecological/evolutionary studies (e.g. Triantafyllidis et al., 1999; Buonaccorsi et al., 2001); and identify fishery stocks (e.g. Bowen & Avise, 1990; Gold & Richardson, 1991, 1993; Gold et al., 1993, 1999; Tringali & Bert, 1996). RAPD (random amplified polymorphic DNA) fragments are molecular markers that are generated by the amplification of random DNA segments using single, 10-bp primers of arbitrary nucleotide sequences (Williams et al., 1990). As with ascnDNA techniques, RAPD markers are randomly drawn from throughout the nuclear genome. Theoretically the number of arbitrary 10-bp primers that can be generated is 410; thus, an enormous number of primers that can be used to generate different patterns of fragments for the DNA of a single individual are available. To generate the fragments, only one primer is used in a PCR in which either total genomic DNA (Williams et al., 1990) or pure mitochondrial DNA (Seyoum et al., in review) is the target DNA. Since the primer is short, there is a high probability that the DNA will have several to numerous priming sites on one strand of the double-stranded molecule and other priming sites in an inverted orientation on the complementary strand. Regions between pairs of these priming sites are amplified in the PCR of the target DNA. This produces several to numerous fragments of different sizes. The amplified DNA is then passed through an agarose gel under specific electrophoretic conditions for a defined period of time. The fragments separate according to their sizes. They are

58

Ecological Aquaculture

typically visualized by staining the gel with ethidium bromide. The pattern produced by the fragments is the RAPD profile of the organism for that particular primer. The pattern can be characteristic of species, populations, or individuals. RAPD genetic tags usually are composed of the profiles generated by several primers, used separately. At the highest degree of resolution, these profiles are specific to individuals. RAPD molecular markers can easily be obtained at low expense and they have advantages over other molecular markers in that RAPD analysis can be performed on a small quantity of template DNA and without prior DNA sequence information (Hadrys et al., 1992). However, great care must be taken to determine the ideal concentration, purity, and uniformity of reagents and conditions in the PCR. A slight change in any of these steps will usually lead to less resolution or different RAPD profiles. RAPDs have been used to analyze inter- and intraspecific population diversity (Dawson et al., 1993; Gibbs et al., 1994), distinguish species and subspecies in tilapia (Bardakci & Skibiniski, 1994), ascertain outcrossing in freshwater bryozoans (Jones et al., 1994), determine paternity and kinship relationships in conifers (Carlson et al., 1991), generate specific probes for taxonomic or RFLP analyses in soybeans (Williams et al., 1990), and develop genetic tags in mollusks (Seyoum et al., in review). Descriptions of the laboratory procedures and recipes for the stains needed to visualize allozyme loci can be found in Selander et al. (1971) and Richardson et al. (1986), and in the appropriate scientific literature that describes allozyme studies and was published in the 1970s and, to a lesser degree, the 1980s. In allozyme electrophoresis, tissues or cell-containing fluids (e.g. blood) are homogenized and placed in a semi-solid medium (e.g. hydrolyzed, cooked potato starch). The gel is placed in an assembly with an electrolyte solution and electricity is passed through it. Enzymatic and other proteins (representing `alleles') that have different electrophoretic mobilities travel different distances through the gel. The alleles of a gene coding for a particular protein are visualized by treating a slice of the gel with a staining solution specific to that enzyme. Both alleles of individuals that are homozygous at a specific enzyme locus travel the same distance in the gel, whereas different alleles travel different distances in heterozygous individuals. Thus, in a manner similar to that for single-locus minisatellite and microsatellite alleles, specific allozyme alleles can be identified and genotypes can be established based on the numbers and locations of the bands produced on the stained gels. However, an important difference is that the allozyme alleles (also commonly termed electromorphs) can actually be multiple, different genetic alleles that have the same electrophoretic mobility. Nevertheless, these alleles commonly conform to Hardy±Weinberg equilibrium assumptions and are thus treated as representatives of the genes that presumably code for those electromorphs. Similar to RFLP and VNTR analyses, the enzyme loci of interest are those that yield polymorphic banding patterns, and a battery of enzymes constitutes the allozyme genotype profile of an individual. Innumerable allozyme studies have been conducted during the past 30+ years. Allozymes have been used for studies in phylogenetics, population genetics, ecology, fisheries biology, conservation biology, and aquaculture genetics (see, for example, Ayala, 1977; Avise, 1994; Leberg, 1996, or Hartl & Clark, 1997, and references therein).

Development and Application of Genetic Tags

59

The analysis of all types of molecular data has now been made much easier by several different types of programs that are available through web sites and free through the internet. These programs are usually developed for particular analysis of DNA, RFLP, or microsatellite data. Thus the programs are usually incomplete in one type of analysis or another and the analyst needs to search through the various programs for one that contains the desired types of analyses. Currently, the focus in developing these programs apparently is to generate a program that has the capacity to deal with large data sets and perform all types of analyses. Nevertheless, at this time there is no single software program package that has the capacity to do all possible types of analyses on a particular molecular data set. Useful and comprehensive websites in existence at this time include the following: for population genetic data analysis, http://www.megasoftware.net/ and http://anthropologie.unige.ch/ arlequin; for 193 phylogenic programs, http://evolution.genetics.washington.edu/ phylip/software.html; and for 1100 molecular and biochemistry analysis entries applicable to Windows, Macintosh, Linux and Unix computers, http:// genamics.com/software/.

Uses of genetic tags for aquaculture Intensive aquaculture For aquaculture activities in which the intent is to contain the cultured organisms up to the time of harvest, the principal use of genetic tags is to identify individuals of interest for some specific reason. For example, genetic tags can be associated with and used to track lineages (family members) that have been bred for specific phenotypic, metabolic, or physiological traits. Rapid growth, metabolic efficiency, attractive coloration, high reproductive capacity, and high survival rates in the culture environment are characteristics that are commonly selected for and cultivated in aquaculture operations. These traits are at least partially under genetic control. Genetic tags enable the tracking of lineages with superior performance in these desired traits. Genetic markers linked to these traits, which are usually controlled by multiple quantitative trait loci (i.e. genes that have alleles with stepwise, measurable differences in their contributions to the trait), can be used to identify, inventory, or register strains that have these commercially valuable characteristics (e.g. Herbinger et al., 1995; Jackson et al., 1998; Sakamoto et al., 1999). Related to this use for genetic tags is the determination of heritability of these quantitative characteristics. Rarely are complex expressed traits controlled by single gene loci. Most physical, physiological, and biochemical attributes are the products of a number of genes acting in consort or in a complex cascade of interactions. The degree to which offspring inherit the desired characteristic can vary considerably among lineages, depending upon the specific alleles possessed by the parents for all of the genes that participate in the expression of the characteristic. Levels of heritability for important aquacultural performance traits are reviewed in Tave (1986). Genetic tags can be used in pedigree analysis as a proxy to track the presence of suites of genes manifested as the high performance of a specific trait that is the product of the

60

Ecological Aquaculture

interactions of numerous quantitative trait loci (Tanksley, 1993). For example, Danzmann et al. (1999) identified microsatellite markers linked to presumptive quantitative trait loci (QTL) for upper thermal tolerance, and Sakamoto et al. (1999) identified microsatellite markers linked to QTL for spawning seasonality in hatchery populations of rainbow trout. Microsatellite markers linked to QTL have also been used to determine if sufficient additive genetic variation exists to make it worthwhile for the aquaculturist to perform selective breeding for the desirable trait. Without multiple alleles in at least many of the loci involved in the genetic pathway leading to expression of the desired trait and variation in the singular and collective performance levels of those alleles, little change in the performance of the trait can be achieved (e.g. Moav & Wohlfarth, 1976). For example, Fishback et al. (2000) employed the microsatellite markers developed by Sakamoto et al. (1999) to show that there was insufficient additive genetic variation for further selective breeding for spawning seasonality in a hatchery group of summer-spawning rainbow trout. Because the heterozygosity of the marker loci was significantly higher in the group of spring spawners compared with the summer spawners, the authors posited that greater selective gains could be achieved by focusing on the spring group. When investigating the variation in trait performances that may be influenced by intrinsic factors (e.g. genetic or phenotypic) or extrinsic factors (e.g. water temperature, salinity or soluble chemical concentrations; food amounts, types or additives), the aquaculturist may want to set up an experimental array in which the factor(s) vary among treatments (e.g. among ponds, tanks, or cages) in graduated intervals or in a matrix. In this type of aquaculture, multiple members of several different lineages of the cultured species are placed into each treatment and their performances are tracked over time. If external tagging is impossible, impractical, or may interfere with the performance of individuals in the experiment, lineages can be tracked using genetic tags. Such tracking is particularly amenable to experimental aquaculture. For example, Herbinger et al. (1995) used microsatellite fingerprinting to show that paternal and maternal effects strongly affect offspring growth and survival in hatchery-bred and -reared rainbow trout. In that study, large and highly significant differences in growth were observed among the progeny of specific sires and dams. In contrast, significant differences in survival were observed among the progeny of specific dams but not among the progeny of specific sires; however, there was no correlation between progeny survival and particular dam attributes such as length, age at maturity, or egg size. Similarly, highly significant differences have been observed in progeny survival among dams in communally bred and reared red drum (Tringali & Bert, in prep.). In that study, large and significant differences in progeny survival, detected by mtDNA fingerprinting, accrued after the fourth week and prior to the twentieth week after hatching (i.e. after major morphological development but before complete development). Genetic tags serve another purpose useful to intensive aquaculture: lost or escaped cultured native species can be found or tracked and their escapement rates quantified if they are genetically tagged such that the cultured animals are differentiable from their wild counterparts (see, for example, Crozier, 1993; Clifford et al., 1997, 1998). One of the most environmentally problematic situations for aquaculturists is the

Development and Application of Genetic Tags

61

accidental or purposeful release of the cultured animals into the environment. Feral, cultured, native-species individuals can interbreed with wild conspecific individuals. As a group, feral, cultured individuals are usually depleted or differentiated in genetic diversity compared with the local wild population. Particularly if they have been derived from non-native broodstock, they can carry alleles that are not adapted to local environmental conditions. If the interbreeding is extensive and successful (i.e. numerous, reproductively active individuals in the subsequent generations contain genes derived from the feral, cultured individuals), the local or regional species gene pool can suffer several consequences: (1) at some or many polymorphic loci, alleles (particularly rare alleles) can be lost from the gene pool; (2) allele frequencies can be significantly shifted in the gene pool; and (3) maladapted alleles can be introduced into the gene pool. An extensive literature exists in which the negative consequences of a change in a species' genetic diversity resulting from interbreeding with feral cultured conspecifics are documented and the reasons for the predominance of negative consequences are postulated (Hindar et al., 1991; Ryman et al., 1995). Genetically tagging the cultured individuals would provide a mechanism for distinguishing those individuals from wild conspecifics in stocked populations and, if the specific genetic tags were developed for different aquaculture operations, for identifying the aquaculture facility from which they escaped or were released. Successful interbreeding can also occur between individuals of closely related species and between pure-species individuals and hybrids. Thus, cultured species or cultured hybrids between a native species and an alien species could interbreed with a native species. The consequence in these cases is contamination of the native pure-species gene pool from foreign genes. The problem is that, in any of these situations, the outcome and impact on native species is unknown, can be dramatic, and cannot be predicted with confidence, even from similar situations that have occurred elsewhere. Genetic tags that allow easy identification of these cultured individuals and permit tracking of the introgression of their genes into native populations are particularly important in these situations (see Costa-Pierce & Doyle, 1997). Genetic tagging could benefit both the aquaculturist and those interested in conservation of native gene pools. If loss of the aquacultured animals is unknown to the aquaculturist, it may be to his/her benefit to have this information and thereby take the steps necessary to prevent such losses in the future. Although feral, cultured, aquatic organisms cannot be gathered and placed again into captivity as can feral, ranched, terrestrial organisms, it may be possible to reduce the potential losses that the aquaculturist experiences and the impacts that result from the introduction of these animals into the environment if their rates of introduction and disperal can be monitored and controlled through the use of genetic tags. Finally, genetic tags are necessary for identification and verification of marketed seafood. Currently, forensic identification of aquaculture products is used principally to verify the identity of labeled products produced from wild fisheries (e.g. Kenchington et al., 1993; Baker & Palumbi, 1996; Birstein et al., 1998; Palumbi & Cipriano, 1998; Baker et al., 2000). A future use of genetic tags could be to discriminate between products produced by aquaculture and identical products produced by the harvest of

62

Ecological Aquaculture

wild animals or among identical products produced by different aquaculture operations.

Stock enhancement The purposeful addition of fish or aquatic invertebrates to a natural, conspecific population that has been depleted for some reason is variously termed stock enhancement or stock supplementation (See Drawbridge, Chapter 11). Regardless of the terminology, the objective is singular: to put individuals gathered from another location or spawned and reared to some specified size in a hatchery into an area where the native population has been greatly reduced due to anthropogenic influences or natural environmental changes. Related to this effort is stock restoration, in which wild or hatchery-reared organisms are placed into an area where the local, conspecific population has been extirpated. Stock enhancement and stock restoration are becoming increasingly popular as management tools for replenishing depleted fisheries. The genetic concerns associated with these efforts are complex, and somewhat different between stock enhancement and stock restoration. Here, we will deal with the use of genetic tags in stock enhancement. Bert et al. (in prep.) have developed a multifaceted, comprehensive program to monitor the genetic effects of stock enhancement. They use genetic tags as an integral component of this program. In stock enhancement programs, genetic tags can be used to perform the following tasks: (1) (2) (3) (4) (5) (6) (7) (8)

(9)

(10)

track parentage and determine parental contribution in the hatchery; compare the genetic diversity of the cultured animals with that of the recipient wild population; estimate effective population size of the broodstock (and consequently of the broods); determine if genotype-specific selection is occurring in the hatchery or after release of the cultured animals; track dispersal and survival of the cultured animals after release; compare growth rates and other life history characteristics of the cultured animals after release with those of wild conspecifics in same cohort; determine the contribution of hatchery-derived animals to local and regional populations; track the level of interbreeding between wild individuals and released cultured animals (with some decay factor related to the inheritance pattern of the gene used); estimate the change in effective population size of the stocked population if interbreeding between the released, cultured animals and wild population occurs; and determine the rate of loss of physical tags.

Determining, in as much detail as possible, information about the actual contribution of each broodstock individual to the broods is a valuable exercise for the aquaculturist and any individuals interested in mimicking, in the hatchery, the genetic

Development and Application of Genetic Tags

63

diversity of the natural population. Because gene alleles are passed from the parents to their offspring essentially without mutation, a genetic tag that enables the researcher to distinguish each broodstock individual from all others can be used to identify the parentage of each brood individual (or a subsample of the brood). Thereby, the percentage of offspring that are derived from each parent can be calculated. In some breeding strategies used by aquaculturists, the broodstock individuals that actually contribute to the broods are unknown. Through the use of a genetic tag, not only can the contribution of each broodstock individual be estimated, but also the individuals that did not contribute to the brood can be identified. In some culture situations, the aquaculturist may choose to either reuse those individuals in other broodstock groups or to discard those individuals. Because numerous types of negative effects can result from decreases in genetic diversity, and because genetic diversity can be severely reduced in stocked populations if large numbers of cultured individuals interbreed with the recipient wild population, it is important to compare the genetic diversity of the cultured animals with that of the recipient wild population prior to release of the cultured animals. Breeding strategies can be devised, within the limits of the capabilities of the aquaculture facility, that produce broods in which the genetic diversity of the recipient wild population is mimicked as best as is possible. Some potential always exists for interbreeding between cultured animals and wild animals of the same or a closely related species in stocked population situations; thus, maintaining the wild-stock genetic diversity as well as possible is always a good precautionary measure. Effective population size (Ne) is a theoretical number generated through an equation that takes into consideration not only the number of broodstock in a population but also the relative contribution of each breeder to the next generation. (For a more complete description, see Sugg & Chesser, 1994, and references therein). Genetic diversity is directly related to broodstock size. Thus, small effective broodstock population sizes generate broods with small Nes and low genetic variation. If, in turn, those broods interbreed at high levels with the recipient wild population, the genetic variation of the stocked population can be greatly reduced, and the attendant concerns about the manifestations of reductions in the fitness and adaptive potential of the stocked population may need attention. When reproductive contributions are not systematically controlled (e.g. through strip-spawning or paired matings), genetic tags must be used to estimate the Ne of the broodstock, the broods, and the stocked population. Selection for particular genotypes can be very strong in aquaculture facilities (Kohanne & Parsons, 1988). The aquaculture environment is usually very different from the natural environment in which the species lives. Genotypes that may be adaptive in the habitats occupied by a species may be maladapted to the culture situation, and the reverse situation can occur. Thus, in a brood, many of the individuals that are ultimately released may be poorly adapted for life in the wild (e.g. Fleming & Einum, 1997). Survival, growth, or other important life-history components may be impaired, ultimately reducing the positive effects of the stock enhancement efforts. In addition, selection in the aquaculture environment can negatively impact the genetic diversity of the broods and reduce the brood Ne despite

64

Ecological Aquaculture

the best attempts to maintain the wild-population level of genetic diversity through the breeding strategy. By periodically assaying a sample of the broods and checking for directional changes in allele frequencies, genetic tags can be used to check for selection in the aquaculture environment over the time that the broods are reared in the aquaculture facility. Ideally, the aquaculturist or researcher can develop a genetic tag that will discriminate cultured animals (and, with some decay factor, their offspring) from wild conspecifics with a high degree of confidence. Then, by systematically sampling (often non-destructively, e.g. by taking fish fin-clips) individuals captured from the stocked population and identifying them as cultured or wild, a number of important lifehistory attributes can be assessed independently for the cultured animals (e.g. dispersal from the release site, survival over time) or compared between the cultured animals and their wild conspecifics (e.g. growth rates, size at reproduction). For example, both to the aquaculturist or scientist responsible for the stock enhancement program and to the funding source, an important question related to the stock enhancement effort is, `Are the cultured animals surviving after release?' Or, knowledge of whether or not the released, cultured animals integrate well into the recipient wild population and how far they disperse over time can be important when considering the genetic effects of the stock enhancement or planning future releases. The location of the release site, timing of the release, and size of individuals to be released can be important when considering the stock enhancement effort from an economic perspective. All of these factors should be optimized such that they yield the highest survival rates of the cultured animals after release and allow for good integration of the cultured and wild individuals. These are factors that can be tested in a preliminary phase of the stock enhancement effort; they can be determined by structured sampling of the stocked population periodically after release of the cultured animals and systematic assaying of the individuals collected to determine their origin (cultured or wild). Similarly, life-history characteristics important for survival of the cultured animals can be compared with those of their wild counterparts. For example, it may be important for released, cultured fish to maintain the same growth rate as conspecific wild fish of the same year class in order to avoid selective predation on the cultured fish or to insure that the cultured fish attain the same reproductive potential as that of the wild fish. The success of the stock enhancement endeavor may depend on the similarity of certain life-history characteristics between the released, cultured animals and wild conspecifics. Regardless of the objectives of the stock enhancement effort, the contribution of the hatchery-derived animals to the recipient population is important. The contribution can be defined as a percentage or as an absolute number of hatchery-derived animals that occur in a specific cohort. The time at which that assessment is made differs depending on those objectives. For example, in a put-and-take stock enhancement program, the objective may be to increase the harvestable component of the population. In that case, an assessment made after the animals reach legal harvestable size is the most important assessment for declaring the stock enhancement effort a success or failure. If the objective of the stock enhancement effort is to provide additional individuals to the reproductive population, an assessment made

Development and Application of Genetic Tags

65

after the animals attain reproductive maturity is the most important assessment. By coupling genetic tag identification of individuals captured from the stocked population with the appropriate life-history data, these types of question can be answered. If the objective of the stock enhancement program is to supplement the reproductively active component of the wild population, then the ultimate measure of success is the level of interbreeding between the enhancement animals and wild conspecifics. However, because an objective for a stock enhancement program is to release native species individuals into locations where the population is depleted compared with its former density, the potential for genetic alteration of the remnant wild population is high if both the ratio of released individuals to wild individuals and the rate of interbreeding between the wild individuals and released individuals are high. The alteration usually comes in the form of changes in the genetic diversity that negatively affect the stocked population. For example, the fitness of the population could be lowered because adaptations to local environments would break down via the introduction of maladapted `exotic' genes (outbreeding depression; Templeton, 1986). Genetic tags can assist with monitoring the rate of interbreeding between the wild stock and released animals (see, for example, Doyle et al., 1995; Tessier et al., 1997). Genetic tag identification can be coupled with age or size information to identify individuals that have a high likelihood of being progeny of hatchery-derived individuals. If the genetic tag is an mtDNA tag, all of the offspring of genetically tagged, released, cultured females (but not males) can be identified. Since males do not transmit their mtDNA to their offspring, the rate of identification decays by 50% with each subsequent generation (assuming that, collectively, the sex ratio of the progeny produced by the hatchery-derived animals each generation is 1:1). If the genetic tag is of nuclear DNA, all of the first-generation offspring can be identified. With subsequent generations, the inheritance patterns can become very complex and are dependent upon the composition of the genetic tag. If interbreeding is successful, almost invariably, the Ne of the stocked population will decrease over subsequent generations (see Tringali & Bert, 1998, and Bert & Tringali, in press). As mentioned above, this can potentially result in a suite of problems in fitness-related life history or morphological components that reduce the probability that the population will survive over ecological or evolutionary time periods. An objective of any stock enhancement program should be to minimize the probability of these types of problems by limiting the contribution of cultured animals to at or below percentages that will cause significant reductions in the stocked population Ne and, to the extent possible, insuring that the recipient wild population genetic diversity is mimicked in the cultured broods. Genetic tags can be used to estimate the Ne produced by a group of broodstock individuals and to estimate the Ne of the stocked population after interbreeding between the cultured animals and wild animals (see, for example, Waples & Do, 1994, for methods). Finally, some cultured animals (e.g. fish) are released at sizes sufficiently large to permit physical tagging of the individuals (e.g. using coded-wire tags or streamer tags). Some degree of tag loss always accompanies any type of physical tagging. Coupling a genetic tag with a physical tag insures that all released, cultured

66

Ecological Aquaculture

individuals are correctly identified when they are captured after release and, simultaneously, the rate of loss of the physical tags can be calculated.

Genetic tags and aquaculture ± the need and the future In the past 10 years, global production of farmed fish and invertebrates (principally shellfish) has more than doubled in both weight and value, and one-quarter of the fish and invertebrates consumed by humans worldwide is derived from aquacultured organisms (Naylor et al., 2000). As fisheries based on wild populations continue to decline (Botsford et al., 1997; Vitousek et al., 1997; Pauly et al., 1998) and as the growing human population searches for efficient and reliable ways to produce affordable, high quality food products, the prospect for continued and accelerated increase in aquaculture-for-food activities is high. Thus, aquaculture will continue to be one of the most rapidly growing sectors of the food production industry for the foreseeable future. In many countries, the culture of aquatic products (both marine and freshwater) is seen as a food production panacea for a rapidly growing population. Development of large scale and small scale operations for both intensive aquaculture and stock enhancement/restoration will at least be maintained at current levels, and will probably accelerate. Genetics is important to this industry, and in many ways and with continued development of the industry, the role of genetics will become increasingly important. Some aquaculturists conducting intensive commercial aquaculture select each succeeding generation of broodstock from the brood generated by the previous generation of broodstock individuals. This practice, although convenient and important when selecting for desirable traits such as elegant color, rapid growth, or high fecundity, can lead to a decrease in the fitness of the broods, principally due to alterations in the genetic diversity of the cultured animals over time (see, e.g. Bert & Tringali, in press). Tracking the genetic diversity of the broods over generations and determining parental contribution to each generation will allow the aquaculturist to determine whether it would be beneficial to introduce individuals from outside of the culture operation (i.e. individuals with `new' genotypes) into his/her broodstock in order to maintain an acceptable level of fitness in the cultured population over generations. Some aquaculturists conduct intensive aquaculture in areas where there is a chance that the cultured animals will escape and interbreed with local wild populations (e.g. open pen aquaculture in lakes, rivers, or the open ocean or facility-based aquaculture that uses flow-through water obtained from local streams, rivers, or estuaries). These aquaculturists may want or need to quantify the rate of escape, level and pattern of dispersal of the escaped individuals, and degree of interbreeding between wild populations and the escaped, cultured animals. The ability to distinguish cultured animals from their wild counterparts and to track the introgression of the genes of the cultured animals into wild populations is essential to this type of monitoring. Finally, the objective of stock enhancement or restoration activities is to produce animals that will be released into areas where the native conspecific population is depleted or extirpated. For many and varied reasons that are dependent on the

Development and Application of Genetic Tags

67

objectives of the activity, it is important to track the cultured individuals after their release and sometimes also to identify their progeny. Genetic monitoring is the only way to track the fate of cultured individuals that are released when they are too small to physically tag and it is the only way to estimate the reproductive contribution of the released, cultured animals. Even if the released individuals are physically tagged, using a genetic marker as a backup insures that the number of cultured animals identified is the maximum possible (physical tags always have a loss rate). Genetic tags are instrumental for all of these tracking and monitoring needs, and they can be retrieved non-destructively because any small amount of tissue ± fresh, preserved, or dried ± can yield DNA. The diversity in the types of genetic markers available, increasing ease of finding adequate markers and assaying large numbers of individuals, and decreasing cost of genetic monitoring make the inclusion of a genetic component in aquaculture operations feasible. The information provided to any individual or group interested in the outcome or impacts of the aquaculture operation (including the aquaculturist) through genetic monitoring based on genetic tags make the inclusion of a genetic component in aquaculture operations essential and indispensable. The future of aquaculture is inextricably linked to genetic tags and genetic monitoring.

Acknowledgments We thank Barry Costa-Pierce for inviting us to write this chapter and gratefully acknowledge Anne McMillen-Jackson and Brent Winner for valuable comments about the manuscript; Anne McMillen-Jackson and Bill Arnold for providing important references; and Kristine Johnston and the Florida Marine Research Institute editorial staff for their superb editorial assistance. This work was supported by funding from the Department of the Interior, US Fish and Wildlife Service, Federal Aid for Sport Fish Restoration Grant Number F-69 to T. M. Bert and M. D. Tringali and by the State of Florida.

References Adamkewicz, L. & Castagna, M. (1988) Genetics of shell color and pattern in the bay scallop Argopecten irradians. Journal of Heredity, 79, 14±17. Avise, J.C. (1994) Molecular Markers, Natural History and Evolution. Chapman & Hall, New York. Ayala, F.J. (1977) Molecular Evolution. Sinauer Associates, Sunderland, Massachusetts. Baker, C.S. & Palumbi, S.R. (1996) Population structure, molecular systematics, and forensic identification of whales and dolphins. In: Conservation Genetics, Case Histories from Nature (eds J.C. Avise & J.L. Hamrick), pp. 10±42. Chapman & Hall, New York. Baker, C.S., Lento, G.M., Cipriano, F., Dalebout, M.L. & Palumbi, S.R. (2000) Scientific whaling: source of illegal products for market? Science, 290, 1695. Bardakci, F. & Skibinski, D.F. (1994) Application of the RAPD technique in tilapia fish: species and subspecies identification. Heredity, 73, 117±123.

68

Ecological Aquaculture

Bentzen, P., Taylor, E.B. & Wright, J.M. (1993) A novel synthetic probe for DNA fingerprinting salmonid fishes. Journal of Fish Biology, 43, 313±316. Bentzen, P. & Wright, J.M. (1993) Nucleotide sequence and evolutionary conservation of a minisatellite variable number tandem repeat cloned from Atlantic salmon, Salmo salar. Genome, 36, 271±277. Bentzen, P., Morris, D.B. & Wright, J.M. (1994) Development and use of variable number tandem repeat markers for population and aquacultural genetics of salmonids. NOAA Technical Memorandum. NMFS±NWFSC, 17, 87±90. Bentzen, P., Taggart, C., Ruzzante, D. & Cook, D. (1996) Microsatellite polymorphism and the population structure of Atlantic cod (Gadus morhua) in the Northwest Atlantic. Canadian Journal of Fisheries and Aquatic Sciences, 53, 2706±2721. Bentzen, P., Wright, J.M., Bryden, L.T., Sargent, M. & Zwanenburg, K.C.T. (1997) Tandem repeat polymorphism and heteroplasmy in the mitochondrial control region of redfishes (Sebastes: Scorpaenidae). Journal of Heredity, 89, 1±7. Bert, T.M., McCarthy, K.J., Cruz Lopez, H. & Bogdanowicz, S. (1996) Character discriminatory power, character set congruence, and the classification of individuals from hybrid zones: an example using stone crabs (genus Menippe). Evolution, 50, 655±671. Bert, T.M. & Tringali, M.D. (in press) The effects of various aquacultural breeding strategies on the genetic diversity of successive broods. Jurnal Biosains (Malaysia). Bert, T.M., Tringali, M.D., Crawford, C. & Seyoum, S. (in preparation) A comprehensive genetic monitoring program for hatchery-based stock enhancement or restoration. In: Ecological and Genetic Implications of Aquaculture Activities (eds T.M. Bert & J. Nielsen), Kluwer Academic Publications, New York. Bilodeau, A.L., Lankford, W., Kim, T., Felder, D.L. & Neigel, J.E. (1999) An ultrasensitive method for detection of single crab larvae (Sesarma reticulatum) by PCR amplification of a highly repetitive DNA sequence. Molecular Ecology, 8, 683±684. Birstein, V.J., Doukakis, P., Sorkin, B. & DeSalle, R. (1998) Population aggregation analysis of three caviar-producing species of sturgeons and implications for species identification of black caviar. Conservation Biology, 12, 766±775. Botsford, L.W., Castilla, J.C. & Peerson, C.H. (1997) The management of fisheries and marine ecosystems. Science, 277, 500±515. Bowen, B.W. & Avise, J.C. (1990) Genetic structure of Atlantic and Gulf of Mexico populations of sea bass, menhaden, and sturgeon: influence of zoogeographic factors and lifehistory patterns. Marine Biology, 107, 371±381. Bruford, M.W., Chasman, D.J., Coote, T. et al. (1996) Microsatellites and their application to conservation genetics. In: Molecular Genetic Approaches in Conservation (eds T.B. Smith & R.K. Wayne), pp. 278±297. Oxford University Press, New York. Buonaccorsi, V.P., Starkey, E. & Graves, J.E. (2001) Mitochondrial and nuclear DNA analysis of population subdivision among young-of-the-year Spanish mackerel (Somberomorus maculatus) from the western Atlantic and Gulf of Mexico. Marine Biology, 138, 37±45. Burke, T., Hanotte, O. & Van Pijlen, I. (1996) Minisatellite analysis in conservation genetics. In: Molecular Genetic Approaches in Conservation (eds T.B. Smith & R.K. Wayne), pp. 298± 313. Oxford University Press, New York. Carlson, J.E., Tulsieram, L.K., Glaubitz, J.C., Luk, V.W.K., Kauffeldt, C. & Rutledge, R.

Development and Application of Genetic Tags

69

(1991) Segregation of random amplified DNA markers in F1 progeny of conifers. Theoretical and Applied Genetics, 83, 194±200. Castagna, M. & Manzi, J.J. (1989) Clam culture in North America: hatchery production of nursery stock clams. In: Clam Mariculture in North America (eds J.J. Manzi & M. Castagna), pp. 111±126. Elsevier Press, New York. Clifford, S.L., McGinnity, P. & Ferguson, A. (1997) Genetic changes in an Atlantic salmon population resulting from escaped juvenile farm salmon. Journal of Fish Biology, 52, 118± 127. Clifford, S.L., McGinnity, P. & Ferguson, A. (1998) Genetic changes in Atlantic salmon (Salmo salar) populations of northwest Irish rivers resulting from escapes of adult farm salmon. Canadian Journal of Fisheries and Aquatic Sciences, 55, 358±363. Colbourne, J.F., Neff, B.D., Wright, J.M. & Gross, M.R. (1999) DNA fingerprinting of bluegill sunfish (Lepomis macrochirus) using (GT)n microsatellites and its potential for assessment of the relative success of male reproductive strategies. Canadian Journal of Fisheries and Aquatic Sciences, 53, 342±349. Coltman, D.W., Bowen, W.D. & Wright, J.M. (1998) Male mating success in an aquatically mating pinniped, the harbour seal (Phoca vitulina), determined by microsatellite DNA markers. Molecular Ecology, 7, 627±638. Costa-Pierce, B.A. & Doyle, R.W. (1997) Genetic identification and status of California regional strains in southern California. In: Tilapia Aquaculture in the Americas (eds B.A. Costa-Pierce & J. Rakocy), pp. 1±17. Louisiana World Aquaculture Society Books, Baton Rouge, Louisiana. Craxton, M. (1991) Linear amplification sequencing: a powerful method for sequencing DNA. Methods, 3, 20±24. Crozier, W.W. (1993) Evidence of genetic interaction between escaped farmed salmon and wild Atlantic salmon (Salmo salar L.) in a Northern Irish river. Aquaculture, 113, 19±29. Crozier, W.W. & Moffett, I.J.J. (1995) Application of low frequency genetic marking at GPI-3 and MDH-B1, 2 loci to assess supplementary stocking of Atlantic salmon, Salmo salar L., in a Northern Irish stream. Fisheries Management and Ecology, 2, 27±36. Danzmann, R.G., Jackson, T.R. & Ferguson, M.M. (1999) Epistasis in allelic expression at upper temperature tolerance QTL in rainbow trout. Aquaculture, 173, 45±58. Dawson, I.K., Chalmers, K.J., Waugh, R. & Powell, W. (1993) Detection and analysis of genetic variation in Hordeum spontaneum populations from Israel using RAPD markers. Molecular Ecology, 2, 151±159. Dowling, T.E., Moritz, C., Palmer, J.D. & Riesenberg, L.H. (1996) Nucleic acids III: Analysis of fragments and restriction sites. In: Molecular Systematics, 2nd edn (eds D.M. Hillis, C. Moritz & B.K. Mable), pp. 249±320. Sinauer Associates, Sunderland, Massachusetts. Doyle, R.W., Herbinger, C., Taggart, C. & Lochmann, S. (1995) Use of microsatellite DNA polymorphism to analyze genetic correlations between hatchery and natural fitness. In: Uses and Effects of Cultured Fishes in Aquatic Ecosystems (eds H.L.J. Schramm & R.G. Piper), pp. 205±211. American Fisheries Society, Bethesda, Maryland. Edwards, S.V. & Potts, W.K. (1996) Polymorphism of genes in the major histocompatibility complex (MHC): implications for conservation genetics of vertebrates. In: Molecular Genetic Approaches in Conservation (eds T.B. Smith & R.K. Wayne), pp. 214±237. Oxford University Press, New York.

70

Ecological Aquaculture

Fishback, A.G., Danzmann, R.G. & Ferguson, M.M. (2000) Microsatellite allelic heterogeneity among hatchery rainbow trout maturing in different seasons. Journal of Fish Biology, 57, 1367±1380. Fleming, I.A. & Einum, S. (1997) Experimental tests of genetic divergence of farmed from wild Atlantic salmon due to domestication. ICES Journal of Marine Science, 54, 1051±1063. Frischer, M.E., Danforth, J.M., Tyner, L.C., Leverone, J.R., Marelli, D.C., Arnold, W.S. & Blake, N.J. (2000) Development of an Argopecten-specific 18S rRNA targeted genetic probe. Marine Biotechnology, 2, 11±20. Garcia De Leon, F.J., Dallas, J.F., Chatain, B., Cannone, M., Versini, J.J. & Bonhomme, F. (1995) Development and use of microsatellite markers in sea bass, Dicentrarchus labrax (Linnaeus, 1758) (Perciformes: Serranidae). Molecular Biology Biotechnology, 4, 62±68. Garcia De Leon, F.J., Canonne, M., Quillet, E., Bonhomme, F. & Chatain, B. (1998) The application of microsatellite markers to breeding programmes in the sea bass, Dicentrarchus labrax. Aquaculture, 159, 303±316. Gharett, A.J. & Seeb, J.E. (1990) Practical and theoretical guidelines for genetically marking fish populations. American Fisheries Society Symposium, 7, 407±417. Gibbs, H.L., Prior, K.A. & Weatherhead, P.J. (1994) Genetic analysis of populations of threatened snake species using RAPD markers. Molecular Ecology, 3, 329±337. Gilbert, D.A., Lehman, N., O'Brien, S.J. & Wayne, R.K. (1990) Genetic fingerprinting reflects population differentiation in the California Channel Island fox. Nature, 344, 764±766. Gilbert, D.A., Packer, C., Pusey, A.E., Stephens, J.E. & O'Brien, S.J. (1991) Analytical DNA fingerprinting in lions: parentage, genetic diversity and kinship. Journal of Heredity, 82, 378± 386. Girman, D.J. (1996) The use of PCR-based single-stranded conformation polymorphism analysis (PCR-SSCP) in conservation genetics. In: Molecular Genetic Approaches in Conservation (eds T.B. Smith & R.K. Wayne), pp. 167±182. Oxford University Press, New York. Gjetvaj, B., Cook, D. & Zouros, E. (1992) Repeated sequences and large-scale size variation of mitochondrial DNA: a common feature among scallops (Bivalvia: Pectinidae). Molecular Biology and Evolution, 9, 106±124. Gold, J.R. & Richardson, L.R. (1991) Genetic studies in marine fishes IV: An analysis of population structure in the red drum (Sciaenops ocellatus) using mitochondrial DNA. Fisheries Research, 12, 213±241. Gold, J.R. & Richardson, L.R. (1993) Genetic distinctness of red drum (Sciaenops ocellatus) from Mosquito Lagoon, eastern Florida. Fisheries Bulletin, 92, 58±66. Gold, J.R., Richardson, L.R., Furman, C. & King, T.L. (1993) Mitochondrial DNA differentiation and population structure in red drum (Sciaenops ocellatus) from the Gulf of Mexico and Atlantic Ocean. Marine Biology, 116, 175±185. Gold, J.R., Richardson, L.R. & Furman, C. (1999) Mitochondrial DNA diversity and population structure of spotted seatrout (Cynoscion nebulosus) in coastal waters of the southeastern United States. Gulf of Mexico Science, 1, 40±50. Hadrys, H., Balick, M. & Schierwater, B. (1992) Mini review ± applications of random amplified polymorphic DNA (RAPD) in molecular ecology. Molecular Ecology, 1, 55±63. Harris, A., Bieger, S., Doyle, R. & Wright, J.M. (1991) DNA fingerprinting of tilapia, Oreochromis niloticus, and its application to aquaculture genetics. Aquaculture, 92, 57±163.

Development and Application of Genetic Tags

71

Harris, A. & Wright, J.M. (1995) Nucleotide sequence and genomic organization of chiclid fish minisatellites. Genome, 38, 177±84. Hartl, D.L. & Clark A.G. (1997) Principles of Population Genetics, 3rd edn. Sinauer Associates, Sunderland, Massachusetts. Hedrick, P.W. (1994) Evolutionary genetics of the major histocompatibility complex. American Naturalist, 143, 945±964. Hedrick, P.W., Hedgecock, D., Hamelberg, S. & Croci, S.J. (2000) The impact of supplementation in winter-run Chinook salmon on effective population size. Journal of Heredity, 91, 112±116. Herbinger, C.M., Doyle, R., Pitman, E. et al. (1995) DNA fingerprint based analysis of paternal and maternal effects on offspring growth and survival in commercially reared rainbow trout. Aquaculture, 137, 245±256. Herbinger, C.M., Doyle, R., Taggart, C. et al. (1997) Family relationships and effective population size in a natural cohort of cod larvae. Canadian Journal of Fisheries and Aquatic Science Supplement, 1, 11±18. Hillis, D.M., Mable, B.K, Larson, A., Davis, S.K. & Zimmer, E.A. (1996) Nucleic acids IV: Sequencing and cloning. In: Molecular Systematics, 2nd edn (eds D.M. Hillis, C. Moritz, & B.K. Mable), pp. 321±381. Sinauer Associates, Sunderland, Massachusetts. Hindar, K., Ryman, N. & Utter, F. (1991) Genetic effects of cultured fishes on natural populations. Canadian Journal of Fisheries and Aquatic Sciences, 48, 945±957. Humphrey, C.M. & Walker, R.L. (1982) The occurrence of Mercenaria mercenaria form notata in Georgia and South Carolina: calculation of phenotypic and genotypic frequencies. Malacologia, 23, 75±79. Jackson, T.R., Ferguson, M.M., Danzmann, R.G. et al. (1998) Identification of two QTL influencing upper temperature tolerance in three rainbow trout (Oncorhynchus mykiss) halfsib families. Transactions of the American Fisheries Society, 129, 469±475. Jeffreys, A.J., Wilson, V. & Thein, S.L. (1985a) Hypervariable `minisatellite' regions in human DNA. Nature, 314, 67±73. Jeffreys, A.J., Wilson, V. & Thein, S.L. (1985b) Amplification of human minisatellites by the polymerase chain reaction: towards DNA fingerprinting of single cells. Nucleic Acids Research, 16, 10953±10971. Jones, C.S., Okamura, B. & Noble, L.R. (1994) Parent and larval RAPD fingerprints reveal outcrossing in freshwater bryozoans. Molecular Ecology, 3, 193±199. Karl, S. (1996) An application of anonymous nuclear loci to conservation biology. In: Molecular Genetic Approaches in Conservation (eds T.B. Smith & R.K. Wayne), pp. 38±53. Oxford University Press, New York. Kenchington, E., Naidu, K., Roddick, D., Cook, D. & Zouros, E. (1993) Use of biochemical genetic markers to discriminate between adductor muscles of the sea scallop (Placopecten magellanicus) and the Iceland scallop (Chlamys islandica). Canadian Journal of Fisheries and Aquatic Science, 50, 1222±1228. King, T.L., Ward, R. & Blandon, I.R. (1993) Gene marking: a viable assessment method. Fisheries (Bethesda), 18, 4±5. Knut, J.E., Paulsen, O.I., Naevdal, G. & Thorkildsen, S. (1994) Genetic studies of cod, Gadus morhua L., in Masfjorden, western Norway: comparisons between the local stock and released, artificially reared cod. Aquaculture and Fisheries Management, 25, Suppl. 1, 77±91.

72

Ecological Aquaculture

Kocher, T. & Stepien, C. (Eds) (1996) Molecular Systematics of Fishes. Academic Press, San Diego. Kocher, T.D., Thomas, W.K., Meyer, A. et al. (1989) Dynamics of mitochondrial DNA evolution in animals: amplification and sequencing with conserved primers. Proceedings of the National Academy of Sciences USA, 86, 6196±6200. Kohanne, M.J. & Parsons, P.A. (1988) Domestication: evolutionary change under stress. Evolutionary Biology, 23, 31±48. Leberg, P.L. (1996) Applications of allozyme electrophoresis in conservation biology. In: Molecular Genetic Approaches in Conservation (eds T.B. Smith & R.K. Wayne), pp. 87±103. Oxford University Press, New York. Lehman, N., Eisenhawer, A., Hansen, K. et al. (1991) Introgression of coyote mitochondrial DNA into sympatric North American gray wolf populations. Evolution, 45, 114±119. Lewontin, R.C., & Hubby, J.L. (1966) A molecular approach to the study of genetic heterozygosity in natural populations. II. Amounts of variation and degree of heterozygosity in natural populations of Drosophila pseudoobscura. Genetics, 54, 595±609. MaKinster, J.G., Roberts, J.E., Felder, D.L., Chlan, C.A., Boudreaux, M. & Neigel, J.E. (1999) PCR amplification of a middle repetitive element detects larval stone crabs (Crustacea: Decapoda: Menippidae) in estuarine plankton samples. Marine Ecology Progress Series, 188, 161±168. Marchalonis, J.J. & Schluter, S. (1990) Origins of immunoglobulins and immune recognition molecules. Bioscience, 40, 758±768. Mathias, J.A., Babaluk, J.A. & Rowes, K.D. (1985) An analysis of the 1984 walleye, Stizostedion vitreum vitreum, run at Crean Lake in Prince Albert National Park, Saskatchewan, Canada, with reference to the impact of spawn-taking. Canadian Technical Report Fisheries and Aquatic Science, 1407, i±iv, 1±34. McConnell, S., Hamilton, L., Morris, D. et al. (1995) Isolation of salmonid microsatellite loci and their application to the population genetics of Canadian east coast stocks of Atlantic salmon. Aquaculture, 137, 19±30. McConnell, S., Ruzzante, D., O'Reilly, P.T., Hamilton, L. & Wright, J.M. (1997) Microsatellite loci reveal highly significant genetic differentiation between Atlantic salmon (Salmo salar L.) populations from the East Coast of Canada. Molecular Ecology, 6, 1075±1089. Moav, R. & Wohlfarth, G.W. (1976) Two-way selection for growth rate in the common carp (Cyprinus carpio L.). Genetics, 82, 83±101. Moritz, C. (1994) Applications of mitochondrial DNA analysis in conservation: a critical review. Molecular Ecology, 3, 401±411. Morris, D., Richard, K. & Wright, J.M. (1996) Polymorphic microsatellites from rainbow trout (Oncorhynchus mykiss) and their use for genetic study of salmonids. Canadian Journal of Fisheries and Aquatic Sciences, 53, 120±126. Murphy, B.R., Nielsen, L.A. & Turner, B.J. (1983) Use of genetic tags to evaluate stocking success for reservoir walleyes, Stizostedion vitreum vitreum. Transactions of the American Fisheries Society, 112, 457±463. Naylor, R.L., Goldburg, R.J., Primavera, J.H., Kautsky, N., Beveridge, M.C.M., Clay, J., Folke, C., Lubchenco, J., Mooney, H. & Troell, M. (2000) Effect of aquaculture on world fish supplies. Nature, 405, 1017±1024. Nyman, L. & Ring, O. (1989) Effects of hatchery environment on three polymorphic loci in

Development and Application of Genetic Tags

73

Arctic char (Salvelinus alpinus) species complex. Nordic Journal of Freshwater Research, 65, 34±43. O'Connell, M. & Dillon, M.C. (1997) Microsatellite DNA in fishes. In: Reviews in Fish Biology and Fisheries (eds G.R. Carvalho & T.J. Pitcher), pp. 331±363. Chapman & Hall, New York. O'Connell, M., Dillon, M.C., Wright, J.M., Bentzen, P., Merkouris, S. & Seeb, J. (1998) Genetic structuring among Alaska herring (Clupea pallasi) populations identified using microsatellite variability. Journal of Fish Biology, 53, 150±163. Ohland, D.P., Harley, E.H. & Best, P.B. (1995) Systematics of cetaceans using restriction site mapping of mitochondrial DNA. Molecular Phylogenetics and Evolution, 4, 10±19. O'Reilly, P., Hamilton, L., McConnell, S. & Wright, J.M. (1996) Rapid detection of genetic variation in Atlantic salmon (Salmo salar) by PCR multiplexing of dinucleotide and tetranucleotide microsatellites. Canadian Journal of Fisheries and Aquatic Sciences, 53, 2292±2298. O'Reilly, P., Herbinger, C. & Wright, J.M. (1998) An analysis of parentage determination in Atlantic salmon (Salmo salar) using microsatellites. Animal Genetics, 29, 363±370. O'Reilly, P. & Wright, J.M. (1995) The evolving technology of DNA fingerprinting and its application to fisheries and aquaculture. Journal of Fish Biology, 47, 29±55. Packer, C., Gilbert, D.A., Pusey, A.E. & O'Brien, S.J. (1991) A molecular genetic analysis of kinship and cooperation in African lions. Nature, 351, 562±565. Paetkau, D. & Strobeck, C. (1994) Microsatellite analysis of genetic variation in black bear populations. Molecular Ecology, 3, 489±495. Paetkau, D., Waits, L.P., Clarkson, P.L., Craighead, L. & Strobeck, C. (1997) An empirical evaluation of genetic distance statistics using microsatellite data from bear (Ursidae) populations. Genetics, 147, 1943±1957. Palsboll, P.J. (1999) Genetic tagging: contemporary molecular ecology. Biological Journal of the Linnaean Society, 68, 3±22. Palumbi, S.R., Martin, A.P., Romano, S., McMillan, W.O., Stice, L. & Grabowski, G. (1991) The simple fool's guide to PCR. Special Publ., Dept. Zoology, University of Hawaii, Honolulu. Palumbi, S.R. (1996) Nucleic acids II: The polymerase chain reaction. In: Molecular Systematics, 2nd edn (eds D.M. Hillis, C. Moritz & B.K. Mable), pp. 205±247. Sinauer Associates, Sunderland, Massachusetts. Palumbi, S.R. & Cipriano, F. (1998) Species identification using genetic tools: the value of nuclear and mitochondrial gene sequences in whale conservation. Journal of Heredity, 89, 459±464. Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. & Torres, Jr, F. (1998) Fishing down marine food webs. Science, 279, 860±863. Perez-Enriquez, R. & Nobuhiko, T. (1999) Use of microsatellite DNA as genetic tags for the assessment of a stock enhancement program of red sea bream. Fisheries Science (Tokyo), 65, 374±379. Potts, W.K. & Wakeland, E.K. (1990) Evolution of diversity of the major histocompatibility complex. Trends in Ecology and Evolution, 5, 181±187. ProdoÈhl, P.A., Taggart, J.B. & Ferguson, A. (1995) A panel of minisatellite (VNTR) DNA locus specific probes for potential application to problems in salmonid aquaculture. Aquaculture, 137, 87±97. Richard, K.R., Whitehead, H. & Wright, J.M. (1996) Polymorphic microsatellites from sperm

74

Ecological Aquaculture

whale and their use in genetic identification of individuals from naturally sloughed pieces of skin. Molecular Ecology, 5, 313±315. Richardson, B.J., Baverstock, P.R. & Adams, M. (1986) Allozyme Electrophoresis: A Handbook for Animal Systematics and Population Studies. Academic Press, Sydney. Ruzzante, D., Taggart, C.T. & Cook, D. (1996) Spatial and temporal variation in the genetic composition of larval cod (Gadus morhua) on the Scotian Shelf: cohort contribution and genetic stability. Canadian Journal of Fisheries and Aquatic Sciences, 54, 2700±2708. Ruzzante, D., Taggart, C.T. & Cook, D. (1998) A nuclear DNA basis for shelf- and bank-scale population structure in NW Atlantic cod (Gadus morhua) stocks: Labrador to Georges Banks. Molecular Ecology, 7, 1663±1680. Ryman, N., Utter, F. & Hindar, K. (1995) Introgression, supportive breeding, and genetic conservation. In: Population Management for Survival and Recovery: Analytical Methods and Strategies in Small Population Conservation (eds M. Gilpin & T. Foose), pp. 341±365. Columbia University Press, New York. Saiki, R.K., Walsh, P.S., Levenson, C.H. & Erlich, H.A. (1989) Genetic analysis of amplified DNA with immobilized sequence-specific oligonucleotide probes. Proceedings of the National Academy of Sciences, USA, 86, 6230±6234. Sakamoto, T., Danzmann, R.G., Okamoto, N., Ferguson, M.M. & Ihssen, P.E. (1999) Linkage analysis of quantitative trait loci associated with spawning time in rainbow trout (Oncorhynchus mykiss). Aquaculture, 173, 33±43. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor Laboratory Press, New York. Sanger, F., Nicklon, S. & Coulson, A.R. (1977) DNA sequencing with chain termination inhibitors. Proceedings of the National Academy of Sciences, USA, 74, 5463±5467. Selander, R.K., Smith, M.H., Yang, S.Y., Johnson, W.E. & Gentry, J.B. (1971) Biochemical polymorphism and systematics in the genus Peromyscus. I. Variation in the old-field mouse (Peromyscus polionotus). Studies in Genetics VI. University of Texas Publication, 7103, 49±90. Seyoum, S. (1990) Molecular identification of the tilapiine species introduced into Lake Langano, Ethiopia. SINET: Ethiopian Journal of Science, 14, 29±40. Seyoum, S. & Kornfield, I. (1992) Taxonomic notes on the Oreochromis niloticus subspeciescomplex (Pisces: Cichlidae), with a description of a new subspecies. Canadian Journal of Zoology, 70, 2161±2165. Seyoum, S., Bert, T.M., Wilbur, A., Arnold, W.S. & Crawford, C. (in review) Development, assessment, and application of a mitochondrial DNA genetic tag for the bay scallop (Argopecten irradians). Seyoum, S., Tringali, M.D., Bert, T.M., McElroy, D. & Stokes, R. (2000) An analysis of genetic population structure in red drum (Sciaenops ocellatus) based on mtDNA control region sequences. Fisheries Bulletin, 98, 127±138. Smith, T.B. & Wayne, R.K. (1996) Molecular Genetic Approaches in Conservation. Oxford University Press, New York. Stabile, J., Waldman, J.R., Hart, J. & Wirgin, I. (1996) Stock structure and homing fidelity is high in Gulf of Mexico sturgeon (Acipenser oxyrinchus desotoi) based on restriction fragment length polymorphism and sequence analyses of mitochondrial DNA. Genetics, 144, 767±775. Stepien, C. (1995) Population genetic divergence and geographic patterns from DNA

Development and Application of Genetic Tags

75

sequences: examples from marine and freshwater fishes. American Fisheries Society Symposium, 17, 263±287. Stepien, C. (1999) Phylogeographical structure of the Dover sole (Microstomus pacificus): the larval retention hypothesis and genetic divergence along the deep continental slope of the northeastern Pacific Ocean. Molecular Biology, 8, 923±939. Sugg, D.W. & Chesser, R.K. (1994) Effective population sizes with multiple paternity. Genetics, 137, 1147±1155. Taberlet, P. (1996) The use of mitochondrial DNA control region sequencing in conservation genetics. In: Molecular Genetic Approaches in Conservation (eds T.B. Smith & R.K. Wayne), pp. 125±142. Oxford University Press, New York. Taggart, J.B. & Ferguson, A. (1984) An electrophoretically detectable genetic tag for hatcheryreared brown trout (Salmo trutta). Aquaculture, 41, 119±130. Taggart, J.B. & Ferguson, A. (1986) Electrophoretic evaluation of a supplemental stocking program for brown trout (Salmo trutta). Aquaculture and Fisheries Management, 17, 155±162. Takagi, M., Taniguchi, N., Cook, D. & Doyle, R. (1996) Isolation and characterization of microsatellite loci from red sea bream (Pagrus major) and detection in closely related species. Journal of the Japanese Society of Fisheries Science, 63, 199±204. Tanksley, S.D. (1993) Mapping polygenes. Annual Review of Genetics, 27, 205±233. Tassanakajon, A., Tiptawonnukul, A., Supankul, P. et al. (1998) Isolation and characterization of microsatellite markers in the black tiger prawn, Penaeus monodon. Molecular Marine Biology and Biotechnology, 7, 55±61. Tave, D. (1986) Genetics for Fish Hatchery Managers. AVI Publishing Company, Inc., Westport, Connecticut. Templeton, A.R. (1986) Coadaptation and outbreeding depression. In: Conservation Biology: the Science and Scarcity of Diversity (ed. M.E. SouleÂ), pp. 105±166. Sinauer Associates, Sunderland, Massachusetts. Tessier, N.T., Bernatchez, L.B. & Wright, J.M. (1997) Population structure and impact of supportive breeding program inferred from mitochondrial DNA and microsatellite DNA analyses in sympatric landlocked Atlantic salmon populations (Salmo salar L.). Molecular Ecology, 6, 750±753. Triantafyllidis, A., Abatzopoulos, T.J. & Economidis, P.S. (1999) Genetic differentiation and phylogenetic relationships among Greek Silurus glanis and Silurus aristotelis (Pisces, Siluridae) populations, assessed by PCR±RFLP analysis of mitochondrial DNA segments. Heredity, 82, 503±509. Tringali, M.D. & Bert, T.M. (1996) The genetic stock structure of common snook (Centropomus undecimalis). Canadian Journal of Fisheries and Aquatic Sciences, 53, 974±984. Tringali, M.D. & Bert, T.M. (1998) Risk to effective population size should be an important consideration in fish stock-enhancement programs. Bulletin of Marine Science, 62, 641±659. Tringali, M.D. & Bert, T.M. (in prep.) Measuring gene correlation among F1 progeny in a complex system of captive breeding. Tringali, M.D., Bert, T.M. & Seyoum, S. (1999a) Genetic identification of Centropominae fishes. Transactions of the American Fisheries Society, 128, 446±458. Tringali, M.D., Bert, T.M., Seyoum, S., Bermingham, E. & Bartolacci, D. (1999b) Molecular phylogenetics and ecological diversification of the transisthmian fish genus Centropomus (Perciformes: Centropomidae). Molecular and Phylogenetic Evolution, 13, 193±207.

76

Ecological Aquaculture

Turner, T.F., Richardson, L.R. & Gold, J.R. (1998) Polymorphic microsatellite DNA markers in red drum (Sciaenops ocellatus). Molecular Ecology, 7, 1771±1788. Vitousek, P.M., Mooney, H.A., Lubchenco, J. & Meilillo, J.M. (1997) Human domination of earth's ecosystems. Science, 277, 494±499. Waples, R.S. & Do, C. (1994) Genetic risk associated with supplementation of Pacific salmonids: captive broodstock programs. Canadian Journal of Fisheries and Aquatic Sciences, 51, 310±329. Watkins, D.I., McAdam, S.N. & Strang, L.X. (1992) New recombinant HLA-B alleles in a tribe of South American Amerindians indicate rapid evolution of MHC class I loci. Nature, 357, 329±333. Wayne, R.K. & Jenks, S.M. (1991) Mitochondrial DNA analysis implying extensive hybridization of the endangered red wolf Canis rufus. Nature, 351, 565±568. Williams, J.G.K., Kubelik, A.R., Livak, K.J., Rafalski, J.A. & Tingey, S.V. (1990) DNA polymorphisms amplified by arbitrary primers are useful genetic markers. Nucleic Acids Research, 18, 6531±6535. Wilson, R.R. Jr, Donaldson, K.A., Frischer, M.E. & Young, T.B. (1997) Mitochondrial DNA control region of common snook for use as a genetic tag. Transactions of the American Fisheries Society, 126, 594±606. Woods, J.G., Paetkau, D., Lewis, D., McLellan, B.N., Proctor, M. & Strobeck, C. (1999) Genetic tagging of free-ranging black and brown bears. Wildlife Society Bulletin, 27, 616± 627. Wright, J.M. (1993) DNA fingerprinting of fishes. In: The Biochemistry and Molecular Biology of Fishes (eds T. Mommsen & P.W. Hochachka), pp. 57±91. Elsevier Press, New York. Zouros, E., Ball, A.O., Saavedra, C. & Freeman, K. (1994) An unusual type of mitochondrial DNA inheritance in the blue mussel Mytilus. Proceedings of the National Academy of Science, USA, 91, 7463±7467.

Chapter 4

Aquaculture Escapement, Implications and Mitigation: The Salmonid Case Study C.J. Bridger1 and Amber F. Garber2 1 College of Marine Sciences, University of Southern Mississippi 2 Department of Zoology, North Carolina State University Introduction Cultured fish occur in the wild from intentional release ± for stock enhancement and sea ranching purposes ± or escapees (individuals escaping to the wild from aquaculture grow-out facilities). Stock enhancement managers wish to supplement wild populations of a species for commercial and/or recreational fisheries, or to mitigate threats of biological extinction. Maintenance of wild genetic composition is a necessity in stock enhancement programs to minimize adverse genetic impacts to wild recipient populations, and is accomplished by utilizing an adequate number of local broodstock (donor spawning individuals) (Tringali & Bert, 1998; Tringali & Leber, 1999). To this end, aquaculture plays a vital role in the restoration of collapsing fisheries (Drawbridge, Chapter 11, this volume). In contrast to stock enhancement, most aquaculture species descend from long-term breeding programs selecting for economically important traits (e.g. rapid growth, late maturation, disease resistance, increased flesh quality, and improved survival within high density rearing environments) (Friars et al., 1996; Gjùen & Bentsen, 1997). Aquaculture net pens are frequently thought of as separate entities placed in the water by man. To the contrary, aquaculture facilities should be viewed in an holistic manner taking into account the role of the facilities as part of a complex ecosystem enhancing the form and function of the natural environment (Costa-Pierce & Bridger, in press). Each component of the `aquaculture ecosystem' must be evaluated on the basis of its sustainability; issues concerning escapees are one such component. Sustainability may be considered three-pronged ± social, economic and environmental ± all of which are very interrelated and require comparable attention for sustainable industry development and operation (Fig. 4.1). Aquaculture facilities are likely to lose domestic fish at some stage during production, especially with the utilization of sea cages. Aquaculture escapement may be classified as either chronic or acute. Chronic leakage is the slow, continuous loss of fish resulting from improper farm practices, small holes in the containment netting from general wear and tear, or localized, small scale predation. Rapid loss of fish due to storm damage or predation from destructive predators, such as seals or tuna, is considered acute escapement. Acute escapement may result in the loss of one to

78

Ecological Aquaculture

Economic sustainability

• Decrease economic loss associated with escapement Social sustainability Sustainable aquaculture

• Minimize impacts of escapees on wild fish stocks

• Decrease environmental group backlash • Decrease conflicts with wild fishers

Environmental sustainability

Fig. 4.1 Benefits of utilizing sustainable aquaculture practices by eliminating fish escapement from aquaculture facilities to the wild.

several cages of fish in a relatively short period of time and may be economically devastating to an aquaculture venture. To ensure social sustainability (and therefore general acceptance) of aquaculture, elimination of escapees to the environment will be required to decrease both public backlash and conflicts with wild fishers. Concerns of potential interactions between escapees and wild fish stocks are legitimate and require focused scientific research to decrease the impacts (disease, ecological and genetic) that escapees may have on wild stocks. Economic sustainability is of the greatest concern to the fish farmer, and is best attained by minimizing (or completely eliminating) revenue losses associated with escapement from cages. This chapter will focus on issues regarding escapees and environmental sustainability.

Occurrence of salmonid aquaculture escapees in the wild The salmonid aquaculture sector has received the most attention and has the highest public profile with regards to escapement from sea cages. Despite increased farm operator awareness and implementation of rigid industry Codes of Conduct and government regulations to mitigate farm fish loss, reports of high aquaculture escapement still occur (Table 4.1). Global occurrence of escapees in wild marine and freshwater fisheries is up to 48% of the total catch with a large proportion (up to 45%) of the sexually mature salmonids being escapees. Escapees can be differentiated from wild conspecifics (counterparts) by external appearance, presence of synthetic flesh colorants from salmonid farm feed, or genetic tags (Table 4.2).

Effects of domestication To increase profits, aquaculture operations selectively breed a fish strain for high growth rates in environments with no predatory stress, while using high density

Aquaculture Escapement, Implications and Mitigation

79

Table 4.1 Occurrence of escapees in wild fisheries and freshwater systems throughout the world Locations

Occurrences of escapees

References

Norway

Escapees in 70% of monitored rivers with 15% of total samples in southern and 1% in northern Norway (greater salmon farming volume also in the south)

Gausen & Moen, 1991

29.1% of marine catch and 5.8% of freshwater catch in 1989

Lund et al., 1991

1 steelhead trout smolt captured migrating to sea

Jonsson et al., 1993

45% of female Atlantic salmon spawners

Sñgrov et al., 1997

22% of fishery in 1990

Webb & Youngson, 1992

After 1986, reared salmon detected frequently (ranged up to 38%) near salmon farms but irregularly (<0.5%) in distant rivers

Youngson et al., 1997

Iceland

30.1% in catch, almost all sexually mature

Gudjonsson, 1991

Faroe Islands

25±48% of fishery

Hansen et al., 1993

West Greenland

1.1% (1991) and 1.4% (1992) escapees in the wild fishery

Hansen et al., 1997

British Columbia

188 Atlantic salmon caught in freshwater from 1990 to 1995 with most displaying gonadal development and external coloration indicating sexual maturation

McKinnell et al., 1997

210 escaped Atlantic salmon adults captured in 29 streams in 1996

McKinnell & Thomson, 1997

20% of redds created by females of definite cultured origin

Carr et al., 1997

51.0±67.2% of smolts were juvenile escapees

Stokesbury & Lacroix, 1997

Scotland

New Brunswick

formulated feeds to yield a food conversion ratio close to 1:1 (i.e. 1 kg dry feed producing 1 kg of wet fish flesh). A large body of literature is available regarding the effects of domestication (process of adaptation to an artificial environment) on the genetics, behavior, morphology and physiology of fish species. We will outline some of the key concerns of domestication and also provide numerous studies that are contradictory, illustrating the complexity of fish behavior and problems associated with sweeping generalizations across species. Many of the examples provided discuss how escapees of hatchery origin may impact wild salmonid populations. Farmed Atlantic salmon, from the Norwegian salmon breeding program, have been selected for growth performance (body weight at slaughter), delayed sexual maturation, flesh color, body composition or fat content (total fat content and amount of fat tissue), and resistance against diseases (principally furunculosis and infectious salmon anaemia (ISA)) (Gjùen & Bentsen, 1997). A common assumption is that genetic variation may be lost in hatchery populations, using a small number of broodstock fish (Verspoor, 1988a). The Norwegian breeding program strives to maintain additive genetic variation through careful record keeping, use of pedigrees

80

Ecological Aquaculture

Table 4.2 Methods to identify and separate escapees and wild conspecifics Individuals identified

Methods of identification

References

Recently escaped individuals

Determine isomeric ratio of natural and synthetic astaxanthin present in muscle tissue

Lura & ékland, 1994

Recently escaped individuals

Carotenoid pigmentation

Craik & Harvey, 1987

Escapees of all ages

Scale characteristics associated with growth

Lund & Hansen, 1991; Hiilivirta et al., 1998

Juvenile escapees

Otolith analysis

Hindar & L'AbeÂe-Lund, 1992

Post-vaccinated individuals

Visual observation of postvaccination intra-abdominal adhesions

Lund et al., 1997

Adult escapees

Discriminate analysis based on fin size measurement as a result of fin erosion in captivity

Potter, 1987

Eggs and alevins from escapee females

Astaxanthin transferred from flesh to gonadal tissue thereby allowing use of isomeric ratio of natural and synthetic astaxanthin to determine parent origin

Lura & Sñgrov, 1991a

Escapees of all ages

Genetic tags

Crozier, 1993; Clifford et al., 1998a, b

(a genealogical tree using symbols to represent sex and presence of a desired trait), and rotating a large number of broodstock fish. Broodstock originates from diverse base populations and stocks are rotated between breeding stations on a regular basis (Gjùen & Bentsen, 1997). Selective breeding for desirable traits in a rather structured `aquaculture environment' may also inadvertently influence escapee behavior patterns (such as predator avoidance, aggression, habitat selection), morphology and physiology. Escapee survivorship and interactions with wild conspecifics may be impacted by selective breeding. Survival of hatchery salmonids in the wild tends to be less than that of wild salmonids (Reisenbichler & McIntyre, 1977; Fraser, 1981). Reduced survival could mitigate adverse effects to wild stocks. Time of escapement also influences survivorship. Escapee salmon smolts entering the wild in spring have higher survival rates than smolts escaping in autumn or winter (Hansen & Jonsson, 1989). Greater survival of hatchery coho salmon exists following release in an estuary as opposed to release within a river (Solazzi et al., 1991). Increased river mortality may be associated with suboptimal river position and decreased predator avoidance. Lower survival has also been attributed to improper spawning behavior (e.g. adequate redd digging and defending, reproductive timing) by domestic fish which, in turn, affects survival rates of newly hatched eggs. Berejikian et al. (1997) observed that merely captive-rearing fry (collecting wild individuals, raising them in artificial environments, and later

Aquaculture Escapement, Implications and Mitigation

81

releasing them back to the river system for spawning) affected subsequent adult spawning behavior and success. Competitive inferiority of escapees is expected for adults during the spawning season. Hatchery coho salmon (Oncorhynchus kisutch) are competitively inferior to wild coho. Hatchery male coho have a 20% lower spawning success than hatchery females (Fleming & Gross, 1992, 1993). Hatchery males are considered less effective at accessing spawning opportunities. Hatchery females had poor site selection, inadequate redd preparation and unsuccessfully defended fertilized eggs (Fleming & Gross, 1993). Hatchery females also have an 8% lower fecundity than wild females (Fleming & Gross, 1992). In the wild, feeding is often limited (or postponed) due to the presence of predators. Selection for high growth rate in captivity (with ad libitum feeding and absence of predators) may inadvertently select for decreased predator avoidance and result in reduced escapee survival. Johnsson et al. (1996) observed reduction in predator avoidance by juvenile domestic brook trout, coupled with an increase in growth rate and food conversion ratio. Similar behavior has also been documented for domestic steelhead trout (Berejikian, 1995). Decreased predator avoidance could serve as a natural method to remove escapees from the wild. However, following escapee spawning in the wild, hybrid (domestic 6 wild) steelhead individuals also display decreased predator avoidance (Johnsson & Abrahams, 1991). Altered hybrid behavior could have negative implications to highly adapted wild salmonid populations. In contrast to these studies, Hvidsten & Lund (1988) observed similar predation rates on hatchery-reared and wild Atlantic salmon smolts. Feeding ad libitum combined with the high stocking densities experienced in artificial environments may affect the degree of aggression and schooling behavior expressed by escapees. Much debate exists, however, as to whether these behaviors will increase or decrease with domestication (reviewed by Ruzzante, 1994). Fenderson et al. (1968) for Atlantic salmon Salmo salar, Swain & Riddell (1990) for coho salmon, and Mesa (1991) for cutthroat trout Oncorhynchus clarki all provide evidence for elevated aggression in hatchery salmonids. Escapees may become more aggressive and display less schooling behavior in an attempt to increase their nearest neighbor distance (spatial distance between competing individuals for limited resources), thereby decreasing competition for limited food in the wild compared with captivity. Steelhead hatchery fry are more aggressive in situations with low population density and low food availability (Berejikian et al., 1996). Increased aggression may not be adequate to ensure escapee survival but may result in decreased predator avoidance. Wild conspecifics, accustomed to natural foraging and not hand-fed ad libitum, should perhaps outcompete escapees of the same size for limited food and habitat (Berejikian et al., 1996). However, owing to aquaculture trait selection for increased growth rate, seasons/habitats with adequate food supply effectively decreasing competition could provide the impetus for escapees to become larger than wild conspecifics during the same period of time. As expected, when given a size advantage, hatchery steelhead fry dominate wild fry (Berejikian et al., 1996). Artificial selection and productive `natural systems' that are underutilized could therefore favor escapees.

82

Ecological Aquaculture

Juvenile habitat selection and subsequent ability to defend a territory may be altered through domestication. Optimal stream position enhances survivorship and increases bioenergetic efficiency when fish remain in the stream area of least resistance. Suboptimal stream position results from losing dominance in confrontations between individuals for limited optimal space (Fausch, 1984). Hatchery-reared Atlantic salmon tend to maintain stream position at mid-water depth, between juvenile wild Atlantic salmon and brook trout (Dickson & MacCrimmon, 1982). Mid-water stream position will likely increase competition by overlapping juvenile habitat with both wild Atlantic salmon and native brook trout, and decrease escapee foraging success. Poor habitat selection may also be indicative of escapees lacking previous river experience, and occupying fast-flowing stream sections. Hatchery fish spend 43% of their time in riffles versus wild conspecifics spending only 29% in riffles (Mesa, 1991) and utilize physical stream structure less than wild fish (Bachman, 1984). Poor habitat selection may reduce escapee survivorship by providing an increased opportunity to be predated. Hatchery steelhead fry stay at the river water surface during daylight (Jonsson et al., 1993). Mid-water and surface stream position also increases risk of being swept downstream by rapid currents. Fast riffle position requires excessive energy expenditure to maintain river position, and subsequently results in smaller individuals that are subordinate to wild conspecifics. Subordinate individuals will, in turn, lose dominance and increase the likelihood of having to inhabit suboptimal pool and riffle position. Wild fish accurately time river migration with appropriate water flow regime. Following entry to the wild, escapees may be less morphologically adapted to local river systems and inferior to wild conspecifics. In addition, escapees have decreased morphological variation in comparison to wild conspecifics (Taylor, 1986). Decreased morphological variation may be the result of genetic homogeneity ± from using small numbers of broodstock and artificially selecting traits ± and/or the aquaculture growout environment with sufficient feed (Swain et al., 1991). Escapee morphometric changes are substantial following one generation in grow-out conditions, and become more prominent following selective breeding programs over numerous generations (Fleming et al., 1994). Although escapee spawning has been documented, decreased reproductive output may be anticipated owing to morphological deficiencies. Male hump height (Quinn & Foote, 1994), snout length (Fleming & Gross, 1994), and secondary coloration patterns (Hanson & Smith, 1967) all influence salmonid reproductive success. Obvious differences in phenotypic expression could result in lower reproductive output. Wild salmon have larger head dimensions, median fins and body depth than hatchery individuals (Swain et al., 1991; Fleming et al., 1994). Fins are frequently used in salmonid spawning behavior (e.g. female Atlantic salmon primarily use the caudal fin to build redds; Fleming et al., 1996), but fins may be eroded on escapees as a result of being held in net pens with high stocking density (Moring, 1982). Finally, successful escapee spawning with wild conspecifics could genetically alter the body morphology of wild populations (Riddell & Leggett, 1981; Taylor & McPhail, 1985). Hybrids (domestic 6 wild) could display changed phenotypic expression (alteration in body form) that may decrease reproductive success when compared with wild fish, resulting in long-term negative effects to wild populations.

Aquaculture Escapement, Implications and Mitigation

83

Physiological differences between hatchery and wild fish have been attributed to selective breeding programs and artificial rearing environments. During excess confinement, comparable to aquaculture environments, wild trout express considerably higher concentrations of plasma cortisol than domestic individuals (Woodward & Strange, 1987). In the wild, physiological trait selection for increased aquaculture production may decrease escapee fitness. Farmed Atlantic salmon have lower swimming endurance than sea-ranched salmon during the peak spawning season (Thorstad et al., 1997). Reduced endurance may be the result of decreased swimming, foraging, and predator avoidance displayed in captivity. Both hatchery Atlantic salmon (Thorstad et al., 1997) and brook trout (Johnsson et al., 1996) have higher somatic fat contents than wild or hybrid (domestic 6 wild) individuals. Increased somatic fat content may further decrease swimming endurance. Domestication typically selects for increased growth rates. Domestic Atlantic salmon growth is twice the growth of wild conspecifics until smolting (Gjùen & Bentsen, 1997). This could result in larger spawner sizes (Reisenbichler & McIntyre, 1977), and increased food conversion efficiencies (Johnsson et al., 1996). Increased size renders an advantage to escapees when competing for optimal river position, spawning opportunities, and defending redds (assuming escapees receive adequate food and display appropriate spawning behavior).

Potential interactions of escapees with wild conspecifics Decline of wild fish stocks has become a global phenomenon. Observed declines have been associated with climate change, overfishing, and habitat destruction (Beamish et al., 1999; Clark et al., 1999). Some wild stocks are slow to recover, even following fishery closures (Hutchings, 2000). In addition to external pressures affecting wild salmonid stocks, stresses from aquaculture escapees are of concern. As an example, on November 13, 2000, `wild Atlantic salmon' was listed as an endangered species in several Maine rivers. In their decision to protect wild Atlantic salmon, the NOAA National Marine Fisheries Service and the US Fish and Wildlife Service cited diseases, interbreeding and competition from `escapee Atlantic salmon' as potential issues threatening recovery efforts of the wild stock. International symposia have been held to discuss issues regarding interactions between salmonid escapees and wild conspecifics (Hansen et al., 1991; Hutchinson, 1997). In addition, a thorough review of salmonid farming has been completed in British Columbia, Canada, evaluating numerous issues of public concern, including the impacts associated with escapees (Anonymous, 1997). In all cases, disease, ecological, and genetic interactions with wild conspecifics are of greatest concern (Fig. 4.2).

Disease interactions One frequent opposition to fish farming is that the opportunity for disease infection and transfer is increased in an aquaculture setting compared with wild stocks. Rearing

84

Ecological Aquaculture

Fig. 4.2 Escapee interactions ± disease, ecological, and genetic ± with wild salmonid populations and the environment.

conditions could raise the likelihood for opportunistic diseases and/or parasites owing to increased stocking density and associated stress. However, the same disease/parasite agents are present within wild salmonid stocks, as well as other marine fish species (Saunders, 1991; Kent et al., 1998). Apparent increases in diseases and parasites in fish farms may be misleading owing to rigorous observation and documentation in aquaculture that is absent for wild fisheries and stock surveys (McVicar, 1997). Infectious agents may be spread from wild stocks to cultured fish (and vice versa) through horizontal transmission (direct contact either via the water or escapees interacting with wild conspecifics) or vertical transmission (from infected wild broodstock or successful spawning of infected escapees). Both horizontal and vertical diseases require extensive contact for transmission. Therefore, it is possible that disease and parasite agents could be more likely transmitted through stock enhancement programs that promote extensive contact and spawning between hatchery-derived and wild fish to increase depleted wild populations. For example, spread of the monogenean parasite Gyrodactylus salaris has been correlated with the introduction of infected hatchery salmon to Norwegian rivers through stock enhancement programs. Numerous nearby rivers that have never been stocked but are in close proximity also have the parasite present. These fish likely originate from stocked parr migrating to neighboring streams (Johnsen & Jensen, 1986; Lund & Heggberget, 1992). Although stock enhancement programs are considered the primary cause of G. salaris in the wild, infected transport tanks, birds and escapees may also be additional sources of infection.

Aquaculture Escapement, Implications and Mitigation

85

Some diseases of wild stocks ± such as enteric redmouth disease (Yersinia ruckeri) ± have been attributed to transmission of the pathogen from infected salmonid farms. However, no major outbreaks or clinical signs of the disease have been observed in wild fish (HaÊstein & Lindstad, 1991) that are likely serving as carriers of the pathogens. It may therefore be possible that the greatest risk of disease transfer exists from wild carriers to aquaculture stocks (Saunders, 1991). If suboptimal conditions exist for the cultured stock, the same opportunistic bacteria, not causing clinical signs of infection in wild fish, may result in a disease outbreak for cultured fish. To prevent opportunistic disease outbreaks, it is imperative that excellent husbandry practices be followed and optimal growing conditions maintained in culture environments. Following best management practices will decrease the opportunity for pathogen transmission and disease outbreak in aquaculture, which, in turn, would mitigate the opportunity for disease transmission to wild fish should escapement occur.

Ecological interactions Age, season, location of escapement, and juvenile experience influence movement patterns and upstream migration of escapees. Sutterlin et al. (1982) documented successful return of mature salmon to the smolt area of release after 13±25 months at sea. Decreased return can be expected for salmon escaping during winter. Up to 60% straying by salmon smolts released during the winter has been observed (Hansen & Jonsson, 1991). Rate of straying increases with distance between salmonid home-river location and escapement site (Heggberget et al., 1991). Increased straying will result in escapee adults entering rivers at random to spawn (Hansen et al., 1987). Haphazard river entry is permitted by escaped fish having no previous river experience offering discernible signposts (cues that allow return to a specific site) to follow during migration. Reduced homing precision to a river also exists for escapee steelhead trout adults (Jonsson et al., 1993). Juvenile river experience will impact escapee upstream migration behavior. Atlantic salmon will return to the river they left as juveniles, regardless of their genetic origin (Hansen & Jonsson, 1994). Hatchery smolts released in the mouth of a river (but without river experience) will return to that river as mature adults at the same time as wild conspecifics. However, hatchery adults enter the river later in the season illustrating the influence of river experience on behavior (Jonsson et al., 1990, 1991, 1994). Escaped farm salmon having no river experience will also delay river ascent compared with wild conspecifics (Eriksson & Eriksson, 1991; Gudjonsson, 1991; Lund et al., 1991; Heggberget et al., 1993). In contrast, hatchery smolts released within a river enter the river at the same time as wild conspecifics (Jonsson et al., 1994). These results illustrate the importance of river experience ± and aquaculture practices ± on timing of river ascent. Salmonid aquaculture industries typically raise early life stages either in tank facilities or ponds/lakes and transport (by truck) juveniles to an estuary without allowing river experience. Escapees would be expected to return to this estuary, assuming they are raised in the vicinity of the juvenile estuary, and enter rivers at a later time than wild conspecifics, having no signposts to continue the migration upstream.

86

Ecological Aquaculture

Once in a river, escapee and wild salmon have been observed together in spawning areas (ékland et al., 1995; Heggberget et al., 1996; Thorstad et al., 1998). Escapees of both sexes may spawn (apparently in equal numbers) with wild conspecifics (Lura & Sñgrov, 1991b; Webb et al., 1991). However, escapees tend to spawn later in the season (Webb et al., 1991). Having spawned later, escapees may dig redds in areas being used by wild salmon, decreasing wild salmon fitness (Webb et al., 1991). This would also indicate that escapee females with no river experience are able to locate suitable spawning areas and successfully dig redds, assuming fins are not eroded to prevent redd preparation. Youngson et al. (1993) indicated that Atlantic salmon escapees tended to hybridize with brown trout (Salmo trutta) more frequently than wild salmon. This may be attributed to overlapping habitat with brown trout and/or abnormal sexual behavior demonstrated by escapee female salmon decreasing spawning opportunities with wild salmon. Lura & Sñgrov (1993) observed escaped Atlantic salmon spawning earlier than native wild salmon in the River Vosso. Earlier spawning time overlapped with brown trout spawning. Escaped Atlantic salmon reproductive behavior could result in sterile progeny, negatively affecting the brown trout population (see `Genetic interactions' section). Additionally, earlier spawning will extend the growing season of hybrid (Atlantic salmon 6 brown trout) alevins. Assuming hybrids overlap juvenile habitat with wild Atlantic salmon, longer growing seasons will result in larger hybrid juveniles compared with wild salmon, and give a competitive and territorial advantage to hybrid progeny (Berejikian et al., 1996). Although successful spawning has been documented, escapee salmon are reproductively inferior to wild conspecifics. After entering a river to spawn, escapees remain in the river for a shorter period of time, with a notable percentage of both males and females leaving unspawned (Jonsson et al., 1990, 1991; ékland et al., 1995). Escapees in Norway move up- and downstream more often; are observed to be further up the river; and are more frequently injured during spawning than wild individuals (Jonsson et al., 1990, 1991; Heggberget et al., 1996; Thorstad et al., 1998). Increased movement and upstream distribution suggest escapees cannot locate `their' spawning area owing to the absence of juvenile river experience. Escapees tend to spawn in the lower reaches of northern Scottish rivers (Webb et al., 1991, 1993). Webb et al. (1991) noted that later escapee river entry compromised further river ascent owing to decreased water levels later in the season. Inferior escapee female spawning results from reduced spawning behavior, construction of fewer redds, and retention of more eggs than wild females (Fleming et al., 1996). Decreased fecundity and spawning behavior will result in less reproductive output from female escapees. Escapee males also display inappropriate courting behavior resulting in decreased success in entering redds to fertilize the eggs (Fleming et al., 1996). In North America, hatchery steelhead have reduced reproductive success in comparison with wild conspecifics (Chilcote et al., 1986; Leider et al., 1990). Predation and competition decrease the survivorship of offspring from hatchery steelhead. Owing to high mortality, hatchery steelhead contribution to the total population decreases from 85±87% at the egg stage to 42% at the adult stage (Leider et al., 1990).

Aquaculture Escapement, Implications and Mitigation

87

Genetic interactions* Wild salmonid stocks are frequently thought of as being highly adapted to their native stream, existing in reproductively discrete populations (brook charr, Salvelinus fontinalis, Angers et al., 1995; Verspoor, 1997, for review; coho salmon, Small et al., 1998; chinook salmon, Oncorhynchus tshawytscha, Nelson et al., 2001). Localized populations may form demes (species population inhabiting a specific geographic location) that express traits (i.e. behavior, morphology, physiology) reflecting ambient environmental conditions within the home river system. Local adaptation is a process whereby the frequency of beneficial traits to enhance the survival and/or reproductive success of individuals expressing them increases through natural selection (Taylor, 1991). Adaptation of salmonids to specific river systems ensures optimal habitat utilization and spawning for the natural environment, but could be detrimental to the population in times of immense human-induced change to the local habitat (e.g. construction of dams and fish ladders, pollution). Escapee `exogenous' population genetic interactions with wild conspecific `indigenous' populations can be classified as direct or indirect (Krueger & May, 1991; Utter, 1998). Direct genetic effects result from introgression (defined here as the incorporation of genetic material from escapees into the gene pool of an indigenous (native) population following successful spawning between the two strains (interbreeding); also called introgressive hybridization) of escapees with wild conspecifics. Direct genetic effects can have negative or positive genetic implications on a highly adapted indigenous population depending on the degree of additional external pressures experienced by the population. Aquaculture breeding programs artificially select for traits that will be economically beneficial in the rearing environment, but not so in a natural system. Therefore, introgression may cause outbreeding depression resulting in the reduction of fitness (survivorship and transmission of genes) in the following generation. Introduction of exotic individuals to the wild gene pool can modify growth, survival, reproduction and behavior of the wild population (Utter, 1998). Einum & Fleming (1997) concluded hybrid salmonids (escapee 6 wild) are likely to be inferior to locally adapted wild individuals, but superior to escapee fish originating from a seventh-generation broodstock for aquaculture. Given intermediate performance, wild salmonids, adapted to the environment, will outcompete escapee and hybrid (escapee 6 wild) individuals. However, hybrids (escapee 6 wild) can replace indigenous wild populations (hybrid swarms; Utter, 1998), especially in regions having a small, locally adapted, indigenous salmonid population experiencing a large accidental escapement of cultured stock (escapees). After an escape of farmed Atlantic salmon, samples taken from a river in Northern Ireland indicated an allelic shift in 0+ age salmon toward the farmed strain with a loss of genetic diversity (using electrophoretic variation) occurring (Crozier, 1993). Using mitochondrial DNA and minisatellite markers, Clifford et al. (1998a, b) suggested escaped adult farmed Atlantic salmon were interbreeding with wild salmon in * See Bert et al., Chapter 3, this volume, for explanation of terminology not defined throughout this section.

88

Ecological Aquaculture

northwest Irish rivers. Escaped juvenile farmed Atlantic salmon had completed their life cycle; and, upon return to the river, escapees bred and interbred with the wild fish. The markers (genetic tags) used in these studies tracked escapee fish and not fitness of the hybrids. However, both studies indicated farmed salmon had a reduced genetic diversity compared with wild individuals. In the absence of other external pressures, hybrid individuals could have negative consequences to a wild population owing to decreased genetic variability and local adaptation. Numerous examples exist illustrating resistance differences of salmonids to various disease/parasitic agents (e.g. vibrio, Gjedrem & Aulstad, 1974; myxosporean parasite Ceratomyxa shasta, Hemmingsen et al., 1986; furunculosis, Bailey, 1987; monogenean Gyrodactylus salaris, Bakke et al., 1990). Disease-related genetic effects result following reproduction, altering the degree of pathogen resistance of local salmonid conspecifics. In such cases, intermediate susceptibility could result and serve as a method to select against hybrids in the wild (Hemmingsen et al., 1986). Of course, most aquaculture stocks are selected for disease resistance traits which could potentially be transmitted through introgression, with the creation of a hybrid (escapee 6 wild) strain having superior disease resistance compared with the original wild stock. Wild salmonid populations can be reduced by high fishing pressures, often artificially selecting for larger individuals, and human-induced environmental changes (e.g. pollution, dam construction). As a result, the reduced wild population can undergo inbreeding depression. Inbreeding depression is decreased fitness from crossing closely related individuals and subsequent reduction in traits (e.g. fertility, growth rate, disease resistance) necessary for survival. When considering the depleted numbers of wild salmonid stocks in some river systems and resultant inbreeding, introgression of escapee stocks could have positive effects on the wild population by increasing the overall genetic variability, adding new alleles to the population (Thorpe, 1991) to cope with the immense human-induced changes occurring. To prevent or relieve inbreeding depression, Wohlfarth (1986) suggested stocking programs should use crossbred fish (domestic strains 6 wild strains) to produce greater vitality (heterosis or hybrid vigor). Crossbreeds could have increased heterozygosity (allelic diversity) owing to crossing of two inbred lineages (in this case wild 6 domestic). Researchers stocking cutthroat trout (wild 6 domestic), brook trout (wild 6 domestic) and pink salmon (wild 6 wild) observed higher returns indicative of better survival, faster growth, or both (Wohlfarth, 1986). However, hybrid vigor may not be experienced by all species having exogenous alleles entering the gene pool (e.g. steelhead trout; Wohlfarth, 1986). A potential drawback of hybrid vigor is that, although improvement of hybrid (escapee 6 wild) strains may be evident, subsequent generations may be less fit. For long-term improvements, Wohlfarth (1986) suggested domesticated fish of one sex be stocked periodically to establish a stable resident population. Indirect genetic effects alter the genetic composition within or among a population without introgression (Utter, 1998). Some examples of indirect genetic effects are introduction of pathogens/parasites by resistant carriers, competitive displacement (for limited habitat or food), and non-introgressive hybridization (introgression that

Aquaculture Escapement, Implications and Mitigation

89

does not go beyond the first generation) (Krueger & May, 1991; Utter, 1998). Indirect genetic effects can cause a reduction in the number of wild individuals in the population which increases the rates of genetic drift (random fluctuations of gene frequencies in a population) and inbreeding (spawning of closely related individuals), artificially changing allele frequencies (Krueger & May, 1991). Non-introgressive hybridization is introgression that does not go beyond the first generation (Utter, 1998; classified as `infertile interspecific hybrid' direct genetic effect by Krueger & May (1991)). Non-introgressive hybridization occurs when escaped Atlantic salmon and wild brown trout interbreed (e.g. Verspoor, 1988b; de Leaniz & Verspoor, 1989; Youngson et al., 1993; Galbreath & Thorgaard, 1995). Hybrid offspring (F1, Atlantic salmon 6 brown trout) can produce viable F2 (hybrid 6 hybrid) and backcross (hybrid 6 parent) progeny (Verspoor, 1988b). However, viable Atlantic salmon 6 brown trout hybrids are functionally sterile owing to major karyotypic (chromosomal complement of ± in this case ± a species) differences between the species (Galbreath & Thorgaard, 1995). Non-introgressive hybridization can displace wild populations (e.g. brown trout) through gamete wastage to produce sterile progeny (Utter, 1998).

Life history strategies of escapees There have been a large number of escapees recovered in wild fisheries; a substantial body of knowledge describing the effects of domestication on fish behavior, physiology and subsequent survival in the wild; and numerous accounts of escapee and wild stock interactions. But what do escapees do and where might they go? What could be a life history strategy for escapees? In 1997, a consortia of Canadian industry and research organizations implemented a biotelemetry study in Bay d'Espoir, Newfoundland, Canada, to monitor movement of domestic triploid steelhead trout in the wild (escapees), following intentional release from aquaculture grow-out and overwintering sites. Escapees were tracked subsequently throughout the Bay d'Espoir study area for 6 months with an advanced fixed data-logging and manual tracking biotelemetry system (Fig. 4.3) (Bridger et al., 2001b). A high degree of fidelity (to remain within the vicinity of the location of escapement) to the grow-out site was observed for individuals released on site during the summer growing season (Fig. 4.4a). Escapees also exhibited directed homing back to the grow-out site when released off-site (Fig. 4.4b), with subsequent fidelity to the grow-out site (Fig. 4.4c). Site fidelity percentage decreased with time but followed with attraction to other aquaculture sites as escapees displayed a directed migration upstream (Fig. 4.5a) (Bridger et al., 2001a). Steelhead outside of a cage in Bay d'Espoir are assumed to be escapees because the species is exotic to the region and raised only as sterile individuals. From stomach content analysis, escapees angled outside the grow-out cages consumed excess farm food falling through the cages (Table 4.3). Less site fidelity was observed for winter escapees (Fig. 4.4d), with a more immediate response to direct movement upstream (Fig. 4.5b) (Bridger et al., 2001a).

90

Ecological Aquaculture

4

SRX

5 Lab

1 2 Bimodal environment barrier (e.g., halocline or depth) 1. 2. 3. 4. 5.

Radio transmission to fixed station Wireless hydrophone (WHS) Acoustic transmission to WHS Satellite uplink Laboratory remotely collecting data

3

Fig. 4.3 Components of the remote, combined acoustic/radio biotelemetry system based upon a fieldproven radio telemetry data-logging receiver augmented with combined acoustic and radio transmitters (CART), remote data links and wireless hydrophone systems (WHS) to allow tracking studies of diadromous species in remote locations having enormous stretches of water (Bridger et al., 2001b). Table 4.3 Stomach contents of escaped domestic steelhead trout angled near aquaculture cages in Bay d'Espoir, Newfoundland, Canada, during the summer grow-out season and those angled near the mouth of the hydroelectric spillway during the winter season. F = full; E = empty stomachs. Fish

1 2 3 4 5 6 7 8 9 10

Fullness F F F F F F F F F F

Summer

Contents

Fish feed Fish feed + mussel spat Fish feed Fish feed Fish feed + mussel spat Fish feed Fish feed Fish feed Fish feed Fish feed + mussel spat

Fullness E F 1 4 1 4 1 4

E E <14 F 1 4

Winter

Contents

± Rocks + branches + bait Rocks + bait Rocks + branches + trichopteran larvae Rocks + branches ± ± Rocks Rocks + branches + bait Rocks + branches

During winter, less feeding opportunities existed and angled individuals (caught upstream) had empty stomachs or contents comprised of bottom debris (Table 4.3). In both seasons, upstream movement continued with a final destination to the hydroelectric spillway, also the location for the aquaculture industry hatchery (Fig. 4.5). Escapees had the opportunity to ascend a natural salmonid river system, the

% remaining

Aquaculture Escapement, Implications and Mitigation

100

100 75 54

50

off site 74%

37 6

0

0 wk 4 wk 8 wk 12 wk

on site 26%

on site 65%

off site 35%

4 hrs

48 hrs

14 wk

N=68

N=66 1 kilometer

1 kilometer

(a)

(b)

100 100

100 66

50 0

100 67

% located

% remaining

91

51

42 5

0 wk 4 wk 8 wk 12 wk

60 47

30 16

15

13 6 7

14 wk

0

13 6

10

6

0 days 2 days 4 days 6 days 16 days

N=66 1 kilometer

(c)

(d)

Fig. 4.4 Site fidelity of domestic triploid steelhead trout in Bay d'Espoir, Newfoundland, Canada: (a) to the summer grow-out site following an on-site release, (b) return to the grow-out site following an off-site release, (c) subsequent fidelity of returned off-site released steelhead to the grow-out site, and (d) fidelity of winter released steelhead trout to the overwintering site (adapted from Bridger et al., 2001a).

Conne River. However, individuals migrated up the hydroelectric spillway and remained in the vicinity for the remainder of the winter season. A lucrative recreational fishery for steelhead trout, not native to Bay d'Espoir waters and presumed to have escaped from aquaculture cages, exists near the entrance to the spillway each winter season. Directed movement to the spillway may either be a homing response to

92

Ecological Aquaculture

Spillway

Spillway

Farm site Fish location

5 kilometers

5 kilometers (a)

(b)

Fig. 4.5 Distribution of domestic triploid steelhead trout in Bay d'Espoir, Newfoundland, Canada: (a) determined from last tracked location for each individual from the summer release site, and (b) winter release site. Note the high degree of aquaculture site fidelity of summer released individuals during directed upstream movement (Bridger et al., 2001a).

their natal water system (flowing through the hatchery and out of the spillway) or simply an olfactory response to the presence of fish food odor exiting the hatchery and flowing out of the spillway. Upstream spawning migration by sterile escapees would illustrate the strong genetic control to spawn having no previous river experience and descendent from artificially selected broodstocks (Lura et al., 1993). The latter explanation accounts for apparent attraction of escapees to other aquaculture sites during summer upstream migration (Fig. 4.5a). Cooper & Scholz (1976) demonstrated steelhead trout imprint their homestream odor as juveniles and use this to return to the natal river to spawn. In this case, escapees may have imprinted the odor of fish feed and used it as a cue to direct movement in the wild. Atlantic salmon will also artificially imprint to a chemical added to the water ± morpholine ± as demonstrated by Sutterlin et al. (1982). From this research, what could be a complete life history strategy for escaped steelhead in Bay d'Espoir? It would seem likely that cultured fish, spending most of their lives in captivity, cannot adequately forage in the wild. During the summer growing season, escapees could direct movement to active aquaculture cages by means of an olfactory response to cues provided by feeding. These fish would consume excess food falling through the cages the same as if they were still held in the cages. Reduced feeding opportunities exist outside the cages, however, and competition is considerable. Escapees may disperse from the aquaculture site, even during the summer grow-out season and more abundant feeding opportunities, to increase its

Aquaculture Escapement, Implications and Mitigation

93

nearest neighbor distance and decrease competition. Steelhead eventually migrate further upstream, in response to hatchery wastewater laden with the odor of artificial feed, and feeding would cease for the winter season. Cessation of feeding in winter is comparable to that experienced in the local aquaculture industry with minimal feed provided to maintain metabolic activity. Movement between aquaculture sites in summer and the spillway in winter would therefore provide a successful strategy for escapees to survive. These observed behaviors can have both positive and negative implications to the aquaculture industry which must be addressed within future aquaculture management plans. On the positive side, aggregating escapees can be recaptured and removed from the environment (Anonymous, 1999). Escapees could be returned to the cages for further grow-out, decreasing economic losses experienced by the aquaculture investor associated with escapement, and eliminating concerns associated with wild stock interactions. Further research may develop advanced recapture methodologies that, for example, utilize escapee attraction to fish feed odors. A downside to strong site fidelity to the cages and eventual dispersion is an increased risk of anglers capturing escapees that have consumed excess medicated feeds intended for caged fish. Medicated feed usage is on the decline within the aquaculture industry with the advent of vaccination techniques and better management practices but may still be necessary in times of disease outbreaks (Willoughby, 1999). Drug residues have been documented in wild fish near aquaculture facilities (BjoÈrklund et al., 1990; Ervik et al., 1994) and with such high fidelity to the cages, and stomach contents depicting high dependence of escapees on farm feeds, risks to the public may be increased. In addition, risks of transferring diseases and parasites from wild carriers to the aquaculture stock may increase with the presence of escapees directly outside cages that can co-exist with both wild fish populations and aquaculture fish. Farm operators reported very close escapee association with aquaculture cages. In fact one operation recaptured one experimental fish during a routine net change! (C.J. Bridger, personal observation). The Newfoundland study may have several limitations for formulating a set of global assumptions to describe salmonid escapee behavior. A large portion of salmonid aquaculture does not occur in enclosed embayments such as Bay d'Espoir, which may have influenced escapee movement. What will escapees, of any salmonid species, do in an open environment that is not `funneled' to a river system? Experimental fish in this study were all triploid females and may be expected to display a different behavior than that of diploid, mixed sex populations (Cotter et al., 2000). Whether escapees were moving upstream as a biological spawning response ± unlikely because they were all sterile triploids ± or simply displaying a homing response to the location of the local industry hatchery and in essence to their `home stream' using the feed odor as a cue for directed movement, is uncertain. Atlantic salmon has displayed similar directed movement toward an aquaculture hatchery (Webb et al., 1991). Escapees were observed near the local industry hatchery and, in fact, some escapees were captured within the hatchery, gaining access when the discharge pipes were submerged. What would be expected of escapees from a region when no hatchery is in its vicinity?

94

Ecological Aquaculture

There exists much debate and continual concern over escapee interactions with wild conspecifics. The research of Bridger et al. (2001a, b) attempts to determine the life history strategy of escaped triploid steelhead trout. Much more extensive research is required to address the questions concerning escapee behavior for other size classes, different salmonid species, sexual condition, and implications to both aquaculture and natural ecosystems.

Mitigation of aquaculture escapees in the wild As a percentage of the total production level, escapement of aquaculture salmonids has been reduced in recent years. This decline has been the result of several pressures imposed on the aquaculture industry by consumers, environmental organizations, government policies and aquaculture economics. In a time of fluctuating prices and decreasing profit margins ± owing to flood of product into the marketplace and increasing production costs ± economic losses associated with large escapes are too costly for the industry to shoulder. Numerous governments and industries have adopted regulations and a Code of Conduct as proactive approaches to mitigate potential interactions between escaped salmonids and wild conspecifics, and to decrease environmental group and consumer concerns. In Norway, salmon farming is prohibited near important salmon rivers, and authorities have introduced strict regulations for the transport of fish (Bergan et al., 1991). In both stock enhancement and aquaculture, use of locally adapted salmonid strains is preferred to minimize negative genetic effects. Arguably, local strains, without selective breeding programs, may not be economically feasible for aquaculture using contemporary operational practices in the current market environment. However, in the end, both environmental sustainability and economic feasibility will have to be considered to ensure that responsible aquaculture operations exist which enhance the form and function of the natural ecosystem, including wild salmonid stocks, and the coastal economy of traditional fishing communities. Just as organic certification fetches premium prices for aquaculture products, environmental certification should require use of natural strains that will also fetch premium prices and ultimately offset economic losses incurred from raising inferior local strains. Sterilization methods, such as triploidy, may be utilized to eliminate genetic risks to wild populations. Performance of triploids will require improvement to achieve comparable growth and survival to diploid individuals (O'Flynn et al., 1997). Escaped triploids will still result in ecological and indirect genetic interactions with wild salmonid populations (including competition) that may effectively decrease the natural breeding population and possibly eliminate the effective wild spawning stock leaving only sterile fish. To this end, the most effective method to decrease interactions between escapees and wild fish is to completely eliminate escapement. Landbased and closed bag containment strategies are being examined for their economic and technical feasibility. In the event of escapement using conventional cages, it is imperative to recapture escapees to remove them from the wild (Anonymous, 1999). Successful recapture

Aquaculture Escapement, Implications and Mitigation

95

requires knowledge of the movement and dispersal patterns of escaped species and strains, and new technologies, necessitating additional research to complete the expensive, but extensive, scientific and engineering studies essential to attain adequate information for escapee removal. When a non-indigenous species, or even an exotic genetic strain, is introduced to an area, it is unknown as to what effects introductions might have on native populations. Introduced species or strains did not evolve in sympatry with native populations and therefore competition for limited natural resources may be unavoidable. It can be assumed that the wild, indigenous population is better adapted to its natural environment than an introduced species or strain, and is therefore expected to displace the exotic population. However, what constitutes a `natural' environment is very questionable today with numerous human perturbations, all of which threaten natural populations. Ironically, with highly adapted wild salmonid stocks declining owing to overfishing, climate change and habitat destruction, it seems that hatchery fish from stock enhancement and aquaculture sources could be all that can survive such drastic human-induced perturbations. Successful spawning between escapees and wild populations will result in an allelic shift creating a hybrid `natural' population that may be better able to survive in a harsh, quickly changing ecosystem.

Acknowledgments This chapter represents the culmination of the past three years of research involving CJB to determine the behavior of escapees in the wild. During the field component of the research, CJB is forever indebted to Richard Booth for teaching the intricate techniques for effective telemetry research, a memorable field experience, and continued friendship. The research was funded through the Ocean Technology Fund (OTF) and the Canadian Centre for Fisheries Innovation (CCFI), and directed by a committee comprised of Scott McKinley (Waterloo Biotelemetry Institute), Dave Scruton (Department of Fisheries and Oceans), Nigel Allen (Marine Institute of Memorial University of Newfoundland), Gary Hoskins (Conne River Aquaculture), Steve Moyse (Newfoundland Salmonid Growers Association), and Keith Stoodley and George Niezgoda (Lotek). Last, but certainly not least, we would like to thank Barry Costa-Pierce for inviting us to contribute to this timely product for the next era of aquaculture development. Barry, `Long may your big jib draw'.

References Angers, B., Bernatchez, L., Angers, A. & Desgroseillers, L. (1995) Specific microsatellite loci for brook charr reveal strong population subdivision on a microgeographic scale. Journal of Fish Biology, 47 (Supplement A), 177±185. Anonymous (1997) Salmon Aquaculture Review: Report of the Environmental Assessment Office. Summary plus Vols 1±5. BC Environmental Assessment Office, Vancouver.

96

Ecological Aquaculture

Anonymous (1999) Project Summary: Salmonid Recapture Technology for Aquaculture Operations. Canada/Newfoundland Agreement on Economic Renewal ACERA #1. Bachman, R.A. (1984) Foraging behavior of free-ranging, wild and hatchery brown trout in a stream. Transactions of the American Fisheries Society, 113, 1±32. Bailey, J.K. (1987) Differential survival among full sib Atlantic salmon (Salmo salar) families exposed to furunculosis. Salmon Genetics Research Program Report, Series No. 58. Bakke, T.A., Jansen, P.A. & Hansen, L.P. (1990) Differences in the host resistance of Atlantic salmon, Salmo salar L., stocks to the monogenean Gyrodactylus salaris Malmberg, 1957. Journal of Fish Biology, 37, 577±587. Beamish, R.J., Noakes, D.J., McFarland, G.A., Klyashtorin, L., Ivanov, V.V. & Kurashov, V. (1999) The regime concept and natural trends in the production of Pacific salmon. Canadian Journal of Fisheries and Aquatic Sciences, 56, 516±526. Berejikian, B.A. (1995) The effects of hatchery and wild ancestry and experience on the relative ability of steelhead trout fry (Oncorhynchus mykiss) to avoid a benthic predator. Canadian Journal of Fisheries and Aquatic Sciences, 52, 2476±2482. Berejikian, B.A., Mathews, S.B. & Quinn, T.P. (1996) Effects of hatchery and wild ancestry and rearing environments on the development of agonistic behavior in steelhead trout (Oncorhynchus mykiss) fry. Canadian Journal of Fisheries and Aquatic Sciences, 53, 2004±2014. Berejikian, B.A., Tezak, E.P., Schroder, S.L., Knudson, C.M. & Hard, J.J. (1997) Reproductive behavioral interactions between wild and captively reared coho salmon (Oncorhynchus kisutch). ICES Journal of Marine Science, 54, 1040±1050. Bergan, P.I., Gausen, D. & Hansen, L.P. (1991) Attempts to reduce the impact of reared Atlantic salmon on wild in Norway. Aquaculture, 98, 319±324. BjoÈrklund, H., Bondestam, J. & Bylund, G. (1990) Residues of oxytetracycline in wild fish and sediments from fish farms. Aquaculture, 86, 359±367. Bridger, C.J., Booth, R.K., McKinley, R.S. & Scruton, D.A. (2001a) Site fidelity and dispersal patterns by domestic triploid steelhead trout (Oncorhynchus mykiss Walbaum) released to the wild. ICES Journal of Marine Science, 58, 510±516. Bridger, C.J., Booth, R.K., McKinley, R.S., Scruton, D.A. & Lindstrom, R.T. (2001b) Monitoring fish behavior with a remote, combined acoustic/radio biotelemetry system. Journal of Applied Ichthyology, 17, 126±129. Carr, J.W., Anderson, J.M., Whoriskey, F.G. & Dilworth, T. (1997) The occurrence and spawning of cultured Atlantic salmon (Salmo salar) in a Canadian river. ICES Journal of Marine Science, 54, 1064±1073. Chilcote, M.W., Leider, S.A. & Loch, J.J. (1986) Differential reproductive success of hatchery and wild summer-run steelhead under natural conditions. Transactions of the American Fisheries Society, 115, 726±735. Clark, W.G., Hare, S.R., Parma, A.M., Sullivan, P.J. & Trumble, R.J. (1999) Decadal changes in growth and recruitment of Pacific halibut (Hippoglossus stenolepis). Canadian Journal of Fisheries and Aquatic Sciences, 56, 242±252. Clifford, S.L., McGinnity, P. & Ferguson, A. (1998a) Genetic changes in Atlantic salmon (Salmo salar) populations of Northwest Irish rivers resulting from escapes of adult farm salmon. Canadian Journal of Fisheries and Aquatic Sciences, 55, 358±363. Clifford, S.L., McGinnity, P. & Ferguson, A. (1998b) Genetic changes in Atlantic salmon population resulting from escaped juvenile farm salmon. Journal of Fish Biology, 52, 118±127.

Aquaculture Escapement, Implications and Mitigation

97

Cooper, J.C. & Scholz, A.T. (1976) Homing of artificially imprinted steelhead (rainbow) trout, Salmo gairdneri. Journal of the Fisheries Research Board of Canada, 33, 826±829. Costa-Pierce, B.A. & Bridger, C.J. (in press) The role of marine aquaculture facilities as habitats and ecosystems. In: Responsible Marine Aquaculture (eds R.R. Stickney & J.P. McVey). CAB International, London. Cotter, D., O'Donovan, V., O'MaoileÂidigh, N., Rogan, G., Roche, N. & Wilkins, N.P. (2000) An evaluation of the use of triploid Atlantic salmon (Salmo salar L.) in minimizing the impact of escaped farmed salmon on wild populations. Aquaculture, 186, 61±75. Craik, J.C.A. & Harvey, S.M. (1987) A biochemical method of distinguishing between wild and farmed salmonid fishes by their carotenoid pigmentation. Journal of the Forensic Science Society, 27, 47±55. Crozier, W.W. (1993) Evidence of genetic interaction between escaped farmed salmon and wild Atlantic salmon (Salmo salar L.) in a Northern Irish river. Aquaculture, 113, 19±29. de Leaniz, C.G. & Verspoor, E. (1989) Natural hybridization between Atlantic salmon, Salmo salar, and brown trout, Salmo trutta, in northern Spain. Journal of Fish Biology, 34, 41±46. Dickson, T.A. & MacCrimmon, H.R. (1982) Influence of hatchery experience on growth and behavior of juvenile Atlantic salmon (Salmo salar) within allopatric and sympatric stream populations. Canadian Journal of Fisheries and Aquatic Sciences, 39, 1453±1458. Einum, S. & Fleming, I.A. (1997) Genetic divergence and interactions in the wild among native, farmed and hybrid Atlantic salmon. Journal of Fish Biology, 50, 634±651. Eriksson, T. & Eriksson, L.-O. (1991) Spawning migratory behavior of coastal-released Baltic salmon (Salmo salar). Effects on straying frequency and time of river ascent. Aquaculture, 98, 79±87. Ervik, A., Thorsen, B., Eriksen, V., Lunestad, B.T. & Samuelsen, O.B. (1994) Impact of administering antibacterial agents on wild fish and blue mussels Mytilus edulis in the vicinity of fish farms. Diseases of Aquatic Organisms, 18, 45±51. Fausch, K.D. (1984) Profitable stream positions for salmonids: relating specific growth rate to net energy gain. Canadian Journal of Zoology, 62, 441±451. Fenderson, O.C., Everhart, W.H. & Muth, K.M. (1968) Comparative agonistic and feeding behavior of hatchery-reared and wild salmon in aquaria. Journal of the Fisheries Research Board of Canada, 25, 1±14. Fleming, I.A. & Gross, M.R. (1992) Reproductive behavior of hatchery and wild coho salmon (Oncorhynchus kisutch): does it differ? Aquaculture, 103, 101±121. Fleming, I.A. & Gross, M.R. (1993) Breeding success of hatchery and wild coho salmon (Oncorhynchus kisutch) in competition. Ecological Applications, 3, 230±245. Fleming, I.A. & Gross, M.R. (1994) Breeding competition in a Pacific salmon (coho: Oncorhynchus kisutch): measures of natural and sexual selection. Evolution, 48, 637±657. Fleming, I.A., Jonsson, B. & Gross, M.R. (1994) Phenotypic divergence of sea-ranched, farmed, and wild salmon. Canadian Journal of Fisheries and Aquatic Sciences, 51, 2808±2824. Fleming, I.A., Jonsson, B., Gross, M.R. & Lamberg, A. (1996) An experimental study of the reproductive behavior and success of farmed and wild Atlantic salmon (Salmo salar). Journal of Applied Ecology, 33, 893±905. Fraser, J.M. (1981) Comparative survival and growth of planted wild, hybrid, and domestic strains of brook trout (Salvelinus fontinalis) in Ontario lakes. Canadian Journal of Fisheries and Aquatic Sciences, 38, 1672±1684.

98

Ecological Aquaculture

Friars, G.W., Bailey, J.K. & O'Flynn, F.M. (1996) Applications of selection for multiple traits in cage reared Atlantic salmon (Salmo salar). Aquaculture, 138, 213±217. Galbreath, P.F. & Thorgaard, G.H. (1995) Sexual maturation and fertility of diploid and triploid Atlantic salmon 6 brown trout hybrids. Aquaculture, 137, 299±311. Gausen, D. & Moen, V. (1991) Large-scale escapes of farmed Atlantic salmon (Salmo salar) into Norwegian rivers threaten natural populations. Canadian Journal of Fisheries and Aquatic Sciences, 48, 426±428. Gjedrem, T. & Aulstad, D. (1974) Selection experiments with salmon. I. Differences in resistance to vibrio disease of salmon parr (Salmo salar). Aquaculture, 3, 51±59. Gjùen, H.M. & Bentsen, H.B. (1997) Past, present and future of genetic improvement in salmon aquaculture. ICES Journal of Marine Science, 54, 1009±1014. Gudjonsson, S. (1991) Occurrence of reared salmon in natural salmon rivers in Iceland. Aquaculture, 98, 133±142. Hansen, L.P., Dùving, K.B. & Jonsson, B. (1987) Migration of farmed adult Atlantic salmon with and without olfactory sense, released on the Norwegian coast. Journal of Fish Biology, 30, 713±721. Hansen, L.P. & Jonsson, B. (1989) Salmon ranching experiments in the River Imsa: effect of timing of Atlantic salmon (Salmo salar) smolt migration on survival to adults. Aquaculture, 82, 367±373. Hansen, L.P., HaÊstein, T., Nñvedal, G., Saunders, R.L. & Thorpe, J.E. (eds) (1991) Interactions between cultured and wild Atlantic salmon. Proceedings of the Symposium Hosted by the Directorate for Nature Management and Norwegian Institute for Nature Research, Loen, Norway, 23±26 April 1990. Aquaculture, 98, v±x + 324 pp. Hansen, L.P. & Jonsson, B. (1991) The effect of timing of Atlantic salmon smolt and post-smolt release on the distribution of adult return. Aquaculture, 98, 61±71. Hansen, L.P., Jacobsen, J.A. & Lund, R.A. (1993) High numbers of farmed Atlantic salmon, Salmo salar L., observed in oceanic waters north of the Faroes Islands. Aquaculture and Fisheries Management, 24, 777±781. Hansen, L.P. & Jonsson, B. (1994) Homing of Atlantic salmon: effects of juvenile learning on transplanted post-spawners. Animal Behavior, 47, 220±222. Hansen, L.P., Reddin, D.G. & Lund, R.A. (1997) The incidence of reared Atlantic salmon (Salmo salar L.) of fish farm origin at West Greenland. ICES Journal of Marine Science, 54, 152±155. Hanson, A.J. & Smith, H.D. (1967) Mate selection in a population of sockeye salmon (Oncorhynchus nerka) of mixed age-groups. Journal of the Fisheries Research Board of Canada, 24, 1955±1976. HaÊstein, T. & Lindstad, T. (1991) Diseases in wild and cultured salmon: possible interaction. Aquaculture, 98, 277±288. Heggberget, T.G., Hvidsten, N.A., Gennerùd, T.B. & Mùkkelgjerd, P.I. (1991) Distribution of adult recaptures from hatchery-reared Atlantic salmon (Salmo salar) smolts released in and offshore of the River Surna, western Norway. Aquaculture, 98, 89±96. Heggberget, T.G., ékland, F. & Ugedal, O. (1993) Distribution and migratory behaviour of adult wild and farmed Atlantic salmon (Salmo salar) during return migration. Aquaculture, 118, 73±83. Heggberget, T.G., ékland, F. & Ugedal, O. (1996) Prespawning migratory behavior of wild

Aquaculture Escapement, Implications and Mitigation

99

and farmed Atlantic salmon, Salmo salar L., in a north Norwegian river. Aquaculture Research, 27, 313±322. Hemmingsen, A.R., Holt, R.A., Ewing, R.D. & McIntyre, J.D. (1986) Susceptibility of progeny from crosses among three stocks of coho salmon to infection by Ceratomyxa shasta. Transactions of the American Fisheries Society, 115, 492±495. Hiilivirta, P., Ikonen, E. & Lappalainen, J. (1998) Comparison of two methods for distinguishing wild from hatchery-reared salmon (Salmo salar Linnaeus, 1758) in the Baltic Sea. ICES Journal of Marine Science, 55, 981±986. Hindar, K. & L'AbeÂe-Lund, J.H. (1992) Identification of hatchery-reared and wild Atlantic salmon juveniles based on examination of otoliths. Aquaculture and Fisheries Management, 23, 235±241. Hutchings, J.A. (2000) Collapse and recovery of marine fishes. Nature, 406, 882±885. Hutchinson, P. (ed.) (1997) Interactions between salmon culture and wild stocks of Atlantic salmon: the scientific and management issues. Proceedings of the ICES/NASCO Symposium, Bath, England, 18±22 April 1997. ICES Journal of Marine Science, 54, 963±1227. Hvidsten, N.A. & Lund, R.A. (1988) Predation on hatchery-reared and wild smolts of Atlantic salmon, Salmo salar L., in the estuary of River Orkla, Norway. Journal of Fish Biology, 33, 121±126. Johnsen, B.O. & Jensen, A.J. (1986) Infestations of Atlantic salmon, Salmo salar, by Gyrodactylus salaris in Norwegian rivers. Journal of Fish Biology, 29, 233±241. Johnsson, J.I. & Abrahams, M.V. (1991) Interbreeding with domestic strain increases foraging under threat of predation in juvenile steelhead trout (Oncorhynchus mykiss): an experimental study. Canadian Journal of Fisheries and Aquatic Sciences, 48, 243±247. Johnsson, J.I., Petersson, E., JoÈnsson, E., BjoÈrnsson, B.T. & JaÈrvi, T. (1996) Domestication and growth hormone alter antipredator behavior and growth patterns in juvenile brown trout, Salmo trutta. Canadian Journal of Fisheries and Aquatic Sciences, 53, 1546±1554. Jonsson, B., Jonsson, N. & Hansen, L.P. (1990) Does juvenile experience affect migration and spawning of adult Atlantic salmon? Behavioral Ecology and Sociobiology, 26, 225±230. Jonsson, B., Jonsson, N. & Hansen, L.P. (1991) Differences in life history and migratory behaviour between wild and hatchery-reared Atlantic salmon in nature. Aquaculture, 98, 69±78. Jonsson, N., Jonsson, B., Hansen, L.P. & Aass, P. (1993) Coastal movement and growth of domesticated rainbow trout (Oncorhynchus mykiss (Walbaum)) in Norway. Ecology of Freshwater Fish, 2, 152±159. Jonsson, N., Hansen, L.P. & Jonsson, B. (1994) Juvenile experience influences timing of adult river ascent in Atlantic salmon. Animal Behaviour, 48, 740±742. Kent, M.L., Traxler, G.S., Kleser, D. et al. (1998) Survey of salmonid pathogens in oceancaught fishes in British Columbia, Canada. Journal of Aquatic Animal Health, 10, 211±219. Krueger, C.C. & May, B. (1991) Ecological and genetic effects of salmonid introductions in North America. Canadian Journal of Fisheries and Aquatic Science, 48 (Supplement 1), 66±77. Leider, S.A., Hulett, P.L., Loch, J.J. & Chilcote, M.W. (1990) Electrophoretic comparison of the reproductive success of naturally spawning transplanted and wild steelhead trout through the returning adult stage. Aquaculture, 88, 239±252. Lund, R.A. & Hansen, L.P. (199l) Identification of wild and reared Atlantic salmon, Salmo salar L., using scale characters. Aquaculture and Fisheries Management, 22, 499±508.

100

Ecological Aquaculture

Lund, R.A., ékland, F. & Hansen, L.P. (1991) Farmed Atlantic salmon (Salmo salar) in fisheries and rivers in Norway. Aquaculture, 98, 143±150. Lund, R.A. & Heggberget, T.G. (1992) Migration of Atlantic salmon, Salmo salar L., parr through a Norwegian fjord: potential infection path of Gyrodactylus salaris. Aquaculture and Fisheries Management, 23, 367±372. Lund, R.A., Midtyng, P.J. & Hansen, L.P. (1997) Post-vaccination intra-abdominal adhesions as a marker to identify Atlantic salmon, Salmo salar L., escaped from commercial fish farms. Aquaculture, 154, 27±37. Lura, H. & Sñgrov, H. (1991a) A method of separating offspring from farmed and wild Atlantic salmon (Salmo salar) based on different ratios of optical isomers of astaxanthin. Canadian Journal of Fisheries and Aquatic Sciences, 48, 429±433. Lura, H. & Sñgrov, H. (1991b) Documentation of successful spawning of escaped farmed female Atlantic salmon, Salmo salar, in Norwegian rivers. Aquaculture, 98, 151±159. Lura, H. & Sñgrov, H. (1993) Timing of spawning in cultured and wild Atlantic salmon (Salmo salar) and brown trout (Salmo trutta) in the River Vosso, Norway. Ecology of Freshwater Fish, 2, 167±172. Lura, H., Barlaup, B.T. & Sñgrov, H. (1993) Spawning behavior of a farmed escaped Atlantic salmon (Salmo salar). Journal of Fish Biology, 42, 311±313. Lura, H. & ékland, F. (1994) Content of synthetic astaxanthin in escaped farmed Atlantic salmon, Salmo salar L., ascending Norwegian rivers. Fisheries Management and Ecology, 1, 205±216. McKinnell, S. & Thomson, A.J. (1997) Recent events concerning Atlantic salmon escapes in the Pacific. ICES Journal of Marine Science, 54, 1221±1225. McKinnell, S., Thomson, A.J., Black, E.A. et al. (1997) Atlantic salmon in the north Pacific. Aquaculture Research, 28, 145±157. McVicar, A.H. (1997) Disease and parasite implications of the coexistence of wild and cultured Atlantic salmon populations. ICES Journal of Marine Science, 54, 1093±1103. Mesa, M.G. (1991) Variation in feeding, aggression, and position choice between hatchery and wild cutthroat trout in an artificial stream. Transactions of the American Fisheries Society, 120, 723±727. Moring, J.R. (1982) Fin erosion and culture-related injuries of Chinook salmon raised in floating net pens. The Progressive Fish-Culturist, 44, 189±191. Nelson, R.J., Small, M.P., Beacham, T.D. & Supernault, K.J. (2001) Population structure Fraser River Chinook salmon (Oncorhynchus tshawytscha): an analysis using microsatellite DNA markers. Fisheries Bulletin, 99, 94±107. O'Flynn, F.M., McGeachy, S.A., Friars, G.W., Benfey, T.J. & Bailey, J.K. (1997) Comparisons of cultured triploid and diploid Atlantic salmon (Salmo salar L.). ICES Journal of Marine Science, 54, 1160±1165. ékland, F., Heggberget, T.G. & Jonsson, B. (1995) Migratory behaviour of wild and farmed Atlantic salmon (Salmo salar) during spawning. Journal of Fish Biology, 46, 1±7. Potter, E.C.E. (1987) Discrimination between wild, farmed and stocked Atlantic salmon using fin measurements. ICES CM M:21. Quinn, T.P. & Foote, C.J. (1994) The effects of body size and sexual dimorphism on the reproductive behaviour of sockeye salmon, Oncorhynchus nerka. Animal Behaviour, 48, 751±761.

Aquaculture Escapement, Implications and Mitigation

101

Reisenbichler, R.R. & McIntyre, J.D. (1977) Genetic differences in growth and survival of juvenile hatchery and wild steelhead trout, Salmo gairdneri. Journal of the Fisheries Research Board of Canada, 34, 123±128. Riddell, B.E. & Leggett, W.C. (1981) Evidence of an adaptive basis for geographic variation in body morphology and time of downstream migration of juvenile Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences, 38, 308±320. Ruzzante, D.E. (1994) Domestication effects on aggressive and schooling behavior in fish. Aquaculture, 120, 1±24. Sñgrov, H., Hindar, K., Kalas, S. & Lura, H. (1997) Escaped farmed Atlantic salmon replace the original salmon stock in the River Vosso, western Norway. ICES Journal of Marine Science, 54, 1166±1172. Saunders, R.L. (1991) Potential interactions between cultured and wild Atlantic salmon. Aquaculture, 98, 51±60. Small, P., Withler, R.E. & Beacham, T.D. (1998) Population structure and stock identification of British Columbia coho salmon, Oncorhynchus kisutch, based on microsatellite DNA variation. Fishery Bulletin, 96, 843±858. Solazzi, M.F., Nickelson, T.E. & Johnson, S.L. (1991) Survival, contribution, and return of hatchery coho salmon (Oncorhynchus kisutch) released into freshwater, estuarine, and marine environments. Canadian Journal of Fisheries and Aquatic Sciences, 48, 248±253. Stokesbury, M.J. & Lacroix, G.L. (1997) High incidence of hatchery origin Atlantic salmon in the smolt output of a Canadian river. ICES Journal of Marine Science, 54, 1074±1081. Sutterlin, A.M., Saunders, R.L., Henderson, E.B. & Harmon, P.R. (1982) The homing of Atlantic salmon (Salmo salar) to a marine site. Canadian Technical Report of Fisheries and Aquatic Sciences, 1058, iii + 6 pp. Swain, D.P. & Riddell, B.E. (1990) Variation in agonistic behavior between newly emerged juveniles from hatchery and wild populations of coho salmon, Oncorhynchus kisutch. Canadian Journal of Fisheries and Aquatic Sciences, 47, 566±571. Swain, D.P., Riddell, B.E. & Murray, C.B. (1991) Morphological differences between hatchery and wild populations of coho salmon (Oncorhynchus kisutch): environmental versus genetic origin. Canadian Journal of Fisheries and Aquatic Sciences, 48, 1783±1791. Taylor, E.B. (1986) Differences in morphology between wild and hatchery populations of juvenile coho salmon. The Progressive Fish-Culturist, 48, 171±176. Taylor, E.B. (1991) A review of local adaptation in Salmonidae, with particular reference to Pacific and Atlantic salmon. Aquaculture, 98, 185±207. Taylor, E.B. & McPhail, J.D. (1985) Variation in body morphology among British Columbia populations of coho salmon, Oncorhynchus kisutch. Canadian Journal of Fisheries and Aquatic Sciences, 42, 2020±2028. Thorpe, J.E. (1991) Acceleration and deceleration effects of hatchery rearing on salmonid development, and their consequences for wild stocks. Aquaculture, 98, 111±118. Thorstad, E.B., Finstad, B., ékland, F., McKinley, R.S. & Booth, R.K. (1997) Endurance of farmed and sea-ranched Atlantic salmon Salmo salar L. at spawning. Aquaculture Research, 28, 635±640. Thorstad, E.B., Heggberget, T.G. & ékland, F. (1998) Migratory behaviour of adult wild and escaped farmed Atlantic salmon, Salmo salar L., before, during and after spawning in a Norwegian river. Aquaculture Research, 29, 419±428.

102

Ecological Aquaculture

Tringali, M.D. & Bert, T.M. (1998) Risk to effective population size should be an important consideration in fish stock-enhancement programs. Bulletin of Marine Science, 62, 641± 659. Tringali, M.D. & Leber, K.M. (1999) Genetic considerations during the experimental and expanded phases of snook stock enhancement. Bulletin of the National Research Institute of Aquaculture, Supplement 1, 109±119. Utter, F. (1998) Genetic problems of hatchery-reared progeny released into the wild, and how to deal with them. Bulletin of Marine Science, 62, 623±640. Verspoor, E. (1997) Genetic diversity among Atlantic salmon (Salmo salar L.) populations. ICES Journal of Marine Science, 54, 965±973. Verspoor, E. (1988a) Reduced genetic variability in first-generation hatchery populations of Atlantic salmon (Salmo salar). Canadian Journal of Fisheries and Aquatic Sciences, 45, 1686±1690. Verspoor, E. (1988b) Widespread hybridization between native Atlantic salmon, Salmo salar, and introduced brown trout, S. trutta, in eastern Newfoundland. Journal of Fish Biology, 32, 327±334. Webb, J.H., Hay, D.W., Cunningham, P.D. & Youngson, A.F. (1991) The spawning behaviour of escaped farmed and wild Atlantic salmon (Salmo salar L.) in a northern Scottish river. Aquaculture, 98, 97±110. Webb, J.H. & Youngson, A.F. (1992) Reared Atlantic salmon, Salmo salar L., in the catches of a salmon fishery on the western coast of Scotland. Aquaculture and Fisheries Management, 23, 393±397. Webb, J.H., Youngson, A.F., Thompson, C.E., Hay, D.W., Donaghy, M.J. & McLaren, I.S. (1993) Spawning of escaped farmed Atlantic salmon, Salmo salar L., in western and northern Scottish rivers: egg deposition by females. Aquaculture and Fisheries Management, 24, 663±670. Willoughby, S. (1999) Manual of Salmonid Farming. Fishing News Books, London. Wohlfarth, G.W. (1986) Decline in natural fisheries ± a genetic analysis and suggestion for recovery. Canadian Journal of Fisheries and Aquatic Sciences, 43, 1298±1306. Woodward, C.C. & Strange, R.J. (1987) Physiological stress responses in wild and hatcheryreared rainbow trout. Transactions of the American Fisheries Society, 116, 574±579. Youngson, A.F., Webb, J.H., MacLean, J.C. & Whyte, B.M. (1997) Frequency of occurrence of reared Atlantic salmon in Scottish salmon fisheries. ICES Journal of Marine Science, 54, 1216±1220. Youngson, A.F., Webb, J.H., Thompson, C.E. & Knox, D. (1993) Spawning of escaped farmed salmon (Salmo salar): hybridization of females with brown trout (Salmo trutta). Canadian Journal of Fisheries and Aquatic Sciences, 50, 1986±1990.

Chapter 5

Farming Systems Research and Extension Methods for the Development of Sustainable Aquaculture Ecosystems Barry A. Costa-Pierce Rhode Island Sea Grant College Program The importance of indigenous knowledge Every culture has an indigenous bank of knowledge. Indigenous knowledge is the information derived from intimate, day-to-day interactions between people and their environments. Indigenous knowledge includes history, art, economics, linguistics, science, engineering, medicine, politics and psychology, as well as agriculture, fishing, hunting, and gathering, plus the activities of trade, economics and all other areas of human enterprise. Indigenous knowledge is present in both traditional and urban societies, and is taught to children through the family unit and the various institutions in society. For example, in Bali, Indonesia, an ancient system of rice cultivation involves terracing, irrigation and dikes that move the island's water to thousands of rice paddies situated at all elevations on the mountainous island. Gadgil & Guha (1992) stated that technical evolution of the Bali rice agroecosystem could only take place if the indigenous knowledge was in harmony with a cultural evolution having a common world vision. The Bali rice paddies are interwoven into a complex institutional system of common property, collective action, and Hindu spirituality. Sustainability of the rice agroecosystem has been achieved because the maintenance of the irrigation systems and food production is an institutionalized social responsibility of all participants. Traditional land and water use practices are changing rapidly. Indigenous knowledge built up over centuries about farming and fishing ecosystems is being lost (Roling & Jiggens, 1998). The wisdom of rural societies and the benefits of nature's good and services to society are being lost as the world grows more urbanized. Holistic knowledge and the wisdom that incorporates the spiritual world and binds it to the physical realities of everyday life are also endangered.

The need to evolve sustainable aquaculture ecosystems Similar to sustainable agroecosystems like those in Bali, aquaculture has traditional knowledge systems and cultural, family, and community roots. States Borgese (1980):

104

Ecological Aquaculture

`That aquaculture has a philosophical base in the East and a scientific base in the West has far-reaching implications. In the East, it is culture, it is life: culture to improve life by providing food and employment. It is embedded in the social and economic infrastructure. All that science can and must do is to make this culture more effective. In the West, aquaculture is science and technology, embodied in industry and providing profits: money. It has no social infrastructure. In this, the West has much to learn from the East.' While aquaculture has cultural roots in the East and is most developed there, it is still an uncommon occupation in Asia (Edwards, 1993). However, there are no mysterious roots to the evolution of aquaculture. Aquaculture originated in China when population densities exceeded the carrying capacities of the natural, oceanic, and agricultural environments to support them in historical times. In order to support these people, new, labor-intensive agricultural innovations such as intensive, small scale agriculture involving intercropping, multistory agriculture, and a tight recycling of nutrients through ecologically sophisticated composting and waste recycling regimes became common (and are now endangered). Aquaculture, one of the most complex and sophisticated of all possible agriculture innovations, arose to meet the protein needs of many Asian societies who reached high population densities in historical times (and possibly European and Hawaiian societies, see Beveridge & Little, Chapter 1, and Costa-Pierce, Chapter 2, this volume). Aquaculture arose as an integral part of a complex polyculture of agricultural systems, and was intricately entwined into the existing water management system for mixed agriculture. Aquaculture has traditional, community roots in Asia. In the West, population densities exceeding environmental and agricultural and aquatic carrying capacities have only recently reached critical stages, and aquaculture has been adopted. However, the rich West has nowhere near the pressing food needs of Asia, with numerous and abundant `protein choices'. As such, commercial aquaculture in the West has evolved similar to industrial agriculture to provide a diversity of products for high priced, luxury protein markets. Western aquaculture is perceived by some as an `industrial' development rather than a `community-based' development. In some cases, the very communities that host new aquaculture operations consider industrial aquaculture operations `outsiders'. In the last decade there have been radical (and reasoned) concerns about aquaculture developments from the very communities that should be its natural supporters ± scientists, rural communities, and environmentalists (Barinaga, 1990; Black, 2000). While the technical issues of pollution control can be resolved by incorporating ecological or ecosystems technologies into aquaculture, public opposition to aquaculture will be a more difficult problem to solve if aquaculture in the West remains an `industry' producing `wastes' and requiring evolution of a new regulatory structure (Fig. 5.1). Industrial agriculture is the largest source of pollution in the over-industrialized countries. Dent & Anderson (1971) stated that agriculture in these countries has `ignored the larger ecological framework in which farming is conducted and as a result agricultural production has often exploited the natural environment'.

Methods for the Development of Sustainable Aquaculture Ecosystems

105

Fig. 5.1 Ensuring the evolution of sustainability in aquaculture will require that the `industrial' model of intensive aquaculture (a) that produces new sources of aquatic pollution, causing degradation of ecosystem services and a new regulatory structure, will evolve into socially and environmentally sustainable `aquaculture ecosystems' (b) that turn wastes into resources using ecological engineering and systems approaches that lead to technical and community sustainability and environmental enhancement (Costa-Pierce, 1996).

Developments in agricultural research towards evolving more sustainable approaches, and new marketing trends towards certified organic products produced from sustainable farming systems, provide the background for the evolution of sustainable aquaculture ecosystems. Like agriculture, aquaculture's evolution towards social, economic, and ecological sustainability depends on innovative farmers, supportive policy makers and social institutions, educators, researchers, bankers, and consumers working together to influence adoption and ensure sustainability (CostaPierce, 1998). Pioneering agriculture research in agroecology, agroecosystems, and farming systems; and the participatory technology development framework embedded in the `farmer first' extension and outreach methodologies provide a road map for evolving sustainable aquaculture ecosystems (Conway, 1986; Alteri, 1987; Chambers et al., 1989; Scarborough et al., 1997; Gliessman, 1998). Clearly, the public will not tolerate the addition of any new sources of pollution or the further degradation of the natural environment which is perceived to come at the expense of the degradation of the quality of life. An increasingly skeptical public determined to fight `the experts' is making connections between the disruption of our environment and human health. In many cases the simple implication of the presence of a chemical implies a hazard and a threat to human health. In order to change the public perception of aquaculture as `outsiders' or `industrial polluters', intensive aquaculture operations must plan to become part of a community and a region, and have a wider plan for community development that works with leaders to provide needed inputs, to recycle wastes, to create a diversity of unprocessed and value-added products, to provide local market access, and to plan for job creation and environ-

106

Ecological Aquaculture

mental enhancement on local and regional scales. When viewed from these community development perspectives, aquaculture and the public it intends to serve have many common objectives. Edwards (1993) states that the `problem for developing countries is essentially how to stimulate agricultural (and aquaculture) productivity and profitability without further environmental degradation, in contrast to the need to reduce the level of intensification to a sustainable level in the developed world.' As an infant enterprise the world over, aquaculture can ill afford to recreate the evolution of commercial agriculture in the mid-twentieth century where huge, toxic, nutrient and chemical loading were washed down a primitive path of the `solution to pollution is dilution'. Rather, modern aquaculture should adopt a new strategy, a model of `community-based, ecologically sustainable aquaculture' that produces certified organic produce, similar to a strategy promoted in agriculture and industry called `input management' (Odum, 1989). Folke & Kautsky (1991) state, `One must expand the boundaries and one's actions far beyond the cultivation site, and realize that there is an unavoidable complementarity between the life-support environment and aquaculture production'. In addition, it is also essential to consider the social ecology of aquaculture developments from the outset in order to articulate what are the most important development goals for sustainable aquaculture systems, and to define what will be required socially in order to develop community-based aquaculture ecosystems and `green' marketing approaches (ecolabelling, sustainable certification, etc.). In order to ensure ecological sustainability of aquaculture, researchers must have new methodologies available to capture both the global (macro) and the farm-level (micro) social ecological processes occurring in order to determine the appropriate paths for research and extension interventions. Technical adaptations alone are inadequate to direct sustainability of complex, new agricultural enterprises such as aquaculture. An improved and more participatory aquaculture research process can stimulate a greater momentum for change and increase the effectiveness of aquaculture extension approaches.

Farmer participatory approaches to development of sustainable aquaculture ecosystems Scientific knowledge is a knowledge base developed and recorded by scientists. Indigenous knowledge encompasses all aspects of life in an environment. Farmers everywhere experiment. They adapt, innovate, and observe the results of their work, and have been doing so for centuries. It is only recently that `farmer led' research processes of agriculture and aquaculture development have been superseded by scientist-directed agricultural and aquatic research. Farmers have invaluable indigenous knowledge and experiment using scientific principles, oftentimes without recognizing their experimentation as science (Kansing & Kremer, 1995). McCorkle (1994) found that farmers design, implement and evaluate farm trials by gathering background information, selecting sites, identifying variables, and monitoring and evaluating these trials. Today, increasing numbers of scientists are acting as facili-

Methods for the Development of Sustainable Aquaculture Ecosystems

107

tators of collaborative research with farmers, in equal partnerships with farmers. Farmers must be able to adapt to continuously changing conditions in order to evolve sustainability. It therefore becomes critical that farmers be able to analyze, monitor, adapt and innovate. Farmers know best about their own land and water use systems, and their social and economic realities. Farmers are skilled innovators who have developed ways of experimenting through trial and error. Trust must be built up with farmers by respecting local values and working together in the spirit of equality. For research scientists this can be a challenge and a revelation since they have to reorient their world views about the meaning of knowledge, science, the economy, gender roles and relations, and the required methodologies needed in order to get buy-in to participatory research and learning processes. In participatory technology development (PTD), farmers can design and experiment using strategies they have developed themselves which they feel are appropriate to conditions they experience on their own farms. Such participatory experimentation promotes empowerment and accountability. These processes lead to institution building, market reforms, and the farmerbased advocacy needed to secure policy reforms and rural economic development (Veldhuizen, 1998). PTD approaches have been used to initiate farmers and fishers who pursue conventional terrestrial agriculture or capture fisheries to incorporate aquaculture into their farming/fishing systems. The overall objective of PTD is to elicit evolutionary change towards sustainability in farming/fishing systems by merging aquatic and terrestrial production systems; by using ecological principles as the basis for new designs for food production systems; and by incorporating simple ecological modeling into holistic systems analyses of production and natural systems. The approach is to use the wisdom of ecology (Odum, 1975) and its underlying principles of hierarchies, complementarity, redundancy, cycling, and diversity to not only meet environmental goals, but also to improve farmer livelihoods by increasing whole farm efficiencies and product values. PTD seeks to demonstrate that the complementarity of systems and enhancement of recycling pathways of on-farm resources will lead to greater resource efficiencies, long-term sustainability, and environmental protection (Lightfoot, 1990). The principal idea behind a participatory PTD approach is that farmers have to be involved in the process of technology research development and dissemination from the outset (Scarborough et al., 1997). Instead of scientists developing one fixed set of techniques in isolation on a research station, then disseminating them as a `technology package' to farmers, ideas from farmers are elicited first; then researchers and farmers work to perfect technologies that suit the farming systems of the target group. Farmers are invited to take part in the technological identification, research and development and extension processes from the outset of the process in contrast to being a passive recipient of technologies developed elsewhere. Farmers can comment on and criticize as much as they want; they can test new technologies on their own farms or at research stations; and they can modify technologies if they think necessary as long as the process of technological modifications/innovations is monitored and recorded. Scarborough et al. (1997) summarized the need for the PTD approaches on the basis of:

108

(1) (2) (3) (4)

Ecological Aquaculture

What works in one place, time and circumstance will not necessarily work in another. What suits one farmer may not suit another with different ideas and constraints. The complexity of a farming situation and livelihoods affects the adoption of interventions. The message-based approach is the least effective teaching method.

An important characteristic of the PTD approaches is that the changes towards sustainability are gradual. A new technology is an addition to or a modification of an existing system, so that adoption is not a big step. This approach to sustainability is rooted in indigenous farming communities, and is a long-term `evolutionary' approach to aquaculture development. It is not a short-term `revolutionary' approach with high capital costs and many exterior inputs. Pretty (1995) described adoption of innovations as a three-step process facilitated by PTD: (1) (2) (3)

Evolutionary learning: step-by-step, cumulative participatory learning by stakeholders. Multiple perspectives: many ways of describing a situation. Iterative group learning: the complexity of the world can only be learned by an iterative process of group inquiry and learning.

Another characteristic of the PTD approach is that the responsibility for adopting a new technology rests entirely with the farmer. The farmer decides whether or not to try a new technology on the farm. Farmer refusal to implement a technology is an important signal to the research or extension worker that something is wrong. Farmers do not receive any financial assistance or other subsidies besides information and engagement by the research and extension workers, and are not in a dependent position. The relationship between farmers, extension agents and researchers is more on an equal footing. The goals of the PTD approaches are to: (1)

(2) (3) (4)

find methods to elicit farmers' farming, aquaculture, fishing, gathering, etc. knowledge systems and the constraints these pose to the adoption of aquaculture; transfer results of collaborative on-station research to aquatic farmers; obtain feedback from farmers in order to regularly revise on-station research agendas; identify areas for collaborative research with farmers on their farms, or together on research stations, or both.

Participatory assessment and planning methods Assessment is needed to evaluate objectives and needs, changes in production, and impacts of collaborative experiments. Assessment involves analysis of the evolution

Methods for the Development of Sustainable Aquaculture Ecosystems

109

of farming ecosystems to explain movements towards ± or away from ± sustainability. Conventional assessments are made in economic terms only, and these usually externalize social and environmental costs. A complete assessment methodology must involve not only production but also impacts of innovations on processing, trade, transport, communications, and consumers (Fig. 5.2). Because such a comprehensive assessment can be very costly of time and money, choices have to be made so that a focus is on key issues of direct importance to farmers/fishers. Participatory assessment supports farmers' analyses of factors that directly affect them, and which they can influence. Wider assessments of factors which farmers are unlikely to influence such as institutional factors are left to researchers and policy-makers.

Fig. 5.2 The evolution of aquatic and terrestrial farming ecosystems does not occur in a vacuum, nor do biotechnical approaches completely resolve constraints. Farming ecosystems are influenced by the physical, sociocultural and political environments in which they are located (modified from FAO, 1990).

Conventional scientific research in aquaculture and agriculture tends to have longterm goals, more generic applications, and more methodological rigor. Participatory technology development provides rapid results to site-specific conditions and provides farmers with better tools to sustain the process of adapting to rapid change. This contrasts with the top-down transfer of technology in which the farmer is perceived as a passive receptor of technologies generated elsewhere on research stations (Netting, 1993). The success of PTD is not measured by outcomes since new problems arise continually. The real success of PTD is its ability to build the farmers' capacity to incorporate concepts of natural and social ecology, systems analyses, and ecological

110

Ecological Aquaculture

economics into the farming family. Questions have been raised, however, about the universal applicability, rigor of research findings, and abilities of PTD to make meaningful policy changes (Scarborough et al., 1997). PTD ecosystems diagnoses and ecosystem modeling approaches allow the needed academic rigor and new methods of communication to track the development agenda of change towards sustainability (Roling, 1994; Dalsgaard et al., 1995).

PTD methodologies to develop sustainable aquaculture ecosystems PTD has also been called `participatory learning for action'. PTD uses assessment tools to enable farmers to analyze their own social ecological situation and to develop a common perspective on natural resources management and food production at the local level. PTD empowers farmers by creating a new information database that they need for action. PTD involves: group facilitation methods; methods for interviewing and dialogue; and visualization/diagramming methods (Pretty, 1995). PTD uses a dynamic set of assessment tools applicable to both researchers and extension agents in order to gather information about an area quickly without getting involved in expensive, large-scale, or highly technological (e.g. GIS) surveys. PTD tools are used by these workers who go into a new area to get acquainted with the situation, then return again and again to perform and update previous assessments. Information on all aspects of society is gathered, such as crops, livestock, businesses, social customs, soils, etc. Complementing the resource surveys are separate interviews to complete gender analyses (GAs). GA analyzes and monitors the roles of men and women in the evolution of farming systems development. It focuses on social relationships and division of resources within social units in order to distinguish impacts of agricultural innovations such as aquaculture on gender activities, aspirations, needs and interests (Feldstein & Poats, 1989). These tools are described below. All of them depend on farmer interviews during which the following data are gathered: a target group of farmers is identified; the order of questions is determined by the flow of conversation and not by an artificial list of ordered questions; l questions are asked in the field because farmer answers are more detailed when they can show what is happening directly; l researcher questions focus on farm typology, descriptions of farming and ecological processes and material flows; rationales and difficulties; and key biological and economic parameters. l l

The International Union for the Conservation of Nature (IUCN) uses a participatory approach to assess progress towards sustainability (IUCN, 1997). These involve methods that assess the systems, the farmers, and the collaborations. The following questions are at the heart of this approach:

Methods for the Development of Sustainable Aquaculture Ecosystems

111

What is the condition of the environment and the people? What are the nature of the people and their environment? l What motivates people to do what they do? l What actions are required to improve their situation and that of their environment? l How do people know if things are getting better or worse? l l

PTD uses tools such as aqua/agroecosystem mapping, pictorial modeling, problem diagramming, and system diagramming to obtain information needed to answer the questions posed by IUCN (1997).

Aqua/agroecosystem mapping Farmers routinely classify areas of their land and aquatic holdings as well as crop and fish varieties in unique ways that attempt to decrease risks (Richards, 1986). Aqua/ agroecosystem mapping enables researchers and extension workers to identify suitable farmers, land types and enterprise mixtures that are most likely to benefit from exposure to aquaculture as a new enterprise. Mapping also helps in identifying current aquatic farmers who may benefit from technologies developed by the sustainable aquaculture group of collaborative researchers, extension agents, and interested farmer investigators. Aqua/agroecosystem mapping methodology is based on Lightfoot et al. (1989a, b; 1990, 1991). Maps are produced for an area illustrating topography and hydrology and distribution of agricultural enterprises. Next, line transects of aqua/agroecological zones are drawn to show the entire range of farm/fishing enterprises, soil types, and problems which occur in each zone. A topography and hydrology map is made by eliciting from farmers each land type they distinguish and then placing these on a spatial map that has roads, homes, watercourses and social functions such as churches/temples, etc. (Fig. 5.3). Areas of standing water, directions of drainage and drainage patters and areas that flood are also indicated. Defining boundaries requires a considerable amount of ground-truthing and interviewing. Aqua/agroecosystem transects are constructed by making a composite section through each land type that is defined by the farmer (Fig. 5.4). For each land type, a local name/soil type, crops, trees, livestock and other enterprises are listed.

Pictorial modeling Pictorial modeling has been developed to help farmers/fishers understand the way in which the farmers/fishers manage inputs, outputs and the recycling of materials between various farm/fishing enterprises, and the household. Pictorial modeling has also been shown to aid greatly in technology transfer to new farmers/fishers who are in the process of adopting aquaculture (Noble et al., 1990; Strong & Arrhenius, 1993). Pictorial modeling is a process in which farmers are asked to draw diagrams of their

112

Ecological Aquaculture

Fig. 5.3 Aqua/agroecosystem map and transect indicating topography and hydrology and transportation access. Transects across each land/water habitat the farmer/fisher identifies are constructed with each enterprise system indicated. Problems and opportunities are also noted during the participatory assessment process (from Lightfoot et al., 1989b).

crop, livestock, home and other farm enterprises, and the resource flows between them. During the aqua/agroecosystem mapping exercises, fishers/farmers are asked to identify problems and opportunities in each land or enterprise system type they identify. Pictorial modeling is done as a group activity with several farmers/fishers participating. The exercise enables farmers to visualize their own farming systems as integrated units, and to see where resource links are lacking between enterprises. In addition, farmers/fishers exchange knowledge concerning their activities with others, and illustrate these pictorially (Fig. 5.5). Researchers and extension workers can identify from farmers' diagrams where technological help might improve the overall farming/fishing ecosystem and how aquaculture might assist in enhancing the productivity and efficiency of resource flows (Figs 5.6, 5.7). Pictorial models can also be combined with aqua/agroecosystem maps (Fig. 5.8).

Problem diagramming Problems are identified from farmer interviews undertaken during the aqua/agroecosystem analyses. During these sessions, farmers work with scientists/extension agents in a two-part process to diagram interactions between biophysical causes and socio-economic constraints for each central problem identified (Lightfoot et al.,

Methods for the Development of Sustainable Aquaculture Ecosystems

113

Fig. 5.4 Aqua/agroecosystem transect across an integrated aquaculture ecosystem with multiple farmer enterprises in southern Malawi, Africa. Local names in the indigenous languages are used along with English characterizations to describe natural and cultivated habitat types (from Lightfoot, 1990).

1989b). The first exercise is to construct a problem tree (Fig. 5.9). The objective of this exercise is to fully understand the farmers' complex system of activities associated with a problem, and to identify constraints that may be solved by researchable areas for on-farm and/or on-station experimentation, or socio-economic research (Lightfoot et al., 1989b). The farmers'/fishers' central problem is put in the center of a piece of paper, then they are asked to identify the primary causes ± both biophysical and socio-economic ± for the problem (Fig. 5.9). Next, farmers/fishers are asked the secondary causes of the primary causes and so on until all linkages are established. By the time the exercise is completed, constraints at the tertiary and higher levels are often the greatest opportunities for interactions with scientists' and extension agents' work with farmers/fishers. A result from the field is shown in Fig. 5.10. Aqua/agroecosystem analysis and systems diagnoses of problems help identify farmers' problems on individual farms. The biological and socio-economic constraints underlying each problem are first outlined with farmers/fishers. Combining results across an area can allow researchers and extension agents to identify if certain problems are widespread among farmers/fishers ± systematic constraints ± allowing greater in-depth analyses. Oftentimes, there is a set of underlying reasons for a problem. Once problem diagrams are collected from a number of farmers in an area, problems can be ranked considering the number of farmers affected, the importance of the farm enterprise involved, and the consequent losses of production and income

114

Ecological Aquaculture

Fig. 5.5 Bioresource modeling of products and waste flows from numerous farm enterprises on an integrated aquaculture ecosystem in Uttar Pradesh, India (modified from Lightfoot et al., 1991).

entailed (Fig.5.11). Such an exercise provides an opportunity for generating ideas for intervention points where specific technological assistance may be required. Farmers respond well to these problem analysis sessions because their problems are always on their minds and they enjoy talking to someone who expresses a real interest in their problems. Indeed these sessions are invaluable in that they can find the root causes of persistent and long-term problems to which collaborative solutions can be found. In addition, because farmers are involved in the problem analysis, they are often very motivated to participate in finding and trying out a solution.

Systems diagramming A systems diagram is constructed by placing the farmer's problem in the center of a circle, then surrounding it with primary biophysical and socio-economic constraint circles (Fig. 5.12). The size of the box circles depends on the frequency of responses received as to the weight of the causes.

Methods for the Development of Sustainable Aquaculture Ecosystems

115

Fig. 5.6 Modeling of biological and social resources at the coastal land±water interface. Exchanges of land and sea products, cash and land tenure systems can be analyzed holistically using participatory technology development tools (from Padilla & De Los Angeles, 1992).

The objective of a systems diagram is to order root causes of farmer problems in a simple manner so that opportunities arise to solve the central priority issue across many farmers/fishers and a large area or region. Once common problems/issues are found, fishers/farmers are prompted to suggest what experiments, ideas or experience they have to solve the issue, and scientists can suggest what technologies or new management options are available. Such interaction leads farmers and research directly into the experimental design for new collaborative research.

The role of the aquaculture research station The overall objective of research in aquaculture is to evolve sustainable farming systems (economically, environmentally, and socially sustainable). PTD methodologies are used due to their low costs, simplicity, and sustainability over the long term with few external subsidies. Adoption of low cost PTD methodologies could save funds, which could be devoted to accelerate research needed to evolve economically viable ecologically sustainable aquaculture ecosystems. PTD has not only the capability of fundamentally revising the means by which researchers and

116

Ecological Aquaculture

Fig. 5.7 Incorporating aquaculture into traditional farming systems offers farmers more options for sustainability and income generation, turning exploitative land uses into sustainable, regenerative integrated aquaculture±agriculture farming ecosystems (from Lightfoot, 1990).

Fig. 5.8 Bioresource flows shown on an aqua/agroecosystem transect of an integrated aquaculture ecosystem in Uttar Pradesh, India (modified from Lightfoot et al., 1991).

Methods for the Development of Sustainable Aquaculture Ecosystems

117

Fig. 5.9 A problem tree for diagramming problems and causes on individual farms and fishing enterprises. Farmers/fishers are asked about the main problems, primary and secondary causes, and causal linkages to the problems. Problems are separated into biophysical and socio-economic problems (from Lightfoot et al., 1990).

extension agents relate to farmers, but also of revising the way research agendas and priorities are generated and funded. Connell (1990) defined an approach to participatory extension, which he termed `minimalist'. The objective of this approach is to `deliver sets of viable technologies to farmers in diverse environments'. The minimalist approach is to offer farmers a `supermarket' (Conway, 1986) or `basket' (Chambers et al., 1989) of technologies at the research station; then to allow farmers sufficient `creative distance' to evaluate the applicability of the technologies to their farming systems; then to encourage farmers to identify areas for collaborative on-farm research. The research station is set up with a continually changing set (the `basket') of technologies that may or may not be of interest to farmers but have been developed collaboratively by researchers and farmers/fishers using PTD methods (Fig. 5.13). Farmers/fishers are invited on a regular basis for `open days' to witness these technologies. Farmers/fishers are encouraged to criticize the demonstrations and comment on better methods to increase production and farm sustainability. Farmers/ fishers are encouraged to return to the research station with ideas, further criticism, or proposals to work with researchers on their farms/boats. Combination of agroecosystem analysis, pictorial modeling, and systems diagnosis

118

Ecological Aquaculture

Fig. 5.10 Example of a problem tree for an integrated aquaculture ecosystem identifying `slender fish' as the main problem in the aquaculture subsystem of a farm in southern Malawi, Africa. Primary and secondary causes are indicated. `Striga' is a local word for rooted aquatic plants. Complete identification of the relationships to central problems creates unique opportunities for extension and research interventions (from R. Noble, unpublished data).

of farmers' problems helps in development of appropriate technology, assists in developing a farmer/fisher-centered research agenda on agriculture/aquaculture experiment stations or research institutes, and facilitates more rapid, efficient, and lower cost transfer of new information to farmers/fishers (Lightfoot & Noble, 1993).

Defining sustainable, ecological aquaculture The Brundtland Commission (WCED, 1987) defined sustainable development as `the ability to meet the needs of the present without compromising the ability of the future generations to meet their own needs.' Some forms of intensive aquaculture development have caused severe social and environmental impacts (Pullin et al., 1993; Black, 2000). The degraded state of most aquatic ecosystems combined with public concerns about adding any `new' sources of aquatic pollution to already overburdened ecosystems will require aquaculture to develop new, ecosystems approaches and sustainable operating procedures; and to articulate a sustainable, ecological pedagogy. In the twenty-first century, aquaculture developers will need to spend as much time on technological advances coming to the field as they do in designing ecological approaches to aquaculture development that clearly exhibit stewardship of the environment. An alternative model of aquaculture development called `ecological aquaculture' is

Methods for the Development of Sustainable Aquaculture Ecosystems

119

Fig. 5.11 Problem ranking table summarizing common problems for groups of farmers/fishers across a region or watershed. Farmers are asked to rank problems they identify as systemic across their enterprises. Problems are ranked using three criteria: (a) the distribution of the problem, (b) the importance of the enterprise to the farming system, and (c) the loss of income for which the problem is responsible (from Lightfoot et al., 1990).

needed that not only brings the technical aspects of ecological principles and systems thinking to aquaculture, but incorporates ± at the outset ± principles of natural and social ecology, planning for community development, and concerns for the wider social, economic, and environmental contexts of aquaculture. Sustainable, ecological aquaculture is the development of aquatic farming systems that preserve and enhance the forms and functions of the natural and social environments in which they are situated. Such a vision incorporates the following ideas about ecological aquaculture systems. First, ecological aquaculture systems are planned using systems ecology approaches that develop aquaculture production networks for various species in a highly diversified, segmented manner, with numerous interconnections that supply inputs and outputs using: (a) local resources, (b) recycled wastes and materials and closing

120

Ecological Aquaculture

Fig. 5.12 A farming ecosystems diagram that links primary and secondary constraints to productivity with biophysical and socio-economic causes. The systems diagram is constructed by arranging primary and secondary constraints into circles surrounding the central problem with biophysical causes on the right and socio-economic constraints on the left. The size of each circle is determined by the number of responses obtained at a group meeting (from Lightfoot et al., 1990).

leaky loops of energy and materials that can potentially degrade natural ecosystems, and (c) planning for maximal job creation and local markets (Costa-Pierce, 1992). Aquaculture depends upon inputs from various food, processing and transportation industries and produces valuable wastewaters, manures and fish wastes, all of which can be a vital part of an ecological system that can be planned and organized for community-based ecosystem rehabilitation, reclamation and enhancement ± not degradation. Secondly, ecological aquaculture treats and recycles its own wastes ± rather than relying upon public subsidies ± and integrates people with technologies in new synergies to create new employment and biotechnical advances with earth-centered knowledge (information science) and appropriate technologies (wind, solar and biomass). Transitions to a sustainable, ecological aquaculture will require a move-

Methods for the Development of Sustainable Aquaculture Ecosystems

121

Fig. 5.13 Participatory technology development (PTD) methodologies help researchers and extension workers define new collaborative research opportunities both on research stations and on-farm/in the field. As a result, the design of proper components to merge in assembling aquaculture ecosystems is relevant to the local farming/fishing situation and the local ecosystems. This integrated aquaculture±silviculture± agriculture ecosystem was implemented on a research station with farmers and eventually spread widely in southern Malawi, Africa (Noble & Costa-Pierce, 1992).

ment from the sewage treatment and assimilative concepts of waste management, towards the concepts of input management and integrated waste treatment technologies. Lastly, ecological aquaculture takes a global view, integrating ecological science and sharing technological information with innovation in the global marketplace, avoiding the proprietary. It is community-based, having positive societal impacts, and is analyzed holistically, incorporating social and environmental costs, not externalizing them. Some characteristics of ecological aquaculture are that it: l l l l l l l l l l l

preserves the form and functions of natural ecosystems; derives most of its energy from renewable sources (solar, wind, water, biomass); relies on waste animal- or plant-based proteins for feeds, decreasing dramatically the use of fish meals and oils; does not produce nutrient or chemical pollution or antibiotics harmful to human or ecosystem health; develops a systems approach to nutrient recycling and regeneration; plans for ecosystem rehabilitation and enhancement; is integrated with agriculture and capture fisheries; uses native or resident species, or has complete containment procedures in place; is integrated with communities to maximize job creation in local industries; produces organic or ecolabelled products using sustainable farming practices; is a global partner, producing information for the world.

Clear, unambiguous linkages between aquaculture and the environment must be created and fostered, and the complementary roles of aquaculture in contributing to environmental sustainability, rehabilitation and enhancement must be developed and

122

Ecological Aquaculture

clearly articulated to a highly concerned, increasingly educated and involved public. Planning for aquaculture development as community development must include planning for aquaculture's vital support industries and for the reuse of aquaculture wastes in agriculture or in environmental enhancement projects. Ecological aquaculture will create new opportunities for a wider group of professionals to get involved in aquaculture since new advances will be needed not only in technology but also in information, community development and facilitation.

Conclusions PTD methodologies are low-cost and sustainable. They put the farmer/fisher first in terms of developing technology appropriate for the evolution of sustainable aquaculture ecosystems. Aqua/agroecosystem analyses, pictorial modeling, problem diagramming, and systems diagnoses are important tools that recognize social and farming systems diversity, and the fact that no single aquaculture technology is universally applicable. There are many aqua/agroecological `niches' and farmers must be allowed to adopt and modify technologies as they see fit to suit their particular circumstances. In farming and fishing systems PTD approaches, the natural and social ecological sciences come together in a very unique form of interdisciplinary environmental scholarship. Fixed answers derived from agriculture/aquaculture experiment stations and passed down to farmers in a top-down manner are too inflexible. These approaches cannot solve the problems of sustainability which require more sitespecific, integrated, social and ecological methodologies. Farmers must be able to choose from a basket of technologies and management approaches. With innovative technologies such as aquaculture, farmers often know as much as researchers ± making the reversal of traditional extension roles a given which should be embraced as a set of unique opportunities for collaborative work, not challenged or fought. By initiating a new era of cooperative research with farmers, a more detailed and very intimate knowledge of the natural and social ecology of farming systems can be combined with useful scientific knowledge to evolve sustainable aquaculture ecosystems.

References Alteri, M. (1987) Agroecology: The Scientific Basis of Alternative Agriculture. Intermediate Technology Publications, London. Barinaga, M. (1990) Fish, money, and science in Puget Sound. Science, 247, 631. Black, K.D. (2000) Environmental Impacts of Aquaculture. CRC Press, Boca Raton, FL. Borgese, E.M. (1980) Seafarm, The Story of Aquaculture. H.N. Abrams, New York. Chambers, R., Pacey, A. & Thrupp, L.A. (1989) Farmer First: Farmer Innovation and Agricultural Research. Intermediate Technology Publications, London. Connell, J. (1990) A `minimalist' approach to participatory extension. Paper presented at the

Methods for the Development of Sustainable Aquaculture Ecosystems

123

Asian Farming Systems Research and Extension Symposium, Bangkok, Thailand, 19±22 November 1990. Conway, G. (1986) Agroecosystem Analysis for Research and Development. Winrock International, Bangkok, Thailand. Costa-Pierce, B.A. (1992) Rice±fish systems as intensive nurseries. In: Rice±Fish Research and Development in Asia (eds C.R. de la Cruz, C. Lightfoot, B.A. Costa-Pierce, V.R. Carangal & M.P. Bimbao), pp. 117±130. ICLARM, Manila, Philippines. Costa-Pierce, B.A. (1996) Environmental impacts of nutrients from aquaculture: towards the evolution of sustainable aquaculture systems. In: Aquaculture and Water Resource Management (eds D.J. Baird, M.C.M. Beveridge, L.A. Kelly & J.F. Muir), pp. 81±113. Blackwell Science, Oxford. Costa-Pierce, B.A. (1998) Constraints to the sustainability of cage aquaculture for resettlement from hydropower dams in Asia: an Indonesian case study. Journal of Environment and Development, 7, 333±363. Dalsgaard, J.P.T., Lightfoot, C. & Christensen, V. (1995) Towards quantification of ecological sustainability in farming systems analysis. Ecological Engineering, 4, 181±189. Dent, J.B. & Anderson, J. (1971) Systems, management and agriculture. In: Systems Analysis in Agricultural Management (eds J. Dent & J. Anderson), pp. 3±14. Wiley, New York. Edwards, P. (1993) Environmental issues in integrated agriculture±aquaculture and wastewater-fed fish culture systems. In: Environment and Aquaculture in Developing Countries (eds R. Pullin, H. Rosenthal & J. Maclean), pp. 139±170. ICLARM, Manila, Philippines. FAO (Food and Agriculture Organization of the United Nations) (1990) Farming Systems Development. FAO, Rome. Feldstein, H.S. & Poats, S.V. (1989) Working Together: Gender Analysis in Agriculture. Vol. 1. Case Studies; Vol. 2. Teaching Notes. Kumarian Press, West Hartford, CT. Folke, C. & Kautsky, N. (1991) Ecological economic principles for aquaculture development. In: Nutritional Strategies and Aquaculture Wastes (eds C.B. Cowey & C.Y. Cho), pp. 207± 222. University of Guelph, Ontario, Canada. Gadgil, M. & Guha, R. (1992) The Fissured Land: An Ecological History of India. Oxford University Press, India. Gliessman, S. (1998) Agroecology: Ecological Processes in Sustainable Agriculture. Ann Arbor Press, Chelsea, MI. IUCN (International Union for the Conservation of Nature) (1997) An Approach to Assessing Progress Toward Sustainability. Tools and Training Series. IUCN, Gland, Switzerland. Kansing, S. & Kremer, J. (1995) A sociological analysis of Balinese water temples. In: The Cultural Dimension of Development: Indigenous Knowledge Systems (eds D. Warren, L. Slikkerveer & D. Brokensha), pp. 258±269. Intermediate Technology Publications, London. Lightfoot, C.W. (1990) Integration of aquaculture and agriculture: a route to sustainable farming systems. Naga, The ICLARM Quarterly, 13, 9±12. Lightfoot, C., Bottrall, A., Axinn, N., Conway, G. & Singh, P. (1989a) Training Resource Book for Agro-Ecosystem Mapping. International Rice Research Institute, Manila, Philippines & Ford Foundation, New Delhi, India. Lightfoot, C., Singh, V.P., Paris, T., Mishra, P. & Salman, A. (1989b) Training Resource Book for Farming Systems Diagnosis. ICLARM Contribution No. 581. International Center for Living Aquatic Resources Management (ICLARM), Manila, Philippines.

124

Ecological Aquaculture

Lightfoot, C., Singh, V.P., Paris, T., Mishra, P. & Salman, A. (1990) Training Resource Book for Farming Systems Diagnosis. International Rice Research Institute, Manila, and International Center for Living Aquatic Resources Management, Manila, Philippines. Lightfoot, C., Axinn, N., John, K.C. et al. (1991) Training Resource Book for Participatory Experimental Design. ICLARM/IRRI, ICLARM Contribution No. 764. Narendra Dev University of Agriculture and Technology, Uttar Pradesh, India; International Center for Living Aquatic Resources Management, Manila, Philippines, International Rice Research Institute (IRRI), Manila. Lightfoot, C. & Noble, R.P. (1993) A participatory experiment in sustainable agriculture. Journal of Farming Systems Research and Extension, 4, 11±34. McCorkle, C. (1994) Farmer Innovation in Niger. Studies in Technology and Social Change 21. Iowa State University Center, Ames, IA. Netting, R.Mc. (1993) Smallholders, Householders: Farm Families and the Ecology of Intensive, Sustainable Agriculture. Stanford University Press, Stanford, CA. Noble, R., Lightfoot, C. & Bage, J. (1990) Training notes to accompany a video on pictorial modeling: a farmer-participatory method for modeling bio-resource flows in farming systems. ICLARM Contribution No. 662. ICLARM, Manila, Philippines. Noble, R. & Costa-Pierce, B. (1992) Aquaculture technology research for smallholder farmers in rural Malawi. In: Aquaculture and Schistosomiasis (ed. National Research Council), pp. 11±25. National Academy Press, Washington, DC. Odum, E.P. (1975) Ecology. Holt-Saunders, New York. Odum, E.P. (1989) Input management production systems. Science, 243, 177±181. Padilla, J.E. & De Los Angeles, M.S. (1992) Economic policies and the sustainable development of coastal resources in the Philippines. Naga, The ICLARM Quarterly, 15, 36±38. Pretty, J.N. (1995) Regenerating Agriculture: Policies and Practice for Sustainability and SelfReliance. Earthscan Press, London. Pullin, R., Rosenthal, H. & Maclean, J.L. (1993) Environment and Aquaculture in Developing Countries. ICLARM Conference Proceedings 31. ICLARM, Manila, Philippines. Richards, P. (1986) Coping with Hunger: Hazard and Experiment in an African Rice Farming System. Allen and Unwin, London. Roling, N. (1994) Platforms for decision-making about ecosystems. In: Future of the Land: Mobilizing and Integrating Knowledge for Land Use Options (eds L.O. Fresco, L. Stroosnijder, J. Bouma & H. van Keulen), pp. 385±393. Wiley, New York. Roling, N.G. & Jiggens, J. (1998) The ecological knowledge system. In: Facilitating Sustainable Agriculture: Participatory Learning and Adaptive Management in Times of Environmental Uncertainty (eds N.G. Roling & M. Wagemakers), pp. 105±125. Cambridge University Press, Cambridge. Scarborough, V., Killough, S., Johnson, D. & Farrington, J. (1997) Farmer±led Extension: Concepts and Practices. Intermediate Technology Press, London. Strong, M. & Arrhenius, C. (1993) Closing linear flows of carbon through a sectoral society: diagnosis and implementation. Ambio, 22, 414±416. Veldhuizen, L. (1998) Principles and strategies of participation and cooperation: challenges for the coming decade. Advances in Geoecology, 31, 979±983. WCED (World Commission on Environment and Development) (1987) Our Common Future. Oxford University Press, Oxford.

Chapter 6

A Market-driven, Social Ecological Approach to Planning for Sustainable Aquaculture: A Case Study of Tilapia in Fiji Barry A. Costa-Pierce Rhode Island Sea Grant College Program Introduction The tilapias have been exported from their native ranges in sub-Saharan Africa to Asia and onwards throughout the tropical world from the Second World War period until the mid-1970s. Tilapia were sent not only for aquaculture development but also for aquatic weed and insect control purposes (Costa-Pierce & Doyle, 1997). This `tilapia craze' is similar to the `carp craze' of the nineteenth century when biologists and fisheries managers seeded the world's aquatic ecosystems with exotic carp species in an attempt to improve them (Cowx, 1997). As a result of the spreading of tilapia, the most widely distributed and established fish worldwide is the Mozambique tilapia (Oreochromis mossambicus) (Pullin & Lowe-McConnell, 1982). O. mossambicus is native to the eastward-flowing rivers of southern Africa from the Shire River in Malawi in the north to almost the tip of the African continent in the south (Trewavas, 1983). The exact origin of O. mossambicus exported from Africa throughout the world is unknown, but likely came from southeastern Africa (Agustin et al., 1997). For many years O. mossambicus was known worldwide as the `Java tilapia'. It acquired this name because a founder stock taken from coastal ponds in Java to a fisheries station in Malacca, Malaysia (then Malaya), was then exported from Malaysia throughout the world where it became firmly established in almost 100 countries (Pullin, 1985). How and when the tilapia got from Africa to Java is shrouded in mystery; but speculation is that the fish was likely an escapee, or was a purposeful `dumping' into the Indonesian environment by an aquarist. Since O. mossambicus was taken from an unknown number of parents and was seeded around the world for decades with no attention to genetic issues, the stocks have deteriorated almost everywhere the fish has escaped. Agustin et al. (1997) found a low amount of genetic diversity in a feral O. mossambicus population compared with wild fish and explained this by a small founder population size resulting in a genetic bottleneck. As a result of genetic deterioration, small-sized, poor quality O. mossambicus lost consumer acceptance, became pests, and acquired a reputation as `trash fish' in many countries (Pullin, 1985). O. mossambicus also impacted negatively many native aquatic ecosystems due to its aggressive, changeable, and non-specific feeding habits; its wide salinity and water quality tolerances (Costa-Pierce & Riedel,

126

Ecological Aquaculture

2000); and its precocious breeding behavior which allowed it to overpopulate many aquatic environments and outcompete native fish. In some locales the fish ± which is known from the wild as feeding on algae and detritus (Bowen, 1987) ± became a predator (on milkfish fry in Kiribati: Lobel, 1980). As a result of these undesirable traits, the Mozambique tilapia has been replaced in nearly all of the major tilapia farming countries by the Nile tilapia (Oreochromis niloticus). The global consensus that the Nile tilapia (O. niloticus) is the best species for aquaculture development is due to at least three important commercial characteristics: (1) (2) (3)

its faster growth rate to larger maximum sizes than other species; its larger size at first reproduction; and its grazing feeding habits and lower trophic position on the aquatic food web.

These characteristics make the Nile tilapia more productive, and cheaper to grow and with less environmental impact than the Mozambique tilapia. However, O. mossambicus is not a pest everywhere. In the impoverished Asian nations after the Second World War (especially Indonesia), the fish has been thought to have saved millions of poor people from starvation and protein malnourishment (DeSilva & Senaratne, 1988). In modern Sri Lanka and Indonesia, the Mozambique tilapia is a prized protein source for the rural poor. And in California, the Mozambique tilapia is grown profitably in intensive farms due to regulations prohibiting importation of new tilapia species such as the preferred Nile tilapia (Zachritz & Rafferty, 2000). It may be premature to label the Mozambique tilapia as inferior to the Nile tilapia for aquaculture development. Wild stocks perform well and show a wide genetic diversity. With proper importations and genetic improvement programs using superior founder stocks the fish can perform well in culture. In California, USA, the Mozambique tilapia has growth rates and yields equal to or better than reported in studies of Nile tilapia (Costa-Pierce, 1997).

The Mozambique tilapia in Fiji The Mozambique tilapia were first introduced to Fiji in the late 1940s from an unknown source. They were introduced again in the mid-1950s as a human food source and as an alternative protein source for pigs (Holmes, 1954). O. mossambicus is called maleya by the local people in Fiji, representing their origin in colonial Malaysia (Malaya). All tilapia stocks imported to Fiji since the mid-1960s have been made with the purpose of developing the potential of tilapia aquaculture. O. mossambicus were stocked during the colonial era into the Singatoka, Rewa and Navua Rivers on Viti Levu, and into rivers on Vanua Levu (Table 6.1). Fiji has large rivers and abundant water resources. It is surprising to see a relatively small island like Viti Levu with a total area of only 10 429 km2 having a number of `continental-size' rivers as long as 260 km, with a few of these over 200 m wide. As a

Planning for Sustainable Aquaculture

127

Table 6.1 Recorded introductions of tilapias to Fiji (from Holmes, 1954; Rawlinson, 1994; S. Lal, personal communication) Species

Country of origin

Year of introduction

O. O. O. O. O. O. O. O.

Unknown Malaysia Malaysia (`Israeli') Taiwan Taiwan Israel Thailand (`Chitrilada') Israel (`ND59')

Unknown, late 1940s 1954 Late 1960s/early 1970s 1980±86 1980±86 1979 1988 1997

mossambicus mossambicus niloticus urolepis hornorum aureus niloticus niloticus niloticus

result, the Mozambique tilapia has flourished in the wild in Fiji, and there is a vibrant and very important artisanal river fishery for tilapia. In addition, the Nile tilapia are stocked in rivers by the Fiji Fisheries Department (FFD). In 1997, the FFD stocked about 76 000 tilapia into rivers (Lal, personal communication, 1997). There are no studies on results of interactions of the Nile and Mozambique tilapias in Fijian rivers, but hybridization of the two species has certainly occurred. Since tilapia have been in Fiji for so long, local people have become accustomed to its taste and appearance (Richards, 1994). Tilapia are one of the most important protein sources for the rural people (along with kai, freshwater mussels [Batissa violacea]). The importance of tilapia river fisheries to the people of Viti Levu has been documented as part of a larger survey of rural subsistence and artisanal fisheries conducted by the FFD and the Commonwealth Scientific and Industrial Research Organization (CSIRO) (Rawlinson, 1994). The United States Peace Corps has been assisting the Government of the Republic of Fiji to develop the rural aquaculture of tilapia since 1982. In 1993, the first commercial tilapia farm was opened near Suva. The Fijian Government views aquaculture as a subsistence protein source for rural communities and an alternative cash crop to kava, coca, vanilla, etc. Tilapia aquaculture in rural areas is well established with a recorded 172 smallholder tilapia farms. National production in 1996 was estimated at 122 metric tons (mt), with smallholder farms producing 74 mt and a semi-commercial sector producing 48 mt, a substantial increase from 1995 when national production was estimated at 68 mt (FFD, 1996). Gross value increased from F$204 000 in 1995 to F$366 000 in 1996 (US$1.40/F$1.00). Most tilapia aquaculture production is centered on Viti Levu Island in Naitasiri, Namosi, Ra and Tailevu provinces. In 1996, 18 new farms started, including a government-run commercial agriculture corporation (Viti Corp.) that opened with the intent of becoming an industry leader in the production of exportable aquaculture products. To support the growth of aquaculture, Fiji has adopted a new `Commodity Development Framework' (CDF). The CDF intends to assist development of a `fully fledged, export-driven agro-industry base' (Ministry of Agriculture, Fisheries, Forests and ALTA [MAFFA], 1996). Under the CDF, the Fijian Government intends to invest F$2.75 million and attract F$1.00 million in private sector money to establish a tilapia aquaculture industry that, it is planned, will produce 1000 mt/year.

128

Ecological Aquaculture

The CDF will assist development of 10 new commercial farms on Viti Levu, three industrial-size tilapia farms in Vanua Levu, and two industrial farms, each vertically integrated with feed mills and processing facilities. Development of aquaculture is expected to save Fiji F$22 million in imports. The objectives of the Fijian Government's aquaculture development are similar to many around the world, so can be used as a case study of the promotion of a new approach to aquaculture development, namely, a market-driven approach to sustainable, ecological aquaculture. In order to do this, I weighed the social, cultural, and economic aspects of expanding tilapia aquaculture production in Fiji; then assessed the current and future demands (volume and price) for domestically produced tilapia in order to make recommendations for technological and development needs ± based upon market information ± for sustainable tilapia aquaculture in Fiji.

Materials and methods A fish marketing survey instrument was designed that was modified from McCoy & Hopkins (1978). Interviews with fish sellers focused on (a) the amount of reef fish in domestic markets, (b) seasonality of supply/demand and price, and (c) price±size relationships of demand. In addition, secondary data were acquired from the Fiji Fisheries Department (FFD) in unpublished internal reports detailing the market structure and additional fisheries marketing data on domestic volumes, prices, and seasonalities. The survey instrument was discussed at a workshop of six Peace Corps and four Ministry of Agriculture, Fisheries and Forests and ALTA (MAFFA) staff, then was tested at the Suva Municipal Market, and finally revised for use. Interviews of fish sellers were conducted by eight Peace Corps staff over an eight-day period in Suva, Nabukalou Creek, Navua, Nausori areas of the Central Region of Viti Levu Island. Municipal markets, roadside stalls, grocers, butchers, and middlemen were interviewed. A total of 31 surveys were completed, comprising 17 fish producers/sellers, eight butchers, four roadside stalls, and two `up market' fish stores.

Results and discussion Fiji domestic fish market structure There was a dramatic change in the domestic marketing structure of seafood in Fiji from 1978 to 1995 (FFD, 1996). In 1978, half of the seafoods sold in Fiji passed through municipal markets and the other half to `outlets' (butchers, supermarkets, roadside stalls, shops, hotels, restaurants, cafes, etc.). By 1995, only 13% of the seafoods were sold at municipal markets; 87% was marketed in the `outlets' (FFD, 1996). The reasons for the change were increased taxes and rental rates on stalls in municipal marketplaces and the growing shortage of fish in coastal areas so that there was no longer any surplus fish to sell in municipal markets. All fish caught was reported to be `staying in the villages'.

Planning for Sustainable Aquaculture

129

Fish consumption by Fijians The majority of Fijians live in inland areas. As a consequence, inland capture fisheries are important sources of subsistence protein. Rawlinson (1994) found that rivers and estuaries were the most important fishing areas on Viti Levu, and that Mozambique tilapia was the principal target species of 22% of 1734 fishing trips monitored (Table 6.2). Over 80% of the inland people surveyed reported targeting Mozambique tilapia for their fishing efforts in estuarine and riverine areas (Rawlinson, 1994). Catching of tilapia in rivers for both subsistence and for sale was dominated by women. Women fishers caught tilapia at rates as high as 12.7±13.9 kg/hour (Rawlinson, 1994). Table 6.2 Frequency of targeted species by fishers on 1734 fishing trips in rivers and estuaries in Viti Levu, Fiji (from Rawlinson, 1994) Species targeted Oreochromis mossambicus Anguilla spp. Kuhlia rupestris Freshwater prawns Lutjanus argentimaculatus Eleotris melanosoma Misc. Carangidae Batissa violacea Leiognathus equulus Palaemon concinnus 37+ other species

Number of fishing trips

Per cent of trips

384 257 173 173 125 136 58 45 37 34 312

22 15 10 10 7 8 3 3 2 2 18

Fish are the most important animal protein sources for Fijians. Studies by Vuki (1991) and Zann (unpublished data cited in Rawlinson, 1994) indicated an annual per capita fish consumption rate of about 40 kg. FFD (1995) stated that the annual fish consumption rate was 50 kg per capita; however, FFD (1995) rates measured fish supply per capita, not rates of fish consumption. Direct fish consumption by Fijians from all ethnic groups in 150 coastal villages at 613 village meals on Viti Levu was studied by Rawlinson (1994). Actual rates of fish consumed averaged 68.2 kg/capita/year. This per capita fish consumption rate is the highest in the world, similar to that of the Japanese, who are reported to consume 68±70 kg/capita/year (New, 1997). A strong weekly pattern in fish consumption exists on Viti Levu. Rawlinson (1994) measured rates exceeding 1.52 kg/ person/meal/day, with a high on a Sunday in one village of 2.8 kg/person/meal/day (Fig. 6.1). The weekly pattern of fish consumption was a repeatable trend in their data, with highest rates of fish consumption always on Sundays. Sundays are important religious holidays, and large family gatherings with generous amounts of fish are consumed on these days. Fishers reported fishing more actively on Fridays and Saturdays to ensure that adequate quantities of fish were available on Sundays (Rawlinson, 1994).

130

Ecological Aquaculture

Fig. 6.1 A strong weekly pattern of fish consumption exists on Viti Levu. Highest rates of fish consumption occur on Sunday, the day of important religious events when large family gatherings consume generous amounts of fish (from Rawlinson, 1994). n Fish consumption in village 1; . Fish consumption in village 2.

Rawlinson (1994) noted that their surveys showed fish supply per capita per year followed closely the estimates of Zann (41.2 vs 40 kg/capita/year reported by Zann). As a result, Rawlinson (1994) estimated a gap between fish supply and fish consumption on Viti Levu at 27 kg/capita/year (68.2741.2 kg/capita/year). They stated that, contrary to widely held opinions that Fijians are self-sufficient in rural fish supplies, villagers were purchasing this fish from markets to make up for the shortfall from their own subsistence catches. While fish purchases are not as important quantitatively as their own catches, it was surprising to see a large amount of fish being bought by the average rural person.

Planning for Sustainable Aquaculture

131

Fish preferences by Fijian ethnic groups Rawlinson (1994) measured the forms of fish eaten at 939 meals on Viti Levu and found that fresh fish were consumed at 549 of these (58%) (Table 6.3). Fijian consumers preferred fresh fish, especially reef fish. The importance of tinned (canned) fish increased inland. Important reef fish species identified as preferred by the Fijian (and Indian) consumers were: kabatia (snappers, Lethrinidae), kake (snappers, Lutjanidae), l ki (small goatfish, Mullidae), l senekawakawa (small groupers, Serranidae), and l kanace (mullet, Mugilidae). l l

Table 6.3 Preference of form of fish eaten at 939 village meals in Viti Levu, Fiji (from Rawlinson, 1994) Form of fish eaten Don't eat fish Fresh fish Tinned fish Totals

Number of meals

Per cent total

330 549 60 939

35 58 6 100

Sizes of reef fish consumed by rural Fijians were small, ranging from 60 to 815 g average weight; but a few of the preferred species were as small as 100±300 g (Table 6.4). During market surveys we witnessed piles of fish labeled `small fish' which were reef fish of 100±300 g average weight being sold at F$4 per kg. These reef fish are the approximate size of tilapia harvested from ponds at the present time in Fiji, and some of them resembled tilapia quite strongly in terms of size and color. Indian consumers comprised two groups, Hindus and Muslims. Vegetarian Hindus, estimated to be about 20% of the Indian population in one interview, do not eat any fish or animal meat products. Muslims were said to prefer fish that does not have Table 6.4 Preference for size of reef fish eaten in two coastal villages in Viti Levu, Fiji (from Rawlinson, 1994) Species of fish

Common names

Lethrinus harak L. mahsena Small Mullidae Epinephelus spp. Mugil spp. Gerres spp. Hemirhamphus spp. Lutjanus spp.

Thumbprint emperor Yellowtailed emperor Goatfishes Groupers Sea mullets Silver body Barred garfish Snappers

Weighs at meals (wet weights, g) 141±261 29±815 86±131 86±131 60±70 75±100 49±421 61±207

132

Ecological Aquaculture

a strong odor or fishy (oily) taste. The small Chinese and Southeast Asian population has a preference for red fish and for live fish. The preferred fish for Chinese was said to be donu (coral trout, Plectroponus sp.). A Chinese seller has pioneered development of a live tilapia market in Suva. Market tests with live tilapia have also been conducted by the FFD at Nausori and Suva (Lal, 1991). Live fish were purchased by all ethnic groups. Indian and Chinese customers preferred plate-sized fish (250±300 g), while no strong size preferences were noted for indigenous Fijians (Lal, 1991). Since tilapia has been in the country for so long all ethnic groups are familiar with it and have become accustomed to the taste and appearance (Richards, 1994). As a result, Fijians have quickly accepted a farmed tilapia. Richards (1994) reported that `batches of 200±300 tilapia sell quickly in retail markets for approximately F$3.00± 4.00 per kg, a price comparable to or slightly higher than the price paid for several species of reef fish'. They also reported that, before 1989, tilapia sold in municipal markets did not exceed 6 mt per year, but this increased rapidly to 20 mt in 1990, then increased further to 72 mt in 1992.

Estimates of fish demand Data from 23 seller interviews is summarized in Table 6.5. Only 18 sellers responded to questions about sales changes over a 1-year period; and only 14 responded adequately about changes they had seen in volumes of fish over the past 5 years. Table 6.5 Changes in fish volumes marketed in Viti Levu, Fiji, from seller interviews in 1997

Number of respondents Median Totals

Weekly sales 7-1992 (kg/week)

1-year volume change (kg/week)

5-Year volume change (kg/week)

Species sold

23

23

23

18

1±11

18±2250

68±3805

60±8830 7852 to 77 75700 to 263

1253

1368

2039

Range 5

5

Weekly sales 7-1997 (kg/week)

Weekly sales 7-1996 (kg/week)

Years in business

14

18

14

740

7100

72790 (145 mt)

76409 (333 mt)

Sellers interviewed had been in the business a median of 5 years, but some had spent 20 years selling fish. Most sold five species of fish. Fourteen of the 18 sellers interviewed reported that their volume of sales of `tilapia-like' reef fish had dropped in the last year, with a median loss of 40 kg/seller. Ten out of 14 reported decreases in the volume of this type of reef fish sold over the past 5 years, with a median loss of 100 kg/ seller. The total amount of reef fish similar to tilapia sold was estimated to have declined by 2790 kg/week for 18 sellers in the 1-year period from 1996 to 1997. Fourteen sellers reported that during the period from 1992 to 1997 the total amount of tilapia-like reef fish they sold was estimated to have declined by 6409 kg/week. Taking this latter estimate of 6409 kg/week of fish `missing from the market' and

Planning for Sustainable Aquaculture

133

extrapolating to 52 weeks (a year), 333 mt less fish are available on the market in 1997 than in 1992, or 67 mt/year. Decreased volumes of fish sales could be due to a variety of factors such as change in the market structure away from the municipal Suva region (where these interviews were conducted) towards roadside stalls; a lack of fish supply due to resource constraints; or due to inaccurate survey data. A search for secondary data on fish volumes and prices was therefore undertaken to confirm/deny the primary data gathered during this study from seller interviews. The FFD has monitored fish volumes and prices in municipal and non-municipal markets since the mid-1980s. A search of their computer database (called `Walu') was accomplished to compile records on changes in fish volumes in both municipal and non-municipal markets and a representative roadside market, all in the critical Central Region of Viti Levu where the primary data from seller interviews was generated. Secondary data was gathered on native Fijian reef fish that resembled tilapia strongly both in appearance and size, e.g. the tilapia could be a `replacement species' for reef fish. It was hypothesized that, because of the similar sizes and appearances of tilapia to the reef fish preferred by consumers, changes over time in volumes of these fish sold could be an important indicator of the total volume of demand. Total fish sales showed no change over the 8-year period (4710 mt in 1987 to 4698 mt in 1995). Most fish (87%) were marketed in informal areas (e.g. roadsides) (Table 6.6). However, large decreases in volume of the most preferred reef fish species occurred: from 966 mt in 1987 to 439 mt in 1995 (Table 6.6). The `missing' volume amounts to 527 mt over the period from 1987 to 1995, or 66 mt/year. Table 6.6 Market volume changes for all fish and selected reef fish species, 1987±95 (metric tons) 1987 market

1987 nonmarket

Total available in 1987

1995 market

1995 nonmarket

Total available in 1995

Total change

Total all fish

969.9

3740.3 (79%)

4710.2

590.2

4108.1 (87%)

4698.3

711.9

Total 5 reef species

172.9

793.0 (82%)

965.9

91.1

348.1 (79%)

439.2

7526.7

Notes: Data from 1987 and 1995 reports of the Fiji Fisheries Department on the `Walu' database and in their annual reports. Five reef fish are: kabatia (Lethrinidae, small snappers); kake (Lutjanidae, small snappers); ki (Mullidae, small goatfish); kanace (Mugilidae, small mullet); senikawakawa (Serranidae, small groupers).

The `missing' volumes of tilapia-like reef fish from both the primary data gathered at Fiji's municipal markets, and the secondary data on tilapia-like reef fish from the FFD's database on all markets, are in agreement. There is approximately 66±67 mt/year loss of fish from Fijian fish markets over a 10-year period from 1987 to 1997. This is evidence to suggest that market volumes of tilapia-like reef fish have decreased in both municipal and non-municipal markets, possibly due to resource constraints.

134

Ecological Aquaculture

There is consensus that expanding reef fish capture fisheries is not an option on Viti Levu (FFD, personal communication). However, due to tilapia's similar size and appearance to the most preferred reef fish, and consumer knowledge and acceptance of tilapia, tilapia may be able to replace the current shortfall and future demand for these small species of reef fish (those detailed in Table 6.6). Bart (1996) estimated that the maximum domestic market for tilapia would be 500 mt per year but did no analyses.

Seasonal variability of demand Secondary data from the Fiji Fisheries Department's `Walu' database was obtained to determine if any strong seasonality in the fish sold to markets in the Central region of Viti Levu was present. Monthly market volumes and fish prices at Nabukulau Creek (Suva), Laqere, and roadside markets were tabulated (Table 6.7). Data was reorganized into ranks by months 1 to 12 for the three markets from 1990 to 1995. There was a significant difference in monthly ranks (Table 6.7). Month of lowest volumes averaged month 4 (April). A histogram plot showed that 51% of all months of lowest sales (17 of 33 months) were the months 1±3 (January±March); and 75% of all months of lowest sales volumes was from January to June. The top month of sales volumes was August. The period of largest volumes was months 10±12 (October to December). During 69% of the 33 months the period of largest volumes sold was from July to December. Table 6.7 Seasonality of fish volume changes in three markets in Viti Levu, 1990±95 Markets

Month of highest volume (N1)

Month of lowest volume (N2)

Nabukulau Creek Laqere Roadside Ranks

12, 9, 9, 9, 11, 12 10, 12, 6, 10 3, 10, 8, 11

3, 4, 1, 8, 1, 4 2, 1, 1, 1 1, 2, 2, 1

P<0.05

Note: Significance determined by use of the Mann±Whitney U-test for two samples of ranked observations, not paired (Sokal & Rohlf, 1969). N1 = 84, N2 = 84.

Data suggests that lower quantities of fish are available in the three markets during the `cyclone season' from January to April. Fish farmers may be better off marketing their products during this period.

Prices Fish prices have increased (with one exception) from 20 to 44% per year for the five most preferred reef fish (and tilapia) from 1987 to 1996 (Table 6.8); with the exception of the increased availability of cheap, imported, frozen jack mackerel. Jack mackerel was the cheapest protein available on the market, selling at retail prices of F$1.69± 1.90 per kg. The mean fish price in Fiji increased from F$2.33 per kg in 1987 to F$3.75 in 1996 (Table 6.8).

Planning for Sustainable Aquaculture

135

Table 6.8 Price changes over time for the most preferred reef fish, 1987±96 ($F/kg) Species

1987

1988

1989

1990

1991

1993

1994

1995

1996 Change %/year

Lethrinidae Serranidae Lutjanidae Mullidae Mugilidae Tilapia Average all

2.35 2.50 2.27 2.08 2.35 2.18 2.33

2.55 2.87 2.25 2.26 2.48 2.38 2.58

2.88 2.98 2.52 2.45 2.77 2.23 2.93

3.06 3.13 2.62 2.85 3.30 2.92 3.28

3.46 3.39 2.80 2.71 3.81 ND 3.73

3.84 4.68 2.49 2.52 4.08 2.66 3.61

3.49 4.05 2.12 2.31 3.92 2.50 3.34

3.73 3.76 2.26 2.79 4.39 3.00 3.61

3.70 3.12 2.14 2.78 4.17 ND 3.75

1.35 0.62 70.13 0.70 1.82 0.82 1.42

36 20 76 25 44 27 38

Notes: There was no data taken in 1992. ND = no data available.

Price±size and price±season relationships There was no strong price±size relationship of demand present in both the primary survey data (Fig. 6.2), and in secondary data from the FFD. Price per kilogram of fish was very variable among the different sizes of fish being sold, with no clear trends. There was no significant relationship of fish price and season (Table 6.9). However, 63% of the months of highest fish prices occurred during October to March when fish availability is lower.

Other outlets 40.5% Non-fish 3.3% 1978

Municipal markets 50.5%

Fish 96.7%

Other outlets 87.4%

1995

Municipal markets 12.6%

Non-fish 27.2% Fish 72.8%

Fig. 6.2 Change in fish marketing structure in Fiji from 1978 to 1995. `Other' outlets refers to roadside stalls, shops, restaurants, cafeÂs, supermarkets, hotels and butchers. `Municipal markets' refers to municipal markets in, principally, Suva.

136

Ecological Aquaculture

Table 6.9 Seasonality of fish price changes in three markets in Viti Levu, 1990±1995

N Creek Laqere Roadside Ranks

Month of highest price (N1)

Month of lowest price (N2)

4, 11, 8, 12, 5, 2 7, 12, 6, 10 11, 7, 1, 1

3, 8±9, 4, 9, 7, 1 2, 1, 1, 1 3, 8, 5, 6

P>0.05

Note: Significance determined by use of the Mann±Whitney U-test for two samples of ranked observations, not paired (Sokal & Rohlf, 1969). N1 = 84, N2 = 84.

Estimated domestic demand for tilapia Interviews with Fijian consumers found that price is the most important determinant of demand. In fish markets, highest demands were for low quality frozen jack mackerel imported from New Zealand at a very low price (F$1.50±1.90 retail in markets). Over 5000 mt of mackerel was imported in 1995 (FFD, 1995). Frozen mackerel is a strong-smelling oily fish distinct from the high quality reef and bottom fishes for which Fiji is famous. Consumers seem to be willing to sacrifice quality for price. It is difficult to estimate the future domestic demands for tilapia but one scenario can be constructed from fish consumption statistics, population projections, and consumer behavior. Fiji is the crossroads of the Pacific, having an ethnically diverse population comprised of native Fijians, Indians, and `others' (East and Southeast Asians, Polynesians, mixed Europeans, and other `mixed'). Since the military coups and political turmoil of the late 1980s, there has been an accelerated immigration out of the country by Indians. By 1996, indigenous Fijians had become the largest population group. In the 1996 census Fiji had 772 655 persons (394 999 Fijians; 336 579 Indians; 41 077 `others'), and population growth rates from 1986 to 1996 for the various ethnic groups were: Fijians (1.8%), Indians (70.4%), and `others' (0.9%). Over the last 10 years Fiji had an overall population growth rate of 0.8%. It is assumed that current per capita fish consumption rates will not change much beyond the current, very high figure of 68.2 kg/capita/year. Fiji's population in 1996 was estimated at 772 655 persons, giving a total seafood demand of 52 695 mt/year. Of the seafood eaten 58% is fresh fish (Table 6.3), so current demand for this product form is 30 353 mt/year. Fiji's population growth rate has been 0.8% over the last 10 years, and if this were sustained the population of Fiji in the year 2006 would be 834 467 persons. Fresh fish demand would rise to 33 008 mt/year. The gap in fresh fish availability would therefore be 2655 mt in 2006 (Table 6.10), or 265 mt/year. Domestic tilapia is sold as fresh fish exclusively in Fiji (fresh dead or live). Domestic tilapia production was estimated at 122 mt in 1996, or 22% of the volume of `tilapialike' reef fish captured (Table 6.5). If we assume 22% is the approximate market share of the fresh fish market in Fiji, and this remains constant to 2006 ± a conservative estimate on both accounts ± in 2006, additional demand for tilapia could equal approximately 58 mt (265 mt 6 0.22). This figure is close to the gap of 66 mt/year in

Planning for Sustainable Aquaculture

137

Table 6.10 Forecast of future demands for fish in Fiji Year

Population

Total fish consumption

Fresh fish

Tinned fish

1996 2006 Gap

772 655 834 467

52 695 56 911 4 216

30 353 33 008 2 655

3 162 3 415 253

Note: Percentage of product forms taken from Table 6.3.

tilapia-like reef fish generated previously using primary and secondary market data. Therefore, it is projected that domestic tilapia farmers in Fiji could safely expand domestic tilapia production from the current 122 mt an additional 58±66 mt to total 180±188 mt by 2006 ± an expansion of 6±7 mt/year each year ± to fill the predicted gap in tilapia-like reef fish marketed as fresh product.

Tilapia retail prices Market structure for tilapia is unorganized and informal at present. The majority of tilapia farmers are smallholders in rural areas who sell their fish fresh to villagers coming on previously announced harvest dates. In Suva, there is just one live market for tilapia operated by a Chinese grower. Fresh tilapia from the only large-scale commercial grower (Viti Corp.) are being marketed whole on ice at F$3.50 per kg and live at F$5.50 per kg at an upscale fish retail store owned by an expatriate Australian. Retail prices for tilapia ranged widely, from F$2.50 at a village pond bank to F$5.50 for live fish at Chandrove Fish Store.

Value-added options The survey team interviewed four small sellers (all women) at roadside markets who were selling small (c. 150±250 g) kabatia and kawakawa as whole fish, fried and smoked. These sellers mentioned that all of their fish was sold quickly in bundles of seven fish for F$5 (equivalent to F$3.57 per kg assuming 200-g fish). These sellers mentioned that the fish were available in abundance only during the period from June to August. The fillet market is very small in Fiji and is not a viable option for any large market development oriented towards the domestic consumer. However, the hotel market for high quality, fresh, Fijian grown and filleted tilapia is attractive, could be used to promote the use of aquaculture as an alternative to coral reef fishing, and may be one way to `ease into the export market'.

Use of tilapia as low cost fertilizer or feed The growing of tilapia and its use and disposal as cheap agricultural fertilizer or as pig feed is an idea that won't quit (Holmes, 1954; E. Stice, personal communication). From my view, this is an idealistic and outmoded idea which should be dispensed of altogether in any sort of planning for tilapia aquaculture development for Fiji,

138

Ecological Aquaculture

including in the planning of development programs for rural smallholder aquaculture farms, for a number of reasons. First, this idea was generated back in the 1950s by westerners (Holmes, 1954) who viewed tilapia as rubbish available for `free' that could be recycled into something worthwhile (= pigs = meat). Tilapia are simply not very good pig food unless they are cooked, which now means they are not `free'. Secondly, aquaculture of tilapia costs money and time of rural people everywhere. Any simple, `back of the envelope' calculation would reveal that feeding table scraps and even pig food is a more economic way of growing pigs rather than growing tilapia to feed pigs. Thirdly, aquaculture development is attempting to improve the image of tilapia as high quality, fresh, and nutritious food for people, especially for people in proteinshort rural and inland areas, and for export. Promotion of any extension efforts using tilapia as low quality food for animals defeats the very purpose of tilapia aquaculture development in both the rural smallholder and commercial industrial sectors. These comments are not meant to say that use of tilapia in integrated systems (like Monfort Boys Town) should not be encouraged. On the contrary, positioning of pigs and other animals above ponds to fertilize them directly with animal manures is a well known pond management option that can grow high quality tilapia from a biotechnical viewpoint. However, it is questionable from a marketing standpoint if this practice would meet the cultural backgrounds and needs of consumers. In addition, it is questionable if any fish grown of any size are `waste', given the propensity of rural people in many fish-eating nations like Fiji to eat fish of any and all sizes.

Conclusions Most fish in Fiji are marketed through outlets, not in city/town centers, and the domestic market demand for tilapia is currently small. It is projected that domestic tilapia farmers in Fiji could safely expand domestic tilapia production from the current 122 mt an additional 58±66 mt to total 180±188 mt by 2006 ± an expansion of 6±7 mt/year each year ± to fill the predicted gap in tilapia-like reef fish marketed as fresh. There is less fish available on the market during the months of January to March (`cyclone season'), and there was some (but not significant) trend of higher fish prices during this period. There was no strong price±size relationship of demand. Current prices for whole tilapia are in the range of the most preferred reef species (F$2.50±5.50/kg). Tilapia prices could not be increased much further unless resource constraints increase with the preferred reef fish species or a new period of economic prosperity occurs and people have more money to buy fish.

Recommendations While Fiji has superb natural resources for growing tilapia, it is recommended that the Fijian government develop a phased, three-step tilapia aquaculture development plan to:

Planning for Sustainable Aquaculture

(1)

(2)

(3)

139

Emphasize the domestic production sector by enhancing market opportunities for the existing rural production sector and offering market incentives to generate cash income from aquacultured fish. Establishment of permanent roadside market stalls in the areas of highest fish production would be one method. Subsidize development and feed costs of a single medium-scale (15±20 ha) commercial tilapia farm that would produce fresh fillets for the high end domestic hotel/tourist industry. Subsidize development and operational costs for a small number (5±10) of small scale (2±5 ha) commercial farms to meet the projected shortfall in supplies of `tilapia-like' coral reef fish and to develop the concept of using sustainable, inland aquaculture development to relieve fishing pressure on the nation's internationally important coral reefs.

The focus of tilapia aquaculture development efforts should be on a `market-driven technology development' approach, not a `technology-driven approach to market development'. As an example of this approach, this study indicates there may be market opportunities for tilapia aquaculture production systems during the `cyclone season' when volumes of the most preferred reef fish are lower and prices possibly higher. A market-driven technology approach to industry development would build a domestic production sector that channels aquaculture products to existing markets as replacement species for overfished reef species, and focuses on using aquaculture to meet social needs. Such an approach investigates further what the market demands, and provides those products to the markets and human needs, rather than producing products, then scurrying around to try to find markets that will take them! Because no strong price±size relationships of demand were found and tilapia production costs are high, it is recommended that farmers continue to produce fish of the same size as at present (150±250 g) and resist consumer and buyer demands to produce larger fish. Under the current market and production circumstances, farmers will likely have difficulty profiting from production of fish larger than at present. In addition, consumers are already traditionally familiar with reef fish of a similar size and appearance to the tilapia currently produced. Since the domestic market is small, Government extension efforts should be focused on the rural smallholder farmers and small-scale commercial family farms in order to increase the social benefits of aquaculture among the inland poor and the indigenous Fijians. It is recommended that Fisheries Department officers obtain aquaculture farming systems and integrated participatory rural development training to evolve inland aquaculture as a more sustainable model. Peace Corps and the Government have a viable rural aquaculture model that is meeting the goals of providing protein and additional income to the rural poor in inland areas of the country. The Monfort Boys Town is another model to emulate, but consumer resistance to eating fish from ponds fertilized with manure should not be overlooked, among consideration of other marketing factors in technology development. Over the long term, Fiji should build integrated aquaculture `farming systems' extension capacity into its education system with the University of the South Pacific. It is recommended that Government assist rural smallholders by developing a rural

140

Ecological Aquaculture

tilapia farmers' association with well publicized annual events (conference, dances, etc.), and organize rural aquafarmers' cooperatives. The biggest constraint to rural aquaculture development is the availability and cost of transportation. It is recommended that Government establish small, roadside marketing stalls in strategic areas near to present centers and to the future growth centers of tilapia production (as has been developed for ginger). A model of these marketing stands and the management of these has already been established by the Suva Municipality at Nabukulau Creek (however, rental fees appear to be too high at Nabukulau Creek). There are truck carriers that come to the inland areas to transport agricultural produce. Fish could be transported in these carriers if a source of ice and sturdy plastic boxes with heavy lids were available. It is recommended that Government study the idea of subsidizing ice production and fish transport from the rural areas to newly constructed fish stands alongside the roadways near the growing centers of fish production. It may be possible that ice capacity available in the capture fisheries sector could be organized to assist the infant aquaculture industry. Development of live haul markets for tilapia, first in areas of concentrated Chinese and Southeast Asian consumers such as Nasori, should be an urgent priority because of the premium prices paid by these consumers. The hotel market for fresh, high quality, Fiji-grown filleted tilapia is one way to `ease into' the export market. It is recommended that Government connect Viti Corp. or one of the government-assisted farms with the tourism sector to market a tilapia fillet product as an `organic' product with no chemicals, locally grown in Fiji, certified, and grown with mountain spring waters, and connect reef conservation with aquaculture in international promotional efforts. Fiji is well placed to be the center of aquaculture research and development for the entire South Pacific. The nation could earn significant revenues from marketing aquaculture goods and services, as well as becoming an important center of aquaculture production. Hawaii is a notable example in this regard. The State earns more income from its aquaculture research, contracting, consulting, and training sectors of the industry than it makes from its aquaculture production sector. As a modern, knowledge-based industry in its infancy in Fiji, aquaculture needs longterm backup in information resources, and applied aquaculture research and development, especially in production technology, post-harvest methods, and marketing.

References Agustin, L., Mather, P. & Wilson, J. (1997) Levels and patterns of genetic diversity in Oreochromis mossambicus: West African vs. introduced feral populations in the Australasian/ Pacific region. In: Tilapia Aquaculture (ed. K. Fitzsimmons), pp. 75±86. Northeast Regional Agricultural Engineering Service, Ithaca, NY. Bart, A. (1996) Commercialization of fish farming (Tilapia niloticus) in Fiji. Peace Corps/ Farmer to Farmer Program Project Report (unpublished mimeo), Suva, Fiji.

Planning for Sustainable Aquaculture

141

Bowen, S. (1987) Composition and nutritive value of detritus. In: Detritus and Microbial Ecology in Aquaculture (eds D. Moriarty & R. Pullin), pp. 192±216. ICLARM, Manila, Philippines. Costa-Pierce, B.A. (1997) The tilapias of the Salton Sea, a marine lake in California. In: Tilapia Aquaculture (ed. K. Fitzsimmons), pp. 595±602. Northeast Regional Agricultural Engineering Service, Ithaca, NY. Costa-Pierce, B.A. & Doyle, R. (1997) Genetic identification and status of tilapia regional strains in California. In: Tilapia Aquaculture in the Americas, Vol. 1 (eds B. Costa-Pierce & J. Rakocy), pp. 1±17. World Aquaculture Society and American Tilapia Association, Baton Rouge, LA. Costa-Pierce, B.A. & Riedel, R. (2000) Fisheries ecology of the tilapias in subtropical lakes of the United States. In: Tilapia Aquaculture in the Americas, Vol. 2 (eds B. Costa-Pierce & J. Rakocy), pp. 1±20. World Aquaculture Society and American Tilapia Association, Baton Rouge, LA. Cowx, I. (1997) Stocking and Introductions of Fish in Freshwater and Marine Ecosystems. Fishing News Books, Surrey. De Silva, S.S. & Senaratne, K. (1988) Oreochromis mossambicus is not universally a nuisance species: the Sri Lankan experience. In: The Second International Symposium on Tilapia in Aquaculture (eds R. Pullin et al.), pp. 445±450. ICLARM, Manila, Philippines. FFD (Fiji Fisheries Department) (1995) Annual Report. MAFFA, Suva, Fiji. FFD (Fiji Fisheries Department) (1996) 1996 Annual Report. MAFFA, Suva, Fiji. Holmes, S. (1954) Report on the possibility of using Tilapia mossambica as animal food. Fiji Agriculture Journal, 25, 79. Lal, S. (1991) Tilapia as a cost-effective sustainable protein source for developing countries. Draft proposal for aquaculture development in Fiji to ACIAR (unpublished). Lobel, P.S. (1980) Invasion by the Mozambique tilapia (Sarotherodon mossambicus; Pisces; Cichlidae) of a Pacific Atoll marine ecosystem. Micronesica, 16(2), 349±355. MAFFA (Ministry of Forestry, Fisheries and ALTA) (1996) Commodity Development Framework (CDF). Executive Summary. MAFFA, Suva, Fiji. McCoy, E. & Hopkins, M. (1978) Method of Conducting a Marketing Study. Department of Fisheries and Allied Aquacultures, Auburn University, AL. New, M. (1997) Aquaculture and capture fisheries ± balancing the scales. World Aquaculture, 28(2), 11±32. Pullin, R. (1985) Tilapias: everyman's fish. Biologist, 32, 84±88. Pullin, R. & Lowe-McConnell, R. (1982) The Biology and Culture of Tilapias. ICLARM, Manila, Philippines. Rawlinson, N. (1994) A survey of the subsistence and artisanal fisheries in rural areas of Viti Levu, Fiji. CSIRO, Cleveland, Australia. Richards, A. (1994) Fiji Fisheries Resources Profiles. Forum Fisheries Agency Report Number 94/4. FFA, Honiara, Solomon Islands. Sokal, R.R. & Rohlf, F.J. (1969) Biometry. W.H. Freeman & Co., San Francisco. Trewavas, E.(1983) Tilapiine Fishes of the Genera Sarotherodon, Oreochromis and Danakilia. British Museum of Natural History, London. Vuki, V. (1991) A fish consumption survey of Dravuni islanders, Great Astrolabe reef. University of the South Pacific, Suva.

142

Ecological Aquaculture

Zachritz, W.H. & Rafferty, K. (2000) Basic considerations in design of geothermal-based tilapia production systems in the United States. In: Tilapia Aquaculture in the Americas, Vol. 1 (eds B. Costa-Pierce & J. Rakocy), pp. 82±90. World Aquaculture Society and American Tilapia Association, Baton Rouge, LA.

Part 3

The Context of Ecological Aquaculture

Chapter 7

Village-based Aquaculture Ecosystems as a Model for Sustainable Aquaculture Development in Sub-Saharan Africa Randall E. Brummett1 and Barry A. Costa-Pierce2 1 International Center for Living Aquatic Resources Management (ICLARM) 2 Rhode Island Sea Grant College Program Introduction Aquaculture in Southern Africa is growing in part due to new approaches to project design and implementation begun in the late 1980s by: the Food and Agriculture Organization of the United Nations (FAO) project `Aquaculture for Local Communities Management', funded by Sweden and Belgium; l the German Agency for Technical Cooperation project `Malawi±German Fisheries and Aquaculture Development'; and l ICLARM's `Research for the Development of Tropical Aquaculture' in Malawi, and `Research for the Future Development of Aquaculture' in Ghana. l

These projects generally followed FAO guidelines described in the Thematic Evaluation of Aquaculture (FAO/NORAD/UNDP, 1987) to develop technologies, to increase awareness, and to reorient extension approaches to make them more participatory. We report here on progress in the development of a village-based approach to integrated aquaculture ecosystems with a fishpond as the focal point for recycling of by-products generated by other farm enterprises. This approach is a potentially important new advance in evolving sustainable aquaculture systems appropriate to the social, environmental and economic situations in rural Africa (Box 7.1). Village aquaculture ecosystems (VAEs) have been shown to improve productivity (Smalling et al., 1996), increase sustainability (Lightfoot et al., 1993), decrease waste, and even rehabilitate degraded rural landscapes (Pullin & Prein, 1995). If applied over larger areas, VAEs have the potential to reduce rural poverty and food insecurity by offering realistic opportunities for diversification and enhanced efficiency to African farmers whose land holdings are shrinking (Dalsgaard et al., 1995; Hoque, 1995; AlarcoÂn & Carls, 1996; Harris, 1996; Smalling et al., 1996).

Box 7.1 Integrated resource management

pumpkin leaf/fruits to pond s Waste Fertile mud f ro m po nds Swee t po tato Rice straw lea ve s

Stream diversion

Waste stovers recycled

Rice-fish

Water Fish

W a t e r s e epag

W

Ma

Waste fruit

e

Rice straw compost

nu

a

te rt fro o liv es m Water live tock sto c food scrapes k

Waste leaf recycled

r

e

Manure to perennial garden

Livestock

eb

Household ran

Ash from cooking fire

was

d

a iz

Maize bran fee

M

Household orchard

Waste stovers recycled

Seasonal garden (pumpkin/maize) Maize

Grain store

te

Pounding of grain to flour

Ecological Aquaculture

Pond dike Banana, sugarcane, vegetable,fruit and leaf waste

146

Since 1985, ICLARM, with financial support from Germany, Denmark and the US, has conducted research into the development of integrated aquaculture techniques for use by African smallholding farmers. In collaboration with researchers, farmers develop maps (such as that shown below) of their farming systems. From this, farmers get new ideas about integrated resource management and often see their farms holistically for the first time. Through on-farm and on-station experiments, researchers, quantify the various resources flows (indicated by arrows) on the map to gain socio-economic and ecological insights into integrated African farming systems.

Village-based Aquaculture Ecosystems in Sub-Saharan Africa

147

The potential impacts of VAEs on African food security are enormous. Small farms account for the vast majority (97% in 1989) of aquaculture production in SubSaharan Africa (King, 1993). According to an FAO study, `There are some 9.2 million km2, equivalent to 31% of the African surface, containing area apt for warm water fish farming at a subsistence level. Of the 48 countries, 40 possess at least some land apt for this use.' (Kapetsky, 1994) Farm trials in Malawi have shown that fish yields on VAEs range from 1000 to 3000 kg/ha/year (Costa-Pierce et al., 1991; Chikafumbwa et al., 1993; Brummett & Noble, 1995a, b). If this production can be replicated on only 1% of the suitable land, 9.2±27.6 million mt of fish per year might be produced. This is between two and six times the catch from Africa's capture fisheries (FAO, 1989). However, extrapolating from land that might be suitable for fish farming to potential economic and environmental impacts of integrated aquaculture on small farmers must be viewed with caution. Despite steady growth, realizing the full potential of aquaculture on Africa's suitable lands has been elusive. Many development projects have relied on `modules' or `technology packages' such as duck±fish or chicken±pig±fish production units. Such technologies are developed and tested for economic efficiency on experiment stations, then promoted among extension agents for transmission to farmers. These technologies are found to be highly lucrative only if implemented in their entirety, and are often beyond the financial means of the majority of the target group. Lack of funds to implement the system leads to a situation where extension agents must travel long distances to interact with those few farmers who are capable of using the experiment station approach. The ways in which rural farmers with small land holdings (`smallholders') make decisions about technology adoption are complex (Brummett & Haight, 1996). Likewise, there is a high degree of variability within and among smallholder farming agroecosystems (ICLARM & GTZ, 1991). To accommodate this variability, technology packages should be customized to fit individual farming situations. The Expert Consultation on Small-Scale Rural Aquaculture (Martinez-Espinoza, 1996) found that, rather than modules, new approaches to problem resolution are needed. ICLARM research in Malawi has demonstrated that VAEs might be developed in the flexible fashion required to achieve the full potential of rural fish farming in Africa.

Research to develop VAEs in Africa Integrated farming research in Africa has evolved from simple animal or crop plus fish modules modified from Asian models, to holistic analyses of complete farming systems (ICLARM & GTZ, 1991; Noble & Costa-Pierce, 1992; Brummett, 1994). ICLARM has developed a Farmer±Scientist Research Partnership (FSRP) approach (Fig. 7.1) (Brummett & Noble, 1995a). The FSRP can be summarized as follows:

148

(1) (2)

(3)

(4) (5)

Ecological Aquaculture

Data on the farm resource base are gathered in a participatory resource mapping exercise. Weekly pond management data are used to set up parallel experimental controls and treatments at the research station and on-farm in order to provide baseline data for both farmers and researchers. Potential new technology is developed in trials conducted on the experiment station using the actual farm resource base and management as experimental controls. Results on-farm and on-station are compared in an interactive session with farmers at the end of the production cycle. Farmers compare results from (2) and (3) above to plan the next round of experimentation.

The FSRP approach to technology development involves farmers throughout and uses actual farm conditions and constraints to design new technology; consequently,

Farmer consultation, discussed and on-farm resource inventory

Choice of experimental topics/protocols with farmers

On-farm experimentation

On-station control via PondSim

Replicated experiments

Discussion of results with farmers Fig. 7.1 ICLARM's Farmer±Scientist Research Partnership (FSRP) approach to the development of integrated aquaculture±agriculture (IAA) technology. PondSim is a spreadsheet package designed by ICLARM to calculate dry matter, organic matter, nitrogen and phosphorus equivalents of pond inputs used by farmers as a means of controlling on-farm conditions during experiment station trials.

Village-based Aquaculture Ecosystems in Sub-Saharan Africa

149

it is able to obviate the necessity of ex post facto adaptation of research-station results to farm conditions. FSRP also establishes the necessary information base on the farming ecosystem and feedback loops for farmers and researchers to trust each other and work together effectively on problems identified collectively (Box 7.1).

Impacts on farm productivity In Malawi, average fish productivity in the VAEs is 1350 kg/ha/year in rain-fed areas and 1650 kg/ha/year in spring-fed areas (Brummett & Noble, 1995a). These are compared to an average of about 900 kg/ha/year for the 48 most productive, nonintegrated fish farms in Southern Malawi (Scholz & Chimatiro, 1996). Differences stem from the range of available pond inputs, and the location of ponds relative to other farm enterprises in the household. On integrated farms, ponds are generally located in vegetable gardens, or vegetable gardens develop around the fish pond to take advantage of emergency irrigation water, and wastes from the garden are used to feed fish (Noble & Chimatiro, 1991; Chimatiro & Costa-Pierce, 1996). Typically, on-farm wastes amount to some 3700 kg of dry matter per year and the material is generated in close proximity to the pond, minimizing the work involved in transportation. To properly feed the typical farm pond, a farmer needs about 522 kg of dry matter (Brummett, 1998). Non-integrated farms, on the other hand, are using maize bran exclusively, as recommended by extension, as the `best' fish food (Kadongola, 1990). Maize bran production averages around 192 kg of dry matter, only 37% of the amount needed to feed a fishpond properly (Noble, 1996). In addition, on a typical Malawi farm, maize bran is produced in the house, often far from the pond, and is also a dry season `emergency food' for humans (Mills, 1991). In contrast, vegetable garden wastes are typically just burned if they are not used in a pond. Integrated fishponds have the potential to profoundly affect the economic and ecological sustainability of small farms. All of the farms involved in research were affected by drought from 1991 to 1995. Yet in all cases, even though maize crops failed and farmers suffered economic losses, integrated pond±vegetable systems kept operating through the drought. Retaining water on the land, ponds enabled farms to sustain their food production and balance their losses on seasonal croplands. For example, in the 1993/94 drought season, when only 60% of normal rain fell, the average net cash income to a study group of rain-fed integrated farms was 18% higher than the non-integrated farms; and this occurred in an area with some of Malawi's severest poverty (Brummett & Chikafumbwa, 1995).

Technology adoption and transmission Virtually all of the ICLARM cooperating farmers who have access to permanent water supplies continue to grow fish and improve their production (Fig. 7.2). Among

150

Ecological Aquaculture

Fig. 7.2 Pond productivity over time in integrated vs non-integrated fishponds in Southern Malawi. The production target of 2500 kg/ha was arrived at by extrapolation of Malawi's national fish production need to the land area available for aquaculture. The technology promoted was too complex for most farmers to fully understand and/or adopt, resulting in declining production as extension support waned. IAA entrylevel technology is much simpler and less productive initially, but evolves on-farm as farmers who understand the technology are able to more efficiently manipulate it to suit their individual situation (Brummett & Williams, 2000).

those farmers with only rain-fed fish ponds, 36% dropped out for one reason or another. Forty per cent of those dropping out did so because of family deaths or illness rather than for any agricultural reason. Those remaining have increased their average pond size from 64 to 88 m2 and new gardens are being planted around the ponds (Brummett & Chikafumbwa, 1995). Of the Malawian farmers who have been exposed to VAE technology through various participatory mechanisms, 86% have adopted at least one of the demonstrated technologies, 76% adopted at least two, and 24% adopted four (Noble & Rashidi, 1990). Interestingly, in follow-up interviews it was discovered that the adopters did not simply copy what they had seen, but rather took the basic ideas and modified them to suit their individual circumstances and farming systems (Brummett & Noble, 1995b). Once in the rural community, VAE technologies spread and evolved without further extension support. A survey found that, within six months of an open day held at the experiment station in May 1990, 46% of adopters in the target area had learned about integrated aquaculture from other farmers. A third of these farmers had adopted two or more technologies from their neighbors. By the end of 1992, almost 80% of the farmers practicing integrated rice±fish farming in Zomba District had never witnessed first hand an extension demonstration (Chikafumbwa, 1994). In Zomba East, where ICLARM worked with 34 farmers from 1991 to 1995 (ICLARM & GTZ, 1991), there are now 225 practicing fish farmers (Scholz et al., 1997).

Village-based Aquaculture Ecosystems in Sub-Saharan Africa

151

Ecological footprints Berg et al. (1996) reviewed the environmental sustainability of various types of aquaculture development by comparing VAEs with pellet-fed cages situated in Lake Kariba, Zimbabwe, based on the concept of ecological footprints (Table 7.1). They estimated the ecosystem support area (or ecological footprint) required for the production of feed and oxygen and for waste nutrient assimilation to support a 1 m2 cage producing 380 g of tilapia/day. Based on the assumption that the fish meal needed would be produced locally from kapenta (Limnothrissa miodon; a pelagic clupeoid), production of 1 kg of caged tilapia would require 4.3 kg of kapenta and 0.9 kg of grain. The fish meal alone required to produce 2000 mt of tilapia in cages would thus appropriate an area almost 21 000 times the area of the cages themselves. Producing 2000 mt of fish in cages would require over 8000 mt of kapenta, corresponding to 40% of the total annual catch in Lake Kariba and a net loss in available fish protein for the human population. Table 7.1 Ecological efficiency of two aquaculture systems (modified from Berg et al., 1996) To support a 1 m2 tilapia cage in Lake Kariba requires: 21 000 m2 of 420 m2 of 160 m2 of 115 m2 of

water area to grow fishmeal crop land to grow grains phytoplankton to produce oxygen benthic community to assimilate waste phosphorus

21 700 m2 `ecological footprint' (6 g fish per m2 of footprint) To support 1 m2 of waste-fed integrated fish pond requires: 0.9 m2 of additional benthic community to assimilate phosphorus 0.9 m2 of phytoplankton to produce oxygen 1.8 m2 `ecological footprint' (264 g fish per m2 of footprint)

Although relatively minor compared with the area needed for food production, the support area for oxygen production and nutrient assimilation was also significant. The surface area of primary production required to produce the oxygen consumed directly by fish in the cages and by the organic waste produced at the farm (fish feces and feed fragments) was 160 times larger than the area of the cages. Assimilation of nutrients released to the environment as a result of feeding and metabolism of the fish required an area 115 times larger than the area of the cages. In contrast, the ecological footprint of VAEs is minuscule. Feed inputs to ponds are based entirely on wastes from agriculture. The relatively low fish production per unit of area in the ponds means the ecological footprint needed for oxygen production and nutrient assimilation can be sustained within the pond system itself. Hence, a minimum of external (e.g. off-farm) life or feed supports are needed to farm tilapia in integrated village ponds.

152

Ecological Aquaculture

This comparative analysis shows that there is a severe risk that cage aquaculture would be a substitute and not a complement to the existing fishery in Lake Kariba. On the other hand, as long as on-farm waste products are the only feeds, the VAEs would complement not deplete the kapenta fishery.

Environment, economics and food security The advantages of VAEs are clear if one only considers longer-term environmental impacts and overall efficiencies of resource use. However, short-term economics for cages often favor the more intensive system, at least on paper. This is because economies of scale for feed processing, storage and marketing favor the more vertically integrated producer. On the other hand, in rural Africa problems such as irregular supply of inputs and spares, unreliable power and unpredictable production and marketing conditions can hit the larger-scale producer proportionally harder than the smaller scale one. The Lake Kariba case study illustrates the food security limitations of aquaculture when it is conceived primarily as a commercial fish production exercise based only on cash-flow models. In the case of the proposed cages, commercial fish farming, although profitable for the investors, would result in a net loss of fish protein available for human consumption. To exacerbate this problem, many commercial aquaculture enterprises in the tropics tend to export their produce to earn the higher profits possible in richer American, European or Japanese markets (Fitzsimmons & Posadas, 1997). All of the larger-scale commercial investments in Southern African aquaculture of which the authors are currently aware, including projects in the Republic of South Africa, Namibia, Tanzania, Swaziland and Zimbabwe, are export-oriented. Most are focusing on high-value species such as shrimp and oysters or on species for which there is little or no local market (e.g. the red swamp crayfish, Procambarus clarki). While providing important investment opportunities, these food production enterprises do not address Africa's quality and quantity food security problems. When locally available materials are used in the manufacture of fish foods, exportoriented commercial aquaculture directly transfers food from poor countries to richer ones. In the best case (i.e. where taxes on exports are used efficiently by government to purchase larger quantities of cheaper food more affordable to poor consumers) lower quality food is being substituted for high quality food. Integrated smallholding-based systems, in contrast, produce direct opportunities for households to address their own food insecurity. Smallholdings sell only a portion of their harvest for cash, keeping a substantial part for household consumption (Brummett & Chikafumbwa, 1995). That portion which is sold is normally transferred to fellow smallholders in the immediate vicinity who also face problems with food quantity and quality (Brummett, 2000). Increases in overall farm productivity associated with the integration of aquaculture can result in large gains in household food production and income generation (Ruddle, 1996). The VAE is clearly the most environmentally friendly, ecologically efficient, and

Village-based Aquaculture Ecosystems in Sub-Saharan Africa

153

socially appropriate aquaculture approach for most of rural Africa. If the ecological footprints for the two systems are inverted, one finds that the cages produce 6 g of fish per m2 of footprint compared with 264 g for the VAE. In addition, having a larger number of widely scattered small producers ensures that microeconomic benefits will accrue to the community directly and will be more equitably distributed. Producing fish in small ponds in small communities also obviates many of the marketing problems that more centralized producers face. It also lowers the overall risk. However, one important question remains: Can small scale producers grow enough fish to make a real difference in living standards? Economically, VAE farms produce almost six times the cash generated by the typical Malawian smallholder (Scholz & Chimatiro, 1996). The integrated pond± vegetable garden is the economic engine on these farms, generating almost three times the annual net income from the staple maize crop and the homestead combined. The vegetable±fish component contributes, on average, 72% of annual cash income (Brummett & Noble, 1995b). On a per area basis, the vegetable garden/pond resource system generates about $14.00 per 100 m2 of land per year compared with $1.00±$2.00 for the maize crop and homestead respectively.

Realizing the potential At the national policy level, resource utilization for aquaculture development impacting the good of the larger human population must come ahead of providing investment advice to individual entrepreneurs interested in intensive aquaculture development (Costa-Pierce & Pullin, 1992). To maximize the positive impact of resource use at the national level, we should compare not only the short-term economics of the two systems, but also how their implementation would effect distribution of wealth, sustainable use of resources and food security (ICLARM & GTZ, 1991). The amount of time and energy involved in ICLARM research efforts has been large relative to the small numbers of farmers involved. Duplicating such efforts on a large enough scale to have widespread regional impact would probably not be costeffective and would therefore be institutionally unsustainable (van der Mheen, 1996). However, from the experience gained over the course of these various projects, ICLARM has extracted what we feel are the key components of a new approach to the evolution of sustainable aquaculture development in rural Africa: `Bottom-up' development planning: knowing the socio-economic, gender, and cultural context of farming households and the national policies that affect them first, before any interventions (ICLARM & GTZ, 1991). l Information transfer: simple messages for farmers generated by research in dynamic consultation with extension. l Sustained adoption: full participation of farmers in the research and development process. l Ecological evolvability: incorporating and utilizing an ecosystems vision of the farm. l

154

l

Ecological Aquaculture

Economic evolvability: incremental, evolutionary increases in cash productivity to overcome recurrent cash-flow constraints.

Each of these components has a technical and social dimension. The technical dimension involves determinates of how productive a new technology is likely to be under a given set of agroecological circumstances. The social dimension involves research±extension±farmer communication and decision-making leading to sustained adoption and transformation to sustainable farming ecosystems.

Information transfer and sustained adoption ICLARM's experience in Malawi is that complex technology packages that hope to address all aspects of integrated fish production are not effective in transferring information to farmers. There are two reasons for this (Brummett & Haight, 1996): (1) (2)

Extension agents are seldom sufficiently proficient in the technologies themselves to clearly communicate them to farmers. Farmers who are generally illiterate and operating at the subsistence level are more interested in risk management than economic analyses concerning single components of their multi-dimensional and highly variable farming ecosystems.

To overcome these problems, it is necessary to distill from farm characterizations and an in-depth understanding of aquaculture technologies a few simple techniques that are low risk in terms of capital requirement and the degree of modification to existing farming practices, but which can significantly address the perceived needs of the farming community. For example, rather than advocating the construction of a `proper' fish pond which drains, ICLARM has successfully advocated the construction of small holes dug into existing boggy spots in farmers' fields. Such a pond is simple to construct, is easily filled by rain- and groundwater, and only costs the loss of production of some of the maize crop which was already producing poorly because of waterlogging; likewise, with harvesting technology (hook and line), production system (partial harvesting of intermediate-sized fish to maximize biomass), and pond feeding strategy (focus on fertilization value of weeds and recyclable by-products). It is not economically or institutionally practical for a team of researchers to be the main mechanism of outreach to farmers. Unfortunately, many extension services are poorly trained and supported and hence not fully capable of choosing and then explaining to farmers the best technology for a particular situation. This problem was discussed in depth at a recent FAO Technical Consultation (van der Mheen, 1996) and several suggestions for their alleviation were put forward: l

Research should provide a `promotion facility' which interacts with extension to analyze information coming from the field, and `experiment stations' to find good matches between needs (both felt and unfelt) and possible solutions.

Village-based Aquaculture Ecosystems in Sub-Saharan Africa

155

To reach large numbers of farmers, a simplified technology should be translated into an extension message which extension and farmers can easily understand. l Extension materials should be generated by researchers and directed at farmers rather than extension agents. l Based on their field experience, extension agents should provide at least initial critical review of the relevance of technologies proposed by scientists. l

Problems in the farmer±extension±research continuum have plagued development efforts for many years. The act of working together to identify problems and design solutions and extension messages is a better way for research and extension to overcome long-standing differences for the benefit of agricultural productivity and rural economic growth. If implemented, these arrangements would, in effect, dissolve the separate identities of the extension and research services (Brummett & Haight, 1996). In so doing, it would bring research and extension together with a single purpose: to improve the management of smallholdings through the introduction of VAEs wherever feasible. Integration of extension and research functions will require a higher level of farmer input to the research program, as well as a higher level of technical flexibility in the extension services. Researchers working in collaboration with farmers and extension agents should increase their sensitivity to the real problems faced in the field. Having extension agents working at the experiment station on research projects should facilitate an improvement in their practical skills and comprehension of the principles of aquaculture. The farmer should be the largest beneficiary of this collaboration by receiving higher quality information and being able to have his or her theoretical questions addressed immediately by research.

Improvements in productivity Simple technologies may be easier to get on to the farm than complicated ones, but they also tend to be less productive. To overcome this problem, ICLARM utilizes an evolutionary approach, the FSRP, to improving overall farm productivity. Experience in Malawi has shown that working with, rather than for, farmers fosters a much more thorough understanding of technology than would be possible by even the most clearly written textbook. For example, the extension services in Malawi have long promoted a set of technologies developed on research stations that can, if properly utilized, produce an average of 2500 kg/ha of fish. These technologies have suffered from being poorly adapted to local farms and misunderstood by both extension agents and farmers. The result is the reported average production by extension-assisted farmers of 900 kg/ha and the high dropout rates witnessed by many projects (Brummett & Haight, 1996). In contrast, ICLARM's proposed technologies produced initially only about 800 kg/ha, but set out new participatory approaches wherein farmers and research personnel learned together how best to grow fish under a particular set of environmental

156

Ecological Aquaculture

and social circumstances. Development of new approaches facilitated the evolution of the growth of pond and farm productivity and increased the number of farmers involved. With this approach, pond productivity in Malawi has grown over time to an average of 1500 kg/ha (Fig. 7.2). In addition, the sense among farmers is that this growth and improvement are something they themselves have accomplished. Farmers have also been key informants in enhancing the increased rate of farmer-to-farmer transfer of technology. To systematically document these evolutionary and new relationships, ICLARM has developed the Research Tool for Natural Resource Management and Evaluation (RESTORE), a participatory method for resource management, monitoring and evaluation. Based on four sustainability indicators (diversity, recycling, capacity and economic efficiency), RESTORE captures key information about how farming systems produce food and wealth within the context of their external and internal environments (Lightfoot & Pullin, 1995). Testing of the tool in Ghana, Malawi and the Philippines has shown RESTORE to be effective in helping farmers to understand potential synergisms within their complex farms (Lightfoot & Noble, 1993). To address the social dimension, the Farmer±Scientist Research Partnership (FSRP) has developed new relationships between farmers, extension agents and researchers.

Economic threshold for commercial transformation With an average annual cash income of less than $10.00 per year, 80% of smallholder farmers in Malawi cannot be considered as part of the cash economy (World Bank, 1996). Macroeconomic policy changes will, consequently, have no effect on small farm productivity except possibly by making the land worth so much that smallholders will be forced to sell out to urban investors who can afford the capital investment required to make the land productive. With no realistic options for employment, the displaced rural poor add to the social crisis now being witnessed in the cities of many developing countries. A change from rural poverty to urban poverty cannot be considered much of an improvement in the lives of these people. The projected additional cost of policing millions of displaced and unemployed smallholders alone ought to encourage a policy of keeping people productively engaged on the land. Because they are based on expensive external inputs, the agricultural technologies of the green revolution and the industrial aquaculture factories of the late twentieth century have not been able to engender significant economic growth on impoverished small African farms. Nor will these technologies help in evolving a solution to sustaining thousands of rural farmers in the twenty-first century. Without massive infusions of external cash, increased economic productivity will have to be based on available resources. Reardon & Vosti (1995) point out the importance of what they term `investment poverty' in rural transformation. When cash flows into a farm household, immediate needs such as hospital or school fees receive first attention. Additional cash inflow might be used to purchase improved housing components (e.g. a plastic liner for a

Village-based Aquaculture Ecosystems in Sub-Saharan Africa

157

thatched roof) or some other coveted item such as a radio. Only after these recurrent cash demands are met will capital be freed for reinvestment in future agricultural production. Because of their critical role in improving the economic performance of the farming system on the basis of materials already available on the farm, and evolving new modes of economic and community sustainability, village aquaculture ecosystems may have a critical role to play in the transition of a farm from `subsistence' to `commercial'. It remains to be proven, through a widely promoted program of developing VAEs and rural self-sufficiency, that it is not too late to save hundreds of millions of Africa's rural poor from a life of urban poverty and social decay.

References AlarcoÂn, E. & Carls, J. (1996) Integration of conservation and sustainable use of biodiversity into international research. In: Agricultural Research and Sustainable Management of Natural Resources (ed. H.-J.A. Preuss), Schriften 66. Zentrum fuÈr Regionale Entwicklungsforschung der Justus-Liebig-UniversitaÈt Giessen, Germany. Berg, H., MicheÂlsen, P., Troell, M., Folke, C. & Kautsky, N. (1996) Managing aquaculture for sustainability in tropical Lake Kariba, Zimbabwe. Ecological Economics, 18, 141±159. Brummett, R.E. (1994) The context of smallholding integrated aquaculture in Malawi. In: Aquaculture Policy Options for Integrated Resource Management in SubSaharan Africa (ed. R.E. Brummett), pp. 3±5. ICLARM Conference Proceedings 46. International Center for Living Aquatic Resources Management, Manila, Philippines. Brummett, R.E. (1998) Making experiment station results more useful to African fish farmers. Naga, The ICLARM Quarterly, 21(2), 19±24. Brummett, R.E. (2000) Factors affecting fish prices in southern Malawi. Aquaculture, 186, 243±251. Brummett, R.E. & Chikafumbwa, F.J.K. (1995) Management of rainfed aquaculture on Malawian smallholdings. In: Sustainable Aquaculture '95, pp. 47±56. Pacific Congress on Marine Science and Technology, Honolulu, Hawaii, USA. Brummett, R.E. & Noble, R.P. (1995a) Farmer±scientist research partnerships and smallholder integrated aquaculture in Malawi. In: The Management of Integrated Agro-Piscicultural Ecosystems in Tropical Areas (eds J.J. Symoens & J-C. Micha), pp. 443±464. Belgian Royal Academy of Overseas Sciences, Technical Center for Agricultural and Rural Cooperation, Brussels, Belgium. Brummett, R.E. & Noble, R.P. (1995b) Aquaculture for African Smallholders. ICLARM Tech. Rep. 46. International Center for Living Aquatic Resources Management, Manila, and Deutsche Gesellschaft fuÈr Technische Zusammenarbeit, Eschborn, Germany. Brummett, R.E. & Haight, B.A. (1996) Research±development linkages. In: Report of the Expert Consultation on Small-Scale Rural Aquaculture (ed. M. Martinez-Espinoza). FAO Fisheries Report 548. Food and Agriculture Organization of the United Nations, Rome. Brummett, R.E. & Williams, M.J. (2000) Aquaculture in African rural and economic development. Ecological Economics, 33, 193±203. Chikafumbwa, F.J.K (1994) Farmer participation in technology development and transfer in

158

Ecological Aquaculture

Malawi: a rice±fish example. In: Aquaculture Policy Options for Integrated Resource Management (ed. R.E. Brummett). ICLARM Conference Proceedings 46. International Center for Living Aquatic Resources Management, Manila, Philippines. Chikafumbwa, F.J.K, Costa-Pierce, B.A., Kadongola, W., Jamu, D.M. & Balarin, J.D. (1993) Investigations on the use of on-farm resources as pond inputs to culture Tilapia rendalli and Oreochromis shiranus on smallholder farms in rural Malawi. Aquaculture, 117, 261±271. Chimatiro, S.K. & Costa-Pierce, B.A. (1996) Waste vegetable leaves as feeds for juvenile Oreochromis shiranus and Tilapia rendalli in monoculture and polyculture. In: The Third International Symposium on Tilapia in Aquaculture (eds R.S.V. Pullin, J. Lazard, M. Legendre, J.B. Amon Kothias & D. Pauly), pp. 183±192. ICLARM, Manila, Philippines. Costa-Pierce, B.A., Lightfoot, C., Ruddle, K. & Pullin, R.S.V. (1991) Aquaculture Research and Development in Rural Africa. ICLARM Conference Proceedings 27. ICLARM, Manila, Philippines. Costa-Pierce, B.A. & Pullin, R.S.V. (1992) Development of smallholder aquaculture in Malawi. World Aquaculture, 23, 43±48. Dalsgaard, J.P.T., Lightfoot, C. & Christensen, V. (1995) Towards quantification of ecological sustainability in farming systems analysis. Ecological Engineering, 4, 181±189. FAO (1989) Fishery Statistics: Catches and Landings, Vol. 68. FAO Statistics Series 98. Food and Agriculture Organization of the United Nations, Rome. FAO/NORAD/UNDP (1987) Thematic Evaluation of Aquaculture. Food and Agriculture Organization of the United Nations, Norwegian Ministry of Development Cooperation and United Nations Development Program. FAO, Rome, Italy. Fitzsimmons, K. & Posadas, B.C. (1997) Consumer demand for tilapia products in the U.S. and the effects on local markets in developing countries. In: Tilapia Aquaculture: Proceedings from the Fourth International Symposium on Tilapia in Aquaculture (ed. K. Fitzsimmons), pp. 613±632. NRAES-106. Northeast Regional Agricultural Engineering Service, Cooperative Extension, Ithaca, NY. Harris, F. (1996) Intensification of Agriculture in Semi-Arid Areas: Lessons from the Kano CloseSettled Zone, Nigeria. Gatekeeper Series 59. International Institute for Environment and Development, London, UK. Hoque, M.T. (1995) Sustainable agriculture: a perspective on fish culture for the small-scale resource-poor farmers of Bangladesh. Journal of Sustainable Agriculture, 5(3), 97±113. ICLARM & GTZ (1991) The Context of Small-Scale Integrated Agriculture±Aquaculture Systems in Africa: A Case Study of Malawi. ICLARM Studies and Reviews 18. International Center for Living Aquatic Resources Management, Manila, Philippines and Deutsche Gesellschaft fuÈr Technische Zusammenarbeit (GTZ), Eschborn, Germany. Kadongola, W.K. (1990) Maize (Zea mays) Bran as a Supplemental Feed in the Culture of Tilapia rendalli (Boulenger) and Oreochromis shiranus (Boulenger). MSc Thesis, University of Malawi. Kapetsky, J.M. (1994) A Strategic Assessment of Warm-Water Fish Farming Potential in Africa. CIFA Technical Paper 27. Food and Agricultural Organization of the United Nations, Rome. King, H. (1993) Aquaculture development and environmental issues in Africa. In: Environment and Aquaculture in Developing Countries (eds R.S.V. Pullin, H. Rosenthal & J.L. Maclean), pp. 116±124. ICLARM Conference Proceedings 31, International Center for Living Aquatic Resources Management, Penang, Malaysia.

Village-based Aquaculture Ecosystems in Sub-Saharan Africa

159

Lightfoot, C. & Noble, R.P. (1993) A participatory experiment in sustainable agriculture. Journal for Farming Systems Research Extension, 4, 11±34. Lightfoot, C., Bimbao, M.A.P., Dalsgaard, J.P.T. & Pullin, R.S.V. (1993) Aquaculture and sustainability through integrated resources management. Outlook on Agriculture, 22, 143±150. Lightfoot, C. & Pullin, R.S.V. (1995) An integrated resource management approach to the development of integrated aquaculture farming systems. In: The Management of Integrated Freshwater Agro-Piscicultural Ecosystems in Tropical Areas (eds J-J. Symoens and J-C. Micha), pp. 145±167. Technical Centre for Agricultural and Rural Cooperation, Wageningen, The Netherlands, and Royal Academy of Overseas Sciences, Brussels, Belgium. Martinez-Espinoza, M. (1996) Report of the Expert Consultation on Small-Scale Rural Aquaculture. FAO Fisheries Report 548. Food and Agriculture Organization of the United Nations, Rome. Mills, G.G. (1991) Entrepreneurship in small-scale farming in the Zomba district, Malawi. In: Aquaculture Research and Development in Rural Africa (eds B.A. Costa-Pierce, C. Lightfoot, K. Ruddle & R.S.V. Pullin), p. 10. ICLARM Conference Proceedings 27. International Center for Living Aquatic Resources Management, Manila, Philippines. Noble, R.P. (1996) Utilization of on-farm resources for aquaculture in rural Africa. In: The Third International Symposium on Tilapia in Aquaculture (eds R.S.V. Pullin, J. Lazard, M. Legendre, J.B. Amon Kothias & D. Pauly), pp. 141±151. ICLARM, Manila, Philippines. Noble, R.P. & Rashidi, B. (1990) Aquaculture technology transfer to smallholder farmers in Malawi, Southern Africa. Naga, ICLARM Quarterly, 13, 14±16. Noble, R.P. & Chimatiro, S. (1991) On-farm biotic resources for small-scale fish farming in Malawi. In: Aquaculture Research and Development in Rural Africa (eds B.A. Costa-Pierce, C. Lightfoot, K. Ruddle & R.S.V. Pullin). ICLARM Conference Proceedings 27. ICLARM, Manila, Philippines. Noble, R.P. & Costa-Pierce, B.A. (1992) Aquaculture technology research for smallholder farmers in rural Malawi. In: Aquaculture and Schistosomiasis. National Academy Press, Washington, DC. Pullin, R.S.V. & Prein, M. (1995) Fishponds facilitate natural resources management on smallscale farms in tropical developing countries. In: The Management of Integrated Freshwater Agro-Piscicultural Ecosystems in Tropical Areas (eds J.-J. Symoens & J.-C. Micha). Technical Centre for Agricultural and rural Co-operation and Royal Academy of Overseas Sciences, Brussels. Reardon, T. & Vosti, S.A. (1995) Links between rural poverty and the environment in developing countries: asset categories and investment poverty. World Development, 23, 1495±1506. Ruddle, K. (1996) The potential role of integrated management of natural resources in improving the nutritional and economic status of resource-poor farm households in Ghana. In: Research for the Future Development of Aquaculture in Ghana (eds M. Prein, J.K. Ofori & C. Lightfoot), pp. 57±85. ICLARM Conference Proceedings 42. International Center for Living Aquatic Resources Management, Manila, Philippines. Scholz, U. & Chimatiro, S. (1996) The promotion of small-scale aquaculture in Southern Region of Malawi: a reflection of extension approaches and technology packages used by the Malawi±German Fisheries and Aquaculture Development Project. In: Report of the

160

Ecological Aquaculture

Technical Consultation on Extension Methods for Smallholder Fish Farming in Southern Africa (ed. H. van der Mheen). ALCOM Report 21. Food and Agriculture Organization of the United Nations, Rome. Scholz, U.F., Chimatiro, S.K. & Hummel, M. (1997) Status and Prospects of Aquaculture Development in Malawi: A Case of MAGFAD. Paper presented at the First Regional Conference on Aquaculture, Bunda College of Agriculture, Lilongwe, Malawi, 17±19 November. Smalling, E.M.A., Fresco, L.O. & de Jager, A. (1996) Classifying, monitoring and improving soil nutrient stocks and flows in African agriculture. Ambio, 25, 492±496. van der Mheen, H. (1996) Report of the Technical Consultation on Extension Methods for Smallholder Fish Farming in Southern Africa. ALCOM Report 21. Food and Agriculture Organization of the United Nations, Rome. World Bank (1996) Malawi Human Resources and Poverty: Profile and Priorities for Action. Report No. 15437-MAI. Southern Africa Department, World Bank, Washington, DC.

Chapter 8

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems William J. FitzGerald, Jr

Introduction Mangrove forests have great ecological and economic potential. The mangroves play an important role in the ecology of the coastal zone area and in support of the marine species that utilize the mangrove ecosystem during part or all of their life cycles. The mangroves' physical location in the coastal zone transitioning between land and sea makes it a unique habitat. This intertidal zone provides physical advantages for aquaculture pond development in that tidal fluctuations allow for filling and emptying ponds without water pumping costs along with natural stocking of cultured species with incoming tides. In addition, the general low economic valuation of this land further makes the area attractive for development. Therefore, this area has been targeted by individuals and corporations for aquaculture development, which is often promoted by governments. This has resulted in heavy utilization, particularly in the past 25 years, for aquaculture development. However, growing awareness of environmental concerns and growing pressure from environmental groups have highlighted cases of misuse and the potential misuse of this unique coastal zone and the unsustainability of some current aquaculture practices. Conservation of the mangrove resource has drawn the attention of national agencies and international environmental non-governmental organizations (NGOs). There is growing pressure from international NGOs to stop the destruction of mangroves. A major target of these NGOs is the aquaculture industry and specifically shrimp culture. Some NGOs are moving towards promoting an international boycott on the importation of cultured shrimp. This international and national NGO pressure along with high profile environmental conferences (e.g. United Nations Conference on Environment and Development in Rio de Janeiro in 1992) are impacting national policies on many environmental issues including preservation of the mangrove forest areas. As pointed out by Primavera (1998a), socio-economic and environmental issues along with policy implementation and coordination need to be revised to produce a sustainable activity regarding shrimp aquaculture. A more `mangrovefriendly' type of aquaculture is identified as one means of contributing to a sustainable system in the conservation and rehabilitation of mangroves. There needs to be a balance between development pressure and conservation of the mangrove resource to allow for sustainable aquaculture development within a

162

Ecological Aquaculture

management framework of resource conservation and utilization. There is increasing pressure on the utilization of coastal resources with expanding population levels in the coastal areas. This population pressure for development and utilization of the coastal resources must be addressed within a realistic framework of conservation, resource management and development. The economic needs of the coastal populations for jobs and income must be addressed. Social issues are closely linked to sustainable development. The mangroves are a coastal resource that has been impacted substantially (e.g. clearing for villages, firewood, construction material, and agriculture) by this growing coastal population pressure. Aquaculture is one of the economic activities that has utilized the mangrove area as a resource. There are a number of ways the further conversion of mangroves to pond aquaculture can be minimized. These include the intensification of aquaculture on existing sites, the promotion of aquaculture which can be developed with minimum impact on the mangrove ecosystem, better pond siting procedures in mangroves, the integrated management of sustainable uses of the mangrove ecosystem, a shift of pond development to outside the mangroves, and non-land-based culture systems (i.e. mariculture). Each of these alternatives potentially introduces their own environmental concerns, if proper siting and best management practices are not applied. Two main sustainable alternatives to aquaculture pond development are silvofisheries and mariculture. Silvofisheries are a form of low-input sustainable aquaculture integrating mangrove tree culture with brackishwater aquaculture. As a system moves along the scale of resource use, the level of efficiency in the use of the resources is determined. The more a cultivation system recognizes and mimics the natural ecosystem functions, the less resource inputs are required and the less negative environmental effects occur (Folke & Kautsky, 1992). Silvofisheries strive to utilize this principle in a culture system with reduced resource use; avoidance of chemicals and medicinal compounds, resulting in less waste generation; and the recycling of nutrients and materials to increase the efficiency of the system. Integrated systems that make use of ecosystems without degrading the resource base on which they depend will be more sustainable and have positive contributions to the surrounding ecosystems and socio-economy. Further extension of aquaculture to meet the needs of the rural poor may be tolerable in the mangrove area, provided it is carried out in a controlled manner outside those areas already heavily exploited, and environmentally sensitive in an integrated program of conservation and utilization, such as silvofishery methods. As sustainable aquaculture can take place at all trophic levels, it is possible that wastes from one type of cultivation can be used as a resource in others, if properly managed. Sustainable food production requires management of resources and ecosystems to satisfy changing human needs, conserve natural resources and maintain or enhance the quality of the environment. The combining of different types of aquatic and terrestrial cultivation systems allows for the fuller use of the resource. This integration of systems and potential mix of products creates a challenge to determine the optimum combination of the various components and the management of the system. The challenge is to stimulate the biotic community structure for full and efficient utilization of materials and energy, so that there will be no unused and

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

163

environmentally degrading waste that accumulates, while still receiving a large harvest of economic value. Furthermore, the challenge is to move away from throughput-based systems to an efficiency-increasing development of coastal aquaculture (Folke & Kautsky, 1992).

SILVOFISHERIES Silvofisheries are a form of integrated mangrove tree culture with brackishwater aquaculture. They are a form of low-input sustainable aquaculture. This integrated approach to conservation and utilization of the mangrove resource allows for maintaining a relatively high level of integrity and biodiversity in the mangrove area while capitalizing on the economic benefits of brackishwater aquaculture. Silvofisheries practices have similarities to rice±fish farming that is practiced in inland areas of Asia (Costa-Pierce, 1992; Koesoemadinata & Costa-Pierce, 1992) in the comanagement of a terrestrial and aquatic crop within an integrated system. There are two basic silvofishery models (Fig. 8.1). One model (Fig. 8.1a, b) consists of mangroves within the pond with a ratio of 60±80% mangrove and 20±40% pond canal culture water area. The second model (Fig. 8.1c, d) consists of the mangroves outside the pond with similar mangrove to water ratio. The pond/mangrove forest (second model) orientation should be constructed with mangrove strips perpendicular to the coastline so that the flow of surface runoff of rainfall is allowed to be transported through the mangroves coastward (not obstructed by pond dikes). The advantages of the latter model, with the mangroves outside the pond, include greater manageability of the brackishwater pond, greater control, greater flexibility of culture practices, higher potential production, and lower construction costs (per production unit). It also avoids the potential toxic levels of tannin from the mangrove trees. In

Fig. 8.1 Silvofisheries models. Cross-hatched areas represent mangrove forest. (a) Standard empang parit silvofishery model with mangrove vegetation on central platform. (b) Modification of (a) with the addition of canals across the central platform area to improve water circulation. (c) and (d) Silvofishery pond unit that separates the open aquaculture area from the mangrove vegetation area. Model (c) maintains a means of water exchange through a gate, while in Model (d) there is no exchange.

164

Ecological Aquaculture

addition, it allows for natural species diversity and flushing of the mangrove vegetation with minimal disruption of the natural drainage flow through the mangroves. The disadvantage would be that the system is more susceptible to development abuse with encroaching on the mangrove area; however, that can be controlled with conditional leases and regulation enforcement. There are a variety of designs within these basic models that attempt to balance the conservation and utilization issues while maximizing economic opportunity. Figure 8.1 illustrates the various basic silvofisheries models. A number of countries are pursuing a form of silvofisheries, including Hong Kong, Thailand, Malaysia, Vietnam, Philippines, Kenya, Tanzania, Senegal and Jamaica. Some of the systems are traditional long-term practices and others are new approaches to utilize the mangrove resource in an environmentally sustainable manner. The productivity of the pond is based essentially on the use of `green manure.' The organic enrichment of the pond is from plant material, in this case mangrove tree debris. The amount of debris varies by the size and density of trees and rate of litter fall. This debris has to undergo a decay process before it becomes usable within the food chain of the cultured species. Different models have been proposed for rehabilitation/reforestation of former mangrove areas that were converted to brackishwater ponds that are not in use or have very low productivity. A discussion of the different models is presented below.

Empang parit model in Indonesia Indonesia's mangrove forests (4.25 million ha) represent approximately 25% of the world's mangroves, and Indonesia is a biogeographical center for a number of mangrove genera. However, this has declined to less than 2.5 million ha in the last 20 years due to coastal population pressures with various uses (Sukardjo, 1999). Various entities from university research programs to national programs within the Ministry of Forestry and the Directorate General of Fisheries of the Indonesian Government have been studying, demonstrating and promoting silvofisheries as part of a reforestation effort and integrated utilization of the resource to produce aquaculture products. These silvofisheries developments range in size from one hectare to thousands of hectares at each site. Empang parit is the traditional application of this integrated aquaculture in the mangrove area. Empang parit is a silvofisheries model (sometimes referred to as `Tambak Tumpang Sari') that is being promoted in Indonesia. The Southern Sulawesi Province Fisheries Office has three demonstration empang parit sites that have been initiated in the Luwu District. This is part of a national program to promote silvofisheries through the Directorate General of Fisheries Office. The Mangrove Rehabilitation and Management Project in Sulawesi (Ministry of Forestry) has empang parit demonstration sites in Luwu and Kwandang. As part of the Island Sustainable Livelihood and Equity Program (University of Hasanuddin), there is a community development project that includes a demonstration empang parit project in Sinjai, Sulawesi. Mangrove reforestation in Lampung consists of silvofisheries projects (Nurdjana,

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

165

1997). In addition, there are large-scale silvofisheries programs in Cikiong and Blanakan in West Java. Unfortunately, controlled production trials, with the collection of production data and inputs and an economic analysis of these systems, have been very limited. Three main silvofishery pond designs are being utilized or promoted in Indonesia (Fig. 8.2). Each design has specific advantages. The empang parit model represents the greatest level of reforestation or maintenance of existing forest to pond area. It represents a resource multi-use that increases food supplies and contributes to the socio-economic well-being of coastal rural populations (Sukardjo, 1989). One model is illustrated in Fig. 8.2a. It essentially consists of an unexcavated central platform pond bottom (80% of total pond area) that alternates between being flooded and exposed as the water of the pond is raised or lowered. A canal that runs adjacent and parallel to the pond dikes surrounds this central platform pond bottom. The canal is normally 3±5 m wide and excavated 40±80 cm below the central platform pond bottom. Fish, shrimp, and crabs are cultured extensively in the canal and can enter the central platform area during periods of flooding. The density of the planted mangrove trees on the platform area ranges from 0.17 to 2.5 trees/m2 in the empang parit system. The density influences the quantity of litter production and organic load in the pond along with other factors of cultivation. This would include the diversity of non-mangrove flora (e.g. algae) and fauna growth that may form an important part of the aquaculture species' diet. The tree density also Design 1 C B

D C A Brackishwater

Design 2 A D

C

C

B A A

A

Brackishwater

Design 3 B A

Fig. 8.2 Illustration of the three main designs for silvofishery ponds in Indonesia. Pond gates (A), pond dikes (B), open water aquaculture areas (C), and the mangrove areas (D) are identified.

D

A

C

Brackishwater

166

Ecological Aquaculture

influences the aquaculture species' production, with farmers preferring a more open density (approximately 0.2 trees/m2) for milkfish. The greater openness of the forested area allows accessibility to the platform area in milkfish culture while a greater tree density can be used for shrimp and mangrove crab culture that prefer additional structural habitat and shelter afforded by the mangroves. The empang parit model has a number of advantages and disadvantages compared with a brackishwater pond or open pond that must be considered in planning, developing, and utilizing silvofishery activities as part of an integrated coastal zone management program including the following: Advantages: low impact on mangrove forests; capturing economic value of the mangrove ecosystem; employment in rural low-income coastal communities; low capitalization and operation costs; a managed sustainable production system; increased efficiency of an integrated system. l Disadvantages: greater construction cost per unit of culture area; greater difficulty to manage; reduced water circulation and greater potential for stagnant areas with low oxygen levels; mangrove trees reduce the penetration of sunlight to the ponds lowering the productivity of phytoplankton and benthic algae; potential toxicity of tannin from mangroves (varies with species); limitation on species cultured (e.g. seaweed would be shaded by trees, reducing growth). l

The second design illustrated in Fig. 8.2 is a modification of the first with the addition of a secondary dike around the central platform. In addition, gates are added to allow for direct control of the inflow/outflow of water from the external water delivery canal. A second dike allows for the inflow/outflow of water to the pond's perimeter canal (open water area for the cultivation of the aquaculture species). This system allows for the separate management of the water level for the mangrove and the open water aquaculture areas of the pond, thereby allowing the optimum level and inundation frequency and duration for both sections. The third design (Fig. 8.2) completely separates the two components of the silvofishery pond. The mangrove area is a separate portion of the pond with its own gates for water level and inundation duration control. Similarly, the open water aquaculture portion has separate gates but shares one common gate with the mangrove component to allow for the flushing of detritus and nutrients from the mangrove area into the open water area. Management options of the open water aquaculture component are increased with similar options afforded by an aquaculture pond.

Sinjai ± South Sulawesi The empang parit operation located in Tongke-Tongke, Samataring, Kecamatan Sinjai Timur (South Sulawesi) was constructed in 1994 (Fig. 8.3(i)). It is a cooperative project of the Ministry of Forestry (Province Office), District Government and the University of Hasanuddin. This evolved out of a community-initiated mangrove replanting program that started in 1984. The replanting consisted mainly of Rhizophora (85%) and minor planting of Avicennia, Bruguiera, and Sonneratia covering an

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

167

Gate Canal

5m Central platform Mangrove trees Mangrove tree seedlings planted to reduce dike erosion

Pen

Mangrove crab pen

Gate Fig. 8.3(i) Empang parit pond layout in Sinjai.

area of 559 ha. The replanting was an effort to stop the increasing coastal erosion of the fishing village. The empang parit pond was constructed within an area that had been part of the planted mangrove. The Rhizophora mucronata previously planted area consisted of 11-year-old trees that were planted 0.5 m apart (Fig. 8.3(ii)). This provided a very dense growth (2.5 trees/m2) that had to be thinned to a density of 0.6 trees/m2 to accommodate the use (pen culture) of the central platform. The Sinjai empang parit design and operation are advanced compared with other demonstration or private efforts implementing the concept of empang parit within the Mangrove Rehabilitation and Management Project in Sulawesi project sites. However, it exemplifies the

Fig. 8.3(ii) Rhizophora mucronata densely planted (2.5 trees/m2) on the central platform of the silvofishery pond in Sinjai. The denseness stunted growth of the 11-yearold Rhizophora and had to be thinned (0.6 trees/m2) to allow for crab pen construction.

168

Ecological Aquaculture

standard or `traditional' empang parit model. The fuller utilization of the pond to maximize the system's production of aquaculture products is being attempted. The pond is 1 ha in size with two wooden gates. The screened gates are left open all the time to allow the water level inside the pond to fluctuate with the tides. The canal that runs around the inner perimeter of the pond is 5 m wide and has a maximum depth (below the central platform area) of 1.0 m and a minimum of 0.7 m (average 0.8 m). The tide range in the pond is 50 cm (relatively small tidal range which reduces pond flushing capacity). The central platform has a water depth range of 50 cm to complete exposure at the lowest tide, but the average depth is 20±30 cm. A cross section of the pond dike, canal and central platform is illustrated in Fig. 8.4.

Maximum water level

Dike

Central platform mangroves

50 cm

80 cm Canal 5 m wide Fig. 8.4 Cross section of empang parit pond at Sinjai.

Potential problems with the system include the following: l

l

l

l l

l

The two gates are located at corners on the same side of the pond. This results in reduced water flushing especially along the opposite canal and would tend to have a greater build-up in organic matter on the bottom and potential stagnation of water with lowered oxygen levels. The mangrove trees are extremely dense in the central platform area. These will contribute a large amount of organic matter to the pond. With the reduced flushing of water in the pond, this has the potential of high BODs and reduced oxygen levels in the pond. The construction of the pens for mangrove crab culture in the entire central platform area will further add to the organic matter and associated decomposition of by-products. The additional input of 5% of the mangrove crab weight daily as trash fish may increase the BOD to a detrimental point and build up hydrogen sulfide in the pond bottom. With the tidal height range of only 50 cm in the pond, this reduces water exchange potential. The pond canals cannot be completely drained, since the bottom of the canal system is below the lowest tide level. This results in greater stagnation potential and eliminates the periodic drying and oxidizing of built-up organics in the pond bottom. There are large amounts of organic debris in the constructed pond dike that make the dike susceptible to shrinkage, leakage, and erosion.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

169

The pond is stocked naturally with juveniles of species entering with incoming tides through the screened gates. The species occurring in the wild stock include siganids (Siganus sp.), mullet (Mugil sp.), milkfish (Chanos chanos), tilapia (Oreochromis sp.), shrimp (Penaeus monodon, P. merguiensis, and Metapenaeus sp.), mangrove crab (Scylla serrata), jacks (Caranx sp.), and seabass (Lates calcarifer). These would be harvested by use of a gill net during the low tide when the fish and shrimp are concentrated in the perimeter canal. Hasanuddin University is conducting a cooperative research project with the Ministry of Forestry and the District Government to develop the aquaculture production of the empang parit system. The Hasanuddin University project is part of a multifaceted community empowerment development program, called Island Sustainability, Livelihood and Equity Program, with international counterparts. One of the innovative approaches to maximize total production from the empang parit system is the addition of mangrove crab (Scylla serrata and S. oceanica) pen culture (stocking density of 3 crabs/m2, with an average seedling size of 70 g). Initially 12 pens (10 m 6 10 m; 100 m2) were constructed within the central platform area (8 pens, Hasanuddin University; 4 pens, District Government). They intend to construct additional pens to fill the platform area. This will result in 36 pens on the platform when completed with a total culture area of 3600 m2 (Figs 8.5, 8.6). The pens are spaced 1 m apart. There are two smaller pens that are constructed in the canal. These two pens are part of a Fishery Office experiment and are stocked at a density of 16.6 crabs/m2. It was indicated that the crab culture in the central platform area provides better water flushing conditions compared with the pens located in the canal which has approximately 80 cm of water depth that is below the lowest tide level and the bottom of the gate. Other options should be explored to maximize production from the system to make it more attractive to private landowners as a viable option for their less productive brackishwater pond areas.

Fig. 8.5 Silvofisheries pond with multiple crab pens on planted mangrove area of central platform, Sinjai.

Two alternative innovative designs of the empang parit model are being planned. One consists of reducing the pond dike construction to a low wall 50 cm in height. On top of the dike a net would be utilized to increase the effective height of the dike to prevent the escape of cultured species while allowing water exchange during high tide. There would be no gates constructed. The low earthen dike would control the minimum water level in the pond. In addition, there would be no canal excavated around the inner

170

Ecological Aquaculture

Fig. 8.6 Crab cage placed in perimeter canal of silvofisheries pond, Sinjai.

wall of the dike. An area would be cleared that would normally be the canal area. The central mangrove planted area would be the same as in the former empang parit design. The advantage of this model is lower construction cost with the decrease in dike height and elimination of the gates. In addition, a greater water exchange and flushing would occur that would minimize the problem of water stagnation and the potential of low oxygen levels resulting in mortality of cultured aquatic species. However, a mechanism for draining the pond (e.g. gate or standpipe) will still be needed. In addition, there will be problems with the extended submergence of the mangrove trees. A second design is to eliminate the pond dikes completely and to construct the bamboo crab pens within the standing mangrove forest. The pens would be for the culture of mangrove crabs as described above. The advantage of this design is the total elimination of the substantial capital investment cost associated with the construction of the pond dikes, canal, and gates. Furthermore, it further reduces the impact on the mangrove forest. The disadvantage would be the elimination of the production associated with the canal area of the impoundment.

Cikiong and Blanakan ± West Java Cikiong and Blanakan are part of a government mangrove management and rehabilitation program (Soemodihardjo & Soerianegara, 1989) that includes silvofisheries, which is administered by the Perum Perhutani (State Forestry Company). They exemplify what can be accomplished in mangrove rehabilitation and management within a controlled and enforced program. All silvofishery farmers must sign a conditional lease. Cikiong has 6600 ha of silvofishery brackishwater ponds with 1508 farmers operating the silvofishery farms. All farms use the traditional empang parit model with an 8:2 ratio. The State Forestry Company has experimental projects that modify the design. The modification consists of an additional dike around the central mangrove platform of the pond with separate gates for the canal and mangrove portion of the pond. This would allow the water level to be controlled separately for the trees (which cannot tolerate being submerged for extended periods) and the canal (to maintain a maximum water level during culture of the fish) (Figs 8.7, 8.8).

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

171

Fig. 8.7 The perimeter canals of two adjacent silvofishery ponds showing the dense mangrove central platform area of a mature silvofishery pond. The canals are the main aquaculture area within the silvofishery system with species entering the mangrove tree platform central area during the tidal flooding periods.

Fig. 8.8 Illustrating a Cikiong (West Java) silvofishery demonstration pond with an elevated walkway through the central platform area of the pond. It shows a mature mangrove forested area forming a complex ecohabitat with the pond's platform bottom and the mangrove vegetation. The area can provide natural food and habitat for the cultured species.

Shrimp are harvested daily using a bamboo trap with a kerosene lantern during the evening hours. Ponds at the Cikiong silvofishery reserve average 1 kg of shrimp per ha per night. This mainly consists of Metapenaeus ensis (Figs 8.9, 8.10). The Blanakan silvofishery has an area of 5300 ha of silvofishery brackishwater ponds with 2060 farmers. The majority are the traditional empang parit; however, also included are silvofishery models similar to Type Ib (Fig. 8.1). There are also ponds with different mangrove to water area ratios up to fully cleared brackishwater ponds.

Fig. 8.9 Placement of shrimp trap in pond by drainage sluice gate. Netting is used to guide shrimp into the trap that contains a kerosene lamp to attract the shrimp to the light in the evening.

172

Ecological Aquaculture

Fig. 8.10 Close-up of shrimp trap construction.

The annual rental fee is US$90/ha (220 000 Rp/ha) for brackishwater ponds (no mangroves), US$37/ha (90 000 Rp/ha) for 1:1, and US$12/ha (30 000 Rp/ha) for 8:2. The main culture species is tilapia. A study (Anon., 1991) on the West Java silvofishery operations compiled production data for the different locations, which is summarized in Table 8.1. Table 8.1 Production (kg/ha/yr) from empang parit locations in West Java (Anon., 1991) Location

Bogor ± Tanggerang Bogor ± Ujung Karawang Purwakarta ± Cikiong Purwakarta ± Pamanukan Indramayu ± Indramayu

Total empang parit area (ha) 1113 7934 6268 4263 6421

Tilapia

Milkfish

Trash fish

Shrimp

Crab

1500 1500

700 600 600 500 500

200 200 250 50 50

200 200 250 300 300

1

Full economic analysis of silvofishery operations is limited. Annual profit from milkfish and shrimp polyculture silvofishery was reported to be US$2091/ha/yr (5 122 000 Rp/ha/yr) (Anon., 1991). However, these figures do not account for all costs of production (e.g. hired labor, annual lease fees, construction or purchase cost depreciation, supplies, and equipment costs); therefore, they provide a misleading, overly optimistic profitability of the silvofishery system. Furthermore, the production levels (production per unit area) that the economic analysis is based on are unrealistic for the stocking level and the area in production under extensive culture practices. This illustrates the need for a careful analysis of silvofisheries operations utilizing

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

173

actual production numbers and all costs of production accounted for to provide a realistic evaluation of the different silvofishery systems. A more complete economic evaluation was conducted by Widiarti & Effendi (1989). This study evaluated silvofishery operations under the Perum Perhutani at Blanakan and Cangkring in West Java. Results of this study provided an annual net income of US$248/ha (608 500 Rp/ha) for farms at Blanakan and US$408/ha (1 000 700 Rp/ha) for farms at Cangkring. These farms utilize the traditional empang parit model with a 8:2 ratio of mangrove to water channels. Therefore, a reasonable expectation of annual net income from this model type would be the average of the above two sites which is US$328/ha/yr (804 600 Rp/ha/yr). Approximately 50% of the farmers' income was attributed to the empang parit in the above study with the balance from agricultural or other sources of income. The difference in net income between the two sites was mainly attributed to the species cultured. Blanakan farms mainly produced tilapia while the Cangkring produced mainly milkfish. Milkfish had a market value approximately 33% higher than tilapia. In a study of former shrimp ponds converted to the silvofisheries system in Central and Western Java, the net profit ranged from $199 to $2,280/ha/yr with the polyculture of milkfish and shrimp being the most profitable (FitzGerald & Savitri, 1997). The Forest Management Division (Bagian Kesatuan Pemangkuan Hutan) of the State Forestry Corporation (Perhutani) conducted a financial analysis for production from different species with an empang parit 8:2 system, which is summarized in Table 8.2. The mangrove crab, seabass (Lates calcarifer), tilapia/chicken and milkfish/ shrimp production analyses are from the Cikiong silvofisheries operation. The milkfish monoculture (for food and bait) is from silvofisheries in Cibuaya operations (Karawang District, West Java). On a per unit production basis, the mangrove crab culture outperformed all other species by a substantial margin. This would justify increased research on the various aspects of mangrove crab culture (e.g. larval culture, grow-out, optimize cost-effective feed conversion, species differentiation and hybridization, etc.). As in the financial analysis reported above, all costs were not included; however, the results are useful to point out the magnitude of difference in the potential economic return among the species considered. This assists in designing a production strategy for an empang parit system that improves the economic return to the farmer. Table 8.2 Value of production (2450 Rp = US$1.00) from empang parit systems, Cikiong and Cibuaya (Anon., 1994, 1995) Milkfish Tilapia with Milkfish and monoculture for Seabass chicken coop shrimp food and bait Mangrove crab (Rp/60 m2 cage) (Rp/ha, empang (Rp/ha, empang (Rp/ha, empang (Rp/ha, empang parit 8:2) parit 8:2) parit 8:2) parit 8:2) Annual net profit (ha/yr)

$1367 3 350 000 Rp

$1347 3 300 000 Rp

$2601 6 372 000 Rp

$2508 6 144 000 Rp

$1322 3 240 000 Rp

Net profit per unit area (m2/yr)

$22.79 55 833 Rp

$0.13 330 Rp

$0.26 637 Rp

$0.25 614 Rp

$0.13 324 Rp

174

Ecological Aquaculture

An abbreviated interview of farmers by the Mangrove Rehabilitation and Management Project in Sulawesi (FitzGerald & Sutika, 1997) at both Blanakan and Cikiong (West Java) silvofishery operations showed an average gross income of US$580/ha/yr (1 419 859 Rp/ha/yr) (ranging from US$313 to US$946/ha/yr (766 200 to 2 318 200 Rp/ha/yr)). The net profit was an average value of US$376/ha/yr (921 632 Rp/ha/yr). The variation in net income per farm was less with a range of US$943± 1558/farm/yr (2 310 400±3 816 400 Rp/farm/yr) (average of US$1283/farm/yr (3 143 454 Rp/farm/yr)). The individual farmers operated between 1.5 to 10 ha of silvofisheries ponds. This may indicate a greater production effort per unit area by farmers with smaller farms. In addition, the individual with 1.5 ha of ponds supplemented this income by obtaining approximately 40% of his total income from rice fields. The individual with 10 ha of silvofishery ponds obtained 100% of his income from the silvofishery operations. This abbreviated assessment of production and costs at least provides a more reasonable evaluation of silvofisheries production potential. However, further research is needed to gather a fuller assessment and an evaluation of the different silvofishery models.

Gei wai model ± Hong Kong Gei wai is a traditional form of silvofisheries used in China and currently maintained at Hong Kong's Mai Po Marshes Nature Reserve administered by the World Wide Fund for Nature, Hong Kong. The system evolved out of brackishwater rice culture, which was a single, low-yield crop. The flooding of fields and harvesting of shrimp was originally a supplemental food source. Modified gei wai ponds are operated at locations in China around Deep Bay. The Mai Po gei wai ponds were constructed in the 1940s. The gei wai ponds in Mai Po Marshes Nature Reserve are shallow intertidal ponds with wide channels 1 m deep with planted island platforms in the central part of the pond. The mangrove area in the gei wais is generally long narrow strips with cleared channels between them. The mangrove density varies and some area is marsh grass (Young, 1996). The ponds are large in size with a length of 1 km by 100 m wide (10 ha pond). This follows the Fig. 8.1b model of silvofishery. It is similar to empang parit; however, canals cross the forested mangrove area within the central platform of the pond. This improves water circulation and reduces the risk of stagnant areas in the pond. It also reduces the mangrove to water area ratio to approximately 2:8 to 2:1 thus providing a larger water culture area (Young & Wong, 1994). There are 24 Mai Po gei wai ponds, with a total area of 272.1 ha. Of this area, mangroves within the gei wai ponds occupy 17.5 ha, reeds 46 ha, and open water 208.6 ha (Young, 1996). This gives an average of 23% of the area vegetated, with the remaining 77% in open water. This is a low proportion of area in macrophytes for silvofisheries ponds (Fig. 8.11). Gei wai ponds were constructed in intertidal wetlands that are enclosed by an earthen dike with a sluice gate at the seaward end. Ponds are typically approximately 10 ha in size. In the pond there are deeper perimeter channels along the dike and channels crossing the central platform area of the pond. Characteristic features of the gei wai are the shallow platform in the center of the pond (characteristics associated

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

175

Fig. 8.11 Aerial view of gei wai ponds at Mai Po Nature Reserve, Hong Kong (photo by William Heering). This shows the diverse shapes and sizes of the platform areas within the gei wai ponds crossed by canals.

with the high intertidal zone of the mangroves), which are specific to this location. This is usually an undisturbed area that maintains its natural level and vegetation (mangroves or reeds), and is subject to submersion and immersion with the tidal inundations controlled through the sluice gate. This vegetated platform area is usually submerged a few centimeters when the pond is filled. The sluice gate is the major management tool used to control the water level, water quality, water temperature, natural stocking of the ponds, exclusion of predatory or undesirable aquatic species, and harvesting of the ponds. There are two different types of gei wai as identified below with associated characteristics (Irving & Morton, 1988). Traditional gei wai: Platform area fluctuates between emersion/submersion with frequent controlled tidal fluctuation of the pond level; frequent partial harvests up to 100 times per year with spring tides; frequent repeated draining and flooding; only natural stocking with tidal inflow; no fertilization; no supplemental feed. l Modified gei wai: Average water level maintained at a greater depth; lower salinity (greater influence of freshwater); ponds are kept semi-permanently flooded; vegetation on central platforms shifts from mangroves to marsh reed, Phragmites communis; ponds are fertilized (e.g. meal of soybean and rolled ground nuts, chicken manure, etc.); ponds are artificially stocked with shrimp postlarvae and fish; shrimp are fed artificial feeds; harvest occurs only twice a year. l

Ponds are stocked naturally by incoming tidal water (no supplemental stocking of seedstock) resulting in a polyculture of shrimp and fish. The shrimp species that occur in the Mai Po gei wai ponds include Penaeus monodon, P. merguiensis, P. penicillatus, Metapenaeus ensis, M. affinis, M. burkenroadi, Macrobrachium nipponense and Palemon orientalis. However, fish species are also introduced with the natural stocking and these include mainly mullet (Mugil cephalus), but also yellow-fin sea bream (Sparus berda), sea bass (Lates calcarifer), bighead carp (Aristichthys nobilis) and grass carp (Ctenopharyngodon idellus) (Young, 1997). Natural pond productivity and debris from the mangroves form the base of the food source. Harvests are divided through the year-long production cycle (50±80 separate partial harvests, April± October). Production previously reached 1900 kg/10 ha/yr (190 kg/ha/yr). However,

176

Ecological Aquaculture

the system has been experiencing declining production as illustrated in Fig. 8.12, which is attributed to agricultural pollutants and deteriorating water supply quality. The decline in production of traditional shrimp gei wai is attributed to the following (Cha et al., 1997; Young, 1997): Conversion of ponds to fish cultivation or urban development activities that were economically more beneficial. l Introduction of tilapia, which were originally a nuisance fish, but are now becoming more accepted commercially. l Siltation of the ponds having the effect of reducing their capacity to produce. l Increased level of pollution in the adjacent bay that killed the juvenile shrimps causing an overall reduction in the number of shrimps entering the gei wai. Pollution also impacted on the harvest, since the sluice gates could not be opened during the initial days of the spring tide because pollutants and debris would be washed into the ponds from the surrounding rivers and estuaries. l

45 40

kg/ha/yr

35 30 25 20 15 10

1990

1991

1992

1993

1994

1995

Year Fig. 8.12 Annual mean shrimp production (kg/ha/yr) from seven gei wai ponds at Mai Po Nature Reserve, Hong Kong (after Cha et al., 1997).

The Mai Po gei wai ponds are characterized by a wide range in salinity from 2.5 to 29%, and seasonal temperatures ranging from an average minimum of 13.5 to an average maximum of 31.28C, large sediment load with sediment accretion of platform area and reduction of canal volume, high zoned mangrove with low frequency and duration of platform tidal submergence, and large allochthonous nutrient input. In addition to the environmental benefit of maintaining a level of mangrove vegetation, the gei wai ponds were reported to enhance wildlife, mainly through providing a nesting area for birds (Cha et al., 1997). A summary of key points in the operation of the Mai Po gei wai follows. l

Nutrient source: Carbon tagging trials were conducted by Dr S.Y. Lee to determine the main source of C in the invertebrate population from the adjacent Deep Bay as well as the shrimp harvested from the gei wai ponds. This was to see if the

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

l

l

l

l

177

hypothesized main source of carbon was being provided by the leaf litter from the mangrove/grass vegetation grown on the raised platform areas within the ponds. The results indicated that the major source of C was from the wastewater source associated with domestic and livestock sewage sources introduced into the water supply. Production: The species of shrimp harvested is Metapenaeus ensis. The gei wai shrimp used to receive a price premium in the Hong Kong market. However, the average inflation adjusted market price has declined since 1990, with increased competition from shrimp production from neighboring parts of China entering the Hong Kong market. In addition, with the pollution of the surrounding water supplies by domestic and livestock sewage the price received for the shrimp has declined in recent years. Harvests: The shrimp harvest from a 10 ha pond under current operations amounts to approximately 15 kg/night. The harvest is for 30±40 nights per year, amounting to 45±60 kg/ha/yr production. Previously, when the gei wai ponds were operated privately they were harvested more frequently, 50±80 nights per year. Fish from the ponds are not harvested for commercial sale (they are often utilized by migratory birds). Production has been declining over recent years due to the increase in pollution levels in the adjacent waters. Management: The staff of the Nature Reserve operate the gei wai ponds. The revenue raised through sales of the production from the ponds covers most of the personnel costs associated with the operation of the ponds. Mangrove/vegetation platform: The majority of the platform areas are covered with the common reed grass, Phragmites communis (Figs 8.13, 8.14). Growth consists of tall dense vegetation covering the raised platform areas of the ponds. In areas where grass is established, mangroves are generally absent and do not establish growths of new trees. The presence or absence of reeds depends on the past management of the ponds, i.e. whether they had at one time been converted to permanent fishponds. The high water level drowning the mangroves was followed by reeds becoming established while accretion and reduced salinity also contributed to the invasive growth of reeds. These ponds were later converted back to shrimp farming but mangroves found it difficult to re-establish because of the thick growth of reeds and lack of a source of propagules.

Fig. 8.13 The proliferation of Phragmites communis provides massive detritus input; however, it depletes the gei wai of open water area and accelerates accretion. This eventually leads to replacement of the mangroves with transitional vegetation associated with freshwater wetlands and upland forests with less tidal inundation influence. Therefore, it impacts on the fishery production and wildlife conservation objectives of the pond. It reduces habitat heterogeneity and has been found to reduce the suitability for waterfowl use in Mai Po Nature Reserve (Lee, 1990b).

178

Ecological Aquaculture

Fig. 8.14 Gei wai, Mai Po Nature Reserve, Hong Kong, showing a modified gei wai that has the central platform area dominated by the marsh reed Phragmites communis.

Siltation and sedimentation: Silt, transported down the Pearl River, is accumulating on the bay edge of the area with expansion of the mangrove growth by approximately 5 m bayward per year. In the most recent decade, the siltation rate in Deep Bay has increased to 1.7 cm per year. This heavy silt load in the water causes problems in the operation of the gei wai ponds, since they fill with the settling of the sediment. This requires major pond rehabilitation with dredging of the perimeter and cross channels of the ponds approximately every 10 years. During this period the growth of the perimeter and platform vegetation encroaches on the pond water channels thus reducing the culture area and subsequently the production from the pond. The sedimentation on the platform area of the pond has a detrimental impact on the mangrove vegetation in that the floor is building up, resulting in less frequent inundation by water and turning the platform into a raised wetland area not suitable for mangroves. Terrestrial species begin to invade these areas. l Exotic species: A problem occurs with climbing vines that were introduced to the Hong Kong area and have become established in the reserve. Invasive climbers such as Derris trifoliata along with exotic climbers such as Mikania micrantha have become established. The vines cover mangrove trees and reduce their growth, and can eventually kill the trees through excessive shading if vine growth is not checked. Without labor-intensive management, the climbers eventually cover and kill the mangrove trees. Current limited effort to control the vines is through removal by hand. l Impact of adjacent area: The development pressure from adjacent areas has impacted on the environmental quality of the Mai Po gei wai. Shenzhen City, which is adjacent to Mai Po on the Shenzhen River and Deep Bay on the China mainland, is a special economic development zone. The Shenzhen Special Economic Zone has developed very rapidly from a town of some 300 000 to a city of 3±4 million. This accelerated growth is a major source of pollution to the Bay, which impacts on the Mai Po Nature Reserve. There is insufficient sewage treatment from Shenzhen and much of the waste from the city goes untreated into Deep Bay. The surrounding area outside the reserve consists of traditionally l

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

179

operated deep-water ponds (culture of fish) and in recent decades they have been filled in for commercial buildings (warehouses, apartment buildings, etc.). Some of the ponds that remain are being intensified. There are no commercially operated private gei wai ponds, since they are no longer considered economically viable. The gei wai ponds at Mai Po are marginally productive of commercial aquaculture species. The production has been declining for a number of reasons as cited above. However, they have also become physically impaired due to the siltation loads essentially elevating the central platform areas over time to the point that tidal inundations are infrequent resulting in minimal export of mangrove detrital material into the pond water column (Lee, 1990a). In addition, the wetland macrophyte vegetation is in a state of transition to a terrestrial form. This all impacts on the productivity of the ponds. The value of the gei wai has shifted from aquaculture to mainly a bird sanctuary and nature habitat reserve that is utilized for the purposes of education and research on mangrove wetlands.

Integrated mangrove/crab pen culture ± Sarawak, East Malaysia In western Sarawak, Trusan Jaya Sematan, a pen culture system for mangrove crabs within the mangroves, has been developed with government assistance (Chang, 1997). Crab farming was originally intended to supplement the income of coastal fishermen; however, some individuals have adopted it as the main income source. The labor requirement to operate a pen is minimal with 1±2 hours per day for inspection, cleaning and water exchange as needed. A farmers' organization was formed, which assists in the purchase of seedstock, trash fish, and marketing, to communally operate a number of the pens while individuals operate other pens. Mangrove crabs are cultured within the intertidal mangrove forest in pens that are 18 m 6 9 m (162 m2) with walls constructed of wooden stakes (local coastal palm ± Oncosperma tigillaria). This is an environmentally low impact system that maintains the natural tidal inundation cycle for the enclosed mangroves. The mangrove vegetation is maintained within the pen except for a perimeter drainage channel (0.3 m wide 6 0.3 m deep). A PVC standpipe serves as the inlet/outlet. The pens are located within the intertidal zone to provide flushing at high tides. During neap tides the water exchange is done with a water pump (Fig. 8.15). The stocking material is caught from the wild. Species cultured are Scylla tranquebarica (dominant) and S. olivaceous. From observations, S. olivaceous has a faster growth rate (Chong, personal communication). The pen is stocked with 1000±1500 crabs (6.2±9.3 crabs/m2). The crabs are fed every other day at high tide with trash fish (approximately 600 kg/cycle). The feed conversion rate ranges from 2.3 to 6. Water is exchanged daily with the tides through the PVC drain and through the wooden stakes on higher tides. Water in the perimeter channel is completely drained and refilled weekly. The grow-out period is 4±7 months with an average of 5.2 months per cycle. There is periodic restocking (2±3 times per month, 50±100 crabs/restocking,

180

Ecological Aquaculture

Fig. 8.15 Crab pens constructed within the standing mangrove forest, Sarawak, East Malaysia (photo by Chanyut Sudtongkong).

approximately 100 g size crabs) to maintain the original stocking density. Partial harvesting is done twice a month during high tides with scoop nets or traps. Crabs are starved for 2 days before harvesting to facilitate capture in traps. Harvest size is 300 g/ crab or greater (3 crabs/kg). Production averages 265 kg/pen/cycle with a mortality rate of approximately 47%. The pen construction cost is $US334 (RM1270; conversion rate of $US1 = RM3.80) with a pen wall life of 5±6 years and a wood plank walkway life of 1±1.5 years. The operating costs per cycle (exclusive of operator's labor) are $US105 (RM400) per cycle. With a farmgate price of $US1.58±1.84 (RM6±7)/kg the gross profit is $US558 (RM2120)/pen/cycle or a net operating profit of $US453 (RM1720)/ pen/cycle ($US86.83 (RM330)/pen/month). One fisherman can operate three or four pens. This would provide an annual income (assuming four pens) of approximately $US4168 (RM15 840). This significantly enhances the economic condition of these coastal resident fishermen which previously would average an income of $US1895 (RM7200) per year. There is no differentiation in the farm-gate price for male or female crabs. However, there is a differentiation based on sex and size in the retail price, which ranges from $US2.63 to 4.21 (RM 10±16)/kg, with females receiving the higher price (Chong, personal communication; and Chang, 1997). Export retail markets in Singapore, Hong Kong and Taiwan are as high as $US14/kg for gravid females (Overton & Macintosh, 1997). The holding and fattening, particularly of females until they carry eggs, is normally carried out as a separate stage along the market channel. It is a step that should be considered for integration into the pen culture system so that the farmer can capture a greater economic benefit. This would be particularly suited to those pens operated under a farmers' cooperative, since the marketing effort could be more efficiently and effectively pursued. The culture system has a minimal impact on the mangrove ecosystem. Mangrove trees are removed only in the area of the pen wall, walkway and canal while maintaining the natural growth of mangrove vegetation within the pen. There is no use of chemicals or other detrimental practices. In addition, the crabs mature and become berried within the pens and it is believed that there is recruitment of crab

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

181

larvae on the outgoing tide to the surrounding mangroves from the pens with the release of larvae from these berried females (Chong, personal communication). The main constraint to the further development of this culture of mangrove crabs is the availability of juvenile crabs. Farmers are encouraged to grow the crabs to a larger size (2±3 crabs/kg). Attempts have been made by the Sarawak Department of Agriculture along with other research centers in Southeast Asia to rear the crabs through the larval stages in a hatchery. However, this has been unsuccessful. Resolving this constraint will open further development opportunities for mangrove crab culture. In addition to the method described here, there was an attempt in central Sarawak to develop a system replacing the wooden pen with the use of netting. However, the project was discontinued due to poaching problems. This system had the advantage of lower capital costs and of eliminating the need to remove any mangrove trees.

Shrimp±mangrove model ± Thailand Thailand is considered to be the most advanced in the overall application of higher intensity shrimp culture in Southeast Asia. However, it has areas of low production or abandoned shrimp ponds. Efforts have been initiated to integrate the recovery of former mangrove areas while utilizing these existing ponds in a productive manner. A silvofishery model utilizing a 3:7 ratio of mangrove to water area is being applied. Since in the Thai shrimp pond design the pond bottom is flat (approximately 1±1.2 m deep) and does not have the central platform area characteristic of Indonesian tambaks, the raised dike bands within the pond have to be added for the planting of the mangrove trees. The latter (usually Rhizophora apiculata) are planted along the side of the dike bank at the water interface approximately 20±30 cm from the dike. This also helps to reduce erosion of the dike bank. The trees add organic matter (litter) to the pond for increased natural food production and provide structural complexity to the dike bank area, which can provide greater shelter to the shrimps, particularly during molting periods. The design of the pond with the planted mangrove area is illustrated in Fig. 8.16. This design is a modification on the basic silvofisheries model previously presented as Type b in Fig. 8.1. The Thai version of the silvofishery pond can be modified to place the narrow raised earthen bands in different arrangements (e.g. parallel bands) from that illustrated in Fig. 8.1. In addition, the design provides an increase in the dike bank ratio to the total water area that is preferred in shrimp culture. The silvofishery pond is operated as a semi-intensive shrimp operation (including aeration and supplemental feeding) instead of an intensive system. The application of this silvofishery model is in the early stage of development and is mainly located in central southern Thailand in the Phetchaburi area. Application of this silvofishery pond design to an integrated shrimp±mangrove model is intended to allow for the gradual integration of mangrove tree cultivation at a lower density and ratio within shrimp ponds and reduce the aversion private farmers might have to a higher level of conversion to mangroves.

182

Ecological Aquaculture

Fig. 8.16 Thailand silvofishery pond model (Tanan & Tansutapanich, 1999).

Shrimp±mangrove model ± Vietnam Impounding of large mangrove areas with natural tidal stocking of wild fish and shrimp is a traditional and widespread technique used by individual farmers as well as by large state farms (Anon., 1998). Several hundred hectares have been established in the Mekong Delta since the late 1970s. State farms are allocating plots of land to households for long-term mangrove±shrimp culture systems with leases of 5±10 years. Canals are dug between plots of forest and are used for traditional or improved extensive brackishwater culture. The majority of the silvofisheries activity is in southern Vietnam with large areas of farm development of several thousand hectares (Vu, personal communication). Farms are usually allocated in 3±10 ha mangrove plots for a 20±25 year lease from the government. The ratio of mangrove to non-mangrove area is 70:30, which is directed from the ministerial level of government; however, this ratio is not always complied with. There are two basic types of silvofishery ponds. Mixed shrimp±forest: This design has a lower tree density of 1000±2000 trees/ha (0.1±0.2 trees/m2). This more open forest allows for greater light penetration for benthic algae and phytoplankton growth. The farm size is 5±10 ha. Hatcheryproduced shrimp (Penaeus monodon) are stocked at a density of 0.5 shrimp/m2 with a culture period of 4±5 months. The production is 200±250 kg of shrimp/ha/ cycle. Secondary species including fish and shrimp from natural wild stock are also harvested. l Mixed mangrove forest±shrimp: This design has a higher tree density of 10 000 trees/ha (1.0 tree/m2). The farm size is 5±10 ha. In addition to the peripheral canals, two canals separated by a dike (approximately 6 m wide) without mangrove trees are included, which allow for increased light to the open water canal area for phytoplankton and benthic algae growth. This central dike can be l

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

183

planted with salt-tolerant vegetables and bananas to augment the income from aquaculture products. Hatchery-produced shrimp (Penaeus monodon) are stocked at a density of 0.5 shrimp/m2 with a culture period of 4±5 months. The production is 200±250 kg of shrimp/ha/cycle. Secondary species including fish and shrimp from natural wild stock are also harvested. The non-mangrove area consists of the water canal, which is approximately 30±40% of the area and about 3 m wide. The two silvofishery pond designs are utilized in the Ca Mau Province of southern Vietnam with a total silvofishery area of 50 000 ha (Luu, personal communication; Johnston et al., 1999, 2000c). The two designs maintain the 7:3 ratio of mangrove area to aquaculture open water area. Both designs are operated as extensive systems with no supplemental feed. Hatchery-reared shrimp postlarvae are used for stocking ponds to compensate for the declining natural wild seedstock and variance in abundance. Even though the physical designs of the `shrimp±forest' and `forest±shrimp' differ, the operation is essentially the same. The planted tree area (usually Rhizophora apiculata) is managed so that it is only inundated at the highest spring tides with a maximum water depth of 20 cm in the treed portion of the pond. The two designs are illustrated in Fig. 8.17. Both include a widened bank area at one end of the pond for the farm owner's house and a vegetable garden, which amounts to approximately 1% (500± 1000 m2) of the land area. This area also acts as a buffer zone between the main source waterway and the pond. A band of mangrove trees are planted along the river/canal banks to prevent erosion. Various degrees of vegetable and fruit cash crop integration are utilized.

Fig. 8.17 Shrimp±forest and forest±shrimp silvofishery pond designs (not to scale) in Vietnam.

184

Ecological Aquaculture

In a study of mixed shrimp±mangrove forest farms (specific type of silvofisheries) located in Ca Mau Province, Southern Vietnam, three factors were identified to be significantly correlated to the shrimp yield in the farms examined (Johnston et al., 2000b). These factors consisted of: l l l l

l l

pond water quality; maximum fluctuation in pond water depth and ammonia concentrations; inadequate management techniques; poor pond design (leaking dikes, accumulation of excavated mud on the raised mangrove planted strips which does not allow periodic flooding of the mangrove tree area, excavated parallel strips in the pond, constricting water circulation and promoting the accumulation of organic matter with anoxic conditions; poor wild seed supply ± dependence on unreliable wild seed recruitment; and reliance on small, low value metapenaeids as the primary culture species.

Production from the ponds consisted mainly of Metapenaeus ensis and M. lysianassa (79±82% of harvest) with minor production of Penaeus indicus (6.7±9.7%). The metapenaeid species represent relatively low valued shrimp species, and Penaeus indicus a high value species. Yields were highly variable with a mean annual production ranging from 12 to 1166 kg/ha/yr, and a derived income ranging from US$54 to US$1626/ha/yr. Secondary fisheries products consisting of fish and mud crabs supplemented total farm production by 24% with a 14% addition to the gross income. This production from these mixed shrimp±mangrove forest systems had no supplementary feeding, aeration, liming or fertilizer and a short 15-day grow-out cycle along with complete dependence on wild stock recruitment for stocking, which all contributed to this low production value. The variation in production ranged from 12 to 1166 kg/ha/yr and in value from US$54 to US$1626/ha/yr. From this wide variation the impact of culture practices applied can be assumed to have a significant role and provides a base for improving production. The study concluded that the key overriding factor that can be modified with positive impact on production is the skill level of management techniques, since this normally impacts to some degree on all of the variables identified above as factors affecting production. A project on traditional shrimp culture in the Mekong Delta, Vinh Chau District, attempts to address areas to improve production (Truong et al., 1992). The productivity in the area has been low with 70±150 kg/ha/yr consisting of mostly small size shrimp (<10 g). This low production was attributed mainly to the declining natural stock of shrimp juveniles (i.e. from over-fishing and destruction of breeding and nursery grounds); therefore, the program focused on providing hatchery-reared seedstock of two commercially valued shrimp (Penaeus monodon and P. merguiensis). P. monodon proved to be most successful with greater tolerance to the marginal culture conditions. There are four traditionally practiced shrimp culture methods in Vietnam (Table 8.3). The project has focused on the first two traditional methods and has not addressed the improvement options for the silvofisheries shrimp culture in the mangrove forest method. However, some of the efforts in improving the other forms of traditional

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

185

Table 8.3 Traditional shrimp culture practices in Vinh Chau District

Traditional culture practice

Area (ha) within Vinh Chau District

Rotation of shrimp (dry season) and rice (rainy season)

2300

Rotation of shrimp (rainy season) and salt (dry season)

1300

Shrimp culture in the mangrove forest operating year round

2600

Specialized shrimp culture originating from the first area, but the land has been heavily affected by the seawater

3000

shrimp aquaculture could be applied to silvofisheries ponds. This would include the selective use of aquaculture species that provide the highest economic return. P. monodon supplemental stocking with the elimination or reduction of less economically desirable species from the pond improved the economic return to the farmer. In addition, improved water management of the ponds resulted in improved production. An integrated shrimp±mangrove pond culture system located in Ngoc Hien District, Vietnam, has large variations in shrimp production from 100 to 600 kg/ha/yr. Mangroves are planted on the central platform of the pond. The ponds are extensive low-input systems. Stocking is by natural tidal inflow with no supplemental stocking of the pond. An economic analysis found that ponds with mangrove coverage of 30± 50% of pond area had the highest economic return. Net profit ranged from US$109 to US$412/ha/yr (Binh et al., 1997). An Integrated Management of Coastal Wetlands program in the Tien Hai District was initiated in 1994. The total area is 12 000 ha. The restoration is limited to 1100 ha and will be utilizing the gei wai silvofishery model. The project targets community groups and the rural poor (Duc, 1996). Oxfam (Mauny, personal communication) is initiating a demonstration project in silvofisheries in Duyen Hai District, Tra Vihn Province, within the Mekong Delta. The project attempts to address the issues of sustainability of replanting efforts while providing an income source to local communities. The objective is to integrate mangroves with aquaculture to provide a stable source of income so that the local community supports the conservation efforts of the program on a sustained basis. A project in Dong Rui Island, Tien Yen District, Quang Ninh Province of northern Vietnam, focuses on preserving mangrove areas mainly through developing local awareness of the environmental value of the mangroves (Vu, personal communication). It also encourages mangrove-friendly aquaculture practices. A 140-ha managed natural mangrove area excludes further shrimp pond development, and has a replanting component for deforested and abandoned ponds. There is interest in shrimp and mangrove crab culture; however, current efforts are focused on restoration and management of the reserve area. Mangrove crab pen and cage culture as described for Indonesia and Malaysia is practiced in Vietnam; however, it is not usually integrated into the silvofishery method. Small coastal ponds (500±1200 m2) for crab culture are used in areas like Thai Binh Province in northern Vietnam where 70% of the families participate to

186

Ecological Aquaculture

some degree in crab culture (Overton & Macintosh, 1997). The next main economic activity in this area is salt making; however, in comparison the annual income from raising crabs can be 5 million VND, but only 1.8 million VND from salt-making (Overton & Macintosh, 1997). The crabs (5 g size) are stocked at 10 crabs/m2 over a 1-year period in the 1200 m2 ponds. This crab culture practice can easily be transferred to a silvofisheries system that maintains the mangrove trees within the enclosure. Scylla serrata culture is carried out in silvofishery ponds in southern Vietnam. Two predominant methods are used that differ in one being more intensive than the other and having less of the pond area in mangroves (Hung, personal communication). The extensive silvofishery crab pond system consists of ponds that are usually 5±10 ha in size with a stocking density of 0.25±0.33 crabs/m2. The mangrove area to open water area ratio is 8:2. There is no supplemental feeding. Production is 80±100 kg/ha per 5to 6-month grow-out cycle with a survival rate of 40±45%. The more intensive silvofishery crab pond culture is carried out in smaller ponds of 0.5 ha with a stocking density of 2±3 crabs/m2. The mangrove area to open water area ratio is 2:8 (inverse of the extensive system). Supplemental feed of trash fish is added. The production is 200±300 kg/ha per 6- to 7-month grow-out cycle with a survival rate of 60±70%. This higher survival rate in the more intensive system is attributed to greater predation in the extensive system by snakes and carnivorous fish (e.g. sea bass, grouper). The crabs are most susceptible to predation during molting periods. The mangrove area (hiding area for snakes and other predators) is decreased in the more intensive system and greater management effort is made in the operation of the pond, which includes predator control. In a study of silvofisheries ponds in southern Vietnam (Johnston et al., 2000b), the use of secondary cash crops (bananas, taro, pineapples, and cherry trees) cultivated along the pond dikes increased total farm production. This practice along with the supplementation of pond production with fish (e.g. sea bass, mullet) and crabs diversified products produced along with broadening their income base and consequently lowered the risk to farmers.

Integrated silvofisheries with agriculture ± Philippines Silvofisheries or aquasilviculture ponds are similar to those in Indonesia and use an 8:2 or 7:3 mangrove to open water area ratio (e.g. Catanauan, Quezon); however, the extent of development is less. Emphasis is on the cultivation of selected species of high economic value; therefore, there is a greater use of purchased hatchery or wild caught seedstock than dependence on natural stocking with tidal water entering the pond. The greater utilization of the various potential land-based products associated with the silvofishery pond including the harvesting of the mangroves and products of the mangroves, and the cultivation of vegetables and fruit along the dikes or central platform area to broaden and expand the economic return from the system is emphasized. In addition, silvofisheries are expected to increase their role in the rehabilitation of abandoned or less productive ponds, since past aquaculture pond

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

187

construction has been largely in previous mangrove forests (Baconguis, 1991; Primavera, 1995). Examples of integrated silvofisheries in the Philippines (Puerto Galera, Mindoro Oriental; and Hinobaan, Negros Occidental) have been developed. A silvofisheries project consisted of integrated ponds with a peripheral canal representing 20% of the pond area for stocking of milkfish or tilapia with an elevated central portion planted with nipa (harvested for construction material, alcohol, and vinegar). In addition, agricultural crops (salt-tolerant strains of banana, cacao, jackfruit, beans, eggplant, corn) were planted in the central area and along the dikes (Primavera & Agbayani, 1996). The Southeast Asian Fisheries Development Center, Aquaculture Department (SEAFDEC-AQD ) operates mangrove crab culture projects in Kalibo (Buswang), Ibajay (Bugtungbato), and Aklan provinces utilizing a modified silvofishery pond system. This system consists of low earthen dike walls around the perimeter of the pond site to maintain a minimum suitable water depth for the cultivation of the crabs. On top of the dike, a 1.8±2 m high net (buried approximately 50±60 cm into the dike, and plastic sheeting added at the top to prevent escape by the crabs over the top of the net) is used to retain the stocked crabs, while allowing for the exchange of water during flood tides. Within this low earthen-diked pond area (4000 m2), it is further divided into cultivation cells (20 units) using the netting material to separate the cells (10 m 6 20 m, 200 m2). This allows for water exchange between the cells of the pond through the netting material. The cells can be managed separately in regards to stocking density and staggering the stocking periods so that a more continuous production can be obtained, thereby stabilizing the supply to market. Furthermore, different aquaculture species can be stocked into the individual cells. Trials are being initiated for grouper and shrimp culture in separate cell units within this system. Two gates are used for water exchange in the system. In addition, a smaller PVC pipe outlet is used to relieve some of the pressure on the small dike during ebb tides when water would be retained in the pond. Mangrove trees are retained or planted within the pond area except in the excavated water canals that are dug to maintain a minimum water depth of 80±100 cm even during low tides. The trees add shelter and some nutrients for natural productivity in the pond. Trial plantings of Gracilaria in crab culture ponds to provide increased structural diversity and shelter on the pond bottom have been carried out by SEAFDEC-AQD. In a study of polyculture systems in Indonesia, the cultivation of Gracilaria with mangrove crabs positively impacted on the economic return (Cordero et al., 1999) (Fig. 8.18). This system is well suited to a cooperative operation arrangement with individuals operating individual cells within the pond. For example, at the Buswang Kalibo, Aklan site, it is operated by a cooperative of 30 individuals. A second demonstration site (Ibajay, Aklan), with a total pond area of 900 m2 and three individual cell units (10 m 6 30 m, 300 m2), is also operated as a cooperative. SEAFDEC-AQD also has a mangrove-friendly aquaculture project (Bugtungbato, Ibajay) that consists of three ponds (a, 300; b, 700; and c, 1300 m2) with traditional high pond dikes. These silvofisheries ponds varied in type and layout with the two smaller ponds (a and b) with a central platform planted with Rhizophora seedlings and stocked with hatchery-

188

Ecological Aquaculture

(a) Fig. 8.18 Low dike/net impoundment system in the Philippines. Crab pens in the Municipality of Kalibo, Aklan, consisting of approximately 4000 m2. (a) Walkway leading to the 20-unit pen/ pond area that was constructed within a previously replanted mangrove area. (b) A view down the perimeter walkway with the individual pen/pond units extending out to the left.

(b)

produced Penaeus monodon and Chanos chanos. The larger pond (c) consists of oldgrowth mangrove, with large Avicennia, canals (50 cm deep), earthen mounds of the mud lobster Thalassina, and stocked with Scylla olivaceous and S. traquebarica. This highlights the potentially diverse application of silvofisheries systems (Fig. 8.19). An additional joint community and SEAFDEC silvofishery project in the Municipality of Ibajay consists of the conversion of abandoned and low productivity ponds (total area of 3000±3500 m2) to silvofishery ponds. These are small, individually gated, earthen ponds of approximately 350±1500 m2 and irregular in shape. The ponds are operated as polyculture systems with one pond a polyculture of mudcrab (S. serrata) and shrimp (Penaeus monodon) at stocking densities of 0.6 crabs/m2 and 5.8 shrimp/m2. The production was 70 kg crabs per cycle (0.05 kg/m2/cycle, 2.5 cycles/ yr, and 40% survival). This silvofisheries pond has a 50:50 ratio of mangrove to open water area. The mangrove areas within the ponds consist of small raised islands that are rarely submerged. The irregular shape of the ponds and the scattered small island structure of the very low density mangroves do not afford an efficient operation or adequate energy flow from the mangrove detritus to support the polycultured shrimp and crabs. Supplemental feeding is required and consists of a snail (golden apple snail, Pomacea sp.) gathered from nearby rice fields. Future plans for the site includes incorporation into an ecotourism zone (Fig. 8.20). Production of mangrove crabs from the low dike/net system is 890±1150 kg/ha/ cycle, with a stocking density of 0.5±1.5 crabs/m2 and an average survival rate of 33± 56% in a 160-day culture period (Trino & Rodriguez, 1999). Monosex stocking has been adopted to reduce the mortality level. Male crabs grow faster than female crabs and the size difference is reported to compensate for the higher prices female crabs tend to receive, particularly for berried females. Three species consisting of S. serrata, S. tranquebarica and S. olivaceous are reported to make up the population. A species preference for cultivation has not been established. Seedstock are obtained from the wild and are caught using crab traps in estuaries of the mangroves. However,

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

189

(a) Fig. 8.19 Low dike/net crab impoundment system in the Municipality of Ibajay, Bugtong Bato, the Philippines. The unit consists of three cages 10 6 30 m constructed within a lowdensity mangrove area. (a) Entering crab pen/ponds that were constructed in a natural mangrove area with low tree density. (b) Within the crab pen/pond area with a walkway between the ponds. Note the low dike with the netting structure forming the (b) walls of the pen/ponds.

SEAFDEC-AQD has had success in rearing the crab larvae through to juveniles with a 2% survival rate. This provides an important technical support to the expansion of sustainable mangrove crab culture and to eventually reduce the pressure on the natural wild population of crab seedstock. An economic assessment of pond monoculture of mangrove crabs (Scylla serrata) was conducted in the Philippines (Samonte & Agbayani, 1992). They found that a stocking density of 5000 crabs/ha (0.5 crabs/m2) was the most economically feasible with the highest annual net income, return on investment (ROI), return on equity (ROE), net present value (NPV), benefit cost ratio (BCR), and internal rate of return (IRR) compared with stocking densities of 10 000, 15 000, and 20 000 juvenile crabs/ha. Stocking densities of 15 000 and 20 000/ha were found to be not economically viable. Samonte & Agbayani (1992) found the ROI and ROE to decrease as stocking density increased above 5000 crabs/ha (ROI was 66% with a 2-crop/year cycle). Agbayani et al. (1990) obtained an ROI of 124% for the same stocking density (5000/ha) with a 3-crop/ year cycle along with the highest mean weight, survival, relative growth, feed conversion efficiency, and gross production. Seville (1987, as cited by Samonte & Agbayani, 1992) obtained an ROI of 44% for a crab farm stocked at 30 000/ha (3 crabs/ m2). These stocking densities are substantially lower than that used in the silvofishery practices applied in the Sarawak pen culture (6.2±9.3 crabs/m2) or cage/pen culture in

190

Ecological Aquaculture

(a)

(b)

Fig. 8.20 Silvofishery ponds in the Municipality of Ibajay, Bugtongbato, the Philippines. (a) The drained pond shows the intermittent island structure with macrophytic mangrove vegetation. (b) Silvofishery pond with netting along the inner wall of the dike to prevent escape and burrowing into the dike wall by the cultured crab (Scylla serrata) which are polycultured with shrimp (Penaeus monodon).

Sulawesi, Indonesia (3 crabs/m2). It is also much lower than the small pond culture (10 crabs/m2) in Vietnam. Further work needs to be done to more fully evaluate this method of crab culture to determine the most economically beneficial stocking and management practices, especially considering the constraint of available seedstock. Different mangrove crab culture practices are summarized in Table 8.4. This helps to illustrate the variation in methods and results. It also helps to point to areas for improvement in the cultivation process (e.g. feeding rate, stocking density and size, and harvest size) and to modify to make the practice sustainable and profitable.

Mangrove ± mud crab (Scylla sp.) cultivation: planning and implementation guidelines Some precautionary measures should be observed when planning and implementing a mangrove crab (Scylla sp.) culture project. As with any development pressure on a natural resource, it needs to be within the carrying capacity of the system to support the activity and it should not exceed the sustainable yield of the species being cultured and harvested. This is particularly important in the case of crabs, since they are vulnerable to over-exploitation, particularly with harvest pressure on the segment of

Table 8.4 Comparison of silvofishery mangrove/mud crab (Scylla sp.) culture Sulawesi (Sinjai), Indonesia

Sarawak, East Malaysia

Species

S. serrata

Culture structure and size

Philippines

Vietnam, Southeast

Vietnam, Central

S. tranquebarica and S. olivaceous

S. serrata

S. serrata

S. serrata

S. serrata

Pen (10 m 6 10 m; 100 m2)

Pen (9 m 6 18 m; 162 m2)

Pen (10 m 6 20 m; 200 m2)

Pond (150 m2)

Pond (extensive 80% mangrove/20% water canal; 5±10 ha pond)

Pond (improved extensive/semiintensive, 20% mangrove/80% water canal; 500± 5000 m2 pond)

Construction material

Split bamboo

Wood (palm)

Low earth dike and net

Earth

Earth

Earth

Stocking density (number/m2)

3

6.2±9.3

0.5±1.5

0.5±1.5

0.25±0.33 (1/3±4 m2)

2±4

Stocking size (g)

70

100

16±20

7±11

5±10

5±10

Culture period (months)

3

4±7 (avg 5.2)

5.3 (160 days)

4±5

5±6

6±7

Harvest size (g)

200±220

300

300

250±333

250±333

300±400 2

2

35±45%

53%

56% (@ 0.5/m ) 33% (@ 1.5/m2)

98% (@ 0.5/m ) 57% (@ 1.5/m2)

40±45%

40±60%

Feed

Trash fish

Trash fish

Trash fish and brown mussel (25% fish/ 75% mussel)

Trash fish and brown mussel (25% fish/ 75% mussel)

None

Trash fish

Feeding rate (%/ average wt/day)

15%

10% CL < 6 cm 5% CL > 6 cm

8%

NA

15±20% initial 8±10% 100±150 g 5±6% > 150 g

Feed conversion (FCR)

8

2.5±6

5.3 (@ 0.5/m2) 7.6 (@ 1.5/m2)

2.1

NA

7±9

Production (kg/m2/cycle)

0.55 (5500 kg/ha)

3.27 (32 700 kg/ha)

0.089 (@ 0.5/m2); 0.157 (@ 1.5/m2) (890±1150 kg/ha)

0.133 (1333 kg/ha)

0.008±0.01 (80±100 kg/ha)

0.028±0.06 (275±600 kg/ha)

191

Survival

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

Philippines

192

Ecological Aquaculture

the population prior to maturation and having the potential to breed. The technology and reliability of hatchery-reared mangrove crabs are still in the development stage and the wild population will be relied on for stock requirements. Some planning and implementation issues should be carefully considered.

Resource assessment and monitoring It is advisable to conduct a pre-assessment of the mangrove crab population. This provides baseline data prior to crab pen activity as well as providing a preliminary indication of the potential seedstock supply, the specific species of mangrove crab/s present, and potential carrying capacity. An assessment of the availability of a suitable economic feed supply is necessary. Trash fish normally consist of fishery products (or by-products) not directly utilized by humans as food. However, the definition of `trash' can be distorted as it depends on perspective. Utilization of juvenile fish that would not normally be consumed by the local population can impact inshore and offshore fisheries with the removal of juveniles of potentially commercially valuable species (or in some cases part of the food chain supporting the commercial species). Alternative protein sources can include gastropods as well as wastes from livestock slaughterhouses. In addition, there has been some work with the use of pelleted feeds. The crabs are omnivorous and can utilize a wide range of foods. A monitoring program that monitors the environment and natural resources inputs as well as the outputs from crab culture should be carried out on a consistent basis. This will assist in identifying potential problems in the early stage and allow for corrective measures.

Local community involvement User conflicts should be avoided, if at all possible, or at least minimized. User conflict can arise out of competing uses of the resources (e.g. subsistence fishermen trapping crabs to sell at the market). Traditional users of the mangrove area may lose access to all or part of the resource that may impact on their livelihood and well-being. Therefore, community involvement should be included in the earliest stages of planning. Regulations and guidelines should be developed in concert with the community and government fisheries offices. Community participation and sense of ownership are critical to sustainable resource conservation. Limiting the development of crab pens in a given area will also be important in the long-term management, and relevant environmental issues need to be impressed upon the community at the grassroots level from the beginning of a project. This will also require community education and awareness programs so that the issues are understood and to avoid misconceptions. The community role is critical, since the ultimate monitoring and enforcement of the regulations will be through community diligence.

Technical support Technical support in the proper siting, construction and operation of the crab culture facilities will be important in the success of the practice. This will involve not only the initial stages with demonstrations, but also long-term technical support to assist

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

193

farmers with specific problems that may arise. This will also be an important part of monitoring the crab culture through collecting data on production inputs and outputs over time. The use of NGOs will be a critical component as part of a long-term program. These services will most likely be integrated into a broader community development program. NGOs often deal with issues at the grass-roots level of the community and with individual farmers where government services often do not. The NGOs will usually serve as the liaison between government fisheries office, community and farmers.

Grow-out and market coordination Lower stocking rates and growing them out to a larger size would be part of a strategy to reduce the demand on the wild seedstock. The harvesting of small crabs (e.g. 150 g) as well as fattening practices all place a high demand on the wild seedstock. In addition, the practice of producing soft-shell mangrove crabs is very demanding on the wild seedstock supply, since the crabs are usually held approximately one month at very high densities and tend to be small (75±150 g). This is practiced in Thailand and Indonesia as well as other locations. As with any production activity, market pressure will play a major role in the ultimate production practices. In the case of crabs, if there is a lucrative market for smaller crabs or crabs at certain stages (e.g. females with roe, soft-shell, etc.) this will work its way into the production practices. Environmental education and resource management along with systems management (including hatchery production) will ultimately have to play a key role, if a sustainable practice is to be maintained.

South Asia silvofisheries Sri Lanka A trial/demonstration was initiated in 2001 consisting of three crab pens constructed within a mangrove area for the cultivation of Scylla sp. The total area of the three pens is 194 m2. The Small Fishers Federation of Sri Lanka is conducting the trial in a mangrove area in Chilaw Lagoon near the Pembala Mangrove Resource Center. They are following basic guidelines established for crab pens in Sarawak, Malaysia, and by SEAFDEC in Iloilo, Philippines, and are using conservative stocking densities (1±3/m2).

India A traditional canal cultivation/fishing method utilized in Muthupet mangrove in Southeast India, locally called vaaikkal meanpedippu, consists of a number of long canals about 1.5±2 km long and 1.5±2 m wide dug across the mangrove wetland from sea to landward. The canals end landward in a trough-shaped area 1.2±1.5 m deep that is midway between the canals in the mangrove. The seaward end of the canals connects to the sea through a widening mouth. During the north-west monsoon

194

Ecological Aquaculture

season the entire mangrove is submerged allowing for the migration of juvenile prawns and fish to the trough area as well as the canals. Once the water recedes with the end of the monsoon season, the fish and shrimp are trapped in semicircular canals built along the main canal. Selvam et al. (2001) indicated an income per fisherman of US$550±650 for the 3-month period from a single canal. This traditional method has had success in mangrove rehabilitation in degraded areas utilizing a participatory local community program with government assistance (e.g. a 100 ha plot in a mangrove reserve forest). The system has environmentally beneficial impacts on the habitat for mangrove vegetation in that it improves soil salinity and moisture levels as well as facilitating natural regeneration of mangrove plants.

Silvofisheries in Africa Silvofisheries are in the planning stage in a few African countries. The Mangrove Action Project (NGO based in Washington, USA) has been active in promoting the concept with local coastal resource management programs and NGOs in Kenya, Tanzania, and Senegal. Most of the silvofisheries planned activity is for small-scale demonstration rural community based development programs. The initial targeted projects involve the cultivation of mangrove crabs within existing mangrove forests. Planning for a crab pen culture silvofishery demonstration has been initiated in the Tanga Delta, Tanzania. The demonstration will consist of four pens constructed within the mangroves. This is a community-based demonstration with participation by fishermen groups from the community in the construction and operation of the crab pens. It is intended to supplement the income of fishermen. In Kenya, a silvofisheries project is in the planning stage. The project development (Tsunza Village, Mombasa) is being coordinated with the East Africa Wildlife Society and intends to establish a crab pen culture silvofishery. The Tsunza Conservation and Development Programme is in the process of establishing a working model of a community-based crab silvofisheries project. The program consists of establishing a demonstration pen modeled after that used by SEAFDEC in the Philippines and the integrated crab pen/mangrove system used in Sarawak, Malaysia. This is being approached as an integration of socio-economic and conservation issues. The intention is to provide alternative income sources for coastal rural communities. An NGO organization, IDEE/Casamance, has begun experimenting with crab silvofishery enclosures (bamboo wall construction) in the Casamance region of southern Senegal, West Africa. The crab silvofishery systems being utilized or considered at all sites are modeled after those in Sulawesi, Indonesia, and Sarawak, Malaysia.

Brackishwater pond/mangrove alternations The staggering of a brackishwater pond with an area of mangroves (outside the pond) is an alternative method of integration of brackishwater ponds into the mangrove

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

195

forest with minimal degradation (Fig. 8.1d). This model would have a ratio of mangrove forest area alternating with a brackishwater pond area of 8:2 to 6:4. This would be a more manageable method of brackishwater pond production, since some of the negative aspects that the empang parit method has on pond management would be eliminated. The alternating areas of brackishwater ponds and mangrove forest strips should be in a perpendicular orientation to the coastline to allow for flow of freshwater to the mangroves in the green belt area. Furthermore, this alternating brackishwater pond and mangrove forest has the advantage of maintaining a greater natural biodiversity of species, since ponds with the mangroves inside are limited to species that tolerate longer submerged periods (e.g. Avicennia, Rhizophora). The appropriate species to utilize when the area will be submerged for extended periods is critical, since it needs to withstand extended pond flooding periodically. Most species require an alternation between flooded and dry periods to prevent mortality.

ENERGY FLOW AND SYSTEM MANAGEMENT Mangroves are solar-based ecosystems. Energy is transformed through autotrophic primary production of the mangrove vegetation. This constitutes the major organic energy input source with shedding of a portion of the net primary production (gross primary production 7 respiration) as litter. This mangrove litter is utilized to support the various biotic and abiotic components within the system. The litter production may be less than 50% of the total primary production and is strongly influenced by environmental conditions (Redfield, 1982b). In silvofisheries systems, the mangrove vegetation species serve as the major primary producers and their production rate will depend on a number of factors including salinity, nutrient supply, accretion rate, interspecific competition, and the degree and duration of disturbances. The primary production of mangroves varies considerably with production declining with increasing latitudes. Variations in the level of contribution made by mangroves to near-shore food chains and nutrient recycling at all latitudes was noted by Alongi (1990a). The extent of energy export or outwelling is predominantly influenced by the geomorphology of the tidal basin, tidal amplitude and water motion, and the ratio of areal extent of the vegetation to the receiving open ocean area (Odum et al., 1979; Twilley, 1985; Twilley et al., 1986a). The general consensus is that there is an outwelling of dissolved nutrients, and particulate and macro detritus matter from mangrove forests. Mangrove ecosystems operate as an energy trap, excess energy being utilized by permanent and temporary residents. Migratory animals (e.g. fish, shrimp, birds) remove a significant portion of the assimilated energy of a mangrove ecosystem. Mangrove litterfall serves as the main source of organic carbon to the system. The estimated value of energy exported from mangrove forests varies considerably, with organic carbon varying from 1.9 to 420 g C/m2/yr (Golley et al., 1962; Heald, 1969; Odum & Heald, 1972; Lugo & Snedaker, 1974; Boto & Bunt, 1981a, b; Twilley, 1985; Woodroffe, 1985; Robertson, 1986; Lee, 1989a; Gong & Ong, 1990; Hemminga et al., 1994). Classic studies on

196

Ecological Aquaculture

mangrove-based food chains in Florida (e.g. Heald, 1971; Odum, 1971; Odum & Heald, 1972, 1975; Odum et al., 1982) suggested that most mangrove litter was flushed from fringing and riverine forests by tidal action. However, more recent measurements of primary production and the degree of energy exported from mangrove systems tend toward the greater capture and utilization of energy within the mangroves, thus a lower level of export than previously considered. This reflects a continuum in refinement of the measurement procedures and a fuller understanding of the complexity of the ecosystem with interaction among the organisms in the system and physical factors influencing energy fluxes (tidal amplitude, site-specific geomorphic characteristics, seasonal influences, etc.).

Energy flows in silvofisheries ecosystems Energy flow through the mangrove biotic community is a complex process of interrelated and dependent inputs and outputs from primary production to the upper trophic levels of the system (Teal, 1962; Odum & Heald, 1972; Fell et al., 1975; Pearson & Rosenberg, 1978; Odum, 1980; Boto, 1982; Clough & Attiwill, 1982; Redfield, 1982a; Tenore et al., 1982; Saenger et al., 1983; Jones, 1984; Lee, 1990a; Heymans & Baird, 1995).

Energy flow Energy enters the silvofisheries system by photosynthesis measured as the net primary production (NPP) and flows between the various biotic components within the system. The silvofisheries model utilizes the mangrove ecosystem's characteristic of transferring synthesized energy (photosynthesis) in the form of litter production, through the processing of the litter by decomposition into an energy form captured by heterotrophs. The basic conceptualization of the energy flow is through linear food chains or feeding relationships. The transfer of food energy is from the source to subsequent trophic levels (e.g. plants to herbivores). However, the energy flow is actually a complex interwoven process involving multiple steps within and among trophic levels instead of a unidirectional progressive linear flow terminating at an upper trophic level in a food chain. It also includes recycling components of captured energy (e.g. decomposers ± bacteria, fungi) along with utilizing allochthonous materials from outside the system (e.g. elements from upstream terrestrial runoff). This process is best described as a food web. Odum (1971) identified the main flow of energy in the food web along the route of mangrove leaf detritus ? bacteria and fungi ? detritus consumers ? middle carnivores ? higher carnivores. Figure 8.21 provides a summary illustration of this energy flow and food web process. Energy input to the system comes from vegetative matter mainly in the form of litter from the mangrove trees, but also includes algae and phytoplankton along with external input of nutrients (dissolved and particulate matter with incoming tides). These energy sources enter the system through different processes and become incorporated into the various components within the food web and recycled.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

197

HERBIVORES (crabs, fish insects)

MANGROVE LITTER and other emergent macrophytic material

MACROALGAE AND PHYTOPLANKTON

MICROBIAL AND PROTOZOAN ORGANISMS activity forming detrital complex

EXTERNAL NUTRIENT INPUT (dissolved and particulate nutrients)

NUTRIENT EXPORT (during drainage or tidal exchange)

DETRITIVORES AND HERBIVORES (fish, shellfish, shrimps, snails, copepods, amphipods, polychaetes etc.)

OMNIVORES/ CARNIVORES (fish, mangrove crabs-Scylla sp., shrimp, birds)

ORGANIC BYPRODUCTS AND WASTES nutrient cycling aerobic/anaerobic process

Fig. 8.21 Silvofisheries pond with energy flow and food web processes.

Management of this energy flow in a silvofisheries system is in the areas of energy input (mangrove density, species and age), aquatic environment (water level, quality, submergence/emergence, and residence time), and in the aquaculture species stocked (utilization of the food web). The food web revolves around the energy transfer to higher trophic levels from the primary production. The knowledge of the various pathways and the associated organism communities along with the production and utilization by species is important in the management of silvofishery systems. There is energy loss (i.e. respiration metabolism) with the transfer from one trophic level to the next. Christensen & Pauly (1993) in a modeled analysis of trophic interactions and flow in aquatic systems found an average trophic transfer efficiency of 10±11% for herbivores/detritivores and first-order predators. Some species feed at more than one trophic level. For example, tilapia are very adaptive not only in being euryhaline but also in their ability to utilize numerous food sources including algae, detritus, and small invertebrates (e.g. amphipods and chironomids); they are therefore regarded as both primary and secondary consumers. Other species are more limited in their potential food source. These food sources and utilization are important criteria in the selection of the species to be cultured. The majority of the aquatic products harvested from silvofisheries systems come from the detritivore and herbivore as well as omnivore categories. It is preferable to utilize species that are lower on the trophic level, since this will allow for a more

198

Ecological Aquaculture

efficient utilization of the system's energy. Some of the cultured species (e.g. shrimp, tilapia, mullet) would be included in the omnivore category that utilizes a wider range of foods. It is possible to culture carnivores especially to control pest species or control over-population of other species. However, this would be a less efficient use of the system's energy flow and therefore should be kept to a minimum. This impact on production is illustrated in a limited set of data from gei wai harvests presented in Fig. 8.22. A significant reduction in the production of carnivorous fish allowed a greater total production of herbivorous fish species. Herbivorous fish contributed on average 86.2% of the fish production over the 3year sampling period (Cha et al., 1997). The organisms of the silvofisheries system can be categorized into ascending trophic levels. The energy stored at each ascending trophic level sequentially declines and is traditionally illustrated as a biomass pyramid for the different trophic levels. These trophic levels would consist of the following. Primary producers ± The major primary producers are photosynthetic: mangroves, algae (e.g. Enteromorpha spp., Cladophora spp.), and phytoplankton, with chemosynthetic organisms (e.g. bacteria) as a minor component. l Primary consumers ± This would include detritivores (e.g. bacteria, fungi) and herbivores (e.g. crabs, Chanos chanos), omnivores (e.g. Mugil cephalus, shrimp, tilapia, mud crabs, polychaete worms, snails, bivalves, amphipods, zooplankton, insects). l Secondary consumers ± This is the first level of carnivores (e.g. shrimp, mud crabs, fish). l Tertiary consumers ± This represents an upper level of carnivores (e.g. fish ± Lates sp., birds). l

100

Shrimp** Carnivorous fish* Herbivorous fish*

kg/ha/yr

80

60

40

20

0 1993

1994 Year

1995

Fig. 8.22 Annual mean production (kg/ha/yr) by species group from gei wai ponds, Mai Po Nature Reserve, Hong Kong (after Cha et al., 1997). * Fish production was measured from one pond. ** Shrimp production is the mean from 7 ponds.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

199

This energy flow within the system is diverted with the harvesting of the aquaculture products. Therefore, this rather complex natural process of energy flow will be a major management component of a silvofisheries system. Proper management practices (e.g. density and species of mangroves, flooding frequency and duration of vegetated platform areas, etc.) need to be implemented to ensure that adequate energy continues to flow into the system to support the food web for the selected aquaculture species.

Key energy transfer components Mangroves serve as a major carbon dioxide sink and provide the basic functions similar to tropical rainforests. Sibert & Naiman (1980) noted the increased appreciation of the role of bacteria in the detritus food chain and viewed estuarine ecosystems to have two productive bases consisting of detritus processing by microbes and photosynthesis with both constituting a balance between autotrophy and heterotrophy and import/export of materials. Microbial reprocessing ensures that detritus resulting from non-grazed plant material is also available to consumers. The flow of energy in the form of carbon follows a general sequence of carbon budget ? standing stock of main ecosystem constituents (density of trees) ? rates of net primary production ? rates of litter production ? consumption by heterotrophs ? storage and export ? higher consumers. Carbon represents 45% of the dry weight of senescent Rhizophora mangle leaves with approximately half of this carbon being leached and half becoming particulate detritus (Fell & Master, 1980). Soluble carbon enters the food chain either as uptake by heterotrophs and consumed by filter feeders or by direct ingestion after flocculation (Cooksey & Cooksey, 1978). Odum (1971) identified the following four ways in which the heterotrophic community may utilize fresh mangrove leaf litter. (1) (2) (3) (4)

Dissolved organic substances ? microorganisms ? higher consumers Dissolved organic substances ? sorption on sediment particles ? higher consumers Leaf material ? higher consumers Leaf material ? bacteria and fungi ? higher consumers

The first two routes of energy exchange are based upon the rapid loss of watersoluble organic substances, which can be used by bacteria and other microorganisms directly. In the third route, higher consumers consume the leaf material directly. The fourth route is the most important means of energy transfer in a detritus system. Odum (1971) further breaks down the trophic levels into a series of subcategories of trophic groupings based on the specific composition of the material ingested. The total nutritional needs of a detritivore could best be supplied from a variety of sources within a detritus pool rather than a single source. Besides the contribution of microalgal to benthic food chains, macroalgae blooms (e.g. Ulva, Enteromorpha), which readily decompose and provide an organic contribution, are rich in available energy substrate and nitrogen content. The more refractive component of a detritus

200

Ecological Aquaculture

pool is thus made available to benthos over a long time period (months to years) and can provide a buffering food supply compared with the pulses of easily assimilated, quickly dissipated phytoplankton and seaweed components (Tenore, 1988). Herbivory is another pathway of energy transfer apart from detritivory and includes consumption of fresh mangrove leaves and algae. Loss of leaf area (Kandelia candel) in a Hong Kong mangrove from insect herbivory was estimated at 11%, but amounted to the consumption of only 2.8±3.5% of the net above-ground primary production (Lee, 1986, 1991). Chow & Bacon (1992) emphasized the potential importance of crab canopy herbivory activity impacting on the energy flow of mangrove derived Aratus sp. fecal pellets through the detritus food web. Even though it is a relatively small quantity, it represents an accelerated pathway to the detrital energy pool. They estimated a 10-fold increase in the rate of processing to the detrital pool, representing a faster rate of energy transfer from vascular plants into the marine food web. Furthermore, the crab fecal pellets are rich in nutrients (carbon and nitrogen). Camilleri & Ribi (1986) suggest that the utilization by copepods and amphipods of leached organic nutrients from mangrove leaves as food, with the formation of particulate matter colonized by microbial organisms, provides a more direct energy pathway. It also provides a relatively rapid energy transfer compared with the plant± grazer±carnivore pathway. Such an alternative food path in mangroves and specifically in a silvofisheries system could affect the turnover rate and the aquatic species composition and production. In ponds, the energy flow in the food web is enhanced by heterotrophic activity, with the microbial process providing nutritious feed to aquaculture species directly or indirectly and by providing nutrients for algal photosynthesis. Pruder (1987) proposes a fuller utilization of this food web interaction with these components of the natural web enhancing aquaculture production. Shankar (1999) further suggests the potential for exploiting heterotrophic production through the efficient harvesting by fish (e.g. tilapia) of bacterial biofilm (comprised of bacteria, algae, and fungi) that develops on substrates. Fish growth was found to be 50% higher in ponds with biofilm grown on biodegradable plant wastes and can provide an alternative to expensive artificial feeds (Shankar, 1999). Colman & Edwards (1987) also emphasized the importance of the interdependence of autotrophic and heterotrophic pathways in the utilization of waste-fed aquaculture. Anderson (1987) emphasized the links between organic matter and inorganic nutrient fluxes as keys to the potential for manipulating these transfers in the aquatic environment with decomposer and herbivore systems integrated to effectively transfer microbial production to higher trophic levels. Analysis and quantification of key processes and their interactions will be important in determining the flow and utilization of energy and matter. The sum of activities including production, leaching, decomposition, and nutrient exchange (import and export) determines the amount of detrital material available to higher trophic levels. The overall quantification of this process rate and its regulation provide an important management tool in planning and operating silvofishery systems. The magnitude of net productivity in a wetland is a balance between import and export, and between autotrophy and heterotrophy (Sibert & Naiman, 1980).

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

201

Decomposition The decomposition process is a critical step in breaking down the major primary production products that serve as the energy base to the silvofisheries system. The resulting nutrients and products of the decomposition process further serve as building blocks to the food web that supports the aquaculture species. The various avenues of decomposition and factors that influence it are discussed below. The influence of the quality of primary production products on this process and its importance in the food web are also highlighted. Decomposition of the mangrove litter by fungi and bacteria results in proteinenriched fragments of detritus. The role of bacteria in transforming detritus into nutritional particles is through mineralization of organic matter, production of ectocrine substances such as vitamins and antibiotics, and the production of food (Sibert & Naiman, 1980). This detrital complex (inclusive of all forms of organic carbon lost by nonpredatory means from any trophic level) forms a major food source for the biomass within the mangrove ecosystem (Odum, 1980; Saenger et al., 1983; Fell et al., 1984; Heymans & Baird, 1995). Detritus formation consists of three primary processes; the initial rapid loss of soluble organic compounds, the colonization of the leaf substrate by bacteria, fungi, and protozoans, and physical and biological breakdown and fragmentation. Microbial enrichment of the detritus encourages macrofaunal consumption. The feces of detritivores further serve as substrate for microbial growth with acceleration of the decomposition process. Decomposers play a major role in energy transfer and recycling through this process. The use of plant material as a nutrient base to the food chain in aquaculture is primarily dependent on production of heterotrophic detritus and autotrophic algae stimulated by leaching nutrients from the mangrove litter. The growth of bacteria, attached and free, probably represents the major pathway of the production of high quality (low C:N) biomass potentially available to the grazers. Bano et al. (1997) found that bacterial production was carbon rather than N or P limited. Mangrove detritus provides most of the energy for bacterial production, which in turn is a significant source of high quality food for grazers, particularly via ingestion of attached bacteria. Mangrove plant litter (leaves, twigs and small branches) undergoes a lengthy fragmentation and mineralization process requiring nearly one year to become available in the detritus pool (Levine, 1981; Chow 1989; Lee, 1989a, b; Gong & Ong 1990). The rate of decomposition is influenced by a number of factors including physical, chemical and biological. Decomposability is mainly a function of the chemical composition of the source material with those high in soluble ash, organic nitrogen and other hydrolyzable components easily degraded. Algal tissue decomposes faster than mangrove litter (Rice & Tenore, 1981). The relative decomposition rate for four groups of plant tissue is phytoplankton > macroalgae > submersed macrophytes > emergent macrophytes. This rate is directly proportional to the initial nitrogen content of the plant tissue as indicated by the C:N ratio, which also likely reflects differences in structural and secondary compounds (Twilley et al., 1986b). The lower the C:N ratio of the leaf litter material the more rapidly would

202

Ecological Aquaculture

decomposition occur in general, since relative availability of N for microbial assimilation influences the rate of decomposition (Rice & Tenore, 1981; Twilley et al., 1986a, b; Robertson, 1988a). It was noted by Twilley (1982) that as tidal inundation frequency decreases in mangrove forests, the litter dynamics are influenced more by the decomposition and respiration characteristics of the forest floor rather than by export. Factors that influence sediment microbial metabolism are tidal inundation frequency, duration of water cover, crab bioturbation, oxygen level, temperature, quantity of organic input, and quality of organic input (Alongi, 1990b; Kristensen et al., 1992, 1994; Alongi, 1994; Nedwell et al., 1994). The intermittent exposure of the substrate with tidal fluctuation influences decomposition differently according to the site's characteristics. Degradation rate of litter is highly variable and is affected by where the litter falls. Odum (1971) found that litter falling on dry ground decomposed slowly, while in freshwater it decomposed somewhat more rapidly, and in seawater the rate of decomposition was the most rapid. Decomposition of litter is generally faster under conditions of continuous submergence (Steinke & Ward, 1987; Robertson, 1988a; Angsupanich et al., 1989). This is attributed to the mechanical breakdown by grazers (e.g. amphipods), greater microbial activity, rapid leaching of labile contents, water movement, and greater environmental stability. However, the decomposition rate can be accelerated in the emerged areas of the intertidal zone of the mangroves in relation to the degree of mechanical breakdown of the litter associated with the presence of grazing activity by fauna in the forest (e.g. crabs, snails, insects). Exposure to wetting action of the tides is an important factor affecting decomposition. Decomposition of litter will be influenced by site-specific conditions and their interactions (e.g. tidal frequency and duration, presence of litter-grazing fauna, temperature, aerobic conditions, etc.) that control the level of microbial activity; therefore, there can be significant differences between sites. However, in a silvofisheries pond a number of the environmental conditions are subject to a degree of management, particularly through controlling the duration and frequency of flooding. Decay of detritus deposits were 4±5 and 8 times faster in the high intertidal zone (Phuket, Thailand) compared to the mid and low intertidal zones, respectively (Kristensen et al., 1994). This accelerated rate of decay in the upper intertidal zones was partly attributed to the increased crab consumption of litter material and bioturbation (feeding and burrow construction), since activities of burrowing animals can stimulate benthic metabolism and cause a reduction of organic matter (Kristensen, 1988). However, excessive desiccation can reduce overall microbial respiration (Kristensen et al., 1992). In addition, decomposition of many organic compounds can be orders of magnitude faster in aerobic than anaerobic mangrove sediment. They concluded that low nitrogen content in sediment detritus was also a limiting factor for microbial activity. Twilley et al. (1986a) noted the decomposition rate was faster in wetter environments. They also found a higher rate of decomposition for leaves with higher nitrogen concentration and lower C:N ratios, suggesting decomposition rate may be related to nitrogen content and therefore its food value. Leaching of labile materials and microbial population establishment are accelerated during water

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

203

soaking. Robertson (1988a) indicated that subtidal decomposition rates are faster because there is a more constant environment for heterotrophs to facilitate decomposition. However, this can be influenced by supratidal biological grazing activity on the mangrove litter by snails and crabs. Lee (1988, 1990b) found litter decomposition to be influenced by temperature and tidal inundation regime with a significantly higher rate of decomposition of the reed, Phragmites communis, in areas with 52.3% air exposure than in areas exposed only 13.5 and 2.5% of the time. This was attributed to algal growth on the litter thus lowering the C:N ratio and attracting grazing gastropods. This illustrates the importance of secondary factors such as algal growth stimulating grazing by fauna, which results in accelerated decomposition. Mackey & Smail (1996) emphasized the importance of temperature as affected by latitude and season in the dynamics of organic cycling in mangrove forests, with higher temperatures accelerating decomposition. This included the effect on decomposers and shredder populations that are active throughout the year in the tropics and contribute to the accelerated decomposition and organic export at a stable rate. As discussed, decomposition of mangrove detritus is essentially a microbially mediated process; however, crabs (sesarmids) can contribute significantly in the initial stages of decomposition. Camilleri (1989) concluded leaf processing by crustaceans shortens the time span between leaf fall and consumption of leaf material by organisms (facilitates utilization by deposit-eating detritivores) and contributes to the retention of litter biomass and nutrients within the mangrove forest. Litter turnover rate is influenced by inundation frequency and crab (Chiromanthes) consumption, with the crabs consuming >55% of the daily litter production (Lee, 1988). Tidal gei wai ponds act as sediment traps and accelerate accretion, leading to progressive increase in elevation of the mangrove stands. Less inundated sites allow more litter accumulation to be processed by crabs (Perisesarma spp. and Parasesarma spp.) in the stands, thus leading to little export (Li & Lee, 1998). Fauna that graze on or shred the litter increase the rate of breakdown of detritus (Tenore et al., 1982). Jones (1984) noted a number of species of crabs that are abundant in tropical Asia± Pacific mangroves that consume significant quantities of litter. Lee (1989b) found that crab (Chiromanthes spp.) consumption of mangrove leaves was a major factor in forming an energy sink and limiting litter export within shrimp tidal ponds (a type of silvofishery) in Hong Kong. Crabs fragmented leaf litter and returned the nutrients as fecal pellets or shredded pieces, which facilitated further degradation. In a study by Emmerson & McGwynne (1992) on consumption of mangrove (Avicennia marina) leaf litter, they found the crab Sesarma minerti consumed 44% of the annual leaf litter, resulting in feces production, and facilitating the leaf turnover an estimated four-fold. Robertson & Daniel (1989) estimated that leaf processing by crabs turns over litter at more than 75 times the rate of microbial decay alone. Similarly Lee (1993) found a selective preference by the mangrove crabs Chiromanthes bidens and C. maipoensis for senescent leaves in the order of Avicennia marina > Kandelia candel > Aegiceras corniculatum which correlated with a preference for leaves with lower C:N ratios (greater nutritive value). In addition A. marina had the lowest tannin level of the three mangrove trees. Lee (1989b) reported that sesarmine crabs were capable of

204

Ecological Aquaculture

consuming >57% of the daily leaf litter. Such in situ consumption by crabs reduces tidal export and may also initiate further processing of mangrove-derived organic carbon by way of coprophagous food chains based on crab feces (Li & Lee, 1998). Mangrove litter dynamics in the New World tropics (Ecuador) were also found to be influenced by mangrove crabs, Ucides occidentalis, that were able to remove daily additions of leaf litter material within one hour except during a limited period (August±October) of crab inactivity (Twilley et al., 1997). Robertson (1986) demonstrated that leaf-consuming crabs (Sesarma spp.) played a significant role in litter turnover of mangrove ecosystems in the Indo-West Pacific. In a study of leafconsuming crab Sesarma messa consumption of leaves of Rhizophora stylosa, consumption amounted to 28% of the annual leaf fall (Robertson, 1986). Robertson (1986) documents the important link sesarmid crabs make between mangrove primary and secondary production with the consumption and retention of a large proportion of the annual leaf fall within mangroves. This represents an important pathway of litter processing not accounted for in previous mangrove food chain models, particularly in the Indo-Pacific mangroves. In a study of a low-lying Rhizophora mucronata mangrove stand in East Africa, litter removal was mainly attributed to the crab Sesarma guttatum and not tidal action, while in an elevated Ceriops tagal area, which is flooded only during spring tides, the detritivorous snail Terebralia palustris served as the major macrobenthic organism responsible for litter removal (Slim et al., 1997). Water availability from either rain or tidal inundation was a determining factor in the amount of litter consumed by T. palustris, since potential desiccation of the snail restricted its foraging movements. In both locations diurnal fluctuation had a significant impact on the litter removed (under favorable conditions 25.2% by day and 41.6% by night for T. palustris, and 40.3% by day and 21.7% by night for S. guttatum). Decomposition rate is influenced by the species-specific characteristics of the vegetation. Tam et al. (1990) found the rate of leaf decomposition and the amount of nutrients released to be species specific and related to the chemical composition of the leaves. Decomposition rates in descending order were Kandelia candel > Avicennia marina > A. corniculatum with 85% to less than 50% of the leaf matter lost in an 8week incubation period. Varying concentrations of C and N occurred over the initial 4 weeks and stabilized at approximately 24:1, mainly due to an increase in %N (Tam et al., 1990). Twilley (1982) found that decomposition and organic carbon leaching rates were much higher in Avicennia than Rhizophora leaves. Twilley et al. (1986a) also noted the effect on litter dynamics from species differences in leaf decomposition rate with Rhizophora mangle decomposing more slowly than Avicennia germinans. Other studies that examined the rate of decomposition similarly found Avicennia to decompose in half of the time or less compared with other mangrove vegetation (Albright, 1976; Boonruang, 1978; Goulter & Allaway, 1979; Steinke et al., 1983). The following relative decomposition rates for species have been identified as Avicennia marina > Rhizophora apiculata (Boonruang, 1978); A. marina >Bruguiera gymnorrhiza (Steinke & Ward, 1987); A. marina > R. stylosa > Sonneratia alba > B. gymnorrhiza (Angsupanich et al., 1989); and A. marina > Ceriops tagal > R. stylosa (Robertson, 1988a). A. marina consistently had the fastest rate of decomposition of

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

205

the species considered and this was up to a factor of approximately three-fold faster decomposition. This would provide an advantageous characteristic (along with low C:N and low tannin content) in the use of A. marina as the main litter source in a silvofisheries system, since it would accelerate the incorporation and utilization of the detrital material into the food web. This has a significant advantage in shortening the period of time to make the nutrients available for use by heterotrophs and the eventual recycling; thereby accelerating the energy turnover. The initial availability and the subsequent role of microbes depend on the detritus source (Tenore, 1988). For example, seaweed-derived detritus is typically high in available calories and nitrogen content that can be directly utilized by macroconsumers while those from vascular plants are typically low in both available caloric and nitrogen content and require an extended time period before becoming available for utilization by macroconsumers. During the latter stages of the slow decay of refractory material, the microbial biomass, and more importantly their nitrogen-rich exudation products, accumulate and can increase the absolute nitrogen mass of the detritus (Hobbie & Lee, 1980; Rice & Hanson, 1984). The physical structure and chemical composition influence the rate of degradation of mangrove litter. For example, leaves covered with a thick cuticle would impede the entry of water and degradative organisms. Avicennia leaves are thin and sink in water while Rhizophora or Sonneratia leaves are thicker and are buoyant initially in water; thereby, contributing to the more rapid decomposition of Avicennia (Wafar et al., 1997). Plant materials high in crosslinked celluloses and lignins resist decay while those high in soluble ash, organic nitrogen and other hydrolyzable components are easily degraded (Twilley et al., 1986b). The level of tannin content in leaves similarly has been associated with impeding microbial activity and grazing by macrofauna which hinders degradation (Alongi, 1987; Alongi et al., 1989; Emmerson & McGwynne, 1992) as well as inhibiting microbial activity in sediments with high concentrations of tannins (Boto et al., 1989). Tannin levels in mangrove leaves of Avicennia marina and Bruguiera gymnorrhiza show significant decreases after 48 h of submergence with a 74±85% reduction after 14 days (Steinke et al., 1993). The physical structure, the nutrient level reflected in the C:N ratio, and the level of tannin in the litter products influence their food value and the decomposition of the materials into the detrital pool of energy. Table 8.5 provides a comparison of selected mangrove species C:N and tannin levels. A. marina has the most rapid rate of decomposition and recycling compared with other mangrove tree leaf litter (Van der Valk & Attiwill, 1984; Steinke & Ward, 1987; Robertson, 1988a; Angsupanich et al., 1989; Lee, 1993). Cundell et al. (1979) found the C:N ratio of senescent Rhizophora mangle leaves to decrease from 90.6 to 40.7 over a 70-day period of immersion. This was attributed to the loss of readily leachable carbohydrates and the increase in nitrogen from microbial activity. The microbial population slowly increased on the leaf with leaching of the tannin compounds. Tannin content in R. mangle senescent leaves was recorded at 5.2% initially, which declined to less than 1% after 35 days of water immersion (Cundell et al., 1979). The antimicrobial and enzyme inhibitant characteristics of tannin are assumed to delay colonization of senescent leaves by bacteria

206

Ecological Aquaculture

Table 8.5 Carbon, nitrogen and tannin content (based on dry weights) of selected mangrove species (Cundell et al., 1979; Robertson, 1986; Lee, 1993; Wafar et al., 1997) Species Avicennia marina Kandelia candel Aegiceras corniculatum Rhizophora stylosa R. mangle R. apiculata R. mucronata Sonneratia alba Avicennia officinalis

% Organic carbon

% Organic nitrogen

C:N

% Soluble tannins

34.5 36.7 54.6 ± 46.2 43.4 43.8 42.4 41.9

1.26 0.75 0.79 ± 0.51 0.98 0.66 1.15 0.95

27.4 49.1 69.1 70.0 90.6 44.3 66.4 36.9 44.1

0.86 2.35 1.95 17.00 5.20 ± ± ± ±

and fungi. In the decomposition of R. apiculata leaves, Raghukumar et al. (1994) similarly observed a sequence in colonization by fungal species and fungi and bacteria biomass peaking at 21 and 35 days respectively. The sequence of events in the decomposition process of Rhizophora mangle and R. apiculata leaves is identified in Table 8.6. The caloric content fell during day 28 through 49 and then increased as microbial biomass accumulated.

Recycling of nutrients/biochemical cycles There is a continuous process of exchange and assimilation of energy fixation, accumulation of biomass, decomposition of dead organic material and mineral cycling within the mangrove ecosystem. There is also a recycling of the nutrients into new growth of the mangrove trees. The nutrient flux through a mangrove ecosystem is illustrated in Fig. 8.23. Holguin et al. (2001) provide an overview of the role of microorganisms in the mangrove ecosystem and conclude that there is a close microbe±nutrient±plant relationship that functions to recycle and conserve nutrients and consequently plays an important role in the productivity and sustainability of the ecosystem. There is a beneficial relationship with plant-growth-promoting bacteria and nutrient recycling, which supports production of plant-root exudates that serve as a food source for

INORGANIC NUTRIENT INPUT • Rainfall • Freshwater runoff • Nitrogen fixation • Mineralization • Tidal borne • Chemical release from fixed states in soil • Man-made influences (agriculture, sewage, etc.) • Upland input

MANGROVE SYSTEM Recycling

NUTRIENT OUTPUT • Tidal transport • Leaching • Denitrification and volatization • Immobilization of inorganic nitrogen • Leaching of soils by freshwater • Active transport by temporary inhabitants of mangroves (e.g., penaeid shrimp, fish, etc.)

Fig. 8.23 Nutrient flux through mangrove forest (after Boto, 1982).

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

207

Table 8.6 Sequence of events in the decomposition of Rhizophora mangle and R. apiculata

Time 1±14 days

Events Rhizophora mangle Cundell et al., 1979 Rapid leaching of reducing sugars and tannin

Caloric content, cal (ash-free)/g

Events Rhizophora apiculata Raghukumar et al., 1994

5200

0±28 days

Initial leaching period followed by rapid microbial decomposition. Species of fungi specific to initial colonization. Rapid cellulose decrease during first 21 days

14±28 days

Depletion of leachable reducing sugar. Leaching of tannin. Initial colonization by micro-organisms especially bacteria

5351

28±49 days

Colonization of outer leaf surface by bacteria, fungi, pennate diatoms, and stalked protozoa

5254

28±60 days

A shift in fungal species to Hyphomycete XVII. Maximum growth of `late colonizer' fungi. Increased growth of thraustochytrids. A more rapid degradation of xylem

49±70 days

Erosion of leaf surface. Rich microflora, cellulolytic bacteria and fungi, stalked diatoms and tube worms

4217

70+ days

Fragmentation of the leaves

5429

microorganisms and production of other plant material serving as a food source for larger organisms (e.g. crabs). The magnitude and direction of net material flux are determined by a combination of physical parameters, such as geomorphology, tidal inundation regime and topography, and biotic factors such as the species and growth form of vegetation, seasonal growth patterns and rates of primary production, and the development stages of the wetland±estuarine system (Odum, 1969; Dame & Lefeuvre, 1994). The nutrient pool of the mangrove ecosystem is regulated by five interacting processes (Saenger et al., 1983). The last process is concerned with the recycling and conservation of the nutrient pool and is of particular interest in the management of silvofisheries systems. These processes are as follows: Freshwater or tidal flooding introduction of organic material and inorganic mineral ions. l Sedimentation-introduced inorganic mineral ions. l

208

Ecological Aquaculture

Wind-introduced inorganic mineral ions. Depletion of nutrient pool with the flooding of freshwater and tidal action. l Microbial decay of organic material providing for internal cycling of mineral ions. l l

Mangroves have been considered as a significant universal exporter of organic matter to coastal waters. However, more recent studies indicate that this role is not nearly as great as previously considered. Estimates of organic carbon from intertidal wetlands range from 45% (Teal, 1962; Twilley, 1985) to less than 1% (Heinle & Flemer, 1976) of their net production, while some intertidal wetlands may even import organic carbon (Woodwell et al., 1977; Lee, 1990a). Mangrove production contributed only 1.8% to the total carbon available in Deep Bay, Hong Kong (Li & Lee, 1998). Heinle & Flemer (1976) found that fluxes of detritus from stable tidal marshes subjected to modest tidal flooding are less than 1% of the maximum areal standing crops and the tidal marsh serves mainly as a nutrient sink. In a study of a South China mangrove it was found to be in equilibrium with no net import or export of N and P (Li, 1997). In Deep Bay, Hong Kong, the outer mangrove is a net exporter of organic carbon while the landward gei wai, a semi-enclosed tidal pond, acts as a net importer. This difference is attributed mainly to the tidal inundation regime (Li & Lee, 1998). Therefore, the energy flow in mangroves is mainly contained within the various subsystems in the mangroves. In a silvofisheries pond the litter and nutrient dynamics can approach that of a closed system. The traditional silvofisheries pond (i.e. no fertilizer or supplemental feed inputs), with modified and controlled tidal flushing along with harvesting of cultured products from the system, is normally a very low exporter or net energy importer with an influx of inorganic nutrients from the adjacent environments and juveniles of species cultured (natural stocking). Gong & Ong (1990) identified the macronutrient levels of N, P, K, Ca, Mg, and Na that are released annually from a managed Malaysian mangrove forest (Matang Mangrove Forest, 40 800 ha). A total of 656 kg dry wt/ha of macro-nutrients with 46% from litter, 44% from dead trees, and 10% from slash are released annually. In Table 8.7 these values have been calculated based on an annual per hectare basis. Table 8.7 The amount of nutrients and organic matter (dry wt) released from small litter (leaves, twigs and fruit), dead trees, and slash (cuttings of trees harvested) from the Matang Mangroves, Malaysia (after Gong & Ong, 1990)

Litter contribution (kg/ha/yr) Biomass N P K Ca Mg Na Organic matter

9 726 54.2 6.3 28.0 132.5 39.9 38.3 *9 050.0

Dead tree contribution (kg/ha/yr) 13 713 73.9 3.8 17.5 36.9 41.8 117.0 12 787.9

* Estimation based on similar percentage for dead trees and slash.

Slash/cuttings contribution (kg/ha/yr) 1 462 11.1 1.6 9.7 10.0 15.7 17.8 1 349.2

Total (kg/ha/yr) 24 901 139.3 11.7 55.1 179.4 97.3 173.1 23 187.0

Per cent of Total

100 0.56 0.05 0.22 0.72 0.39 0.70 93.12

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

209

In general, nitrogen is considered to be the major nutrient limiting production in the marine environment. The nitrogen cycle similarly has a functional role in the food web as a major building component of organic matter. Nitrogen is usually considered a limiting growth factor for primary producers. In enriched waters the primary producers grow much more abundantly and the food chain is contracted (multiplicative effect with higher productivity 6 higher ecological efficiency) so that more of the primary production ends up in the upper level consumers. The main source of nitrogen is through the breakdown of organic matter and the oxidation of inorganic forms of nitrogen. In addition, in coastal land based transitional zones, such as the mangroves, a major source of nitrogen is also introduced through the runoff from the land. There has been found (Potts, 1984) to be a significant input of nitrogen from benthic microbial populations in mangroves associated with periods of high tide, since the tidal cycle is a major factor impacting on the desiccation of communities of cyanobacteria (activation of nitrogenase activity after wetting with the incoming tide). Similarly, Twilley et al. (1986a) noted that nitrogen fixation may be a major source of nitrogen in mangroves and is influenced by desiccation (tidal inundation duration) reducing the nitrate reductase enzyme activity. In tropical conditions with year-round warm temperatures, N fixation by bluegreen algae in fishponds is considered to be a significant factor in aquaculture pond fertilization (Egna & Boyd, 1997). Commonly nitrogen limitation has been found throughout the intertidal zone and phosphorus limitation was also evident at the higher elevation areas within the mangroves (Boto & Wellington, 1983). Boto & Wellington (1983) found a highly significant correlation between mature Rhizophora leaf nitrogen and phosphorus levels with soil ammonium and extractable phosphorus; therefore suggesting the use of mangrove leaves as indicators of mangrove forest nutritional status. It was further noted in this study that there was significant mangrove growth response after ammonium enrichment at a lower intertidal site and at an upper site with the addition of phosphate enrichment. Kristensen (1988) found benthic fauna to have a significant role in nutrient cycling. Below the oxic zone a decline in the nutritional quality of the organic matter with depth and age in the sediment slows the rate of mineralization by anaerobic processes (e.g. fermentation, denitrification, and sulfate reduction). However, the exchange of nitrate between the sediment and the overlying water is affected considerably by the burrow-dwelling infauna. Burrows serve as an extension of the sediment±water interface. Ventilation of burrows and tubes is a major factor controlling biogeochemical processes occurring in sediments (Kristensen, 1984, 1985). In addition, during feeding (i.e. deposit or filter feeding) the majority of infaunal animals selectively concentrate organic rich material into fecal pellets and these pellets are sites of high microbial activity. Microenvironments, such as fecal pellets and intermittently ventilated burrows, create a close spatial and temporal coupling of both the nitrification and denitrification processes. Therefore, the benthic animals can increase the nutrient turnover in coastal ecosystems and improve primary production through stimulation of the mineralization process. The role of benthos fauna in cycling nitrogen is influenced by the basic nutritional

210

Ecological Aquaculture

needs of the benthos, the food resources, and the feeding strategies that the benthos organisms use to exploit the available resources. The benthos includes a variety of feeding types consisting of suspension and deposit feeders and scavenging/carnivorous organisms. The potential food sources of benthic deposit feeders include benthic microalgae, detritus, fecal pellets and microbenthos (i.e. bacteria, fungi, and protozoa). However, refractory detritus (e.g. vascular plant material ± mangroves, seagrass), which is typically low in nitrogen and composed of structural materials not directly assimilable by detritivores, must be depolymerized by microbial decomposition with the resultant microbial biomass or transformation products being available to deposit feeders. The microbial decomposition of fecal pellets (coprophagy) and detritus from vascular plants' refractory nature is necessary before the nutrients become available to macroconsumers as part of the detritus food chain. The process increases the nutritional value and particularly the relative nitrogen content of the detritus (Tenore, 1988). In addition to being an essential part of the food web, bacteria are important mineralizers of organic detritus and recyclers of essential nutrients (e.g. carbon, nitrogen and phosphorus), which are particularly important functions in mangroves. Sedimentary bacteria populations play a significant role in energy flows and cycles of tropical mangroves, and are mainly controlled by temperature and tidal inundation frequency (Boto et al., 1989). Bacteria account for a disproportionate share of nutrient uptake to the extent that bacterial communities act as a sink for carbon, processing most of the energy and nutrients in tropical aquatic systems, thereby serving as a basic driving force for aquatic food webs (Alongi, 1994). This function is heightened in the tropics. Bacteria enrich the protein content of detrital plant material and fecal pellets by decomposing refractory components over time. Detritivores derive a significant amount of their nutrition from digestion of this enriched material with bacteria and their mucus. Bacterial abundance is usually controlled by the carrying capacity of the system, nutrient supply and environmental conditions with the microbial food web serving as a sink for energy and major nutrients (Alongi, 1994).

Selection of appropriate mangrove species There are six recognized major groups of mangrove species based on geographical regions with different degrees of speciation; however, Rhizophora and Avicennia are considered pan-world (Chapman, 1984). Not only are species differentiated by habitat preference and tolerances, but there are also characteristics of the various species that are important to a silvofisheries system. Selection of the most appropriate mangrove species is particularly important in replanting projects as part of a silvofisheries system where a greater degree of control can be exercised. Some of the desired characteristics in selecting appropriate mangrove tree species include: l l

tolerance of trees to extended submergence; rapid growth and high litter production (i.e. high value of the litter/production ratio);

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

211

high nutrient value of litter products; rapid decomposition of litter; l mixed population of vegetation to maximize production; and l openness of canopy to allow light penetration to the mangrove floor ± enhancing algae production. l l

Lugo et al. (1988) identified a few environmental core factors (hydrological and nutritional) that are responsible for the productive characteristics of wetland forests (fresh and saltwater) and a grouping of wetland forests in three types, which consist of riverine, basin and fringe types. However, the effect of the core factors can be modified by cumulative secondary environmental factors. This serves as a helpful guide in identifying characteristics of potential silvofisheries sites. The core factors that define the fundamental niche are classified as: kinetic energy of water flow (e.g. waves, tides, water runoff); l predominant direction of water flows (whether water flows through the wetland unidirectionally as in river floodplains, bidirectionally as in fringe forests, or fluctuating vertically as in basin forests); l hydroperiod (duration and frequency); and l nutrient supply (nutrient quality of site's sediments and waters). l

In summary, structural complexity and rate of ecosystem processes usually follow the order riverine > fringe > basin. General biotic responses of freshwater and saltwater wetland environments in relation to the basic core factors consist of the following: Forest structure ± The average number of tree species decreases as follows: riverine > basin > fringe. The number of tree species decreases with increasing intensity of hydroperiod and hydrologic energy. Salinity also decreases species richness (mangroves having fewer species than freshwater wetlands). Tree density is higher in basin forests than riverine. Younger stands have higher tree densities (hurricane/typhoons maintain younger forests). Increasing hydrologic energy affects tree density. l Primary productivity and evapotranspiration ± Net primary productivity is higher (lower respiration) in saltwater forested wetlands. Riverine forests are always more productive (up to twofold). l Litter dynamics ± Average rate of litter fall is higher in saltwater than in freshwater wetlands. The rates of saltwater wetland forests follow the ranking of riverine > fringe > basin based on hydrologic energy and nutritional factors. Rates of litter decomposition are much higher in saltwater than in freshwater forests. Litter fragmentation and transport by tidal forces is partly responsible. Mangrove forests have higher total organic carbon and higher rates of carbon export than freshwater-forested wetlands. l Nutrient dynamics ± The ratio of mass of organic matter to mass of nutrients in litter fall provides insight into how much carbon is returned to the forest floor per l

212

Ecological Aquaculture

unit of nutrient mass. Mangroves return more organic matter to the forest floor per unit of N and Ca nutrient return than do freshwater wetland forests. Nutrient use efficiency by litter fall and litter turnover is higher in tidal saltwater wetlands than in freshwater wetlands. The ratio utilized of mangrove to open water area in silvofishery ponds varies substantially among the countries practicing silvofisheries. Table 8.8 illustrates this variation. Table 8.8 Comparison of the predominant ratio used of mangrove to open water area in silvofishery ponds Mangrove area

Open water area

80% 80% 30% 70% 20%

20% 20% 70% 30% 80%

Indonesia Philippines Thailand Vietnam China, Hong Kong

These predominant ratios by country vary within the countries also. This ratio is a critical factor in the basic energy dynamics of the extensive silvofishery pond system, since the mangroves are normally the major source of nutrients and organic matter supporting the pond's food web. Therefore, it will have a significant impact on aquaculture production from the system. There needs to be further research into this ratio to determine an optimum range under different conditions. It has been reported (Soewardi et al., 1996; Takashima, 1999) that productivity of shrimp increased with the increase in mangrove area within the pond (Table 8.9). However, these values are based on a production area limited to the open water portion of the total silvofishery pond (i.e. 20% in an 8:2 silvofishery pond) and then projected to a production value reflecting an area that is completely open water (100% of the pond). Furthermore, this does not account for the utilization during periods of flooding of the mangrove area by the aquaculture species. This extrapolated value can be misleading in terms of total actual production from a given land area. Production values, as well as other pond area related values (e.g. stocking density) should be based on the total land area Table 8.9 Productivity of shrimp ponds with different levels of mangrove area within the pond (Soewardi et al., 1996; Takashima, 1999)

Amount of pond area with mangroves (%) 0 40±60 70±80 80

Production projectedbased on open water area (kg/ha/yr)

Reported value adjusted to reflect production based on actual land area of silvofishery pond ± inclusive of mangrove and open water area (kg/ha/yr)

171.2 181.0 335.0 413.5

171.2 72.4±108.6 67.0±100.5 82.7

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

213

utilized in the pond system (including mangrove area). Since this will allow for a comparative analysis of different systems (e.g. silvofisheries, extensive, semi-intensive, and intensive aquaculture systems), it is essential for a realistic evaluation of the system's comparative economic values as well as a benefit/cost analysis comparison. The silvofisheries system can also have the addition of integrated products (e.g. construction material, honey, alcohol, vegetables, fruits, etc.) to those produced by the aquaculture component that would contribute to the total economic value of the system and therefore should be reflected in the total quantity of land utilized by the system. A comparison of production from silvofisheries ponds of two categories of mangrove density and age and stocking practice (natural ± with tidal influx of wild stock, and purchased ± supplemental stocking of seedstock) resulted in net revenue differing by a factor of approximately 1.8-fold increase in ponds with the lower mangrove density and stocked milkfish (Rusdi & Jasin, 1994). This further reflects the importance of the mangrove area to open water area ratio and density along with the management practices applied to the ponds' operation on the level of financial return the farmer obtains. There needs to be a balance in the diversity, area and density of mangroves that addresses not only the requirements of the silvofishery pond system, but also the broader management of the mangrove forest so as to maintain its functions and biodiversity. Caution needs to be exercised in the development of silvofishery activities, particularly in large developments, that utilize monospecies planting (e.g. Rhizophora sp.) on the overall viability of the mangrove forest. As in managed timber forests that cultivate large tracts of single tree species, this can impact on the diversity and survival of plant and animal species that are dependent on the forest system directly and indirectly.

Survival (submergence/emergence) Tolerances of mangrove species to submergence will be of particular interest in the management of the silvofisheries system, since in ponds designed with a single control of water level the mangrove trees will be exposed to extended periods of submergence when the ponds are maintained at their upper water capacity for the benefit of the cultured aquatic species. Limited research has been done on specific duration tolerance to submergence for different species of mangrove trees. However, general information can be drawn from the natural distribution of the different species within the land to seaward transition in species, which would reflect increasing exposure to periods of submergence. Waston (1928) delineated five classes based on frequency of inundation: (1) (2) (3)

species growing on land flooded at all high tides (Rhizophora mucronata); species growing on land flooded by medium high tides (Avicennia alba, A. marina and Sonneratia griffithi); species growing where they are flooded by normal high tides (majority of mangroves but dominated by Rhizophora);

214

(4) (5)

Ecological Aquaculture

species growing on land flooded by spring tides only (Bruguiera gymnorrhiza and B. cylindrica); species on land flooded by equinoctial or other exceptional tides only (Bruguiera gymnorrhiza, dominated by R. apiculata and Xylocarpus granatum).

Dagar et al. (1991) identified three conspicuous zones (proximal, middle and distal zones) within mangroves from seaward to landward. The proximal zone (seaward zone) consisted mainly of Rhizophora spp., Avicennia spp. and Sonneratia spp. However, this classification was too broad with excessive overlap of species to be useful in identifying the most appropriate species for silvofisheries other than the genera of species in the proximal zone. Por (1984) suggested further delineation of the zones to a total of 11 zones based on a linear progression from total submergence to emergence quantified in hours per day and days per month. Untawale (1987) identified five zones of mangrove distribution based on salinity: (1) (2)

(3)

(4) (5)

euhaline zone (30±40 ppt, high wave action, rocky and sandy substratum): absence of mangroves; polyhaline zone (18±30 ppt, low wave action and sandy clay substratum): Sonneratia alba, Rhizophora mucronata, R. apiculata, Avicennia marina, A. officinalis, Bruguiera gymnorrhiza, B. parviflora and Acanthus officinalis; mesohaline zone (5±18 ppt, silty clay bottom, feeble wave action); Kandelia candal, Avicennia officinalis, Rhizophora spp., Aegiceras corniculatum and Sonneratia alba; oligohaline zone (0.5±5 ppt, silty substratum): Sonneratia caseolaris, Acrostichum aureum, Scirpus sp., Cyperus sp., Fimbristylis sp.; limnetic zone (<0.5 ppt, gravel and coarse sand substratum).

The above classification by salinity can be useful in limiting species for use in silvofisheries to those found in the higher and mid-salinity ranges (i.e. polyhaline and mesohaline zones), since this would be the area subject to greater tidal influence and subsequently submerged for the greater periods as is likely under silvofisheries practices. Species-specific information assists further in delineating profiles of the most appropriate mangrove tree species. Bruguiera parviflora and Aegiceras corniculatum tolerate waterlogging and can be found in depressions (Chapman, 1976). Avicennia marina tolerates a wide range of submergences (50±707 per annum) and is the species most tolerant to high salinity (Chapman, 1976). Sonneratia species are commonly associated with normal seawater salinity, while Rhizophora mucronata and R. stylosa tolerate up to 55 ppt and require a minimum of 12 ppt (Chapman, 1976). Choy & Booth (1994) studied the effect of prolonged flooding (8 weeks) of an Avicennia marina dominated mangrove, which included a differentiation in survival by size classes. Larger trees showed the greatest ability to survive while shrubs displayed near mass mortality. Pneumatophores that were submerged died back from their tips. The prolonged flooding stimulated flowering and fruiting with large numbers of propagules produced for reseeding. In addition, prolonged flooding can result in loss of

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

215

Avicennia's associated flora and fauna. Dr J.H. Primavera (personal communication, 1998) indicated that Sonneratia alba in the Philippines did not tolerate extended submergence for more than 7 days and seedlings and saplings of S. alba died after 10 days of submergence while trees of Rhizophora and Avicennia tolerated this period of submergence. However, big A. officinalis and A. lanitas trees were vulnerable to longterm (>6 months) flooding with twigs and branches dying. A. alba trees similarly were vulnerable and died within 8 months of inundation. Elster (1998) indicated that in trials on the tolerance of Avicennia germinans, Laguncularia racemosa, and Rhizophora mangle seedlings to total submergence Avicennia and Laguncularia only survived 2±3 weeks and Rhizophora several months of total submergence. Management of tolerance to submergence of different species of mangroves can be partially addressed in the silvofishery pond design. Utilizing a design that incorporates a secondary dike around the perimeter of the mangroved central platform with additional gates for the direct inflow/outflow of water to the external water delivery canal and a gate that drains into the perimeter canal of the pond as previously illustrated in Fig. 8.2 will separate water level control. This system allows for the separate management of the water level for the mangrove and the open water aquaculture areas of the pond so that the optimum level and exchange frequency for both components can be maintained. Figure 8.24 shows a pond with this design.

Fig. 8.24 Modified silvofisheries pond model, which encloses the central platform area with the mangroves by the addition of a dike. Gates are added to allow for direct water level control of the central platform area separate from that of the open water area of the pond. This has the advantage of being able to control the water level to optimize the conditions for the aquaculture species (maximum depth) while allowing for periodic draining of the mangrove area.

An area of consideration is the species of mangrove tree utilized in the silvofishery system. This not only includes the selection of the appropriate species for the specific pond environment (e.g. submerged period), but also the growth and litter production rate of the different species. For example, a study on litter production rate found that Sonneratia alba stands produced 17.6 mt/ha/yr, and Rhizophora apiculata stands produced 14.64 mt/ha/yr of litter (Hayashi, 1994). In addition to litter production quantity, it is important to examine the quality and decomposition rate of the litter (how quickly it is utilized in the food web) (Fig. 8.25).

216

Ecological Aquaculture

Fig. 8.25 The photo shows the conversion of a previously cleared aquaculture pond into the silvofishery system. Rhizophora sp. is a slow growing mangrove species that allows for greater light penetration to the pond when planted in well spaced rows. In the background, right side, the adjacent pond area is planted with a faster growing Avicennia sp.

Excessive sedimentation around the mangroves can eventually build up the substrate to the point of blocking the aeration of the roots causing a root oxygen deficit. This along with extended tidal submergence can result in anaerobic soil conditions causing the death of mangrove trees (Saenger et al., 1983). The vertical structures (e.g. prop roots, pneumatophores, trunks) of densely forested mangroves impede the tidal current causing microturbulences with eddies, jets and stagnation zones. This facilitates the continued suspension of particulate flocs entering with the flood tide, but settling them within the forest during the slack high tide thus actively contributing to the creation of mud banks (Furukawa & Wolanski, 1996). This is an important factor in the management of a silvofisheries system, since the survival of the mangrove vegetation is critical as is reduction of stress on the vegetation that would lower the productivity of the mangroves, which is a major energy input source to the system.

Production (growth rate and litter production) The mangrove litter serves the function of a green manure forming detrital matter and the release of nutrients as a fertilizer in the silvofisheries system. Since the litter mass represents the major form of energy input to a silvofisheries pond, the degree of litter production is of paramount importance. Lee (1988) found faunal biomass positively correlated with detritus-enriched substrates and concluded detritus availability was a limiting factor in secondary production in a study of a silvofisheries pond (gei wai). A number of factors influence net primary production and subsequently litter production including tidal flushing, salinity, nutrient status, and substrate characteristics. Season, location, monsoon storms, and deforestation influence the input of mangrove litter for decomposition. Annual litter fall of tropical mangrove forests is higher than that of subtropical mangroves. This has been attributed to a number of factors including difference in the vegetation structure, climatic conditions, forest growth phase, soil fertility, tidal activity and hydrologic conditions (Kusmana et al., 1995). Site location within the intertidal zone influences levels of litter fall (Boto et al., 1984; Kusmana et al., 1995). There is a negative influence associated with increasing latitude, and increasing topographic height in the intertidal zone (Bunt, 1982; Twilley et al., 1992; Osunkoya & Creese, 1997). Salinity influence has been noted by Carter et

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

217

al. (1973) to increase net primary production with salinity to a maximum chloride level of 1±13%, and thereafter decreasing with increasing salinity for Rhizophora, Avicennia and Laguncularia. Boto et al. (1984) noted a significant and negative correlation between biomass and soil salinity with highest net primary production recorded at upstream sites subject to freshwater influence. Seasonal patterns in production of litter have also been noted (Bunt, 1982; Clough & Attiwill, 1982; Rajagopalan et al., 1986; Kusmana et al., 1995). Litter fall is comprised of the various components of the tree that are shed over time. The different components have specific characteristics and undergo decomposition at different rates. The composition of litter fall for selected species is presented in Table 8.10. Table 8.10 Percent contribution of litter components (Wafar et al., 1997) Component

Leaves Floral Fruits Others

Rhizophora apiculata

R. mucronata

Sonneratia alba

Avicennia officinalis

68.2 4.4 17.4 9.7

67.3 4.2 18.9 9.6

57.7 2.9 13.4 25.0

43.0 8.2 9.9 36.9

Litter fall recorded varies widely even for the same species. For example, litter production recorded for Rhizophora apiculata ranged from 3.4 tonnes/ha/yr in Phuket Island, Thailand (Poovachiranon & Chansang, 1982) to 15.8 tonnes/ha/yr in Malaysia (Sasekumar & Loi, 1983). The majority of these wide ranges in litter production can usually be attributed to tree height and latitude (Saenger & Snedaker, 1993; Li & Lee, 1997). The size of the trees will have a significant impact on the quantity of litter entering the system and being made available for decomposition. For example, litter production from Avicennia taller than 10 m was 8 tonnes/ha/yr while Avicennia less than 4 m in height produced 3 tonnes/ha/yr (Woodroffe, 1985). With mangrove replanted silvofisheries ponds, the mangrove vegetation will initially contribute an insignificant amount to the primary production of the pond. However, with increasing age, the initial dominant primary productivity from phytoplankton and macroalgae will switch to the emergent mangrove vegetation. Litter production influences include salinity (high salinity in sediment impairs the flora, i.e. there is low biological activity); light (duration and intensity); temperature; nutrient supply; soil type; tidal inundation (frequency and duration); drainage; age of mangrove stand ± there is a positive correlation with age; and storms. Litter yields can also be attributed to specific growing conditions that are favorable, and mangroves with greater tidal activity and water turnover have greater litter fall than mangroves in areas with stagnant waters (Pool et al., 1975). In contrast, Lee (1990a) has suggested that the restricted tidal inundation (frequency and current flow) leads to higher tree density (by limiting the dispersal of propagules) and also to higher productivity by retaining the majority of the nutrients for in situ recycling. Twilley et al. (1986a) noted an inverse relationship of litter production and average

218

Ecological Aquaculture

soil salinity and a positive correlation of litter production with water turnover resulting in a ranking of riverine > fringe > basin > scrub mangrove types. It has also been found that low availability of soil phosphorus is a major contributing factor to low net primary production (Boto et al., 1984). These nutrient constraints are site specific and affected by other soil factors (e.g. redox potential, salinity). Wafar et al. (1997) modeled litter fall using six environmental parameters with the calculated litter fall value coming within 10% of the actual measured litter fall. These parameters were: l l

l l

l l

Atmospheric temperature (high temperature influences transpiration, stressing the vegetation to shed litter). Rainfall (increased rain suggests lower salinity; increased nutrient supply with freshwater enhances litter production; low rainfall suggests higher salinity, transportation becomes metabolically too expensive resulting in thinning of canopy). Wind speed (litter shedding is accelerated by increased wind). Incident solar radiation (will affect the photosynthesis/respiration ratio (measured as oxygen produced or consumed) and increase turnover with rapid growth). Humidity (higher humidity reduces the desiccation rate and provides a more suitable environment for growth of fungi and bacteria). Evaporation rate (greater evaporation leads to higher salinity, transportation becomes metabolically too expensive resulting in thinning of canopy).

Litter production varies between different species and this is a critical factor in the replanting selection of mangrove trees to maximize litter input to a silvofisheries system. Sasekumar & Loi (1983) measured the litter fall from three mangrove forest zones (Malay Peninsula) with Avicennia > Rhizophora > Sonneratia. Bunt (1982) identified the litter production of different species (off the Australian coast 188 20'S and 1468 10'E) for the following species in declining order of litter production: Rhizophora apiculata > R. stylosa > Bruguiera gymnorrhiza > Ceriops tagal > Avicennia marina > R. lamarckii. Table 8.11 presents the production of three main species. Twilley (1982) found mixed mangrove vegetation composition (Avicennia, Rhizophora, and Laguncularis) to have greater litterfall than in a monospecific (Avicennia) by over 300%. Bunt (1982) noted diversity in the physical character of the mangrove litter, and differences in overall vegetational character from estuary to estuary may Table 8.11 Litter production (after Bunt, 1982) Species

Rhizophora apiculata Bruguiera gymnorhiza Avicennia marina

Litter fall (g dry wt/m2/day)

Range in litter fall (g dry wt/m2/day)

3.05 2.36 2.00

1.68±7.70 1.08±3.21 1.71±2.55

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

219

well exert a substantial influence on the character of dependent food webs and on trophic dynamics. This would also apply to silvofishery systems.

Food value (digestibility, tannin content, protein level) Contribution of mangroves The mangrove vegetation in a silvofisheries system will serve as the base source of nutrients that will support the production of aquaculture products. The quantity and quality of this nutrient source will be critical inputs that go to determining the production capacity of the system. Therefore, an understanding of key factors that contribute to the desirable qualities of the mangrove vegetation will be important in the management of the system. Mangroves are widely considered detritally based systems with mangrove leaves as a major nutrient and food source. The major energy pathway is through the detritus food chain (Odum, 1971; Fell et al., 1984; Reice et al., 1984). A modeling program (Across Trophic Levels System Simulation) of a Florida mangrove ecosystem divides segments of the various input/output components of the ecosystem for analysis with results indicating a significant fraction of the carbon reaching many of the predators originated from detritus (Ulanowicz et al., 1999). Though the detrital food web is considered the main path of energy flow, alternative pathways may also be significant and provide a more efficient and direct utilization of energy by upper trophic levels. A number of factors will influence the value of this detrital-based source of energy including physical structure, rate of decomposition, tannin content, nutritional content, protein level, and digestibility. The nutrient value of plants varies with growth conditions, especially soil fertility. In general, plant material is low in N and P. C:N ratio of detrital material is an important factor influencing its value as a primary food source with a lower ratio being of greater nutritional value. The difference in carbon and nitrogen elements in some common mangrove species is presented in Table 8.12. The plant material from mangrove litter is nitrogen-deficient. Leaf litter is characteristically low in nitrogen, with a C:N ratio >90 common (Twilley et al., 1986a, b; Lugo et al., 1990). Bacteria and fungi are the principal sources of this nitrogen increase. An increase in nitrogen has been identified to be two to three times the Table 8.12 Carbon and nitrogen content (% dry wt) of leaves of mangrove species (after Bunt, 1982)

Species Rhizophora apiculata Rhizophora stylosa Rhizophora lamarckii Bruguiera gymnorhiza Sonneratia alba Ceriops tagal

Carbon mean

Carbon range

Nitrogen mean

Nitrogen range

C:N

45.72 45.72 45.02 53.51 46.10 43.92

43.90±47.84 42.43±48.60 40.33±47.87 42.01±47.47 43.37±50.57 41.01±47.47

0.49 0.34 0.36 0.42 1.04 0.38

0.37±0.59 0.20±0.60 0.25±0.58 0.27±0.74 0.77±1.37 0.13±0.77

93.3 134.5 125.1 126.4 44.3 115.6

220

Ecological Aquaculture

original level during decay of mangrove leaves (Fell et al., 1975; Wafar et al., 1997). The C:N ratio of mangrove litter decreases over time as the decay process progresses. Fell et al. (1975) found considerable leaching of water-soluble materials during the first 2 weeks of submergence. The carbon level was reported to drop while absolute nitrogen values increased over the next 5 weeks of submergence. Microorganisms associated with litter are responsible for the nitrogen increase and nutritional enrichment and are a major source of nutrition in detrital material (Fell et al., 1975; Boesch & Turner, 1984; Tam et al., 1990; Lee, 1995; Wafar et al., 1997). Most consumers have a C:N ratio requirement of 17:1 or lower (Russel-Hunter, 1970). Since the C:N ratio for bacteria and fungi is 12.5 and 15.8 respectively with a caloric content of 5300 and 5523 cal/g ash free dry wt (Ausmus & Witkamp, 1974), they provide significant nutritional enhancement to the detrital material and are likely grazed by aquatic invertebrates as well as fish. In a study of leaf preference (Avicennia marina, Rhizophora stylosa, and Bruguiera gymnorrhiza) for consumption by the crab Sesarma erythrodactyla, A. marina was preferred (Camilleri, 1989). This was attributed to the lower tannin and a high nitrogen content in A. marina. Of the remaining two species, B. gymnorrhiza was preferred. A strong preference was exhibited for aged rather than fresh leaves of R. stylosa. An examination (Queensland, Australia) of consumption by two abundant species of leaf-eating mangrove crabs (S. messa and S. smithii) of leaves from four mangrove species (R. stylosa, A. marina, B. exaristata, and Ceriops tagal) showed a preference for decayed leaves over fresh (Michelli, 1993). Feeding niche characteristics were evident in the two crab species, with S. messa showing no preference distinction for the four species of mangroves, while S. smithii showed a preference for R. stylosa. However, Micheli (1993) indicated that the choice was not influenced by the leaf properties measured (tannin, water content, per cent of organic matter, C:N ratio, and leaf toughness). Nocturnal feeding of S. smithii and S. messa further delineated niche characteristics. Nutritionally, it was suggested that with the high C:N ratio of mangrove leaves, bacteria and microalgae living in the upper layer of the sediment may be the principal source of nitrogen for the crabs. In studies that examined the utilization and subsistence of nematodes on mangrove leaves (Avicennia marina and Rhizophora stylosa) it was found that only fresh A. marina sustained increased nematode populations (Alongi, 1987; Tietjen & Alongi, 1990). Poor nutritional quality of mangrove litter with high tannins and low nitrogen content is attributed to the low abundance of nematodes observed in the study. Mangrove-derived tannins were found to negatively affect meiobenthos in tropical mangroves and potentially play an important role regulating distribution and abundance of meiobenthos (Alongi, 1987). The majority of the primary production is from the emergent macrophytes (e.g. mangroves, marsh reeds). Li & Lee (1998) measured the total net above-ground primary production in a mangrove tidal pond (silvofisheries form) in Hong Kong with 95.5% from mangrove vegetation (mangroves and reeds), 3.3% from macroalgae, and 1.2% from phytoplankton. A limiting factor in phytoplankton productivity was high turbidity. Frequent tidal water exchange (cropping of the phytoplankton population) may also limit productivity, since the highest phyto-

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

221

plankton productivity in coastal ponds was reported with increased water residence time (O'Connell & Andrews, 1987). However, the phytoplankton and macroalgae production has a higher value in direct utilization by the cultured aquaculture species (e.g. shrimp, milkfish). Therefore, even though quantitatively both macroalgae and phytoplankton are less than the emergent macrophytes, they play an important role in the food web supporting the cultured species. Algae provide a high quality food source directly to aquatic herbivores that are potential aquaculture products, in contrast to mangrove vegetation which must go through a degree of decomposition before entering the aquatic food chain. In addition, dead algal cells decompose between five and ten times faster than mangrove leaf tissues (Boonruang, 1978; Rice & Tenore, 1981; Twilley et al., 1986b). The majority of marine phytoplankton production (55%) is utilized by herbivores (Cebrian & Duarte, 1994) and is therefore more directly accessible to higher trophic levels than is the carbon production of mangroves, which is mostly channeled through decomposers (Lugo et al., 1988; Lee, 1990a, b). Furthermore, phytoplankton species have been shown to be able to utilize nitrogen at the various stages of inorganic nitrogen (i.e. ammonia, nitrite, and nitrate) (Cooper, 1933; Harvey, 1939). In contrast, vascular terrestrial plants (e.g. mangroves) mainly utilize the oxidized form of inorganic nitrogen, nitrate. Wafar et al. (1997) quantitatively evaluated the nutrient input to estuarine waters from mangroves and phytoplankton. They found that per unit area of mangrove canopy cover is about twice as productive as that of a phytoplankton water-column area for C, but was only 20% as productive for N and P compared with phytoplankton. This would further illustrate the higher nutritional value of phytoplankton over mangrove litter. Protein content of senescent Rhizophora mangle leaves increased from the initial 3.1% to 22% during a 12-month decomposition period with 95% of the material decomposed (Heald, 1969; Odum, 1971). Similarly Cundell et al. (1979) found the protein value to increase from 3.2% to 5.6% after 70 days of immersion. Waghmode (1987) analyzed the nitrogen and protein levels in four mangrove species. This is compared with the protein level in algae species commonly found in mangroves (Table 8.13). This table shows the nutritional superiority of the algae to the mangrove material. Tam et al. (1990) noted massive growth of algae, Enteromorpha and Ulva, attached to mangrove leaf litter bags. This algal growth was likely stimulated by the nutrients leaching from the litter and the surface serving as attachment substrate. Similarly, algae have been noted growing directly on the litter and mangrove tree roots in other studies (Fortes, 1975, as cited by Primavera, 1996; Davey & Woelkerling, 1985; Tanaka & Chihara, 1987; Micheli, 1993). This stimulation of algal growth serves as a high nutritional food for herbivores. The high protein level of Enteromorpha spp. has been recorded as well as being a highly desired food for herbivorous fish (FitzGerald, 1978). The preference for Enteromorpha in the diet of siganids, which include species that are common to mangroves, has been documented (Tsuda & Bryan, 1973; Tsuda et al., 1974; Bryan, 1975). The limited time frame used in mangrove plant material biomass sampling could potentially contribute to the underestimation of the importance of macroalgae as a

222

Ecological Aquaculture

Table 8.13 Nitrogen content in selected species of mangrove trees and algae (Dawes & Goddard, 1978; FitzGerald, 1978; Tredici et al., 1986; Waghmode, 1987; Brown, 1991; Qasim, 1991; Naidu et al., 1993; Aisha et al., 1995; Canizares-Villaneuva et al., 1995; Manoharan & Subramanian, 1996; Vonshak et al., 1996; Kennish & Williams, 1997; Wahbeh, 1997).

Mangrove species Avicennia marina Sonneratia alba Ceriops tagal Lumnitzora racemosa Algal species Enteromorpha sp. Ulva sp. Caulerpa sp. Chaetomorpha sp. Laurencia sp. Acanthophora sp. Dictyota sp. Spirulina sp. Oscillatoria sp. Nitzchia sp.

Nitrogen content (g/100 g dry wt)

Protein content (g/100 g dry wt)

0.89 2.72 1.14 0.25

5.05 3.25 6.50 1.43

4.50 2.1 3.08 3.8

14±27 13±21 10±18.3 24±37 24.5 21±23 18±27 30±62 30±41 26

3.39±3.72

food source with their characteristic rapid growth and turnover influenced by seasonality and cyclical growth in some species. Ali et al. (1983) attributed higher carbon fixation during summer months to increased algal standing crop. Even though algae are minor contributors in terms of total biomass of the plant material within mangroves at a given time, they are important sources of high quality food for the aquatic species. The significance of algal production in mangrove areas has been pointed out by Rodelli et al. (1984). Sullivan & Moncreiff (1990) noted the general underestimation of the importance of benthic algae in salt-marsh food webs. In a study of a salt marsh, the findings of Sullivan & Moncreiff (1990) strongly suggested that the main food source for the marsh's invertebrates and fish were benthic and planktonic algae. They also noted the food source preference for these algae by lower trophic level consumers (e.g. harpacticoid copepods, nematodes, ostracods, mollusks, juvenile shrimp, and certain fish species). Similarly, Hughes & Sherr (1983) showed a high dependence on benthic algae and phytoplankton by the fauna in a salt marsh, with more than 55% of the body carbon derived from algal carbon. Benthic microalgal production on mudflats has been recognized as an important nutritional component in marshes and off-shore areas from mangroves. It is likely that a benthic microalgal complex plays a significant role in the nutrition of the aquatic fauna of mangroves. Dittel et al. (1997) examined the importance of mangrove particulate detritus in the diet of postlarval Penaeus vannamei. They found that postlarvae reared on zooplankton, detritus and meiofauna diets more than tripled their weight (over a 6-day

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

223

trial period) whereas postlarvae fed a detritus-only diet barely doubled their weight. Their stable isotope analysis indicated that the shrimp postlarvae preyed on organisms that derived their carbon and nitrogen from benthic algae and/or phytoplankton. They concluded that a benthic algae-based food web is a primary source of carbon for the postlarvae. However, the findings also indicated that nitrogen was derived indirectly from the degradation of mangrove-derived organic matter, suggesting nitrogen from mangrove detritus is remineralized and taken up by benthic algae at the base of the food chain. This therefore supports their hypothesis that mangrove detritus is an important source of nitrogen from shrimp food webs in the studied mangrove ecosystem. There is an increasing recognition of the high productivity and nutrient importance of diatoms, cyanobacteria, and filamentous algal communities (Hughes & Sherr, 1983; Sullivan & Moncreiff, 1990; Mallin et al., 1992; Newell et al., 1995). It has been suggested that primary production from periphyton and phytoplankton in mangrove systems can be significant where shading is not severe and the ratio of open water to forest is high (Robertson et al., 1992, as cited in Primavera, 1996). Kristensen et al. (1988) found the primary production of benthic microalgae was lower in the mangrove study area compared with that reported for other intertidal mudflat areas (nonmangrove) and attributed this lower production to shading and low dissolved inorganic nitrogen in the mangrove. Li & Lee (1998) found in Futian Mangrove Nature Reserve (China) that Avicennia usually has a more open canopy allowing greater light penetration, resulting in higher algae production. A diversified flora of blue-green algae was found to develop on aerial roots and the muddy sediment around pneumatophores that are exposed to sunlight, whereas in shaded forest areas they were absent (Dor & Levy, 1984). However, direct exposure to full intensity light will also impact on algal species, particularly through accelerated desiccation. Beanland & Woelkerling (1983) noted the influence of Avicennia tree canopy presence or absence on the algal species frequency distribution on pneumatophores with species preference for shade or sun exposed pneumatophores. Almodovar & Pagan (1971) found that removal of shade caused disappearance of algae from the intertidal portions of stilt roots and lowered the species diversity of sublittoral algae. Therefore, it is important to reach a balance between the density and growth of the macrophyte vegetation with the light intensity requirement for the growth of the algae. The emergent macrophyte growth on the central platform of a silvofisheries pond significantly increases the potential vertical surface area (e.g. pneumatophores, prop roots, etc.) for the growth of algae. The extent of this surface will depend on the tidal flooding and frequency level and duration of exposure that allows for the growth of the algae. Davey & Woelkerling (1985) demonstrated the rapid recolonization of algal species on the denuded intertidal portion of pneumatophores. In addition, the exudates associated with these structures are high in organics that can potentially enhance the growth of microalgae, which would provide an additional potential food source for grazers. Burkholder & Almodovar (1973) noted that mangrove roots provide an excellent supportive substrate and favorable conditions for algal growth. They further noted that this epiphytic community of seaweeds provides a source of food and shelter to numerous invertebrates and vertebrates. They found the algae on

224

Ecological Aquaculture

a mangrove root (Rhizophora prop root) to be equivalent in organic matter productivity to approximately a square meter column of phytoplankton in a moderately rich tropical lagoon. Vertical zonation of algae with species differentiated on mangrove stilt roots, knee roots and pneumatophores is predominantly influenced by water movement, tidal frequency, and emersion duration (Davey & Woelkerling, 1985; Tanaka & Chihara, 1987). Tanaka & Chihara (1987) identified four macroalgae zones within a mangrove forest (Okinawa) consisting of a Rhizoclonium zone, a Bostrychia zone, a Caloglossa zone, and a Catenella±Cladophora zone that respectively represent zones of decreasing emersion. Davey & Woelkerling (1985) identified horizontal zonation through the mangroves related to the degree of emergence with a decrease in frequency in occurrence, relative cover, and biomass moving landward for the three species of red algae examined. Macroalgae growing on mangrove trunks, prop roots and litter fall were common food items for the crab Sesarma messa and possibly other grazers, which indicates the potential significance of benthic algae to the nutrition and energy flow in the mangrove fauna (Micheli, 1993). Utilization by aquaculture species The relative importance of food items ingested does not necessarily correlate with their assimilation or importance as a nutritional source. This is emphasized by Primavera (1996), particularly with shrimp. As noted by Chong & Sasekumar (1981), even though organic detritus frequently appears in the diet of penaeids, it does not necessarily reflect its importance as a food source. The organic detritus probably serves as a supplement when other preferred food items are scarce. In addition, it serves as a substrate for microorganisms, which are of higher food value. Fry (1984) found that seagrass detritus in a seagrass bed was not the dominant carbon source that supported the majority of the associated fauna, but rather it was algal-based food webs. In addition, blue-green algae complexes can form important components of the diet of certain mangrove fauna. This indicates a management role in facilitating the environmental conditions for algal growth in support of aquaculture species commonly used in silvofishery systems. As discussed, there are different sources of primary production and numerous factors that contribute to the nutritional food value of these sources in accordance with their ingestion, digestion, and assimilation. This has led to a wide variation in the level of importance of the dietary components contributing to shrimp nutrition. Newell et al. (1995) found mangrove detritus to contribute directly to the nutrition of juvenile shrimp within the tidal creeks and indirectly through consumption of small detritivorous invertebrates inhabiting mangroves, other trophic intermediaries, and benthic microalgae. Based on stomach content analysis alone (not analyzed for assimilation), organic detritus was assumed to be a very important food source for penaeid shrimp constituting 72±100% of the stomach content (Su & Liao, 1984). In a study of natural food types in an extensive pond system for P. monodon, it was found that detritus ranked highest (gut content analysis) in frequency of occurrence (95± 100%), followed by copepods and other animal remains (Bombeo-Tuburan et al., 1993). They concluded detritus was the primary food. Qasim & Easterson (1974)

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

225

found an average 93% assimilation efficiency of Metapenaeus monoceros when fed exclusively on detritus. An analysis of P. merguiensis stomach contents found that 85% of the guts examined contained mostly decaying mangrove material and the animals were classified as carnivorous detrivores (Chong & Sasekumar, 1981). The animal food chain components would include foraminiferans, copepods, amphipods, gastropods, and polychaetes. In a study of the assimilation efficiencies of seven species of penaeids (Moriarty & Barclay, 1981), the food items consisted predominantly of Foraminifera, small mollusks, crustaceans and polychaetes. Wassenberg & Hill (1993) found a similar dietary composition for P. merguiensis juveniles and adults. They further found that tidal stage, daylight, and size of P. merguiensis affect the composition of the diet. In an Australian mangrove forest habitat (creeks), detrital floc consisting of organic aggregates and small particles from fecal pellets found in a matrix with diatoms and dead algal filaments comprised up to 74% of the gut volume for P. merguiensis small juveniles (Robertson, 1988b). However, the composition of the diet (e.g. detrital floc, crustaceans, bivalves, polychaetes, and insects) varied between the three size groups of shrimp and locations sampled. Even though Robertson (1988b) found that the importance of recognizable mangrove detritus increased with prawn size, it never exceeded 15% of the volume of the diet. Utilizing stable isotopes in the analysis of the material flows in aquatic ecosystems often provides a more definitive identification of the origin of major dietary components actually assimilated (as opposed to just ingested) and the potential trophic level utilization. The isotopic composition of aquatic organisms can provide basic information on their food source and trophic level, with the 13C and 15N content of animals reflecting their diets. 15N enrichment reflects a trophic level effect. 13C is utilized to trace the flow of food items into the target species, with 34S occasionally being used to further clarify the source. Rodelli et al. (1984) identified carbon isotope values consistent with about 65% of the tissue carbon derived from mangrove detritus with the remainder from phytoplankton and benthic microalgae from several species of juvenile prawns collected within tidal creeks draining a mangrove forest. Shrimp were found by use of stable carbon isotope ratios to obtain most of their dietary carbon from benthic blue-green algae (algal mats comprised mainly of Spirulina sp.) rather than mangrove detritus (Stoner & Zimmerman, 1988). Less than 25% of the diets of the three penaeid species (P. notialis, P. brasiliensis, and P. subtilis) in a mangrove fringed lagoon in Puerto Rico were from mangrove detritus. They also noted the shift towards larger prey (e.g. amphipods, polychaetes) as the shrimp size increased (Stoner & Zimmerman, 1988). In a study of nutritional sources for penaeids in a Malaysian mangrove, Newell et al. (1995) using stable isotopes (13C, 15N, and 34S) identified mangrove detritus as contributing to the nutrition of juvenile P. merguiensis living within the tidal creeks, but not in off-shore populations. They also suggested that the mangrove material might be utilized indirectly by feeding on small detritivorous invertebrates. In addition, benthic microalgae was noted as a significant dietary component. Similarly, Loneragan et al. (1997) found a declining role of mangrove detritus in the diets of juvenile penaeids and metapenaeids by location from upstream mangrove creeks to off-shore sites; therefore, the composition of the diets depended on the location of the shrimp. In an analysis of stable carbon and nitrogen

226

Ecological Aquaculture

isotope ratios of primary producers and juvenile shrimp species (Penaeus monodon, P. merguiensis, P. indicus, and Metapenaeus ensis), Primavera (1996) showed that juvenile shrimp in a riverine mangrove derived their carbon from plankton and epiphytic algae rather than directly through the mangrove detrital pathway. In addition, the nitrogen isotope values suggested a 2±3 trophic level shift between phytoplankton and shrimp. This wide range in dietary material and quantity reflects not only species preferences, but also site-specific characteristics and availability of the dietary components; therefore, penaeid shrimp tend to be opportunistic omnivores. Peterson & Fry (1987) highlighted in a case study the difference in diet of the same species of mussel as a function of location in which the organisms grew. The food resources available at various sites within a saltwater marsh ecosystem determined the diet. This was reflected in the isotopic carbon tracer indicating a phytoplankton source for the mussels collected by the ocean, while those collected at the inner marsh area reflected a salt-marsh grass (Spartina) source. These results further emphasize the influence of microhabitat on diet utilization, which would add to the variance of results for shrimp and other species analyzed under different study locations. Loneragan et al. (1997) found in a multiple stable-isotope analysis that the primary source of carbon supporting food webs of several species of juvenile penaeid prawns depended also on the location within the estuary. In their findings, the mangrove/ terrestrial predominance in the food web was limited to the small spatial scale of the mangrove fringe of small creeks, while macroalgae and seston increasingly contributed to prawn nutritional importance in food webs seaward. Chong et al. (2001) attempted to clarify the importance of mangrove detritus as a primary food source of juvenile prawns through the use of dual stable-isotope C and N ratio analyses. They found a declining transition in importance of mangrove detritus (as high as 84% of total assimilated carbon) from the upper estuaries of mangroves (i.e. Malaysian Matang mangrove swamp 8 km inland) moving downstream and increasing distance off-shore. The isotope ratios indicated that prawns assimilated what they consumed, which would also be indicative of key food sources available in their specific habitats. The seemingly conflicting findings between various research workers and studies on the degree of contribution of mangrove detritus to the nutrition of prawns can be partially attributed to study site specific characteristics (e.g. distance upstream, density of mangrove vegetation, species predominance, litter fall quantity, tidal flushing, shading of waters, season, etc.) that change the contribution ratio of key carbon sources available from plankton, macroalgae, or mangrove vegetation. It is important to know the dietary requirements and preferences of the potential aquaculture species to select the most appropriate species when possible to best utilize the mangrove litter energy based silvofishery system. The high protein requirement (34±42%) of penaeid shrimp (Sedgwick, 1979) indicates that detritus is not its primary source of food. The diet of penaeid shrimp is dependent on their microhabitat, abundance of food items, and size of shrimp (Chong & Sasekumar, 1981; Robertson, 1988b; Stoner & Zimmerman, 1988). It was concluded (Minello & Zimmerman, 1991) that juvenile white shrimp (P. setiferus) had a better capacity to directly use plant foods than brown shrimp (P. aztecus), which appeared to be an obligate carnivore

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

227

feeding on epifaunal and infaunal organisms. Sedgwick (1979) found in trials that the rate of food consumption by P. merguiensis was related to the energy content of the diet and implied the protein level required for supporting maximum growth and optimum protein conversion efficiency is also energy dependent, with the energy to protein ratio changing with age. Assimilation of mangrove carbon with an efficiency of 13.4% was found for juvenile P. merguiensis, while it dropped to 2.1% for adults illustrating a possible ontogenetic loss of cellulase enzyme production (Newell et al., 1995). Refinement in the analysis of shrimp diets indicates a relatively low role of mangrove detritus and a greater role of carnivorous prey as the shrimp increase in size. For example, the different life stages of Penaeus merguiensis have the following dietary preferences: (1) (2)

(3)

pelagic postlarvae are carnivores feeding largely on copepods; epibenthic postlarvae and juveniles are carnivorous detritivores feeding mainly on organic detritus along with prey of foraminiferans, copepods, larval bivalves and brachyuran larvae; and adults are detritivorous carnivores mainly feeding on larger crustaceans and lesser amounts of organic detritus (Chong & Sasekumar, 1981).

Other potential aquaculture species including milkfish, tilapia, and mullet are capable of utilizing the detritus as well as primary production (algae and phytoplankton). Mud crabs (Scylla sp.) are opportunistic omnivores (Brown, 1993) and would be likely to utilize some dietary components similar to shrimp. The use of a polyculture stocking strategy would allow a fuller utilization of the food web developed in a silvofisheries pond. The degree of importance in direct consumption of detritus for shrimp will vary with the size of the shrimp (stage in life cycle), species, environmental parameters, and availability of food items in the microhabitat. However, the importance of detritus in the silvofisheries system remains central to the majority of aquaculture species' food items through direct and indirect dependence on the detritus base and consumption at the secondary trophic level. Therefore, the determination of the specific degree of direct consumption and assimilation of detritus by the various potential aquaculture species remains in a varying state of uncertainty as refinement of analytical research techniques progresses. However, the importance of detritus as a direct or indirect contributory base to the natural food web supporting the aquaculture species is well established.

Management of silvofisheries system Management of a silvofisheries system is essentially ecosystem management of the mangrove properties along with production of aquaculture species. It is important to understand the various properties of the mangrove ecosystem and how they will impact on the silvofisheries system over a gradient of conditions (diverse types of mangroves under different ecological and geomorphic conditions) in formulating a management plan.

228

Ecological Aquaculture

Management of the silvofisheries system has a common basis with other aquaculture systems; however, being an extensive system and utilizing co-management practices with silviculture in the management of the mangrove vegetation, it will have some unique characteristics. The management is simplified compared with that of an intensive system. However, at the same time it is more complex, in that to successfully manage the system and to optimize production a fuller understanding of a complex ecological biosystem is preferred. Some of the key points in the management of a silvofisheries system include the following. Pond site selection: evaluation of soil characteristics; evaluation of water quality; evaluation of tidal characteristics; evaluation of availability of seedstock; buffer zone. l Pond design and construction: determination of which silvofisheries model to use; selection of appropriate mangrove species, density and potential composition; design; construction (dike, perimeter and inner channels, sluice gate); pond operation (water management; stocking; grow-out; control of predators; provision of nutrition; harvest). l

This list of key management points is not meant to be comprehensive covering all aspects of the establishment of an aquaculture activity (e.g. harvest±postharvest handling, processing, market identification, coordination and development, etc.), but to highlight areas from a broader programmatic perspective. Some further explanation of these points with regard to their unique application in a silvofisheries project is provided below.

Pond site selection The same basic criteria for site selection that apply to extensive or semi-intensive brackishwater aquaculture ponds located within the mangroves apply to a silvofisheries pond, the exception being that it is essentially a prerequisite to be located within the intertidal mangrove forest for the silvofisheries model. These criteria are identified in Table 8.14. Additional criteria apply in the case of silvofisheries ponds. These include the following: l l

species of dominant mangrove in natural area, and selection of species for planting.

Availability of autotrophic-generated nutrients and the process of regeneration are the major factors in governing productivity in the silvofisheries system. In addition, the availability and species composition of natural stocking material is critical. Siting will be important, particularly with total dependence on natural stocking and tidal flushing. The closeness of the site to natural drainage and waterways within the mangroves will facilitate the movement of natural stocking material. These factors will have an overriding influence on the production of the aquaculture product in this type of system and therefore heighten the importance of

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

229

Table 8.14 Site selection silvofisheries brackishwater ponds criteria for initial screening. Most suitable is >42 (75% or more of possible points); suitable is between 28 and 42 (<75>50%); not suitable is <28 (<50%) (from Hajek & Boyd, 1994) Property

Slight 2

Moderate 1

1.5±2.5

1.0±1.5

Location in intertidal (m above MSL)

0.0±1.5

70.5±1.0

Elevation (m below mean high water level)

0.5±1

Green belt buffer from coast (m) Green belt buffer from river (m)

Location Tidal amplitude (m)

Water source Total dissolved solids (TDS mg/l) Intake source

Water quality Salinity (ppt)

Severe 0

Restrictive factor

<1.0 or >2.5 Sufficient tidal range 0.5±2.0

Pumping required to drain or fill

0.0±0.5

>1.0

Water exchange to allow for filling and complete drain

>150

50±150

<50

Protection from shoreline erosion and storm damage ± site-specific conditions

>25

25±10

<10

Protection from dike erosion and flooding damage ± site-specific conditions

15 000±25 000

5 000±15 000

<5 000

Ocean coast, bay

Estuary, river Upstream river ± large with low± flow moderate flow

Osmoregulation Limited quantity and quality of water supply

15±25

25±35

<15

Low salinity, osmoregulation ± impacts on species selected for culture

Turbidity ± soil particles (NTU)

20±100

100±200

>200

Excessive shading, pond sediment accumulation

Transparency (plankton, cm)

30±40

25±30

>40

Excessive plankton

DO (ppm)

4±7

7±12

<4

Low oxygen

pH

8±8.5

7.5±8

<7.5

Acidity

<10%

10±20%

>20%

Stress on cultured species ± impact on immune system ± mortality

>100

50±100

<50

Potential acidity or toxicity

Thickness of organic soil material (cm)

<50

50±80

>80

Seepage; hard to compact

pH of 50±100 cm layer

>5.5

4.5±5.5

<4.5

Too acid

Parameter fluctuation (temperature, salinity, DO, ammonia, pH) within 48± 72 hr period Soil Depth to sulfidic or sulfuric layer (cm)

Contd

230

Ecological Aquaculture

Table 8.14 Contd Property

Soil (contd) Soil texture

Slight 2

Moderately fine

Moderate 1

Severe 0

Fine

Medium

Slope of terrain (%)

<2

2±5

>5

Depth to rock (cm)

>150

100±150

<100

<2 000

2 000±2 500

>2 500

60

60±150

>150

Frequency of flooding

None

Occasional

Frequent

Temperature average (8C)

28±32

25±28

<25

Climate Rainfall (annual, mm) Rainfall seasonality (consecutive days without appreciable rainfall)

Land use Current use

Vegetation on site

Mangrove density (mature growth) Mangrove species

Ratio of mangrove to open water area

Restrictive factor

Seepage Slope Shallow; seepage Excessive rainfall reducing salinity; osmoregulation High salinity

Flooding Lower temperature ± slow growth

Unused/ abandoned aquaculture ponds and rice fields

Agriculture cropland

Present User conflict, negative mangrove impact on undisturbed preservation mangroves area or within green belt

Diverse mature mangrove forest

Small mangrove trees natural or planted monoculture

Cleared grass, Planting costs and lag shrub period for mangroves to produce adequate organic input

1.0±1.5/m2 Avicennia and/or Sonneratia 70±80%

0.2±1.0/m2

<0.2/m2

Multispecies

Rhizophora

Quality of litter and quality of mangrove habitat

<50%

Quantity of litter and water area for aquaculture species

50±70%

Quantity of litter

proper site selection. Adequate access and security should also be considered in selecting a site.

Pond design and construction The various pond designs and construction have been discussed elsewhere in this chapter. Pond designs will be very site specific and depend on the associated mangrove forest system, along with policies and regulations applied. Some key considerations in design are:

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

l l l l l l l l

231

Mangrove area to open pond water area ratio (40±80%); The ratio of water area to pond dike length (reflects production area to capital cost investment); Gate width ratio (cm/ha) ± important in taking in wild seedstock and nutrient flux. It should be >50 cm/ha; Tidal flushing rate; Tidal amplitude range (1.5±2.5 m); Flow of water within a pond to prevent stagnation (low oxygen level <4 ppm); Depth of water and duration of flooding on the mangrove forested central platform (20 to 50 cm deep; duration dependent on species, but generally < 7 days); Depth and width of perimeter channel (0.8±1.2 m deep; 5±8 m wide).

In pond aquaculture it is usually advantageous (management and production) to use a rectangular design; however, in a silvofishery pond a rectangular shape does not provide these advantages since the water culture area is actually a canal. It is more efficient from a cost perspective to construct a silvofishery pond as a square, since this reduces the linear size of the perimeter dike and the associated costs. One of the important considerations in a silvofishery model is the pond water area to perimeter dike total length. This is a measure of potentially productive pond area to the construction capital cost of the dike. To illustrate the difference between the two basic silvofishery models a 1-ha (200 m 6 50 m) impoundment is used. For an empang parit model (Fig. 8.1a), this would give a ratio of 5:1 (2000 m2 of water culture area)/500 m perimeter dike). This is compared to a ratio of 20:1 (10 000 m2 of water culture area/500 m perimeter dike) for a brackishwater pond in an alternating brackishwater pond/mangrove model (Fig. 8.1d). This would give the latter system a construction cost advantage of a fourfold increase in the ratio with a potentially greater income-generating area for the same pond dike construction cost. However, where replanting of the adjacent area (outside the pond) within the model is necessary, these costs should be included. This would be best suited to privately owned land and to mangrove areas that have been cleared previously so that the cost of clearing the central platform is not included. The size of silvofisheries ponds ranges from less than 1 ha to 10 ha. There is a tradeoff as when the size of the pond increases the management level decreases. However, the major benefit of large ponds is the reduced construction costs (fewer sluice gates and less dike construction) per unit of water area. A pond size of 1±2 ha provides a high level of management while keeping construction costs to a moderate level and is a suitable size for incremental lease units. The tidal flushing rate is dependent on the tidal range, which is a critical criterion under site selection, and is managed with the use of the sluice gate. The sluice gate should be of adequate width to allow for the flushing of the pond as needed. A minimum gate width ratio that should be considered is a value >50 cm/ha. In addition, it is important to properly design the gate with angled entrance wings on both sides of the dike to minimize erosion of the dike from the flow of water around the gate and to provide a more efficient flow of water through the gate by reducing eddies and turbulence as the water enters and exits the gate.

232

Ecological Aquaculture

The perimeter canal should be deep enough so that the water level can be maintained at approximately 0.8±1 m when the central mangrove vegetated area is drained. This is critical to maintain an adequate water volume to support the aquaculture species. If this area is of inadequate depth, the potential is increased of high temperatures and low oxygen levels that are detrimental to the growth and survival of the cultured species. The width of the perimeter canal will be dependent on the mangrove vegetated/open water ratio. However, it is usually at least 5 to 6 m wide. Interior cross canals would be of smaller width and depth (e.g. 1±2 m wide and 50±80 cm deep). The dike wall along the perimeter canal can be planted with a single row of mangrove seedlings to minimize dike soil erosion (Fig. 8.26).

Fig. 8.26 Rhizophora sp. seedlings planted along the dike bank to minimize soil erosion.

The use of channels across the mangrove tree platform area is advisable. To facilitate the movement of detrital energy associated with decomposition of litter around the mangrove trees to the aquatic environment, the addition of deep channels across the vegetated platform was noted to significantly increase the portion of litter entering the aquatic sector of the pond (Lee, 1989a). In addition, cross channels facilitate water circulation in the pond and reduce the occurrence of stagnant or low water movement areas and low oxygen levels. The channels would also allow easier access (both ingress and egress) for the aquaculture species to the platform area for feeding during high tides. The maintenance of a `green belt' or an area of mangrove tree growth to act as a buffer between the coastline and rivers should be perpetuated. A width that is adequate to prevent erosion associated with flooding and wave action should be maintained. The percentage of mangrove area within silvofisheries systems and the tree density vary substantially; however, since a silvofishery system is operated as an extensive system, the reduction in mangrove vegetated area will reduce the potential nutrient input that supports the food web of the system. This is also a balance between the environmental objective of maintaining a maximum level of mangrove vegetated area

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

233

and that of open water area for cultivation of aquaculture species. As a general guide 60±80% of the area for mangrove vegetation should be the target. The tree density and orientation will have an impact on the level of sunlight reaching the mangrove floor and the open water area, which impacts on the production of benthic algae and phytoplankton. The proper orientation of planted tree rows can increase the sunlight penetration by aligning the rows so that they allow for maximum southern exposure (in the Northern Hemisphere) between the tree rows.

Pond operation Maintaining a silvofisheries system requires an adequate supply of water and nutrients, and stability in the system. A stable system that avoids wide fluctuations, which would have a negative impact on the production of the aquaculture products, is beneficial. Management practices can be divided between two main categories of (1) totally extensive with all natural ± no supplementation of seedstock, fertilizers, or feed; and (2) supplemented extensive (extensive plus) with moderate supplementation of natural production ± addition of purchased seedstock, fertilizer, and limited feeding. In the conversion of abandoned or low production aquaculture ponds to silvofishery ponds, there will be a transition phase affecting the management and production of the aquaculture component. This is associated with the period from when the mangrove seedlings are planted until the trees mature with full canopy cover in the silvofishery pond. This transition period affects the open water area management as well as the aquaculture species that can be cultured. The most dramatic impact is on the cultivation of seaweeds (Fig. 8.27).

Fig. 8.27 A silvofisheries pond that was recently converted from an aquaculture pond with low production (Sulawesi, Indonesia). Gracilaria and milkfish are being polycultured. The Gracilaria production will have to be phased out as the pond transitions from this initial conversion with the planting of seedling Rhizophora to a more mature stage in which the Rhizophora canopy cover increases and shades out the pond bottom.

234

Ecological Aquaculture

Water management The major management tool in a silvofisheries pond is the ability to control tidal inundation and water level. Tidal inundation frequency and duration are important in determining the species of mangrove trees suitable for the pond platform area. Furthermore, it also has a major influence on the pathway of litter utilization and the degree of export to the adjacent water column. The purpose of controlling this water level is to facilitate conditions for the rapid decomposition of the mangrove litter and the flushing of the dissolved and particulate organic matter and nutrients into the water canals for utilization in the food web supporting the cultured species. Therefore, management of tidal inundation has a significant impact on how much organic matter enters the water column and becomes available to the cultured aquaculture species and ultimately the quantity of aquaculture products produced. Management of tidal flooding should allow the incoming tide water to enter rapidly, bringing with it suspended nutrients and juveniles of the cultured species of shrimp and fish. The movement of particulate organic carbon (terrestrial, mangrove, and marine origin) is the greatest with large amplitude tidal cycles (Rezende et al., 1990). The holding of the tidal water for a period to allow flocculation of sediments and settlement of the organic nutrients followed by the slow draining of the tidal water to minimize the resuspension and export of sediment-associated nutrients will maximize the potential net import of extraneous nutrients. This procedure would be repeated to increase the net energy flow into the pond. A study of mangroves has shown a substantial importation of suspended particulate carbon and nitrogen during rising tides (Kristensen et al., 1988). The water entering channels at high velocity carried up to twice the level of suspended material, which decreased proportionally with the declining water velocity. Receding tides were low in suspended particles thus indicating the material was deposited within the mangrove system. Water quality and landscape control (sedimentation impacting on drainage, canal depth and volume, height of the central platform that changes over time and its corresponding influences on the vegetation species and production, etc.) are important management elements in a silvofishery system. Accretion of sediment within the mangrove area over time presents a problem, since tidal inundation frequency and duration decrease. This not only impacts on the energy flow from the vegetated platform to the adjacent water column within the pond, it impacts over time on the survival of the mangrove species present. Therefore, this manifests a management problem, particularly in locations where incoming water carries a heavy sediment load. Periodic flushing and dredging of the platform and canals will be required to control this sediment accretion. Tidal ponds (silvofisheries) with low frequency and strength of tidal inundation into the mangrove vegetation stand (high zoned, infrequently inundated mangrove community type) impact on the energy flow to the peripheral water canals. It was found in gei wai tidal ponds in Hong Kong that there is probably less than 1% of the litter produced that becomes available to the water column consumers; however, there is dissolved organic matter (DOM) export (Lee, 1989a). This illustrates the importance in siting and the potential impact of soil sedimentation over time in transforming the vegetated pond platform area from

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

235

strong tidal influence characteristics to one that is in transition to a minimum inundated terrestrial forest and its ultimate impact on energy flow in the system. Similarly, reduction in nutrient export in higher zones with a low frequency of tidal inundation was noted by Twilley et al. (1986a). Stocking/harvesting Mangroves and intertidal wetlands serve as a nursery and feeding ground for fish and prawns (Boesch & Turner, 1984; Gundermann & Popper, 1984; Robertson & Duke, 1987; Morton, 1990; Sasekumar et al., 1992; Chong et al., 1996). Many of these species are suitable for culture in silvofishery systems, including Penaeus spp., Metapenaeus spp., Scylla spp., mullet, milkfish and siganids. In a study of shrimp recruitment into mixed shrimp±mangrove forestry ponds, there was a wide variation in recruitment (Johnston et al., 2000a). The dependence on wild stock recruitment with low net recruitment (0.01±1.6 postlarvae/m2) contributed to the low overall production from the ponds in the study. A number of environmental factors impact on immigration activity of shrimp postlarvae into the mangrove estuaries including tidal cycles, moon phases, rainfall, temperature, and nutrient levels. The following factors were identified to be likely influences on the recruitment in the study; however, analysis of the data from sampled farms was inconclusive. This can be attributed to site-specific characteristics and the interaction of these variables masking clear identification of the influence of each variable, such as: l

l l l l l l

l l

Variation in hydrodynamics of the waterways: inner canals and smaller waterways tend to have lower postlarval migration compared with rivers and subsequently lower recruitment to ponds away from the river or larger canals. Tidal amplitude: less migration in water areas with decreasing tidal amplitude. Management practices: timing of harvests and recruitment with tidal change to maximize each activity. Day vs night spring tide recruitment: tendency to better recruitment at night but not significant. Duration of recruitment: concentrate recruitment period during initial days of each spring tide with peak immigration during first half of each flood tide period. Consecutive days in a recruitment period: densities of recruited shrimp highest on the first day of the spring tide. Seasonal recruitment peaks: two main seasonal recruitment periods with difference in species composition and tendency to variation in size and number of postlarvae. Sluice gate size: the width of the sluice gate in relation to the pond size will impact on water flow efficiency during both harvest and recruitment. Localized conditions: poor water quality, sewage, agricultural fertilizers and pesticides contribute to localized reduction in seed densities.

In a study of juvenile penaeid recruitment to mangroves in the Central Philippines, there was a pattern of twice-yearly recruitment that coincided with warm water

236

Ecological Aquaculture

temperatures and moderate salinity (Primavera, 1998b). It was further noted that restricted distribution of the dominant penaeid species (Metapenaeus ensis, M. anchistus and Penaeus merguiensis) was potentially attributed to spatial positioning based on salinity and/or substrate preference. Complete reliance on wild stock recruitment contributes to the variation in production and increasing risks inherent in production. Replacement of this dependence with the use of hatchery-produced seedstock will reduce the risks in production and make it more manageable and stable. Strong, fast-flowing water current through the sluice gate facilitates the entry of shrimp and other cultured juveniles from the adjacent waters. It also facilitates the harvesting of the grown-out shrimp from the pond, which are caught in a mesh net bag at the sluice gate. Therefore, it is critical to have sufficient tidal amplitude to create these strong flows of water through the sluice gate. In addition the sluice gate needs to be constructed wide enough to allow for the volume of water to flow through. Stocking and harvesting are mainly done on spring tides when this tidal amplitude is at its maximum. When harvesting can be coordinated with night hours, this will be more efficient since the shrimp are generally most active during the night (Edwards, 1978; Garcia & Le Reste, 1981; Vance, 1992; Vance & Staples, 1992). Applying trophic conversions previously discussed to the potential production from a silvofisheries pond can be done to obtain a production target. However, the complexity of the system and interaction of variables that influence production will have a substantial impact. With this in consideration, based on a feed conversion of 50 kg of plant material to produce 1 kg of fish/shrimp silvofishery product (assuming that there is a 10:1 conversion for each trophic level and the fish/shrimp are gaining their nutrition from feeding at the first and second trophic levels above the primary production) within a 1-ha pond and an 8:2 mangrove/canal area ratio, it could potentially produce 900 kg/yr of fish/shrimp with an available litter input rate of 45 000 kg fresh wt/ha/yr. This is based on the mangrove litter being the main source of nutrient input; however, since there are other various inputs the production would be adjusted upward in relation to those additional nutrient inputs. Rajagopalan et al. (1986) applied the conversion of mangrove productivity to the production of aquaculture products. Utilizing a growth efficiency of 10.5 to 35.2% and the assimilation efficiency of 93% reported in laboratory trials (Qasim & Easterson, 1974) on energy conversion of mangrove detritus to Metapenaeus monoceros the potential production of shrimp was calculated. Applying conversion efficiency factors of 50% from litter fall to detritus and 20% from detritus to shrimp, they projected the potential shrimp production and the stocking rate that would be supported by the system (Table 8.15). The above is useful in estimating potential production from a system and the required number of juveniles to adequately stock the pond. Conversion factors will most likely have to be adjusted to the system to which it is being applied. The above calculation is for a 1-year period and it is most likely that the shrimp grow-out will be divided into two or three grow-out periods within a year; therefore, the actual stocking per grow-out cycle would be the calculated based on the annual stocking rate divided by the number of cycles per year. Takashima (1999) indicated that the

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

237

Table 8.15 Calculation of potential production and required stocking rate (after Rajagopalan et al., 1986) Calculated values Quantity of litterfall Conversion efficiency Quantity of detritus Conversion efficiency Quantity of shrimp (weight) Number of shrimp harvested assuming an average size of 12 g Stocking rate (assume 80% survival)

5 000 kg/ha/yr 50% (2:1) 2 500 kg/ha/yr 20% (5:1) 500 kg/ha/yr 41 667 shrimp/ha/yr 52 084 shrimp/ha/yr

correlation between increased stocking and increased production from a silvofishery pond begins to plateau with a stocking biomass greater than 20 g/m2. However, this will vary with the pond's environmental conditions as well as with the species cultured. Supplemental activities within the silvofisheries system to enhance and increase total production from the area over that produced through aquaculture would include: Collecting: crabs, polychaetes, oysters, and other aquatic animals from the mangroves. Sipunculid worms (Phascolosoma arcuatum) are collected in Vietnam for the domestic market and export to China (Vu Do Quynh, personal communication). It should be noted that in locations where the mangroves had been deforested there were no worms, and tree replanting efforts are monitoring for the re-establishment of this commercially valuable worm (Brands, personal communication). l Honey production: Avicennia sp. tends to be a preferred species for honey production. l Raising livestock: chickens, ducks or pigs raised along the banks, or in the canals (ducks). The keeping of bees for the production of honey. This provides an additional harvestable product plus fertilization of the pond to increase production potential of the aquaculture products. Avicennia sp. has been successfully used as cattle fodder in the Philippines (Baconguis, personal communication, 1999). l Harvesting mangrove trees: cropping the mangrove stands for wood products (e.g. charcoal, construction material, etc.), dye, alcohol, vinegar, sugar, medicine and tannin. This has to be carefully managed cropping to maintain an adequate level of organic carbon input to the ponds from the mangrove NPP. l

Grow-out A potential innovative area for enhancing production of silvofishery pond systems is to utilize effluent from adjacent semi-intensive or intensive aquaculture farms. The elevated nutrient levels in these effluents could be utilized to enhance the nutrient level in the silvofishery ponds. The practice of aquaculture effluent treatment in holding

238

Ecological Aquaculture

ponds is well documented (Culley et al., 1981; Pillary, 1992; Bird, 1993; Jenkins et al., 1995; Schwartz & Boyd, 1995). Mangrove sediment microbial communities are capable of depurating added nutrient loads, and mangroves have been suggested for use in nutrient removal from secondary and tertiary sewage effluent (Nedwell, 1974; Soegiarto, 1984; Corredor & Morell, 1994; Wong et al., 1995; Machiwa & Hallberg, in press). However, a non-human waste source would be preferred as the source of added nutrients. In a study that examined the use of mangrove forests as treatment of shrimp pond effluent, it was estimated that 2±3 ha of mangroves would be required per ha of semi-intensive ponds and up to 22 ha for intensive shrimp effluent (Robertson & Phillips, 1995). This wide ratio range is due to the variation in culture practices used in the shrimp ponds, and site characteristics of the mangroves. A conservative usage of effluent would potentially compensate for the generally accepted nitrogen and phosphorus nutrient limited mangroves (Boto & Wellington, 1983; Rivera-Monroy & Twilley, 1996; Kaly et al., 1997). N and P requirements for a Rhizophora forest were estimated to be 219 kg N/ha/yr and 20 kg P/ha/yr (Robertson & Phillips, 1995). Fertilization would increase the growth rate of the Rhizophora trees and increase their nutrient concentration and potential food value. The transformation of the added nutrients into a harvestable product (e.g. aquaculture product, or mangrove wood) would be a means of removing these excess nutrients from the environment in the form of commercial products. This type of integration of systems and further environmental mediation would increase energy utilization and production of a valued product. However, it would be limited to sites where the two system types are within economically acceptable proximity to allow for the use of the effluent. Caution needs to be exercised to avoid over-fertilization that would exceed the carrying capacity of the mangrove±silvofishery system so that there is a shift in the balance between aerobic±anaerobic to one that is anaerobic. Therefore, the quantity of supplemental fertilization used from semi-intensive or intensive aquaculture system in the silvofishery system will be site specific. However, with integrated coastal resource planning this could become a viable practice. In an analysis of nutrient flow through intensive marine shrimp ponds, it was found that 95% of N and 71% of P came from the applied commercial feed (Briggs & Funge-Smith, 1994). Of this amount only 24% of the N and 13% of the P were incorporated into the shrimp, while 35% and 10% of the N and P respectively were ultimately exported to the environment (63±67% through water exchange and the remainder at drainage). The sediments retained 31% and 84% of the N and P respectively. This information would assist in planning the potential utilization of this effluent in silvofisheries ponds to reduce negative environmental eutrophication and the capture of this valuable resource within the products of silvofisheries. The further integration of silvofisheries with livestock or agriculture crops is possible. The integration of poultry and silvofisheries is being evaluated at the demonstration facility in Cikiong, Indonesia (Fig. 8.28). However, livestock and aquaculture integration has been practiced for hundreds of years in China and elsewhere for the controlled eutrophication of fishponds using poultry or other animal manure for production of herbivorous fish. Little & Satapornvanit (1996) review the application of integrated poultry and fish production. Poultry may be integrated with fish culture

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

239

Fig. 8.28 The placement of a poultry house directly over the canal in a demonstration silvofisheries pond to provide an additional source of nutrients, Cikiong (West Java), Indonesia.

to benefit both poultry production and fish/shrimp production. It also provides environmental advantages in processing wastes from the poultry, which are of value as nutrient inputs to fish culture. The placement of the poultry house over the pond facilitates the waste introduction into the pond directly to save labor. Confining the poultry next to or over water can also improve their productivity under tropical conditions. Evaporative cooling reduces heat stress in broilers and access to water improves feather quality of ducks (Edwards, 1986; Theimsiri, 1992). The utilization of the poultry wastes provides the additional nutrients to the silvofisheries system to enhance the production of potential products (aquaculture species and wood). Tilapia aquaculture in ponds is reported to thrive with the integrated culture of 1000 to 1500 egg-laying ducks/ha and yielding 12 MT/ha/yr (Little & Satapornvanit, 1996). The nutrients are released rapidly compared with mangrove litter. This enhances the production of phytoplankton, algae and detritus for consumption by herbivores and detritivores. Nutrient budgets indicate 15±20% and 8±12% of the input nitrogen and phosphorus respectively are incorporated in fish production with most of the nutrients accumulated in the sediments (Edwards, 1993). It is important to balance the production of waste from poultry to prevent excess eutrophication (risk of low oxygen levels) of the fishpond. The quality of poultry wastes used in fish culture varies greatly mainly due to quality and type of feed used. The type of poultry production system (e.g. intensive, extensive) can also greatly influence the amount of fish produced. Similarly, the dynamics of poultry flocks (e.g. number, age, and type) needs to be monitored to control the production of waste.

RECOMMENDATIONS FOR THE FUTURE DEVELOPMENT OF SILVOFISHERIES AQUACULTURE ECOSYSTEMS The application of silvofisheries should be approached with a reasonable measure of caution as with any development in an area as environmentally sensitive as a mangrove ecosystem. The basic principles of sustainable development and management must be part of the considerations prior to development. For example, the availability

240

Ecological Aquaculture

and impact on the natural population of species cultured must be assessed including the availability and impact on seedstock recruitment and survival. Population growth places pressure for development in the mangrove coastal zone and adjacent area. Development in this mangrove coastal zone and adjacent area must take into account environmental, conservation, social, and economic issues. The overall development strategy should be one that integrates all four issues. Development is a dynamic, evolving process that changes over time with expanding activities, research, and emerging technology, and these changes should be reflected in policies and regulations and implementation strategies through periodic revision. Silvofisheries development should be incorporated where appropriate into a comprehensive coastal planning process that identifies the environmental characteristics of specific areas and the carrying capacity of the area. Determination of the baseline information for proper planning and use of the coastal resources on a sustainable basis that takes account of all potential uses as well as conservation should be included. Development should be diverse in nature and serve the fundamental needs of the community while providing an economic base for expansion within an environmentally sensitive framework. The integrated approach will allow for sustained economic activities while implementing a conservation and rehabilitation program for the mangroves. Beneficial contributions to the environment from silvofisheries development need to be considered in the broader evaluation of silvofisheries and planning of the coastal resources use and conservation. An example of this is identified in a study on silvofisheries in the Indramayu area (Western Java), which is a point of feeding and replenishing energy reserves in the north±south migration of birds (FitzGerald & Savitri, 1997). This important environmental value to migratory bird populations has been demonstrated at locations such as the Mai Po Nature Reserve in Hong Kong with its associated silvofisheries ponds (Young & Wong, 1994; Young, 1996). Silvofisheries were also utilized in the conversion of non-productive shrimp aquaculture ponds as part of a community empowerment process in Central and Western Java (FitzGerald & Savitri, 1997). These types of external benefits to the environment need to be considered in a benefit/cost analysis as part of coastal management programs. FitzGerald & Sutika (1997) suggested a conservation model that allowed for restricted mariculture and aquaculture activities, including silvofisheries, in the mangrove area. It consists of four zones across the mangrove area, including the `green belt' that is utilized in many Southeast Asian countries. The width of the zones varies according to site-specific characteristics (i.e. width of mangroves will result in the constriction or elimination of zones landward). This model calls for a shift of new aquaculture pond (non-silvofishery) development to the area behind the mangroves. This model is targeted at areas with a low current level of private development and a high percentage of government land ownership and control over the mangrove area in a zoning pattern with the following characteristics: l

Development of mariculture activities in the near-shore areas adjacent to the mangroves: the area seaward of the mangroves and the adjacent estuaries has potential for mariculture activity (dependent on environmental and physical conditions of the site). Activity would potentially include raft, cage, pen and stick-

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

241

line culture of seaweeds, coral-reef fish, lobsters (mainly holding), sea cucumbers, shellfish, and other potential species as they become technically viable. l Establishment and maintenance of a mangrove coastal green belt that reflects sitespecific environmental conditions and maintains fauna and flora integrity of the mangrove forest. The green-belt area is strictly a preservation area for the mangroves and no aquaculture activity should take place in this zone. l Allowance for integrated aquaculture/forestry activity, silvofishery, in the zone immediately behind the mangrove green belt zone. This zone of integrated nondegrading activity would be of a similar width to the green belt. The conservation area is a transition zone that would allow limited economic utilization of the mangrove area, but in an integrated sustainable manner that preserves the overall integrity of the mangroves. l The area landward of the integrated aquaculture±mangrove zone could be zoned to allow for sustainable aquaculture pond activity up to the semi-intensive culture intensity level (meets site selection criteria and under strictly enforced controlled use regulations). The proposed utilization zone still maintains a level of integrity in the mangroves with the use of silvofisheries. The proposed ratio is 1:1 (mangrove area to water canal) in the silvofishery model design. This will allow for a greater potential production level (input dependent) from the silvofishery ponds in this zone.

Comparative economic assessments It is useful to evaluate the capital costs and the potential value of products produced associated with a silvofisheries pond in comparison to a brackishwater extensive aquaculture pond strictly from an economic perspective. This is of value in decision making for both the community or individual entrepreneur and also for the planning/ policy organizations to make the best informed decisions. It can also provide a measure of the level of incentives that may be necessary to make silvofisheries a comparable economic activity for a private landowner/entrepreneur. Both silvofisheries and brackishwater extensive aquaculture ponds are essentially the same in construction with the major exception being that the central platform area in an extensive aquaculture pond is cleared of all vegetation. The cost associated with the perimeter dike construction would usually be less in a silvofisheries pond than that of the extensive aquaculture pond due to the cost efficiency in the reduced perimeter length associated with a square design, as discussed under the pond design section above. The construction costs associated with the clearing of the central platform would be zero for a silvofisheries pond except when cross channels are constructed through the platform. Therefore, from a construction cost perspective in comparing a silvofisheries and an extensive brackishwater aquaculture pond these costs can be represented as follows: l l

Silvofisheries pond = X(Ds) + Y(CAs) Brackishwater extensive aquaculture pond = X(Da) + Y(CAa), where X is the cost per linear meter of the perimeter dike; D is the linear length of the peri-

242

Ecological Aquaculture

meter dike; Y is the clearing/construction cost per square meter of the central platform; and CA is the area of the central platform cleared. Subscripts `a' and `s' indicate brackishwater extensive aquaculture and silvofisheries ponds respectively. If the difference in annual net operating value of products produced between the silvofisheries and extensive pond (
Ownership Mangroves and tidal wetlands are often traditionally common property resources with rather low perceived value in developing countries. Use or optimum yield from any common property resource must be carefully executed and must consider a mix of benefits, including those of environmental, social and economic significance. In coastal aquaculture, the competition is usually between the traditional users of previously open-access resources and those who are encroaching on and expropriating these resources. This development activity impacts the social, economic, cultural, and environmental status of the area. Therefore, it is critical that the interactions among these factors are fully integrated into the assessment and planning process to increase the level of success in meeting the goals and objectives of land use and development plans and to reduce potential user conflicts. Land ownership is an important decision criterion in the use of the mangrove area. The `no associated land cost' or relatively low cost is a significant incentive for development. Therefore, the use of the mangroves allows the rural, low-income landless to enter land ownership through a lower entry barrier regardless of the long-term development costs (mainly in labor) and environmental costs. This is important from a social aspect in that land ownership brings a higher status in the community, which potentially raises the individual's influence in it. This land ownership not only benefits the individual, but also bestows long-term benefit to his family and heirs.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

243

Silvofishery has been successful in situations where it is a community project on community property or individual/family operations on controlled government land conditionally leased for silvofishery (e.g. Blanakan and Cikiong, Indonesia). However, in areas that are under private ownership, the owner would normally have little incentive to put cleared property back into forest that would be perceived as being of lower value. This emphasizes the importance of maintaining government ownership of mangrove areas. Government should maintain ownership and control of the mangrove area where it has not already been privatized. This will allow for a more controlled utilization through conditional leases in an integrated and environmentally sensitive manner under a land use plan. In the case of private land ownership of the mangroves, an important component that seems to be inadequately considered in the realistic implementation of the various silvofishery models is the capital costs in constructing the pond. The construction cost associated with the dike and gate construction of a silvofisheries pond is similar to that of an aquaculture brackishwater pond. The revenue generation from a silvofishery model would have to justify this type of investment on private property with private capital as opposed to an activity with a greater rate of return on investment (e.g. semi-intensive or intensive aquaculture operation). Underwriting silvofishery demonstration projects that are not carefully designed within basic economic realities by government agencies can create a misleading model that is not viable on privately owned land without substantial government subsidy. Therefore, demonstration projects have to be realistically designed that fulfil the intended longterm implementation objectives of the program. If the mangrove forest is under private ownership and it is considered that the environmentally friendly development is in the public good, government incentives appropriately designed to provide sufficient incentive to private landowners to develop silvofisheries as opposed to less environmentally benign options should be carefully considered. These should be designed to provide a reasonable economic return within an enforceable regulatory framework. An incentive program should consider various options to make the rehabilitation/ reforestation model more attractive to private landowners. Some incentive program options could include: Tax rebates or abatements on property taxes. This could be based on a formula that essentially defrays some of the economic loss the private landowner would incur through reduced production and land value. l Higher property tax on brackishwater ponds in former mangrove areas that are abandoned or low in production. This would be a negative incentive approach that places an increased economic cost to owners to leave the property unproductive. Therefore, they would have the incentive to either reforest the property, possibly utilize silvofisheries, or increase production (improved brackishwater pond condition) to relieve the increased tax expense of the property. l Land exchange with government property that is in a non-critical habitat area of equivalent economic value. l

244

Ecological Aquaculture

The use of both incentives and disincentives can be utilized in the implementation of a policy that has been determined to be desirable (national and local level). The use of `green taxes' has been advocated as a means of addressing environmental externalities. The attributes of green taxes include the fact that they are cheaper to administer, do not distort economic activity, and are fair (Anon, 1996).

System models Silvofishery is a labor-intensive technology appropriate for an individual, family or community operation and can be a viable alternative to brackishwater pond culture. It diversifies products from the land and aquatic production within an environmentally benign framework and is integrated into the mangrove forest ecosystem. The silvofishery model is not suited to commercial large-scale aquaculture activity on privately owned land. The cost of pond construction with reduced management options makes the empang parit silvofishery model not economically attractive without more fully integrated resource utilization into economically valued products. However, empang parit and other silvofishery models can be a useful alternative activity within the mangrove intertidal zone as a subsistence and conservation type of activity. Silvofishery models can also be viable forms of aquaculture in converting abandoned brackishwater ponds into an integrated reforestation and utilization program especially where the cost of pond construction is not included with utilization of pre-existing ponds. Government subsidy in the form of low-cost lease and a package of technical and capital assistance should make it an attractive option to meeting the needs of the rural poor within a program of mangrove rehabilitation. The selection of the most appropriate silvofishery model will be site dependent and influenced by the status of the mangrove ecosystem and the various inputs required. Therefore, no single model is the best for all situations. Silvofisheries development should be integrated into an area-wide integrated approach to coastal zone management. The integrated approach to conservation and utilization of the mangrove resource allows for maintaining a relatively high level of integrity in the mangroves while capitalizing on the economic benefits of brackishwater aquaculture. Silvofishery pond systems utilizing energy derived from mangrove litter and the detritus complex input exclusively can be of particular benefit in locations where there are limited potential feed sources and fertilizers in the community to support aquaculture production methods that require these supplemental inputs. Communities with all the available protein sources utilized as food for human consumption or livestock will benefit from the additional protein supply from a silvofishery pond system that does not utilize these limited energy sources for aquaculture production. In addition, utilizing the mangrove litter as green manure in the pond system will relieve competition for scarce fertilizer resources. The silvofishery model that would be preferred in most applications is one of the integrated mangrove±crab culture systems. The advantages of this system would be its absence of any permanent alteration of the mangrove forest area, low capital investment, low labor requirement, low technology, small unit size, incremental ability for expansion, and production of a high valued product.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

245

The alternation of mangrove with pond area model is a silvofisheries model (Fig. 8.29) that would have a number of advantages as well as some disadvantages that are site dependent. It would maintain the same high maximum ratio of mangrove to pond culture area as any of the other methods while providing superior management control and potential production. Variations in the ratios can be made based on environmental, conservation, development, and policy considerations. It is recommended that units of 2- and 4-ha pond areas be standardized as the model units with an individual pond size of 1 ha. This would mean that for every 2 ha of pond area there would be an 8-ha area of mangroves maintained around the pond unit. For areas under private ownership, the brackishwater pond/mangrove alternation model would be the most appropriate. An advantage of this model is that the mangroves outside the pond would be structurally unaltered and the tidal and groundwater movement would not be constrained. The major disadvantage of this model is that it is particularly susceptible to violation of the mangrove area by encroachment or noncompliance where regulations are not rigorously enforced.

Mangrove area Fig. 8.29 Illustration of 10 ha silvofisheries site with 8:2 mangrove to brackishwater pond area utilizing the brackishwater pond/mangrove alternation system (not to scale).

Pond

A form of silvofisheries that separates the mangrove vegetated area from the aquaculture area is illustrated in Fig. 8.30. Dikes enclose and separate the two compartments, and water can enter both areas through separate gates. In addition, water can enter the aquaculture area from the mangrove compartment bringing nutrients and organic matter from the mangrove forest.

Fig. 8.30 Silvofisheries model (Directorate General of Fisheries, Indonesian Government). A = Gate for water exchange controlling inflow/outflow; B = perimeter dike wall of silvofishery unit; C = open water aquaculture portion of unit; D = mangrove vegetation (planted or natural) portion of unit.

246

Ecological Aquaculture

Research needs Research on the optimization of production within an integrated mangrove/aquaculture silvofishery system should be a priority area. Studies should be conducted to increase the production from the various silvofishery models. This is particularly important for reforesting a private land area. The different silvofishery models, specifically the two main types (mangroves integrated within the pond area and mangroves integrated around the outside of the pond area), should be evaluated. This would include optimizing use of inputs and stocking strategies for different species within a polyculture production system. Density of trees and silviculture management (e.g. trimming, selective cutting, etc.) to maximize the production of litter need to be evaluated as part of an overall optimization of the silvofisheries system. The preferred species of mangrove trees that can be used for specific pond conditions should be determined as well as conducting an analysis on the type and amount of vegetation cover, litter production rate of the different trees and the decomposition rates of the different tree litters. These would be important factors in the food web and support the maximum productivity of the pond. The food web resulting from mangrove vegetation litter transferred to the ponds and optimization of this input should be researched to obtain a better understanding of this process so that appropriate management practices can be developed. The principles of ecological engineering should be applied to the development and refinement of the system (Odum, 1969; Chan, 1993; Mitsch, 1993; Grant, 1998). The role of microbiota (mainly bacteria and fungi) in the decomposition and recycling of nutrients and their importance in the sustainable nature of the mangrove ecosystem's food web is likely underestimated. This represents a major area for further research that can contribute to the development of best management practices for silvofisheries systems as well as the understanding of the mangrove ecosystem and its complex interactions as a whole. Options for enhancing bacterial and fungal activity to accelerate and enhance the decomposition and nutrient recycling process need to be identified. The potential interaction and impact of the silvofisheries system on nutrient recycling efficiency with a possible shift in the balance between aerobic and anaerobic bacteria resulting from a large portion of the system being submerged for extended periods should be evaluated. This may require adjustments in management practices to accommodate the potential influence on recycling of nutrients in the system. Further refinement in the knowledge of the interaction between the various primary producers in the system, and the decomposition of nutrient recycling components and how best to convert this energy into aquaculture products are important considerations. Evaluating the system to better understand the trophic production and food web dynamics will be an area of major importance in moving towards the optimization of the system. The sustainability of a system needs to take into account the ecosystem structure and function along with energy flow through trophic interactions. Conducting input±output, trophic and cycle analysis along with determining the holistic system properties is needed to get a better understanding of critical interactions and energy flow and thereby to best manage the silvofisheries

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

247

system. It is necessary to include all forms of macrophytes and algae to evaluate the importance of different autotrophic compartments in the total primary production, then to study the energy flow and its utilization in the mangrove ecosystem. It will be equally important to determine and document the difference in the food items ingested from those that are actually assimilated. This will assist in selecting where possible the most appropriate potential aquaculture species for the system. A thorough study of the macroalgae and microalgae in mangroves to understand their growth dynamics within the mangrove forest and physical growth parameters and how to maximize their production to benefit production of aquaculture species is essential to maximizing the production efficiency. Changes in the balance of autotrophs in nutrient cycles in the aquatic ecosystem will be a component in maximizing the energy available to the food web supporting the aquaculture species. It will also be important to determine how to best balance mangrove growth (density, height, canopy cover, species, etc.) with production of algae, since excessive shading from maximum growth density of macrophytic vegetation would impair production of the algae within the platform area. These are all part of a requirement to best enhance the quality and quantity of the food web. Silvofisheries have the potential to capture some of the economic benefits of the mangrove areas within an environmentally sensitive framework and a sustainable activity. Improvement in the economic return from this system will be a key factor in the wide acceptance of silvofishery methods as an economically viable activity in the mangroves. Silvofisheries can also provide an alternative economic activity to the rural poor and reduce development pressure on the mangrove forests. Therefore, they should be considered in an overall development and management strategy for the coastal zone and could serve a role in the transition that shifts more intensive aquaculture to areas outside the mangroves.

Acknowledgements I would like to acknowledge with sincere gratitude Dr Jurgenne H. Primavera, Dr Gary D. Pruder, Mike Gawel, Alfredo Quarto, Dr Danielle Johnston, Dr Lew Young, Dr Christina Halling, and Dr Claude E. Boyd for their valuable input and comments. In addition, the assistance of Kris Anderson and Lois Ann Kiehl of PRAISE (University of Hawaii) in obtaining literature is greatly appreciated.

References Agbayani, R.F., Baliao, D.D., Samonte, G.P.B., Tumaliuan, R.E. & Caturao, R. (1990) Economic feasibility analysis of the monoculture of mudcrab (Scylla serrata) Forsskal. Aquaculture, 91, 223±231. Aisha, K.A, Shabana, E.F., El-Abyad, M.S., Kobbia, I.A. & Schanz, F. (1995) Pulse feeding with nitrate and phosphate in relation to tissue composition and nutrient uptake by some macroalgae from the Red Sea at Ghardaqa (Egypt). Journal of Basic Microbiology, 35, 135±145.

248

Ecological Aquaculture

Albright, L.J. (1976) In situ degradation of mangrove tissues. New Zealand Journal of Marine and Freshwater Research, 10, 385±389. Ali, M.A., Krishnamurthy, K. & Jeyaseelan, M. (1983) Energy flow through the benthic ecosystem of the mangroves with special reference to nematodes. Mahasagar Bulletin of the National Institute of Oceanography, 16, 317±325. Almodovar, L.R. & Pagan, F. (1971) Notes on a mangrove lagoon and mangrove channels at La Parguera, Puerto Rico. Nova Hedwigia, 21, 241±253. Alongi, D.M. (1987) The influence of mangrove-derived tannins on intertidal meiobenthos in tropical estuaries. Oecologia, 71, 537-540. Alongi, D.M. (1990a) Effect of mangrove detrital outwelling on nutrient regeneration and oxygen fluxes in coastal sediments of the central Great Barrier Reef lagoon. Estuarine, Coastal and Shelf Science, 31, 581±598. Alongi, D.M. (1990b) Abundances of benthic microfauna in relation to outwelling of mangrove detritus in a tropical coastal region. Marine Ecology Progress Series, 63, 53±63. Alongi, D.M. (1994) The role of bacteria in nutrient recycling in tropical mangrove and other coastal benthic ecosystems. Hydrobiologia, 285, 19±32. Alongi, D.M., Boto, K.G. & Tirendi, F. (1989) Effect of exported mangrove litter on bacterial productivity and dissolved organic carbon fluxes in adjacent tropical nearshore sediments. Marine Ecology Progress Series, 56, 133±144. Anderson, J.M. (1987) Production and decomposition in aquatic ecosystems and implications for aquaculture. In: Detritus and Microbial Ecology in Aquaculture (eds D.J.W. Moriarty & R.S.V. Pullin), pp. 123±147. ICLARM Conference Proceedings 14. International Center for Living Aquatic Resources Management, Manila, Philippines. Angsupanich, S., Miyoshi, H. & Hata, Y. (1989) Degradation of mangrove leaves immersed in the estuary of Nakama River, Okinawa. Nippon Suisan Gakkaishi, 55, 147±151. Anon. (1991) Application `social forestry' strategy by using silvofishery system for supporting national food production. Perum Perhutani Unit III West Java, Bandung. Anon. (1994) Overview of Cikiong project. Bagian Kesatuan Pemangkuan Hutan (BKPH) Cikiong. Anon. (1995) Overview of Cibuaya project. Bagian Kesatuan Pemangkuan Hutan (BKPH) Cikiong. Anon. (1996) Green taxes are good taxes. The Economist (November 16, 1996), p. 20. Anon. (1998) Traditional extensive shrimp culture in Vietnam. http://allserv.rug.ac.be/ *booghe/asrdc/tradshrp.htm. Ausmus, B.S. & Witkamp, M. (1974) Litter and soil microbial dynamics in a deciduous forest stand. US-IBP EDFB Report 73-10, 183 pp. Baconguis, S.R. (1991) Aquasilviculture technology: key to mangrove swamp rehabilitation and sustainable coastal zone development in the Philippines. Canopy International, 17, 4±7. Bano, N., Nisa, M.U., Khan, N., Saleem, M., Harrison, P.J., Ahmed, S.I. & Azam, F. (1997) Significance of bacteria in the flux of organic matter in the tidal creeks of the mangrove ecosystem of the Indus River delta, Pakistan. Marine Ecology Progress Series, 157, 1±12. Beanland, W.R. & Woelkerling, W. (1983) Avicennia canopy effects on mangrove algal communities in Spencer Gulf, South Australia. Aquatic Botany, 17, 309±313. Binh, C.T., Phillips, M.J. & Demaine, H. (1997) Integrated shrimp±mangrove farming systems in the Mekong Delta of Vietnam. Aquaculture Research, 28, 599±610.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

249

Bird, K.T. (1993) Aquatic plants for treatment of aquaculture wastewater. Aquaculture Magazine (January/February 1993), 39±42. Boesch, D.F. & Turner, R. (1984) Dependence of fishery species on salt marshes: the role of food and refuge. Estuaries, 7, 460±468. Bombeo-Tuburan, I., Guanzon Jr, N.G. & Schroeder, G. (1993) Production of Penaeus monodon (Fabricius) using four natural food types in an extensive system. Aquaculture, 112, 57±65. Boonruang, P. (1978) The degradation rates of mangrove leaves of Rhizophora apiculata (BL.) and Avicennia marina (Forsk.) Vierh. at Phuket Island, Thailand. Phuket Marine Biology Center Research Bulletin, 26, 1±6. Boto, K.G. (1982) Nutrient and organic fluxes in mangroves. In: Mangrove Ecosystems in Australia: Structure, Function and Management (ed. B.F. Clough), pp. 239±257. Australian Institute of Marine Science, Canberra. Boto, K.G. & Bunt, J. (1981a) Carbon export from mangroves. In: The Cycling of Carbon, Nitrogen, Sulfur and Phosphorus in Terrestrial and Aquatic Systems (eds I.E. Balbally & J.R. Freney), pp. 105±110. Australian Academy of Science, Canberra. Boto, K.G. & Bunt, J. (1981b) Tidal export of particulate organic matter from a Northern Australian mangrove system. Estuarine, Coastal and Shelf Science, 13, 247±255. Boto, K.G. & Wellington, J. (1983) Phosphorus and nitrogen nutritional status of a northern Australian mangrove forest. Marine Ecology Progress Series, 11, 63±69. Boto, K.G., Bunt, J.S. & Wellington, J. (1984) Variations in mangrove forest productivity in Northern Australia and Papua New Guinea. Estuarine, Coastal and Shelf Science, 19, 321±329. Boto, K.G., Alongi, D.M. & Nott, A. (1989) Dissolved organic carbon±bacteria interactions at the sediment±water interface in a tropical mangrove sediment. Marine Ecology Progress Series, 51, 243±251. Briggs, M.R.P & Funge-Smith, S. (1994) A nutrient budget of some intensive marine shrimp ponds in Thailand. Aquaculture and Fisheries Management, 25, 789±811. Brown, I.W. (1993) Mangrove crabs. In: Nearshore Marine Resources of the South Pacific (eds A. Wright & L. Hill), pp. 609±642. Institute of Pacific Studies and Forum Fisheries Agency, Honiara. Brown, M.R. (1991) The amino-acid sugar composition of 16 species of microalgae used in mariculture. Journal of Experimental Marine Biology and Ecology, 145, 79±99. Bryan, P.G. (1975) Food habits, functional digestive morphology, and assimilation efficiency of rabbitfish Siganus spinus (Pisces: Siganidae) on Guam. Pacific Science, 29, 269±277. Bunt, J.S. (1982) Studies of mangrove litter fall in tropical Australia. In: Mangrove Ecosystems in Australia: Structure, Function and Management (ed. B.F. Clough), pp. 223±237. Australian Institute of Marine Science, Canberra. Burkholder, P.R. & Almodovar, L. (1973) Studies on mangrove algal communities in Puerto Rico. Florida Science, 36, 66±74. Camilleri, J.C. (1989) Leaf choice by crustaceans in a mangrove forest in Queensland. Marine Biology, 102, 453±459. Camilleri, J.C. & Ribi, B. (1986) Leaching of dissolved organic carbon (DOC) from dead leaves, formation of flakes from DOC, and feeding on flakes by crustaceans in mangroves. Marine Biology, 91, 337±344. Canizares-Villanueva, R.O., Dominguez, A.R., Cruz, M.S. & Rios-Leal, E. (1995) Chemical

250

Ecological Aquaculture

composition of cyanobacteria grown in diluted, aerated swine wastewater. Bioresource Technology, 51, 111±116. Carter, M.R., Burns, L.A., Cavinder, T.R., Dugger, K.R., Fore, P.L., Hicks, D.B. et al. (1973) Ecosystems Analysis of the Big Cypress Swamp and Estuaries. United States Environmental Protection Agency, South Florida Ecological Study, Atlanta, Georgia. Cebrian, J. & Duarte, C. (1994) The dependence of herbivory on growth rate in natural plant communities. Functional Ecology, 8, 518±525. Cha, M.W., Young, L. & Wong, K. (1997) The fate of traditional extensive (gei wai) shrimp farming at the Mai Po Marshes Nature Reserve, Hong Kong. Hydrobiologica, 352, 295±303. Chan, G.L. (1993) Aquaculture, ecological engineering: lessons from China. Ambio, 22, 491±494. Chang, W.W.S. (1997) Pen culture of mudcrabs in the mangrove ecosystems in Sarawak (East Malaysia). Aquaculture Asia, 2, 2±5. Chapman, V.J. (1976) Mangrove Vegetation. Strauss & Cramer Publishers, New York. Chapman, V.J. (1984) Mangrove biogeography. In: Hydrobiology of the Mangal (eds F.D. Por & I. Dor), pp. 358±386. W. Junk Publishers, The Hague. Chong, V.C. & Sasekumar, A. (1981) Food and feeding habits of the white prawn Penaeus merguiensis. Marine Ecology Progress Series, 5, 185±191. Chong, V.C., Sasekumar, A. & Wolanski, E. (1996) The role of mangroves in retaining penaeid prawn larvae in Klang Strait, Malaysia. Mangroves and Salt Marshes, 1, 11±22. Chong, V.C., Low, C.B. & Ichikawa, T. (2001) Contribution of mangrove detritus to juvenile prawn nutrition: a dual stable isotope study in a Malaysian mangrove forest. Marine Biology, 138, 44±86. Chow, B.A. (1989) An evaluation of the importance of mangroves to the ecology of West Harbour, Jamaica. M. Phil. Thesis, University of the West Indies, Jamaica. Chow, B.A. & Bacon, P. (1992) Mangrove forests and faecal rain. Caribbean Marine Studies, 3, 57±58. Choy, S.C. & Booth, W. (1994) Prolonged inundation and ecological changes in an Avicennia mangrove: implications for conservation and management. Hydrobiologia, 285, 237±247. Christensen, V. & Pauly, D. (1993) Flow characteristics of aquatic ecosystems. In: Trophic Models of Aquatic Ecosystems (eds V. Christensen & D. Pauly), pp. 338±352. ICLARM Conference Proceedings 26. ICLARM, Manila. Clough, B.F. & Attiwill, P. (1982) Primary productivity of mangroves. In: Mangrove Ecosystems in Australia: Structure, Function and Management (ed. B.F. Clough), pp. 213±222. Australian Institute of Marine Science, Canberra. Colman, J. & Edwards, P. (1987) Feeding pathways and environmental constraints in waste-fed aquaculture: balance and optimization. In: Detritus and Microbial Ecology in Aquaculture (eds D.J.W. Moriarty & R.S.V. Pullin), pp. 240±281. ICLARM Conference Proceedings 14. International Center for Living Aquatic Resources Management, Manila, Philippines. Cooksey, K.E. & Cooksey, B. (1978) Growth influencing substances in sediment extract from a subtropical wetland: investigation using a diatom bioassay. Journal of Phycology, 14, 347±352. Cooper, L.H.N. (1933) Chemical constituents of biological importance in the English Channel, November 1930 to January 1932. Part I. Phosphate, silicate, nitrate, nitrite, ammonia. Part II. Hydrogen ion concentration, excess base, carbon dioxide and oxygen. Journal of the Marine Biological Association UK, 18, 677±753.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

251

Cordero, F.J.M., FitzGerald, Jr, W.J. & Leung, P. (1999) Evaluation of productivity in extensive aquaculture practices using interspatial TFP index, Sulawesi, Indonesia. Asian Fisheries Science, 12, 223±234. Corredor, J.E. & Morell, J. (1994) Nitrate depuration of secondary sewage effluents in mangrove sediments. Estuaries, 17, 295±300. Costa-Pierce, B.A. (1992) Rice±fish systems as intensive nurseries. In: Rice±Fish Research and Development in Asia (eds C.R. dela Cruz, C. Lightfoot, B.A. Costa-Pierce, V.R. Carangal & M.P. Bimbao), pp. 117±130. ICLARM Conference Proceedings 24. ICLARM, Manila, Philippines. Culley, D.D., Rejmankova, E., Kvet, J. & Frye, J. (1981) Production, chemical quality and use of duckweeds (Lemnaceae) in aquaculture, waste, management, and animal feeds. Journal of the World Mariculture Society, 12, 27±49. Cundell, A.M., Brown, M.S., Stanford, R. & Mitchell, R. (1979) Microbial degradation of Rhizophora mangle leaves immersed in the sea. Estuarine and Coastal Marine Science, 9, 281±286. Dagar, J.C., Mongia, A.D. & Bandyopadhyay, A. (1991) Mangroves of Andaman and Nicobar Islands. Oxford & IBH Publishing, New Delhi. Dame, R.F & Lefeuvre, J. (1994) Tidal exchange: import±export of nutrients and organic matter in new and old world salt marshes: conclusions. In: Global Wetlands: Old World and New (ed. W.J. Mitsch), pp. 303±305. Elsevier Science, Amsterdam. Davey, A. & Woelkerling, W. (1985) Studies on Australian mangrove algae. III. Victorian communities: structure and recolonization in western Port Bay. Journal of Experimental Marine Biology and Ecology, 85, 177±190. Dawes, C.J. & Goddard, R. (1978) Chemical composition of the wound plug and the entire plants for species of the coenocytic green alga, Caulerpa. Journal of Experimental Marine Biology and Ecology, 35, 259±263. Dittel, A.I., Epifanio, C., Cifuentes, L. & Kirchman, D.L (1997) Carbon and nitrogen sources for shrimp postlarvae fed natural diets from a tropical mangrove system. Estuarine, Coastal and Shelf Science, 45, 629±637. Dor, I. & Levy, I. (1984) Primary productivity of the benthic algae in the hard-bottom mangal of Sinai. In: Hydrobiology of the Mangal (eds F.D. Por & I. Dor), pp. 179±191. W. Junk Publishers. The Hague. Duc, Le Dien (1996) Integrated coastal management in the Tien Hai District, Thai Binh Province. Paper presented at the Ecotone IV meeting, Ho Chi Minh City, Vietnam. Edwards, P. (1978) The fishery and fisheries biology of penaeid shrimp on the Pacific coast of Mexico. Oceanography Marine Biology Annual Review, 16, 145±180. Edwards, P. (1986) Duck/fish integrated farming systems. In: Duck Production Science and World Practice (eds D.J. Farrel & P. Stapleton), pp. 267±291. Abart Publishing, New York. Edwards, P. (1993) Environmental issues in integrated agriculture±aquaculture and wastewater-fed fish culture systems. In: Environment and Aquaculture in Developing Countries (eds R.S.V. Pullin, H. Rosenthal & J.L. Maclean), pp. 139±170. ICLARM Conference Proceedings 31. ICLARM, Manila. Egna, H.S. & Boyd, C.E. (1997) Dynamics of Pond Aquaculture. CRC Press, New York. Elster, C. (1998) RegenerationsmoÈglichkeiten der Mangrove im Gebiet der Cienaga Grande de Santa Marta (Kolumbien). Okologie Tropischer Waldsysteme. GTZ, Eschborn.

252

Ecological Aquaculture

Emmerson, W.D. & McGwynne, L.E. (1992) Feeding and assimilation of mangrove leaves by the crab Sesarma meinerti de Man in relation to leaf-litter production in Mgazana, a warmtemperate southern African mangrove swamp. Journal of Experimental Marine Biology and Ecology, 157, 41±53. Fell, J.W., Cefalu, R., Master, I. & Tallman, A. (1975) Microbial activities in the mangrove (Rhizophora mangle) leaf detrital system. In: Proceedings of the International Symposium on Biology and Management of Mangroves, 8±11 October, 1974, Honolulu, Vol. 2, pp. 661±679. University of Florida, Gainesville. Fell, J.W. & Master, I. (1980) The association and potential role of fungi in mangrove detrital systems. Botanica Marina, 23, 257±263. Fell, J.W., Master, I. & Wiegert, R. (1984) Litter decomposition and nutrient enrichment. In: The Mangrove Ecosystem: Research Methods (eds S.C. Snedaker & J.G. Snedaker), pp. 239± 255. Monograph on oceanographic methodology 8. UNESCO, New York. FitzGerald, Jr, W.J. (1978) Environmental parameters influencing the growth of Enteromorpha clathrata (Roth) in the intertidal zone on Guam. Botanica Marina, 21, 207±220. FitzGerald, Jr, W.J. & Sutika, N. (1997) Mangrove rehabilitation and management project in Sulawesi: Artisanal Fisheries. Ministry of Forestry, Government of The Republic of Indonesia. FitzGerald, Jr, W.J. & Savitri, N. (in press) Integration of silvofisheries into coastal management and mangrove rehabilitation in Java, Indonesia. Thematic Review of Coastal Wetland Habitats and Shrimp Aquaculture. NACA. Bangkok. Folke, C. & Kautsky, N. (1992) Aquaculture with its environment: prospects for sustainability. Ocean and Coastal Management, 17, 5±24. Fry, B. (1984) 13C/12C ratios and the trophic importance of algae in Florida Syringodium filiforme seagrass meadows. Marine Biology, 79, 11±19. Furukawa, K. & Wolanski, E. (1996) Sedimentation in mangrove forests. Mangroves and Salt Marshes, 1, 3±10. Garcia, S. & Le Reste, L. (1981) Life cycles, dynamics, exploitation and management of coastal penaeid shrimp stocks. FAO Technical Paper. 203. FAO, Rome. Golley, F.B., Odum, H. & Wilson, R. (1962) A study of the structure and metabolism of a red mangrove forest in southern Puerto Rico in May. Ecology, 43, 9±18. Gong, W.K. & Ong, J. (1990) Plant biomass and nutrient flux in a managed mangrove forest in Malaysia. Estuarine, Coastal and Shelf Science, 31, 519±530. Goulter, P.P.E. & Allaway, W. (1979) Litter-fall and decomposition in a mangrove stand, Avicennia marina (Forsk.) Vierh. in Middle Harbour, Sydney. Australian Journal of Marine and Freshwater Research, 30, 541±546. Grant, W.E. (1998) Ecology and natural resource management: reflections from a systems perspective. Ecological Modelling, 108, 67±76. Gundermann, N. & Popper, D. (1984) Notes on the IndoPacific mangal fishes and on mangrove related fisheries. In: Hydrobiology of the Mangal (eds F.D. Por & I. Dor), pp. 201±209. Dr W. Junk Publishers. The Hague. Hajek, B.F. & Boyd, C. (1994) Rating soil and water information for aquaculture. Aquacultural Engineering, 13, 115±128. Harvey, H.W. (1939) Substances controlling the growth of a diatom. Part I. Journal of the Marine Biological Association UK, 23, 499±520.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

253

Hayashi, S. (1994) The development of sustainable mangrove management project. Progress Report of Long Term Expert on Ecosystems. JICA, Tokyo. Heald, E.J. (1969) The production of organic detritus in a south Florida estuary. PhD dissertation, University of Miami. Heald, E.J. (1971) The production of organic detritus in a South Florida estuary. Sea Grant Technical Bulletin No. 6., University of Miami Sea Grant Program (Living Resources), Miami, Florida. Heinle, D.R. & Flemer, D. (1976) Flows of materials between poorly flooded tidal marshes and an estuary. Marine Biology, 35, 359±373. Hemminga, M.A., Slim, R., Kazungu, J., Ganssen, G. & Nieuwenhuize, J. (1994) Carbon outwelling from a mangrove forest with adjacent seagrass beds and coral reefs (Gazi Bay, Kenya). Marine Ecology Progress Series, 106, 291±301. Heymans, J.J. & Baird, D. (1995) Energy flow in the Kromme estuarine ecosystem, St Francis Bay, South Africa. Estuarine, Coastal and Shelf Science, 41, 39±59. Hobbie, J.E. & Lee, C. (1980) Microbial production on extracellular material: importance in benthic ecology. In: Marine Benthic Dynamics (eds K.R. Tenore & B.C. Coull), pp. 341±346. University of South Carolina Press, Columbia. Hoguin, G., Vazquez, P. & Bashan, Y. (2001) The role of sedimentation microorganisms in the productivity, conservation, and rehabilitation of mangrove ecosystems: an overview. Biology and Fertility of Soils, 33, 265±278. Hughes, E.H. & Sherr, E. (1983) Subtidal food webs in a Georgia estuary. Journal of Experimental Marine Biology and Ecology, 67, 227±242. Irving, R. & Morton, B. (1988) A Geography of the Mai Po Marshes. World Wide Fund for Nature Hong Kong, Hong Kong University Press. Jenkins, M.R., Wade, R., Fletcher, J. & Hankins, J. (1995) Economic analysis of nontraditional water resources for aquaculture in West Virginia. Shepherdstown, WV: Conservation Fund, Freshwater Institute. Contract No. 94-101 WV-11529-94-I-302-0411. Johnston, D.J., Clough, B., Xuan, T.T. & Phillips, M.J. (1999) Mixed shrimp±mangrove forestry farming systems in Ca Mau Province, Vietnam. Aquaculture Asia, 4, 6±12. Johnston, D.J., Trong, N.V., Tuan, T.T. & Xuan, T.T. (2000a) Shrimp seed recruitment in mixed shrimp and mangrove forestry farms in Ca Mau province, Southern Vietnam. Aquaculture, 184, 89±104. Johnston, D.J., Clough, G., Xuan, T.T. & Phillips, M.J. (2000b) Shrimp yields and harvest characteristics of mixed shrimp±mangrove forestry farms in southern Vietnam ± factors affecting production. Aquaculture, 188, 263±284. Johnston, D.J., Clough, B., Xuan, T.T. & Phillips, M. (2000c) Mixed shrimp farming±mangrove forestry in the Mekong Delta, Vietnam. Austasia Aquaculture, 14, 55±58. Jones, D.A. (1984) Crabs of the mangal ecosystem. In: Hydrobiology of the Mangal (eds F.D. Por & I. Dor), pp. 89±109. W. Junk Publishers, The Hague. Kaly, U.L., Eugelink, G. & Robertson, A. (1997) Soil conditions in damaged north Queensland mangroves. Estuaries, 20, 291±300. Kennish, R. & Williams, G. (1997) Feeding preferences of the herbivorous crab Grapsus albolineatus: the differential influence of algal nutrient content and morphology. Marine Ecology Progress Series, 147, 87±95. Koesoemadinata, S. & Costa-Pierce, B.A. (1992) Development of rice±fish farming in

254

Ecological Aquaculture

Indonesia: past, present and future. In: Rice±Fish Research and Development in Asia (eds C.R. dela Cruz, C. Lightfoot, B.A. Costa-Pierce, V.R. Carangal & M.P. Bimbao), pp. 45±62. ICLARM Conference Proceedings 24. ICLARM, Manila, Philippines. Kristensen, E. (1984) Effect of natural concentrations on nutrient exchange between a polychaete burrow in estuarine sediment and the overlying water. Journal of Experimental Marine Biology and Ecology, 75, 171±190. Kristensen, E. (1985) Oxygen and inorganic nitrogen exchange in a Nereis virens (Polychaeta) bioturbated sediment±water system. Journal of Coastal Research, 1, 109±116. Kristensen, E. (1988) Benthic fauna and biogeochemical processes in marine sediments: microbial activities and fluxes. In: Nitrogen Cycling in Coastal Marine Environment (eds T.H. Blackburn & J. Sorensen), pp. 275±299. John Wiley & Sons, Chichester. Kristensen, E., Andersen, F.O. & Kofoed, L.H. (1988) Preliminary assessment of benthic community metabolism in a south-east Asian mangrove swamp. Marine Ecology Progress Series, 48, 137±145. Kristensen, E., Devol, A., Ahmed, S. & Saleem, H. (1992) Preliminary study of benthic metabolism and sulfate reduction in a mangrove swamp of the Indus Delta, Pakistan. Marine Ecology Progress Series, 90, 287±297. Kristensen, E., King, G., Holmer, M. et al. (1994) Sulfate reduction, acetate turnover and carbon metabolism in sediments of the Ao Noam Bor mangrove, Phuket, Thailand. Marine Ecology Progress Series, 109, 245±255. Kusmana, C., Takeda, S. & Watanabe, H. (1995) Litter production of a mangrove forest in East Sumatra, Indonesia. Proceedings of the Fifth Seminar on Mangrove Ecosystems. Jember, 3-6 August 1994, pp. 247±265. Lee, S.Y. (1986) The intensity and consequences of herbivory on Kandelia candel (L.) Druce leaves at the Mai Po marshes, Hong Kong. In: Proceedings of the Second International Marine Biological Workshop: The Marine Flora and Fauna of Hong Kong and Southern China, pp. 717±725. Hong Kong University Press, Hong Kong. Lee, S.Y. (1988) The ecology of a traditional tidal shrimp pond in Hong Kong, the production and fate of macrodetritus, and implications for management. PhD Thesis, University of Hong Kong. Lee, S.Y. (1989a) Litter production and turnover of the mangrove Kandelia candel (L.) Druce in a Hong Kong tidal shrimp pond. Estuarine, Coastal and Shelf Science, 29, 75±87. Lee, S.Y. (1989b) The importance of sesarmine crabs Chiromanthes spp. and inundation frequency on mangrove (Kandelia candel (L.) Druce) leaf litter turnover in a Hong Kong tidal shrimp pond. Journal of Experimental Marine Biology and Ecology, 131, 23±43. Lee, S.Y. (1990a) Primary productivity and particulate organic matter flow in an estuarine mangrove-wetland in Hong Kong. Marine Biology, 106, 453±463. Lee, S.Y. (1990b) Net aerial primary productivity, litter production and decomposition of the reed Phragmites communis in a nature reserve in Hong Kong: management implications. Marine Ecology Progress Series, 66, 161±173. Lee, S.Y. (1991) Herbivory as an ecological process in a Kandelia candel (Rhizophoraceae) mangal in Hong Kong. Journal of Tropical Ecology, 7, 337±348. Lee, S.Y. (1993) Leaf choice of sesarmine crabs, Chiromanthes bidens and C. maipoensis, in a Hong Kong mangal. In: Proceedings of the First International Conference on the Marine Biology of Hong Kong and the South China Sea, Hong Kong, 28 October±3 November, 1990, pp. 597±602.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

255

Lee, S.Y. (1995) Mangrove outwelling: a review. Hydrobiologia, 295, 203±212. Levine, E.A. (1981) Nitrogen cycling by the red mangrove Rhizophora mangle, in Joyuda Lagoon, on the west coast of Puerto Rico. MSc Thesis, University of Puerto Rico, Puerto Rico. Li, M.S. (1997) Nutrient dynamics of a Futian Mangrove forest in Shenzhen, South China. Estuarine, Coastal and Shelf Science, 45, 463±472. Li, M.S. & Lee, S. (1997) Mangroves of China: a brief review. Forest Ecology and Management, 96, 241±259. Li, M.S. & Lee, S. (1998) The carbon dynamics of Deep Bay, eastern Pearl River estuary, China; I: a mass balance budget and implications for shorebird conservation. Marine Ecology Progress Series, 172, 73±87. Little, D. & Satapornvanit, K. (1996) Poultry and fish production ± a framework for their integration in Asia. Second FAO Electronic Conference on Tropical Feeds Livestock Feed Resources within Integrated Farming Systems. http://fao.org/ag/aga/agap/frg/conf96.htm/ little2.htm. Loneragan, N.R., Bunn, S. & Kellaway, D. (1997) Are mangroves and seagrass sources of organic carbon for penaeid prawns in a tropical Australian estuary? A multiple stableisotope study. Marine Biology, 130, 289±300. Lugo, A.E. & Snedaker, S. (1974) The ecology of mangroves. Annual Review of Ecology and Systematics, 5, 39±64. Lugo, A.E., Brown, S. & Brinson, M. (1988) Forested wetlands in freshwater and salt-water environments. Limnology and Oceanography, 33, 894±909. Lugo, A.E., Brown, S. & Brinson, M. (1990) Concepts in wetland ecology. In: Ecosystems of the World: Forested Wetlands (ed. A.E. Lugo), pp. 53±85. Elsevier, Amsterdam. Machiwa, J. & Hallberg, R. (in press) An empirical model of the fate of organic carbon in a mangrove forest partly affected by anthropogenic activity. Ecological Modelling. Mackey, A.P. & Smail, G. (1996) The decomposition of mangrove litter in a subtropical forest. Hydrobiologia, 332, 93±98. Mallin, M.A., Burkholder, J. & Sullivan, M. (1992) Contributions of benthic microalgae to coastal fisheries yield. Transactions of the American Fisheries Society, 121, 691±695. Manoharan, C. & Subramanian, G. (1996) Effect of ossein effluent on the biochemistry of Oscillatoria pseudogeminata var. unigranulata Biswas. Journal of Environmental Biology, 17, 157±161. Micheli, F. (1993) Feeding ecology of mangrove crabs in North Eastern Australia: mangrove litter consumption by Sesarma messa and Sesarma smithii. Journal of Experimental Marine Biology and Ecology, 171, 165±186. Minello, T.J. & Zimmerman, T. (1991) The role of estuarine habitats in regulating growth and survival of juvenile penaeid shrimp. In: Frontiers of Shrimp Research (eds P.F. DeLoach, W.J. Dougherty & M.A. Davidson), pp. 1±16. Elsevier, Amsterdam. Mitsch, W.J. (1993) Ecological engineering: a cooperative role with the planetary life-support system. Environmental Science and Technology, 127, 438±445. Moriarty, D.J.W. & Barclay, M. (1981) Carbon and nitrogen content of food and the assimilation efficiencies of penaeid prawns in the Gulf of Carpentaria. Australian Journal of Marine and Freshwater Research, 32, 245±251. Morton, R.M. (1990) Community structure, density and standing crop of fishes in a subtropical Australian mangrove area. Marine Biology, 105, 385±394. Naidu, K.A., Tewari, A., Joshi, H.V., Viswanath, S., Ramesh, H.P. & Rao, S.V. (1993)

256

Ecological Aquaculture

Evaluation of nutritional quality and food safety of seaweeds of India. Journal of Food Safety, 13, 77±90. Nedwell, D.B. (1974) Sewage treatment and discharge into tropical coastal waters. Search, 5, 187±190. Nedwell, D.B., Blackburn, T.H. & Wiebe, W.J. (1994) Dynamic nature of the turnover of organic carbon, nitrogen and sulphur in the sediments of a Jamaican mangrove forest. Marine Ecology Progress Series, 110, 223±231. Newell, R.I.E., Marshall, N., Sasekumar, A. & Chong, V.C. (1995) Relative importance of benthic microalgae, phytoplankton, and mangroves as sources of nutrition for penaeid prawns and other coastal invertebrates from Malaysia. Marine Biology, 123, 595±606. Nurdjana, M.L. (1997) Indonesia strives to turn its shrimp aquaculture environmentally friendly. Aquaculture Asia, 2, 3±8. O'Connell, M.F. & Andrews, C. (1987) Plankton ecology in relation to flushing rate in four Newfoundland ponds. International Revue Gesamen Hydrobiologia, 72, 487±515. Odum, E.P. (1969) The strategy of ecosystem development. Science, 164, 262±270. Odum, E.P. (1980) The status of three ecosystem-level hypotheses regarding salt marsh estuaries: tidal subsidy, outwelling, and detritus-based food chains. In: Estuarine Perspectives (ed. V. Kennedy), pp. 485±495. Academic Press, New York. Odum, W.E. (1971) Pathways of energy flow in a south Florida estuary. University of Miami Sea Grant Technical Bulletin No. 7. Miami, FL. Odum, W.E. & Heald, E.J. (1972) Trophic analyses of an estuarine mangrove community. Bulletin of Marine Science, 22, 671±738. Odum, W.E. & Heald, E.J. (1975) The detritus-based mangrove community. In: Estuarine Research, Vol. 1 (ed. L.E. Cronin), pp. 265±286. Academic Press, New York. Odum, W.E., Fisher, J.S. & Pickral, J.C. (1979) Factors controlling the flux of particulate organic carbon from estuarine wetlands. In: Ecological Processes in Coastal and Marine Systems (ed. R.J. Levington), pp. 69±80. Plenum Press, New York. Odum, W.E., McInvor, C.C. & Smith, T.J. (1982) The Ecology of the Mangroves of South Florida: A Community Profile. US Fish and Wildlife Service, Office of Biological Services, Washington, DC. Osunkoya, O.O. & Creese, R.G. (1997) Population structure, spatial pattern and seedling establishment of the grey mangrove, Avicennia marina var. australasica, in New Zealand. Australian Journal of Botany, 45, 707±725. Overton, J.L. & Macintosh, D.J. (1997) Mud crab culture: prospects for the small-scale Asian farmer. INFOFISH International, 5, 26±32. Pearson, T.H. & Rosenberg, R. (1978) Macrobenthic succession in relation to organic enrichment and pollution of the marine environment. Oceanography Marine Biology Annual Review, 16, 229±311. Peterson, B.J. & Fry, B. (1987) Stable isotopes in ecosystem studies. Annual Review of Ecology and Systematics, 18, 293±320. Pillary, T.V.R. (1992) Aquaculture and the Environment. Fishing News Books, Cambridge, MA. Pool, D.J., Lugo, A.E. & Snedaker, S.C. (1975) Litter production in mangrove forests of southern Florida and Puerto Rico. In: Proceedings of the International Symposium on Biology and Management of Mangroves (eds G.E. Walsh, S.C. Snedaker & H.J. Teas), pp. 213±237. University of Florida, Gainesville.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

257

Poovachiranon, S. & Chansang, H. (1982) Structure of Ao Yon mangrove forest and its contribution to coastal ecosystem. Biotropica, 17, 101±111. Por, F.D. (1984) Editor's note on tides and water levels in the mangal. In: Hydrobiology of the Mangal (eds F.D. Por & I. Dor), pp. 25±26. W. Junk Publications, The Hague. Potts, M. (1984) Nitrogen fixation in mangrove forests. In: Hydrobiology of the Mangal (eds F.D. Por & I. Dor), pp. 155±162. W. Junk Publishers, The Hague. Primavera, J.H. (1995) Mangroves and brackishwater pond culture in the Philippines. Hydrobiologia, 295, 303±309. Primavera, J.H. (1996) Stable carbon and nitrogen isotope ratios of penaeid juveniles and primary producers in a riverine mangrove in Guimaras, Philippines. Bulletin of Marine Science, 58, 675±683. Primavera, J.H. (1998a) Tropical shrimp farming and its sustainability. In: Tropical Mariculture (ed. S.S. DeSilva), pp. 257±289. Academic Press, New York. Primavera, J.H. (1998b) Mangroves as nurseries: shrimp populations in mangrove and nonmangrove habitats. Estuarine, Coastal and Shelf Science, 46, 457±464. Primavera, J.H. & Agbayani, R.F. (1996) Comparative strategies in community-based mangrove rehabilitation programs in the Philippines. Paper presented at the Ecotone V: Community Participation in Conservation, Sustainable Use and Rehabilitation of Mangroves in Southeast Asia, Ho Chi Minh City, Vietnam, January 1996. Pruder, G.D. (1987) Detrital and algal based food chains in aquaculture: a perspective. In: Detritus and Microbial Ecology in Aquaculture (eds D.J.W. Moriarty & R.S.V. Pullin), pp. 296±308. ICLARM Conference Proceedings 14. International Center for Living Aquatic Resources Management, Manila, Philippines. Qasim, R. (1991) Amino acid composition of some common seaweeds. Pakistan Journal of Pharmaceutical Sciences, 4, 49±54. Qasim, S.Z. & Easterson, D.C.V. (1974) Energy conversion in the shrimp Metapenaeus monoceros fed on detritus. Indian Journal of Marine Science, 3, 131±134. Raghukumar, S., Sharma, S., Raghukumar, C., Sathe-Pathak, V. & Chandramohan, D. (1994) Thraustochytrid and fungal component of marine detritus, IV. Laboratory studies on decomposition of leaves of the mangrove Rhizophora apiculata Blume. Journal of Experimental Marine Biology and Ecology, 183, 113±131. Rajagopalan, M.S., Gopinathan, C.P. & Balachandran, V. (1986) Productivity of different mangrove ecosystems. Proceedings of the Symposium on Coastal Aquaculture, 4, 1084±1087. Redfield, J.A. (1982a) Primary productivity and trophic dynamics. In: Mangrove Ecosystems in Australia: Structure, Function and Management (ed. B.F. Clough), pp. 211±212. Australian Institute of Marine Science, Canberra. Redfield, J.A. (1982b) Trophic relationships in mangrove communities. In: Mangrove Ecosystems in Australia: Structure, Function and Management (ed. B.F. Clough), pp. 259±261. Australian Institute of Marine Science, Canberra. Reice, S.R., Spira, Y. & Por, F.D. (1984) Decomposition in the mangal of Sinai: the effect of spatial heterogeneity. In: Hydrobiology of the Mangal (eds F.D. Por & I. Dor), pp. 193±199. W. Junk Publishers, The Hague. Rezende, C.E., Lacerda, L., Ovalle, A., Silva, C. & Martinelli, L. (1990) Nature of POC transport in a mangrove ecosystem: a carbon stable isotopic study. Estuarine, Coastal and Shelf Science, 30, 641±645.

258

Ecological Aquaculture

Rice, D.L. & Hanson, R. (1984) A kinetic model for detritus nitrogen: role of the associated bacteria in nitrogen accumulation. Bulletin of Marine Science, 75, 401±474. Rice, D.L. & Tenore, K.R. (1981) Dynamics of carbon and nitrogen during the decomposition of detritus derived from estuarine macrophytes. Estuarine, Coastal and Shelf Science, 13, 681±690. Rivera-Monroy, V.H. & Twilley, R.R. (1996) The relative role of denitrification and immobilization in the fate of inorganic nitrogen in mangrove sediments. Limnology and Oceanography, 41, 284±296. Robertson, A.I. (1986) Leaf-burying crabs: their influence on energy flow and export from mixed mangrove forests (Rhizophora spp.) in northeastern Australia. Journal of Experimental Marine Biology and Ecology, 102, 237±248. Robertson, A.I. (1988a) Decomposition of mangrove leaf litter in tropical Australia. Journal of Experimental Marine Biology and Ecology, 116, 235±247. Robertson, A.I. (1988b) Abundance, diet and predators of juvenile banana prawns Penaeus merguiensis, in a tropical mangrove estuary. Australian Journal of Marine and Freshwater Research, 39, 467±478. Robertson, A.I. & Duke, N. (1987) Mangroves as nursery sites: comparisons of the abundance and species composition of crustaceans in mangroves and other nearshore habitats in tropical Australia. Marine Biology, 96, 193±205. Robertson, A.I. & Daniel, P.A. (1989) The influence of crabs on litter processing in high intertidal mangrove forests in tropical Australia. Oecologia, 78, 191±198. Robertson, A.I. & Phillips, M.J. (1995) Mangroves as filters of shrimp pond effluent: predictions and biogeochemical research needs. Hydrobiologia, 295, 311±321. Rodelli, M.R., Gearing, J., Gearing, P., Marshall, N. & Sasekumar, A. (1984) Stable isotope ratio as a tracer of mangrove carbon in Malaysian ecosystems. Oecologia, 61, 326±333. Rusdi, H. & Jasin, H.H. (1994) Socio economic condition of `Tambak Tumpangsari' farmer in social forestry of RPH Cibuaya region, BKPH Cikion, KPH Purwakarta, Perum Perhutani Unit III, West Java. In: Ecological Assessment for Management Planning of Segara Anakan Lagoon, Cilacap, Central Java (eds F. Takashima and K. Soewardi). NODAI Center for International Program Tokyo University of Agriculture JSPS-DGHE Program. Russel-Hunter, W.D. (1970) Aquatic Productivity: Introduction to Some Basic Aspects of Biological Oceanography and Limnology. Macmillan, London. Saenger, P., Hegerl, E. & Davis, J. (1983) Factors which maintain ecological process in mangroves. Global Status of Mangrove Ecosystem. Commission on Ecology, paper 3, pp. 29±32. Saenger, P. & Snedaker, S. (1993) Pantropical trends in mangrove above-ground biomass and annual litterfall. Oecologia, 96, 293±299. Samonte, G.PB. & Agbayani, R. (1992) Pond culture of mud crab (Scylla serrata): an economic analysis. SEAFDEC Asian Aquaculture, 14, 3±5. Sasekumar, A. & Loi, J. (1983) Litter production in three mangrove forest zones in the Malay Peninsula. Aquatic Botany, 17, 283±290. Sasekumar, A., Chong, V., Leh, M. & D'Cruz, R. (1992) Mangroves as a habitat for fish and prawns. Hydrobiologia, 247, 195±207. Schwartz, M.F. & Boyd, C. (1995) Constructed wetlands for treatment of channel catfish pond effluents. The Progressive Fish-Culturist, 57, 255±266.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

259

Sedgwick, R.W. (1979) Influence of dietary protein and energy on growth, food consumption and food conversion efficiency in Penaeus merguiensis de Man. Aquaculture, 16, 7±30. Selvam, V., Karunakaran, V.M., Ravichandran, K.K. & Gunalan, R. (2001) A traditional method integrating fisheries development with mangrove regeneration and its application in restoring degraded mangrove wetlands. Proceedings of the International Symposium on Mangroves, University of Tokyo. Seville E. (1987) Pond culture of mud crab (Scylla serrata): an economic analysis. SEAFDEC Asian Aquaculture, 14, 3±5. Shankar, K.M. (1999) Bacterial biofilm in aquaculture. Aquaculture Asia, 4, 47. Sibert, J.R. & Naiman, R.J. (1980). The role of detritus and the nature of estuarine ecosystems. In: Marine Benthic Dynamics (eds K.R. Tenore & B.C. Coull), pp. 31±323. University of South Carolina Press, Columbia. Slim, F.J., Hemminga, M., Ochieng, C., Jannink, N., Cocheret de la Moriniere, E. & van der Velde, G. (1997) Leaf litter removal by the snail Terebralia palustris (Linnaeus) and sesarmid crabs in an East African mangrove forest (Gazi Bay, Kenya). Journal of Experimental Marine Biology and Ecology, 215, 35-48. Soegiarto, A. (1984) The mangrove ecosystem in Indonesia, its problems and management. In: Physiology and Management of Mangroves (ed. H.J. Teas), pp. 69±78. W. Junk Publishers, The Hague. Soemodihardjo, S. & Soerianegara, I. (1989) Country report: Indonesia. The status of mangrove forests in Indonesia. In: Symposium on Mangrove Management: Its Ecological and Economic Considerations (ed. I. Soerianegara). Southeast Asian Regional Center for Tropical Biology, Bogor Indonesia, Aug. 9±11, 1988. Biotrop. Special Publication No. 37, Bogor, Indonesia. Soewardi, K., Saladin, A., Affandi, R. & Takashima, F. (1996) Production of natural penaeid shrimp (Penaeid sp.) in various silvofishery conditions. In: Research on Various Fisheries Aspects with Special References of West Java, Indonesia (eds K. Soewardi & F. Takashima), pp. 45±48. Contribution of Collaborative Research Work between Faculty of Fisheries, Bogor Agricultural University and Tokyo University of Fisheries. Steinke, T.D., Naidoo, G. & Charles, L. (1983) Degradation of mangrove leaf and stem tissues in situ in Mgeni estuary, South Africa. In: Tasks for Vegetation Science (ed. H. Teas), pp. 141±149. W. Junk Publishers, The Hague. Steinke, T.D. & Ward, C.J. (1987) Degradation of mangrove leaf litter in the St Lucia Estuary as influenced by season and exposure. South African Journal of Botany, 53, 323±328. Steinke, T.D., Holland, A. & Singh, Y. (1993) Leaching losses during decomposition of mangrove leaf litter. South African Journal of Botany, 59, 21±25. Stoner, A.W. & Zimmerman, R. (1988) Food pathways associated with penaeid shrimps in a mangrove-fringed estuary. Fisheries Bulletin US, 86, 543±551. Su, M.S. & Liao, I. (1984) Preliminary studies on the distribution and the stomach contents of some common prawns from the coast of Tungkang, Taiwan. In: Proceedings of the ROCJapan Symposium on Mariculture (eds I.C. Liao & R. Hirano), pp. 57±71. Tungkang Marine Laboratory Conference Proceedings 1. Tungkang Marine Laboratory, Taiwan. Sukardjo, S. (1989) Tumpang sari pond as a multiple use concept to save the mangrove forest in Java. Biotrop Special Publication No. 37, 115±128. Sukardjo, S. (1999) Mangroves country report ± Indonesia. SEAFDEC Workshop on Mangrove Friendly Aquaculture, Iloilo City, Philippines.

260

Ecological Aquaculture

Sullivan, M.J. & Moncreiff, C. (1990) Edaphic algae are an important component of salt marsh food-webs: evidence from multiple stable isotope analyses. Marine Ecology Progress Series, 62, 149±159. Takashima, F. (1999) Silvo-fishery ± an aquaculture system harmonized with the environment. Workshop on Mangrove-Friendly Aquaculture. SEAFDEC, Iloilo City, Philippines, 11±15 January. Tam, N.F.Y., Vrijmoed, L. & Wong, Y. (1990) Nutrient dynamics associated with leaf decomposition in a small subtropical mangrove community in Hong Kong. Bulletin of Marine Science, 47, 68±78. Tanaka, J. & Chihara, M. (1987) Species composition and vertical distribution of macroalgae in brackish waters of Japanese mangrove forests. Bulletin of the National Science Museum of Tokyo, Series B, 13, 141±150. Tanan, S. & Tansutapanich, A. (1999) A report of mangrove-friendly shrimp farming in Thailand. Workshop on Mangrove-Friendly Aquaculture, SEAFDEC, Iloilo City, Philippines, 11±15 January. Teal, J.M. (1962) Energy flow in the salt marsh ecosystem of Georgia. Ecology, 43, 614±624. Tenore, K.R. (1988) Nitrogen in benthic food chains. In: Nitrogen Cycling in Coastal Marine Environment (eds T.H. Blackburn & J. Sorensen), pp. 191±206. John Wiley & Sons, New York. Tenore, K.R., Cammen, L., Findlay, S. & Phillips, N. (1982) Perspectives of research on detritus: do factors controlling the availability of detritus to macroconsumers depend on its source? Journal of Marine Research, 40, 473±489. Theimsiri, S. (1992) New concept on poultry breeder management in tropical countries. In Proceedings of Luang Suwan International Poultry Symposium, the Sixth Animal Science Congress of AAAP, 27±28 November, 1992, Bangkok, pp. 61±71. Tietjen, J.H. & Alongi, D. (1990) Population growth and effects of nematodes on nutrient regeneration and bacteria associated with mangrove detritus from northeastern Queensland (Australia). Marine Ecology Progress Series, 68, 169±179. Tredici, M.R., Papuzzo, P. & Tomaselli, L. (1986) Outdoor mass culture of Spirulina maxima in sea-water. Applied Microbiology and Biotechnology, 24, 47±50. Trino, A. & Rodriguez, E. (1999) Mud crab culture in tidal flats with existing mangroves. Workshop on Mangrove-Friendly Aquaculture. SEAFDEC, Iloilo City, Philippines, 11±15 January. Truong, T.N., Vu, N. & Duong, Q. (1992) Improvement of traditional shrimp culture in the Mekong Delta, Vietnam. Final report phase I and II. Canada Fund for Local Initiatives, Sector 04 Fisheries, Project No. G4700-975-91/92/92-08. Artemia Shrimp R & D Center, University of Cantho, Vietnam. Tsuda, R.T. & Bryan, P. (1973) Food preference of juvenile Siganus rostratus and S. spinus in Guam. Copeia, 3, 604±606. Tsuda, R.T., Bryan, P., FitzGerald, Jr, W. & Tobias, W. (1974) Juvenile±adult rearing of Siganus (Pisces: Siganidae) in Guam. Unpublished paper presented to the South Pacific Commission 7th Technical Meeting on Fisheries Nuku'a 'lofa, Tonga. Twilley, R.R. (1982) Litter dynamics and organic carbon exchange in black mangrove (Avicennia germinans) basin forests in a southwest Florida estuary. Ph.D. Thesis University of Florida, Gainesville. Twilley, R.R. (1985) The exchange of organic carbon in basin mangrove forests in a southwest Florida estuary. Estuarine, Coastal and Shelf Science, 20, 543±557.

Silvofisheries: Integrated Mangrove Forest Aquaculture Systems

261

Twilley, R.R., Lugo, A. & Patterson-Zucca, C. (1986a). Litter production and turnover in basin mangrove forests in southwest Florida. Ecology, 67, 670±683. Twilley, R.R., Ejdung, G., Romare, P. & Kemp, W. (1986b) A comparative study of decomposition, oxygen consumption and nutrient release for selected aquatic plants occurring in an estuarine environment. Oikos, 47, 190±198. Twilley, R.R., Chen, R. & Hargis, T. (1992) Carbon sinks in mangroves and their implications to carbon budget of tropical coastal ecosystems. Water, Air & Soil Pollution, 64, 265±288. Twilley, R.R., Pozo, M., Garcia, V.H., Rivera-Monroy, B.H., Zambrano, R. & Bodero, A. (1997) Litter dynamics in riverine mangrove forests in the Guayas River Estuary, Ecuador. Oecologia, 111 (1), 109±122. Ulanowicz, R.E., Bondavalli, C., Heymans, J. & Egnotovich, M. (1999) Network analysis of trophic dynamics in South Florida Ecosystems, FY98: the mangrove ecosystem. Technical Report Series No. TS-191-99 of the University of Maryland System Center for Environmental Science, Chesapeake Biological Laboratory. Untawale, A.G. (1987) Country reports: Asia, India. In: Mangroves of Asia and the Pacific: Status and Management. Technical Report of UNDP/UNESCO Research and Training Pilot Programme and Mangrove Ecosystems in Asia and the Pacific, pp. 51±87. Vance, D.J. (1992) Activity patterns of juvenile penaeid prawns in response to artificial tidal and day±night cycles: a comparison of three species. Marine Ecology Progress Series, 87, 215±226. Vance, D.J. & Staples, D. (1992) Catchability and sampling of three species of juvenile penaeid prawns in the Embley River, Gulf of Carpentaria, Australia. Marine Ecology Progress Series, 87, 21±213. Van der Valk, A.G. & Attiwill, P. (1984) Decomposition of leaf and root litter of Avicennia marina at Westerport Bay, Victoria, Australia. Aquatic Botany, 18, 205±221. Vonshak, A., Kancharaksa, N., Bunnag, B. & Tanticharoen, M. (1996) Role of light and photosynthesis on the acclimation process of the cyanobacterium Spirulina platensis to salinity stress. Journal of Applied Phycology, 8, 119±124. Wafar, S., Untawale, A. & Wafar, M. (1997) Litter fall and energy flux in a mangrove ecosystem. Estuarine, Coastal and Shelf Science, 44, 111±124. Waghmode, A.P. (1987) Preliminary studies in the nitrogen dynamics of estuarine ecosystem. Journal of the Marine Biological Association of India, 29, 362±364. Wahbeh, M.I. (1997) Amino acid and fatty acid profiles of four species of macroalgae from Aqaba and their suitability for use in fish diets. Aquaculture, 159, 101±109. Wassenberg, T.J. & Hill, B. (1993) Diet and feeding behavior of juvenile and adult banana prawns Penaeus merguiensis in the Gulf of Carpentaria, Australia. Marine Ecology Progress Series, 94, 287±295. Waston, J.G. (1928) Mangrove forests of the Malay Peninsula. Malayan Federal Records, 6, 1±275. Widiarti, A. & Effendi, R. (1989) Socio-economic aspects of brackishwater pond forest in mangrove forest complex. Symposium Mangrove Management. Biotrop Special Publication No. 37, pp. 275±279. Wong, Y.S., Lan, C., Chen, G., Li, S., Chen, X., Liu, Z. & Tam, N. (1995) Effect of wastewater discharge on nutrient contamination of mangrove soils and plants. Hydrobiologia, 295, 243±254.

262

Ecological Aquaculture

Woodroffe, C.D. (1985) Studies of a mangrove basin, Tuff Crater, New Zealand I. Mangrove biomass and production of detritus. Estuarine, Coastal and Shelf Science, 20, 265±280. Woodwell, G.M., Whitney, D., Hall, C. & Houghton, R. (1977) The flax pond ecosystem study: exchanges of carbon in water between a salt marsh and Long Island Sound. Limnology and Oceanography, 22, 833±838. Young, L. (1996) Mangrove distribution in the gei wais at the Mai Po Marshes Nature Reserve. In: Proceedings of the International Mangrove Workshop at the Mai Po Marshes Nature Reserve (ed. S.Y. Lee). University of Hong Kong Press, Hong Kong. Young, L. (1997) Mangrove conservation and shrimp aquaculture. In: Proceedings of the First International Symposium on Marine Conservation, Hong Kong, 26±27 October 1996 (ed. B.W. Darvell), pp. 128±155. Hong Kong Marine Conservation Society, Hong Kong. Young, L. & Wong, K. (1994) Conservation importance of the traditional gei wai prawn ponds at the Mai Po Marshes Nature Reserve, Hong Kong. Unpublished report for WWF Hong Kong.

Chapter 9

An Integrated Fish and Field Crop System for Arid Areas James E. Rakocy University of the Virgin Islands

Introduction In arid areas food production often cannot keep pace with demand and large amounts of food must be imported at a substantial cost to local economies. Watershed destruction and shifts in weather patterns are creating dry lands at an increasing rate. The limited water that is available for agriculture in the arid regions should be conserved and reused. The thought of raising fish in arid areas as part of a sound water management plan and a system for enhanced food production seems on the surface to be counterproductive and impractical. On the contrary, there is a fish culture system that can be integral to dry land agriculture while having no net effect on water supplies other than evaporative losses. The system is greenwater tank culture. This system is more productive than standard pond culture techniques while not being overly sophisticated or expensive. Waste nutrients and organic matter from the system are used to grow crops and improve soil structure and fertility. Research on greenwater tank culture systems is conducted at the University of the Virgin Islands (UVI) in response to the lack of freshwater resources and the need for consistent supplies of fresh fish and vegetables in the US Virgin Islands, which are located at 188 north latitude in the tropics. Although the Virgin Islands have annual rainfall of 1 m, there are no persistent rivers or streams, and rainstorm runoff events occur infrequently, less than once per year. Rainfall is uniformly distributed in small amounts that quickly evaporate due to high temperatures, intense solar radiation and strong trade winds. Groundwater is limited and is used primarily for domestic water needs. Although some groundwater is used for livestock watering and microirrigation of fruits and vegetables on a small scale, the supply is not large enough for pond aquaculture. The Virgin Islands represent the high end of the dry land agricultural scale. Although most land areas have harsh conditions, greenwater tank aquaculture is an appropriate technology. In designing dry land aquaculture ecosystems, considerations of water conservation and production intensification are needed. Tank systems conserve water more than pond culture; but they are not as technologically advanced, expensive, or as demanding as standard recirculating systems. Although recirculating systems are

264

Ecological Aquaculture

used increasingly in temperate developed countries to raise high value fish, they are not as common or as economically feasible in developing tropical countries where ambient temperatures are suitable for aquaculture, and systems do not necessitate system enclosure and a high degree of control. The greenwater system developed at UVI represents simpler and less expensive technology that intensifies production of tilapia over that of static pond culture by 25- to 30-fold without employing sophisticated solids removal and biofiltration devices or oxygenation. The system is suitable for continuous production in the tropics or summertime production in the subtropics. Tilapia was the fish used in developing the greenwater system at UVI. Tilapia's ability to tolerate high densities and poor water quality makes it an ideal candidate for this system. Moreover, tilapia's characteristic of swimming near the tank bottom and grazing on detritus is instrumental in moving solid waste to the drain for quick removal from the rearing tank, an important factor for successful operation of this system. Many fish species restrict their movements to the water column and do not `work' the tank bottom. Therefore, the results obtained with tilapia may not transfer to other species. Species with less tolerance to poor water quality may not do as well. For less tolerant species, densities and feeding rates may have to be lowered.

Greenwater tank culture Variations of greenwater tank culture are used in Israel (Avnimelech et al., 1994; Avnimelech, 1998), Louisiana (Lutz, 1998) and California (Schroeder & Serfling, 1989). The process whereby metabolic waste products of fish are treated in greenwater tank culture, allowing productivity to increase, is alternately referred to as a `suspended growth process', an `activated suspension process' or a `bioconversion process'. Regardless of name, the principle is the same and involves the oxidation of toxic ammonia and nitrite to relatively non-toxic nitrate ions by nitrifying bacteria attached to suspended organic matter. Additional removal of ammonia occurs through direct uptake by phytoplankton. Organic wastes are incorporated into bacteria or respired as carbon dioxide. The rich microbial populations of greenwater tank systems provide nutrition to tilapia and improve feed conversion efficiency (Schroeder & Serfling, 1989; Avnimelech et al., 1994; Kochba et al., 1994). Most greenwater systems remove settleable solid waste daily while some add a fixed-film nitrification unit to supplement the biofiltration that occurs in the water column (Lutz, 1998). Greenwater tank systems must be aerated to sustain adequate dissolved oxygen (DO) levels in tanks with elevated organic loading and correspondingly higher biochemical oxygen demand (BOD). A major advantage of greenwater tank systems is that water treatment is based primarily on bacteria. In ponds, water treatment is based mainly on phytoplankton. An algal die-off in ponds leads to oxygen depletion, necessitates emergency aeration and disrupts feeding and production. An algal die-off in greenwater tanks has relatively minor impact. Mechanical aeration continues to supply sufficient oxygen, and bacteria continue to oxidize toxic waste metabolites. Dead algae settle to the tank bottom and are eliminated through daily sludge removal, thereby reducing BOD and avoiding secondary ammonia production.

An Integrated Fish and Field Crop System for Arid Areas

265

The UVI greenwater system At UVI, greenwater tank culture research for grow-out is conducted in circular fiberglass tanks that are 6 m in diameter and 1.22 m in height (Figs 9.1±9.3). Water volume averages 30 m3. The tanks are continuously aerated by a 0.05-hp vertical lift pump and 13 air stones (4.55 cfm total). The rearing tank floor has a 38 slope to a central drain. Water is drawn off the bottom of the rearing tank from a central drain by an airlift pump into a 1.4-m3 cylindro-conical clarifier with a 458 bottom slope. Each clarifier is stocked with 10±15 male tilapia fingerlings to graze algal growth and facilitate the settling of solids. Two baffles are placed perpendicular to the water flow. The central baffle extends the height of the cylinder to slow and disperse the incoming water flow. A second baffle is located near the outlet to retain floating solids. Flow rate is maintained at 20 liters/min so that the entire rearing tank volume passes through the clarifier once every 24 hours. The clarifier residence time is 70 min. Concentrated solids (sludge) are removed and volumetrically measured twice daily from each clarifier (Figs 9.4±9.6).

Fig. 9.1 Six greenwater tank systems for tilapia grow-out. (Photograph by J.E. Rakocy.)

Greenwater tank culture for fingerling production is conducted in circular fiberglass tanks that are 3.66 m in diameter and 1.22 m in height with an average water volume of 11 m3 (Figs 9.7±9.9). Tanks are continuously aerated by 25 air stones with a total air volume of 9.65 cfm. Water is drawn off the bottom of the rearing tank from a central drain by an airlift pump into a 380-liter cylindro-conical clarifier with a 458 bottom slope (Fig. 9.10). Each clarifier is stocked with 10±15 male tilapia and has only one central baffle. Flow rate is maintained at 8 liters/min, which produces one water

266

Ecological Aquaculture

Clarifier (1.4 m3)

Rearing tank 6.1 m

1.1 m

Baffles

Air lift Water volume = 30 m3 Solids

Fig. 9.2 Schematic of 30-m3 greenwater production tank.

Fig. 9.3 Greenwater grow-out tank with clarifier in the foreground. (Photograph by J.E. Rakocy.)

Fig. 9.4 Sludge removal from cylindro-conical clarifier. (Photograph by J.E. Rakocy.)

An Integrated Fish and Field Crop System for Arid Areas

267

Fig. 9.5 Greenwater sludge being poured in a trough. (Photograph by J.M. Martin.)

Fig. 9.6 Greenwater sludge collected in a sump for irrigation of field crops. (Photograph by J.E. Rakocy.)

exchange per day in the fingerling rearing tank and a 47-min retention time in the clarifier. Sludge was removed from the clarifier and volumetrically measured twice daily in trial 1 and four times daily in trials 2 and 3. Six grow-out experiments have been conducted using a strain of red tilapia and Nile tilapia (Oreochromis niloticus, Egypt strain). All experiments except number 5 included two treatments with three replications per treatment. The red tilapia were derived from Florida red tilapia, which underwent 16 generations of selection in Florida and Jamaica before UVI received them 15 years ago, during which time

268

Ecological Aquaculture

380-L clarifier 3.66 m

Air-lift pump

Baffle

5.08 cm drain line 3

Fig. 9.7 Schematic of 11-m greenwater nursery tank.

Fig. 9.8 Nursery tank (11-m3) for greenwater production of tilapia fingerlings. (Photograph by D.S. Bailey.)

Fig. 9.9 Cylindro-conical clarifier for greenwater nursery tank. (Photograph by D.S. Bailey.)

An Integrated Fish and Field Crop System for Arid Areas

269

Fig. 9.10 Airlift pump. Air injected into the drain line lifts water into the clarifier. (Photograph by D.S. Bailey.)

additional selection occurred. The goal of the experiments was to develop an efficient system for producing high densities of tilapia in greenwater tanks. A number of parameters were examined in experiments to determine the effect of stocking rate, water exchange, sludge removal and alum addition. The results of these experiments are summarized in Table 9.1. Three fingerling production trials have been conducted. The goal of these trials is to determine the optimum stocking rate to produce 50-g fingerlings in a 12-week period (Fig. 9.11). Operational procedures are being changed or modified to develop a set of best management practices. The results of these experiments are summarized in Table 9.2.

Fig. 9.11 Target size (50 g) for Nile tilapia fingerling produced in greenwater nursery system. (Photograph by D.S. Bailey.)

Stocking rates Grow-out Stocking rates varied from 8 to 26 fish/m3. The optimum stocking rate for this greenwater system depends on the length of the culture period, the initial size, the final desired size and the feeding rate and method. All the fish in the grow-out experiments were fed a complete diet of floating pellets containing 32% protein. The culture period in the six experiments varied from 20 to 30 weeks. Initial fingerling size varied from 16 to 201 g. In the first four experiments, the fish were fed twice a day at a fixed percentage of their body weight daily, starting with an initial feeding rate of 3 to 4%.

270

Stocking size (g)

Final size (g)

Growth rate (g/d)

Final biomass (kg/m3)

FCR

Survival (%)

8 16

46 46

± ±

2.8a 2.8a

2.9a* 5.9b*

1.7a 1.6a

93a 95a

No water exchange Water exchange-5%/d

12 12

94 94

± ±

3.4c 3.5c

6.5c* 6.8c*

1.9c 1.8c

95c 97c

Nile

Sludge removal No sludge removal

24 24

16.5 16.5

422d 364e

2.4d 2.1e

9.0d 7.6e

1.60d 1.61d

88.3d 86.3d

4

Nile

Sludge removal No sludge removal

26 26

71.5 70.4

514f 417g

2.6f 2.1g

13.4f 10.8g

1.41f 1.51g

99.3f 98.9f

5

Nile

25.8 mg alum/liter/wk 38.7 mg alum/liter/wk 51.5 mg alum/liter/wk

17.8 17.8 17.8

4.35 4.41 4.66

16.0 16.3 17.1

1.79 1.86 1.87

97.6 99.2 98.7

6

Nile

No alum 50 mg/liter alum/wk

24 24

3.4h 2.8h

17.0h 14.5h

1.56h 1.55h

95.3h 96.3h

Experiment

Species

Treatment

1

Red

Stocking rate Stocking rate

2

Red

3

* Net yield.

Stocking rate (number/m3)

194.7 195.1 201.3 40.9 42.1

947 954 1002 744h 629h

Ecological Aquaculture

Table 9.1 Summary of six greenwater tank culture experiments giving tilapia species (red or Oreochromis niloticus), experimental treatment, stocking rate, stocking size, final size, growth rate, final biomass, feed conversion ratio (FCR) and survival. Values within each column of an experiment followed by the same letter are not significantly different (P > 0.05)

An Integrated Fish and Field Crop System for Arid Areas

271

Table 9.2 Summary of three greenwater tank culture experiments for the production of Nile tilapia fingerlings (Oreochromis niloticus), giving stocking rate, stocking size, growth rate, final biomass, feed conversion ratio (FCR) and survival over a 12-week production cycle. Stocking rate (number/m3)

Stocking size (g)

Final size (g)

Growth rate (g/d)

Final biomass (kg/m3)

FCR

Survival (%)

1

200 400 600

0.7 0.7 0.7

58.6 35.1 21.0

0.69 0.41 0.24

9.8 12.9 11.7

1.04 1.04 1.04

83.7 91.7 93.1

2

177 355 532

7.3 7.4 7.3

137.2 90.4 59.2

1.55 0.99 0.62

18.6 19.6 20.6

1.45 1.69 1.65

76.4 61.3 65.5

3

300 400 500

1.2 1.1 1.1

75.1 64.3 49.4

0.88 0.75 0.57

21.0 23.0 21.7

1.06 1.09 1.12

94.7 91.2 89.4

Experiment

The feeding rates were gradually reduced to 0.8±1.4% at the end of production cycle. In experiments 5 and 6, the fish were fed ad libitum twice daily with an amount of feed they could consume in one hour. In the Virgin Islands there are two desired final fish sizes. They are 500-g red tilapia for the West Indian market, which prefers whole fish after scales, guts and gills are removed (Fig. 9.12). The other market is for 900 g or larger Nile tilapia from which skinless fillets are produced (Fig. 9.13).

Fig. 9.12 Red tilapia produced in a greenwater tank system. (Photograph by J.E. Rakocy.)

The results of the experiments were far from conclusive in determining the optimum stocking rates because too many variables changed from one experiment to the other and the greenwater culture technology was continually being refined and improved. However, some general observations and goals can be given. Being situated in the tropics where year-round production is feasible, greenwater systems should be designed to give two crops per year for simplicity of management planning.

272

Ecological Aquaculture

Fig. 9.13 Nile tilapia produced in greenwater tank system. (Photograph by J.E. Rakocy.)

Assuming a 1-week turnaround time for harvesting, marketing, refilling and stocking, the culture period would be 25 weeks or 175 days of feeding. For red tilapia, none of the stocking densities in experiments 1 (20 weeks) and 2 (24 weeks) affected growth. The length of the culture period and the initial size did affect growth rate between the experiments. Starting with 46-g fingerlings, the highest stocking rate (16 fish/m3) resulted in an average growth rate of 2.8 g/day during 20 weeks. Assuming that the experiment had run for 25 weeks and the average daily growth rate increased to 3.0 g/day (as tilapia become larger the daily growth rate increases), the fish would have reached an average size of 571 g. Starting with 50-g fingerlings, a growth rate of 2.57 g/day will produce 500-g fish in 175 days. In later experiments, final densities of 17 kg/m3 were obtained for Nile tilapia with improved management techniques. Assuming that red tilapia can reach this density at a 500-g average size, then the optimum stocking rate would be 34 fish/m3. An experiment is needed for verification. Several densities between 16 and 34 fish/m3 will be tested in future experiments, always starting with advanced fingerlings weighing approximately 50 g. Nile tilapia reached the 900 g to 1 kg size range in experiment 5 at a stocking density of 17.8 fish/m3, a growing period of 25.6 weeks and an advanced fingerling size of approximately 200 g. This is not a practical fingerling size for conditions in the Virgin Islands. As with red tilapia, an initial size of 50 g is more appropriate. Nile tilapia would have to grow at an average rate of 4.86 g/day to reach 900 g in 25 weeks. A growth rate of 4.86 g/day was not approached in experiments 3,4 or 6

An Integrated Fish and Field Crop System for Arid Areas

273

at densities of 24, 26 and 24 fish/m3, respectively. Only in experiment 5 with the large fingerlings and a stocking rate of 17.8 fish/m3 were the growth rates in the range of 4.35±4.66 g/day. Knowing that a final density of 17 kg/m3 is feasible, a stocking rate of 18.9 fish/m3 would produce 900-g fish if the desired growth rate could be obtained. An experiment needs to be conducted at densities ranging from 16 to 20 fish/m3 (e.g. 16, 18 and 20 fish/m3). Although improved management procedures should lead to higher growth rates, a fingerling size of 50 g may be too small, and therefore experiments will be conducted testing 75- and 100-g fingerlings.

Fingerlings In trial 1, the aeration system was inadequate to sustain optimum feeding rates at the high densities, and DO levels were often < 2.0 mg/liter during the last month of the trial. This initial aeration system consisted of 13 air stones (4.5 cfm total) around the tank perimeter. Not only were DO levels insufficient for fish metabolism, they affected nitrification and high levels of ammonia and nitrite developed. Under these conditions only the lowest stocking density (200 fish/m3) exceeded the desired final size (Table 9.2). In trial 2, the aeration system was upgraded to 25 air stones with 9.65 cfm total air volume. As DO levels increased, final size, growth rate and final biomass increased. The results were confounded by having by a much higher initial stocking size (>7.0 g) and much lower survival, which contributed to the high feed conversion ration (1.45±1.69). The cause of high mortality at 177 fish/m3 is unknown but may have been due to bird mortality as no dead fish were found (the tanks were covered by nets, but some holes were noticed in the nets after the experiment). At 355 fish/m3, massive mortality occurred 2 days before the end of the experiment due to bacterial disease, possibly resulting from excessive organic matter in the water column. It is speculated that the density was approximately 30 kg/m3 when the die-off occurred. This density exceeded the capacity of the system's waste treatment process. At 532 fish/m3, massive mortality occurred due to one incidence of nitrite toxicity, which was brought under control by adding calcium chloride. In spite of the problems, the final biomass reached their highest levels (18.6±20.6 kg/ m3), and all treatments exceeded the target fingerling size (50 g) by a wide margin (59.2±137.2 g). In trial 3, stocking rates of 300, 400 and 500 fish/m3 were tested under a new set of management procedures (see feeding and water quality management sections). The initial stocking size (1.1±1.2 g) was appropriate for fry coming off a 28-day sexreversal procedure. Growth was rapid and fish stocked at 300 and 400 fish/m3 exceeded the target size (Table 9.2). There were no replications of stocking rate treatments for advanced fingerling production as only three systems are available (Table 9.2). When best management practices are determined and stocking rates appear to bracket the optimum rate, experimental treatments will be replicated over time.

274

Ecological Aquaculture

Feeding Grow-out All fish in the grow-out phase are fed a complete diet (32% protein, 3% fat) delivered as an 8-mm floating pellet. The daily ration was initially determined by using a feeding rate schedule based on the percentage of body weight, which was determined every month by sampling. The daily ration was divided into two allotments, which were delivered at approximately 0900 hours and 1600 hours. The initial feeding rate of 3±4% of body weight declined to 0.8±1.4% by the end of the growing cycle, which was low for fish ranging from 400 to 500 g in experiments 1±3 (Table 9.1). Daily feed consumption varied and often the fish were overfed as indicated by the relatively high feed conversion ratios (1.6±1.9) in experiments 1 and 2. If the fish could not finish their ration, it remained floating for most of the day, causing cancellation of the afternoon feeding. Although nitrification in a suspended greenwater system proved to be less stable than fixed-film nitrification, water quality variables such as DO, ammonia and nitrite could not always explain the fluctuation of appetite. Starting with experiments 5 and 6, the fish were fed ad libitum. Initially a feeding rate was determined for the week by one day of ad libitum feeding, as determined by all the feed the fish could finish in 1 hour per feeding twice a day. Later the feeding period was reduced to 30 min. Eventually feed rates were adjusted on Mondays, Wednesdays and Fridays. The feeding rate determined that day was used for the next day or the next 2 days after the Friday determination. Feed conversion rates came down dramatically in experiments 5 (1.79±1.87 for fish ranging from 947 to 1002 g) and 6 (1.55±1.56 for fish ranging from 629 to 744 g). Management time increased, but feed costs declined. During the 3 hours required for weekly feed adjustments, other tasks can be performed.

Fingerlings Similar to grow-out feeding, fingerling feeding procedures went through a period of adjustment until a suitable system was developed. The current procedure is to feed a complete diet (50% protein, 17% fat) of sinking granules (1.2 mm) at 10% of body weight daily for the first 2 weeks. The feeding rate is adjusted daily based on a hypothetical (1:1) feed conversion ratio. The daily ration is divided into four feedings at 0800, 1100, 1400 and 1700 hours. A standpipe with a fine-mesh screen is used during the first 2 weeks to draw effluent from the top of the tank and prevent feed loss to the clarifier. After 2 weeks, a complete diet (44% protein, 15% fat) of floating pellets (1.6 mm) is fed ad libitum for 30 min four times daily for 4 weeks. From this point on, the standpipe is removed and effluent to the clarifier is drawn off the bottom of the tank. A medium-mesh plastic screen retains the fingerlings in the rearing tank. During the last 6 weeks of the grow-out period, a complete diet (32% protein, 5% fat) with 4.8-mm floating pellets is used. The fish continue to be fed four times daily ad libitum for 30 min. A maximum feeding rate of 550 g/m3/day has been obtained with the fingerling production system.

An Integrated Fish and Field Crop System for Arid Areas

275

Water quality management Grow-out Continuous aeration and twice-daily solids removal are essential to the operation of the grow-out system. The aeration system has maintained DO levels in the range of 5±6 mg/liter at final densities of 17 kg/m3 and average daily feeding rates for treatments that ranged from 100 to 153 g/m3 in experiments 5 and 6. The maximum daily feeding rate exceeded 200 g/m3. This rate is less than the maximum feeding rate of 360 g/m3/day reported by Avnimelech (personal communication) and by Kahle (personal communication), who was president and general manager of Solar Aquafarms in Niland, California. However, their systems employed substantially more water exchange (Avnimelech) or aeration (Kahle). Diurnal measurements of DO during one day in experiment 4 showed that the DO concentrations reached a peak value (7.8 mg/liter) at 1200 hours and the lowest value (3.7 mg/liter) at 1800, 2000 and 2200 hours (Cole et al., 1997). This diurnal cycle does not occur during a period of algal die-off when the water turns brown in color. During these periods, the system depends totally on mechanical aeration, and the elevated biochemical oxygen demand (BOD) of decaying algal cells decreases DO concentrations temporarily until another algal bloom develops. The fish respond by consuming less feed during these periods. The system depends on continuous aeration. A disruption in aeration during the day does not lead to immediate stress in the fish. However, failure of the aeration system at night will lead to mortality. On two occasions, partial mortality occurred when the vertical-lift pump failed during the night even though diffused aeration via the 13 air stones continued. The clarifiers were very effective in removing solid waste and improving water quality. Over the course of experiment 4, for every 1000 g of feed input, based on dry weight, the clarifiers removed 357 g of total solids, 20 g of total nitrogen, 5.5 g of total phosphorus and 42 g of biochemical oxygen demand (BOD) (Cole et al., 1997). For each 1 kg of net fish production 33 liters of sludge were generated, although the sludge was quite dilute, averaging 1.1% dry weight of solids. Daily production of sludge averaged 65 liters or 0.23% of the rearing tank volume. In addition to feces and detritus, much of the sludge consisted of senescent algal cells, as indicated by its dark green appearance and a 37-fold increase in chlorophyll a concentrations in the sludge compared with the concentrations in the water column. In experiments 3 and 4, the treatments consisted of twice-daily sludge removal compared with no sludge removal. In each experiment, final fish size, growth rate and final biomass were all significantly higher when the sludge was removed (Table 9.1). However, sludge removal did not affect survival in either experiment or feed conversion ratio in experiment 3. The effect of sludge removal on water quality from experiment 4 is shown in Table 9.3. Sludge removal did not significantly affect concentrations of DO, total ammonia nitrogen (TAN) or nitrite nitrogen, variables that are critical to good fish production. Sludge removal did significantly decrease the concentrations of nitrate nitrogen, dissolved orthophosphate, chemical oxygen

276

Ecological Aquaculture

Table 9.3 Mean values of water quality parameters (mg/l) in greenwater tank systems with sludge removal twice daily and no sludge removal, and in the sludge removed from the tanks in experiment 4. The systems were used for the production of Nile tilapia (Oreochromis niloticus) over a 24-week period. Values in each row followed by the same letter are not significantly different (p>0.05)* Parameter Dissolved oxygen Total ammonia nitrogen Nitrite-nitrogen Nitrate-nitrogen Dissolved orthophosphate Chemical oxygen demand Total suspended solids Settleable solids (ml/l)

Sludge removal

No sludge removal

Sludge

6.7a 0.87a 2.69a 109.3a 7.9a 200.3a 168a 6.9a

6.3a 0.77a 3.29a 137.0b 17.3b 337.0b 492b 54.7b

± 1.61 6.10 89.1 10.5 793.8 10 957 644.1

* From Cole et al. (1997).

demand (COD), total suspended solids (TSS) and settleable solids (Fig. 9.13). It is possible that the high levels of suspended solids affected fish growth in the treatment where sludge was not removed daily. The final TSS concentration was 1250 mg/liter in the treatment without sludge removal compared with 368 mg/liter in the sludge removal treatment. The accumulation of organic matter and its decomposition may have produced more intermediate organic breakdown products, possibly decreasing fish growth. When sludge was removed daily, algal populations appeared to be healthier, as dead or clumped algal masses were quickly removed from the system and water clarity was better, which increased the depth of the photic zone. In the treatment without sludge removal, there was more nitrification, as indicated by higher nitrate-nitrogen levels, and more mineralization, as indicated by the high levels of dissolved orthophosphate. The effectiveness of the clarifiers can be assessed from their solids levels, which were 65 times higher for TSS and 93 times higher for settleable solids when compared with levels in the water column. Other means of improving water quality and fish growth were investigated. In experiment 2, in addition to daily sludge removal, daily water exchange of 5% was performed in one treatment. Water exchange did not affect any of the fish production parameters. However, final densities (6.5±6.8 kg/m3) in this experiment were well below the maximum density of 17 kg/m3. At higher densities and feeding rates, water exchange rates may have a positive impact on fish production by reducing excessive levels of nitrates and organic matter. In experiments 5 and 6, alum (aluminum sulfate) was tested as a means of improving water quality by flocculating and increasing the removal of suspended particulate organic matter. To counteract the acidic reaction of alum in water, 0.37 mg/liter of Ca(OH)2 were added for each 1.0 mg/liter of alum addition after total alkalinity decreased to less than 100 mg/liter as CaCO3. Sludge production and water clarity did increase after the addition of alum. Alum reduces phosphate, which forms a precipitate with aluminum. A reduction in phosphate can decrease algal growth if phosphorus becomes a limiting nutrient. In experiment 5 (unpublished data), which was a non-replicated preliminary trial to assess the effect of

An Integrated Fish and Field Crop System for Arid Areas

277

three levels of alum (per week) on tilapia production, it appeared that increasing levels of alum addition did exert a positive effect on growth rate, final size and final biomass (Table 9.1). As alum addition increased, there was an increase in daily sludge production (72, 78 and 106 liters/day of sludge at 25.8, 38.7 and 51.5 mg/liter/week of alum, respectively). Experiment 6 (unpublished data) employed replicated treatments to determine if weekly additions of 50 mg/liter of alum affect tilapia production. The results showed that there was no significant difference (P>0.05) in any of the production parameters (Table 9.1). Comparing the nonalum and alum treatments, there were no significant differences in daily sludge production (62 and 67 liters/day), TAN (2.7 and 3.8 mg/liter) and nitrite nitrogen (2.0 and 4.6 mg/liter), respectively. Nitrate nitrogen was significantly (P<0.05) higher in the non-alum treatment (156.9 mg/liter) than in the alum treatment (104.0 mg/ liter) while DO was significantly lower in the non-alum treatment (5.6 mg/liter) than in the alum treatment (6.5 mg/liter). It appears that the application of alum did increase the removal of suspended organic matter and its associated nitrifying bacteria, resulting in less nitrification and lower levels of nitrate nitrogen. By removing more organic matter and BOD, the alum treatment produced higher DO levels. Based on this experiment, the application of alum is not recommended as a water quality management technique for greenwater systems. Nitrite-nitrogen removal is less efficient with a suspended growth process than TAN removal. However, any harmful effects of nitrite nitrogen were mitigated by high levels of chloride ions (108 mg/liter) in the groundwater source in experiments 1± 4. To control nitrite toxicity a ratio of 6:1 as Cl to NO2 is required. In experiment 6, nitrite nitrogen reached a high of 23.2 mg/liter and it was necessary to increase the chloride concentration by adding 500 mg/liter of calcium chloride. Sodium chloride is not recommended because high sodium levels are harmful to the production of field crops, which are irrigated with the sludge. Biofiltration occurs through the association of nitrifying bacteria attached to suspended organic matter in the water column. As with any form of biological filtration, an acclimation period of 4±6 weeks is required to establish adequate levels of suspended organic matter and associated nitrifying bacteria. With new culture water, the initial feeding levels should be relatively low and ammonia and nitrite levels should be monitored frequently. Stocking 50-g fingerlings within the recommended rate results in a safe feeding level for biofilter establishment. Using recycled culture water with an established suspension of nitrifying bacteria is recommended. A suspended nitrification process is less stable than fixed-film nitrification. The concentration, composition and size of particulate organic matter may fluctuate, affecting the populations of nitrifying bacteria. As an example of the dynamics of suspended solids in greenwater system, clarifiers may yield only 8 liters of sludge daily for several weeks and then suddenly produce more than 400 liters of sludge daily without an apparent algal die-off. After the water column biofilter is established and nitrate levels begin to increase, pH values will decline, and frequent base addition is required to maintain pH at approximately 7.5. The pH is monitored daily and 500 g of Ca(OH)2 is added to the 30-m3 tanks whenever pH decreases to less than 7.5.

278

Ecological Aquaculture

Fingerlings The maximum feeding rate (550 g/m3/day) for the 11-m3 fingerling production tanks was approximately 2.75 times higher than the maximum feeding rate (200 g/m3/day) for the 30-m3 grow-out tanks, and therefore different water quality management procedures were needed. As previously described, the aeration system had to be upgraded from 13 to 25 air stones (9.65 cfm), which increased water mixing considerably over that of the grow-out tanks. Feed was administered four times daily instead of twice as in the grow-out tanks. Even with additional aeration, DO sometimes became limiting after feeding events, as indicated by the fish congregating near the air stones. Therefore, in the middle of trial 2, solids were removed four times daily after each feeding to reduce BOD. Finally, the suspended growth process was so biologically active that pH declined steadily throughout the day. It became necessary to add approximately 100 g of Ca(OH)2 after each feeding to prevent large pH fluctuations and their negative impact on the fish and nitrifying bacteria. Moreover, large base additions (e.g. 400 g of Ca(OH)2) were detrimental to the fish, especially if ammonia was high, because they created `hot zones', areas of very high pH and unionized ammonia, which caused the fish to jump, obviously injuring them and increasing mortality. In trial 2, the systems were filled with recycled greenwater from the grow-out tanks. Initial nitrate-nitrogen levels were approximately 300 mg/liter (Table 9.4). Although this was recycled water, it did not come from a heavily fed tank and populations of nitrifying bacteria were low. As a result, the initial feed rate was too high and some extremely high concentrations of ammonia nitrogen (39.8±91.0 mg/liter) and nitrite nitrogen (57.4±73.4 mg/liter) developed during weeks 3±6. This finding illustrates that recycled water must come from a heavily fed tank with an active population of nitrifying bacteria to circumvent the biofilter establishment period and feed at relatively high levels from the outset of fingerling production. Remarkably, there was no significant fish mortality during this period as chloride levels protected against nitrite toxicity and the pH was low enough (near 7.0) to limit the amount of TAN in the unionized toxic form. Later in the production period, the treatment stocked at 532 fish/ m3 developed extremely high concentrations (121.4 mg/liter) of nitrite nitrogen that exceeded the protective level of the chloride ions, resulting in more than 30% morTable 9.4 Mean values of water quality parameters (mg/l) and daily sludge production (l) in a greenwater fingerling production system (11 m3) during trial 2. The system was used to grow sex-reversed male Nile tilapia (Oreochromis niloticus) to an advanced size (>50 g) during a 12-week production cycle Parameter

177 fish/m3

355 fish/m3

532 fish/m3

Dissolved oxygen Total ammonia nitrogen Nitrite nitrogen Nitrate nitrogen* Sludge

6.6 11.6 16.0 333 (305) 42

4.4 17.1 27.9 324 (294) 49

3.6 30.6 43.7 344 (313) 48

* The systems were started with recycled greenwater. Initial values are in parentheses.

An Integrated Fish and Field Crop System for Arid Areas

279

tality, which was brought under control by adding more CaCl. This event prompted the procedural change whereby base was added frequently during the day with the goal of maintaining pH at 7.5. As a result, nitrite nitrogen decreased to 6.1 mg/liter within a week. The daily water exchange rate ranged from 0.71 to 1.0%. Sludge removal accounted for approximately half of the water exchange and averaged 0.42%. With such a small water exchange rate, dissolved organic matter increased and the intensive aeration system caused a large accumulation of foam in the center of the tank. In trial 3, a new procedure was initiated. Starting week 7, on the same day each week 25% of the system's water was exchanged to prevent nitrate build-up and the accumulation of dissolved organic matter. This procedure has greatly reduced foaming, substantially reduced nitrite levels and improved overall water quality (Table 9.5). However, removing 25% of the water also removes 25% of the biofilter, which has caused ammonia and nitrite values to increase moderately after the dilution. Although increased water exchange is beneficial in some respects, it may be better to apply it when the biofilter is functioning optimally. The amount of water exchange and the interval between water exchanges need to be investigated. Table 9.5 Mean values of water quality parameters (mg/l) and daily sludge production (l) in a greenwater fingerling production system (11 m3) during trial 3. The system was used to grow sex-reversed male Nile tilapia (Oreochromis niloticus) to an advanced size (>50 g) during a 12-week production cycle Parameter

300 fish/m3

400 fish/m3

500 fish/m3

Dissolved oxygen Total ammonia nitrogen Nitrite nitrogen Nitrate nitrogen Sludge

6.0 3.7 19.7 195 50

5.0 2.1 16.6 188 54

5.1 6.7 24.3 175 51

In trial 3 another important consideration became apparent. It is possible that during periods when sludge production is high and sedimentation of organic matter is optimal for some unknown reason, the sludge removed from the system represents a good portion of the biofilter and water quality may deteriorate. This concept requires further study.

Water consumption Total water consumption was 0.15 m3/kg of tilapia production in experiment 6 where a final standing crop of 17.0 kg/m3 was obtained in the non-alum treatment. In the `Dekel' system of Israel, tilapia are intensively cultured in 500-m2 concrete tanks aerated with two or three 1.5-hp paddlewheel aerators and the water is recycled through a 1.2-ha earthen pond stocked with carp at 1 fish/m2. This system achieves total yields (tilapia and carp) of 2.4 kg/m2 and consumes 1.2 m3 of water per kg of

280

Ecological Aquaculture

production (Mires et al., 1990). A pond that produces 5 mt/ha consumes approximately 2±4 m3 of water per kg of production, depending on mean depth and evaporation and seepage rates. In experiment 5, the daily water exchange (make-up) rate was 0.58%, a rate that compares favorably with rates of <0.5% reported by Schroeder & Serfling (1989) for closed greenwater systems at Solar Aquafarms and 0.1±2.4% for the greenwater tank systems of Louisiana (Lutz, personal communication).

Integrated field crop production Experiments 1 and 2 were integrated with the field production of bell peppers planted at a density of 36 470 plants/ha. In experiment 1, bell peppers (Capsicum annuum) were fertilized with water from the low stocking density, the high stocking density and the sludge. There were additional fertilizer treatments consisting of liquid inorganic fertilizer, applied manually and by drip irrigation (fertigation), granular inorganic fertilizer and cow manure (Palada et al., 1999a). The treatments were arranged in a randomized complete block design with four replications. The inorganic fertilizer treatments were fertilized at a rate of 200 kg/ha with N and 100 kg/ha with P and K. Fish culture water and sludge were applied 10 times at weekly intervals throughout the experiment at a rate of 3.8 liters/plant per application, which was equivalent to 138 m3/ha. The liquid inorganic fertilizer was applied seven times. All treatments received the same amount of irrigation water. The total fruit yield (10.6 mt/ha) from the fertigation treatment was significantly higher (P<0.05) than that from all other treatments. There were no significant differences between the other treatments, but production appeared to increase in direct relationship to the strength of the aquaculture effluent: 2.82 mt/ha for low stocking density water; 4.12 mt/ha for high stocking density water; and 6.62 mt/ha for sludge. Based on mean nutrient concentrations, application of 1 cm-ha of low density rearing tank water would result in the addition of 1.73 kg/ha of inorganic N and 0.34 kg/ha of total P, while the high density rearing tank water would result in the addition of 3.62 kg/ha of inorganic N and 0.46 kg/ha of total P. The addition of nitrogen from organic sources, although substantial, was not determined. In experiment 2, the treatments were fish culture water with no daily water exchange and 5% daily water exchange, both applied with two types of drip irrigation emitters (Turbo Key and Hardie), sludge applied by drip irrigation using a Bow Smith Gripper adapter emitter, liquid inorganic fertilizer applied by drip irrigation and granular inorganic fertilizer applied in bands around individual plants. Inorganic fertilization rates were the same as in experiment 1. Fish culture water and sludge were applied once or twice a week for a total of 25 applications over a 3-month growing period. Each application was equivalent to 88±113 m3/ha. All treatments received the same amount of irrigation water. Total fruit yield ranged from 10.5 mt/ ha for fertigation to 15.1 mt/ha for sludge. Total yield from the sludge was significantly (P<0.05) higher than that from fertigation but not significantly different from any of the other treatments. The total yields with fish culture water ranged from

An Integrated Fish and Field Crop System for Arid Areas

281

Fig. 9.14 Water samples in Imhof cones collected from greenwater grow-out tank showing the effects of sludge removal and no sludge removal. (Photograph by W.M. Cole.)

11.3 to 12.98 mt/ha. Based on mean nutrient concentrations, respective applications of 1 cm-ha from 0% water exchange tanks, 5% water exchange tanks and sludge would result in the addition of: 6.07, 3.01 and 3.53 kg/ha of inorganic N, 0.94, 0.28 and 7.34 kg/ha of total P; and 5.94, 1.75 and 5.80 kg/ha of K. The contribution of nitrogen from organic sources was not determined. This study showed that it is possible to grow vegetable crops using effluents from intensive tilapia culture in tanks without external fertilizer inputs. Yields can be maintained at levels comparable to yields using commercial fertilizers. Re-using aquacultural effluent to irrigate vegetable crops and recycling the nutrients in the effluent, thereby eliminating the need for inorganic fertilizers, would increase profits over that of non-integrated operations for fish and vegetable production in arid areas. Aquacultural effluents from experiment 4 were used to irrigate and fertilize two crops of pak choi (Brassica rapa) using three methods of drip irrigation (Palada et al., 1999b). The treatments were: (1±2) fish culture water from the rearing tanks with solids removal and without solids removal using Hardie E-2 emitters; (3) sludge using a micro-tubing adapter; (4) sludge using a 0.5-cm opening on 1.27-cm polyhose; (5) culture water from the solids removal tanks plus sludge once a week with the polyhose; (6) fertigation at 100 kg/ha of N with Hardie E-2 emitters; and (7) granular fertilizer applied as a band at 100 kg/ha of N plus Hardie E-2 emitters solely for irrigation. Since all treatments were given a basal application of P and K, each at 50 kg/ha, this experiment mainly tested the effects of nitrogen application from the treatments on plant growth. In the first crop, production was significantly higher (P<0.05) in the three sludge treatments and fertigation, ranging from 22.1 to 26.1 mt/ ha, with the culture water + sludge treatment being the highest (Fig. 9.15). In the second crop, there were no significant differences in production (40.9±47.4 kg/ha) between fertigation and any of the aquacultural effluent treatments. Production (35.2 kg/ha) was significantly lower (P<0.05) in most cases with granular fertilizer. The highest production occurred again with the culture water + sludge treatment. This experiment showed that the nitrogen, obtained from aquacultural effluent, can produce vegetable yields comparable to or better than that obtained by using inorganic nitrogen from fertilizers.

282

Ecological Aquaculture

Fig. 9.15 Pak choi irrigated with greenwater sludge. (Photograph by J.E. Rakocy.)

An experiment was conducted to determine the effects of greenwater aquacultural effluent from UVI's system on the production of guinea grass (Panicum maximum) managed for hay (Valencia et al., 2001). The experiment was conducted over a 3-year period during the dry season. Aquacultural effluent (diluted sludge) was applied at a total rate of 120, 0 and 60 kg N/ha during an 18-week period during years 1, 2 and 3, respectively, and compared with inorganic fertilizer (ammonium nitrate), applied all 3 years at a total rate of 60 kg/ha/year split into two applications each year, and a control without fertilization (Fig. 9.16). The aquacultural effluent was applied weekly by sprinkling using an irrigation rate of 2 cm-ha per week. The inorganic nitrogen treatment and the control were irrigated at the same rate using well water. Each year there were three forage harvests at 8-week intervals. Total dry matter yield for the 3 years was 14.2, 13.8 and 12.9 mt/ha for the aquacultural effluent, 13.1, 9.9 and 6.4 mt/ ha for inorganic fertilizer and 8.6, 5.7 and 2.6 mt/ha for the control, respectively. Dry matter yield using aquacultural effluent was significantly higher (P<0.05) than that of the inorganic fertilizer treatment and the control in years 2 and 3. The dry matter yield of the control was significantly lower in year 1 than that of the other treatments.

Fig. 9.16 Forage grass plots. Greenwater sludge and rearing tank water improved yield in treatment plots over that of control and inorganic fertilizer plots. (Photograph by J.E. Rakocy.)

An Integrated Fish and Field Crop System for Arid Areas

283

Aquacultural effluent produced better quality hay. The crude protein percentage in years 1 and 3 was 9.6 and 10.3% in the aquacultural effluent treatment compared with 8.6 and 8.3% in the inorganic fertilizer treatment, respectively. The results of this study show that aquacultural effluent is superior to inorganic fertilizer for hay production. Based on this study, the recommended rate for applying aquacultural effluent for grasses such as guinea grass is 60 kg N/ha/year in the seasonally dry tropics. In year 2, the slow release (mineralization) of nitrogen from the residual organic matter of the sludge, applied in year 1, produced higher hay yields than that of the inorganic fertilizer applied in year 2. Due to the slow release of nitrogen from the aquacultural effluent, there is less leachate of nitrogen in the system. Inorganic fertilizers produce rapid forage growth for a short period but do not have a longlasting effect on production and are subject to leaching if a heavy rainfall occurs. Guinea grass hay fields can serve as a nutrient sink for the wastewater from greenwater tank culture operations and provide an environmentally acceptable, profitable and simple method for effluent disposal. Researchers have documented the benefits of aquacultural sludge obtained from other fish species and production systems. Willet & Jacobsen (1986) found that sludge dredged from concrete-lined rainbow trout ponds had moderate neutralizing capacity in strongly acidic soils. Using the sludge deposited under rainbow trout cages, Mazzarino et al. (1997) found the sludge to be superior to traditional inorganic fertilizer regimes for growing ryegrass. Olsen (1992) found aquacultural sludge from trout production raceways to perform as well as commercial fertilizer in the production of sweet corn and spring wheat. Their results depended on the method and rate of application of the sludge. Aquacultural sludge contains no weed seeds, a common problem with manure (Olsen, 1992). There are concerns with the application of aquacultural sludge (Rakocy, 1995). Fish culture sludge stored under anaerobic conditions is malodorous, and liquid sludge often forms a crusty surface if it is not quickly incorporated into the soil (Olsen, 1992). Transporting and applying the sludge uniformly is expensive and difficult. The exact chemical composition and fertilization value of sludge is variable and depends on many factors including fish species, feed quality, culture method, and condition and length of storage (Westerman et al., 1993). Some of these concerns are circumvented if the sludge is used immediately in an adjacent field crop operation, simulated by the UVI experiments. However, applying greenwater tank culture sludge through a drip irrigation system to conserve water in dry areas poses a risk of clogging, even when large-orifice emitters are used.

Future research Additional research will be conducted to determine the optimum stocking rates for the desired final size during a 25-week grow-out period. The length of the culture period for fingerlings will be reassessed. A reduction in the culture period from 12 to 10 weeks would reduce management costs by 17%. The maximum sustainable feeding rate with minimal risk of water quality deterioration needs to be determined. Methods

284

Ecological Aquaculture

should be found to stabilize feeding response and reduce the variability in the population density and performance of nitrifying bacteria. Additional work is needed on determining optimal sludge application rates and delivery methods for a variety of tropical field crops. A major goal is to scale up the grow-out tank to a commercial size for small islands. Work is nearly complete on the construction of a 200-m3 tank that is 16 m in diameter and 1 m in average depth. The tank is being built with a concrete block wall and a high density polyethylene liner that is 0.76 mm (30 mils) thick. A 1-m3 clarifier with a 458 slope is being incorporated into the center of the tank floor, which is sloped at 38 from the wall to the center clarifier. Four 0.75-hp vertical-lift pumps will be employed to provide aeration. One of these pumps will be tilted to create a circular flow pattern to move settleable solids to the clarifier. Solids will be removed by opening an external standpipe. If two crops per year can be produced at a final density of 17.0 kg/m3, this tank will produce 6.8 mt of tilapia annually. This projection needs to be confirmed and methods need to be found to increase production. Another project involving greenwater tank culture is beginning. A 2-ha model integrated farm is being established. It will produce tilapia, livestock, fruit and vegetables. All water for the farm will be provided by a 4000-m2 rainwater catchment and stored in a 500-m3 lined pond that is 3 m deep. There will be seven 80-m3 fiberglass rearing tanks, and production will be staggered so that one tank is ready for harvesting each month for a period of a month so that there will be a continuous supply of tilapia. Total production should be 16.4 mt annually. The aquacultural sludge will be stored in a 150-m3 lined pond. The rainwater catchment and pond liners will consist of 0.76 mm high-density polyethylene material. The sludge will be used to irrigate and fertilize 4000 m2 of fruit crops and 4000 m2 of vegetable crops. The fruit and vegetable plots will be rotated to reduce pest pressure. During fallow periods, cover crops or St Croix white hair sheep will be produced. All plant production from the farm will be irrigated and fertilized by aquacultural effluent. The benefits of integration will be documented and an economic analysis will be performed. This model farm project will determine the economic feasibility of integrated fish and field crop production on a small commercial scale. As populations expand, weather patterns change and resources diminish, there will be greater pressure to find alternative food production systems that conserve and recycle water and nutrients. The integration of greenwater tank culture with field crop production is just one example of the types of systems that will be needed in the future.

References Avnimelech, Y. (1998) Minimal discharge from intensive fish ponds. World Aquaculture, 29, 32±37. Avnimelech, Y., Kochva, M. & Diab, S. (1994) Development of controlled intensive aquaculture systems with limited water exchange and adjusted carbon to nitrogen ratio. Israeli Journal of Aquaculture, 46, 119±131.

An Integrated Fish and Field Crop System for Arid Areas

285

Cole, W.M., Rakocy, J.E., Shultz, K.A. & Bailey, D.S. (1997) Effects of solids removal on tilapia production and water quality in continuously-aerated, outdoor tanks. In: Tilapia Aquaculture: Proceedings of the Fourth International Symposium on Tilapia in Aquaculture (ed. K. Fitzsimmons), pp. 373±384. Northeast Regional Agriculture Extension Service Publication 106. Cooperative Extension, Ithaca, New York. Kochba, M.S., Diab, S. & Avnimelech, Y. (1994) Nitrogen modeling transformation in intensively aerated ponds. Aquaculture, 120, 95±104. Lutz, C.G. (1998) Greenhouse Tilapia Production in Louisiana. Louisiana State University Agricultural Center, Publication No. 2705, Baton Rouge, Louisiana. Mazzarino, M.J., Walter, I., Costa, G., Laos, F., Roselli, L. & Satti, P. (1997) Plant response to fish farming wastes in volcanic soils. Journal of Environmental Quality, 26, 522±528. Mires, D., Amit, Y., Avnimelech, Y., Diab, S. & Cochaba, M. (1990) Water quality in a recycled intensive fish culture system under field conditions. Bamidgeh, 42 (4), 110-121. Olsen, G.L. (1992) The use of agricultural by-products as a fertilizer for Idaho crops. University of Wisconsin Sea Grant Technical Report No. WISCU-W-91-001, Madison, Wisconsin. Palada, M.C., Cole, W.M. & Crossman, S.M.A. (1999a) Influence of effluents from intensive aquaculture and sludge on growth and yield of bell peppers. Journal of Sustainable Agriculture, 14, 85±103. Palada, M.C., Crossman, S.M.A., Cole, W.M. & Kowalski, J.A. (1999b) Application of recycled aquaculture water for production of leafy vegetables. In: Proceedings of the Twentyeighth National Agricultural Plastics Congress (ed. K.D. Batal), pp. 10±15. American Society for Plasticulture, State College, Pennsylvania. Rakocy, J.E. (1995) The role of plant crop production in aquacultural waste management. In: Aquacultural Engineering and Waste Management (ed. M.B. Timmons), pp. 349±364. Northeast Regional Agricultural Engineering Service Publication 90. Cooperative Extension, Ithaca, New York. Schroeder, G.L. & Serfling, S. (1989) High-yield aquaculture using low-cost feed and waste recycling methods. American Journal of Alternative Agriculture, 4, 71±74. Valencia, E., Adjei, M. & Martin, J. (2001) Aquaculture effluent as a water and nutrient source for hay production in the seasonally dry tropics. Soil Science and Plant Analysis, 32, 1293±1301. Westerman, P.W., Hinshaw, J.M. & Barker, J.C. (1993) Trout manure characterization and nitrogen mineralization rate. In: Proceedings of the Aquacultural Engineering Conference on Techniques for Modern Aquaculture (ed. J.K. Wang), pp. 35±43. American Society of Agricultural Engineers, St. Joseph, Michigan. Willet, I.R. & Jakobsen, P. (1986) Fertilizing properties of trout farm waste. Aquacultural Wastes, 17, 7±13.

Chapter 10

Sustainability of Cage Aquaculture Ecosystems for Large-scale Resettlement from Hydropower Dams: An Indonesian Case Study Barry A. Costa-Pierce Rhode Island Sea Grant College Program Introduction Due to increased demands for reliable supplies of electric power, irrigation, and drinking water, the number of new hydropower reservoirs is increasing dramatically, especially in Asia. According to the last report of the International Commission on Large Dams, the total number of dams on Earth grew from about 5000 in 1950 to more than 40 000 in 1986, with China the home to about 50% of these (McCully, 1996). Dams continue to be one of the only means to increase humanity's access to more of the Earth's runoff for new cities and expansion of irrigated agriculture, but they are increasingly expensive socially, environmentally, and economically (Postel et al., 1996). For example, in arid, rapidly urbanizing southern California, USA, the new Eastside Reservoir will flood just 1800 ha but will cost some $2 billion. In addition, there is a trend towards construction of larger dams having greater combined costs. Construction of dams with elevations greater than 100 m rose by 27% between 1991 and 1993. More than half of these dams were in China, India, and Turkey (Gardner & Perry, 1995). Large hydropower reservoirs have caused massive social disruption, increased incidences of water-borne diseases, erosion, and other social and environmental degradation (Petr, 1978; Hunter et al., 1983; Lelek, 1984; Cernea, 1988, 1997; McCully, 1996) (Fig. 10.1). There is a need to develop new, more sustainable environmental planning and policy approaches that integrate social and ecological concerns in hydropower projects worldwide. These social ecological approaches would formulate and carry out long-term rehabilitation efforts with rural societies to restore damaged aquatic environments from hydropower projects. New approaches are especially needed in densely populated areas of Asia where the pace of dam construction is accelerating (Sutandar et al., 1990; Alam et al., 1995; Cernea, 1997; Costa-Pierce, 1997). It has been estimated that annual inland fish production in Asia is 5.5 million metric tons (mt), comprising 57% of the world's inland fish production (De Silva, 1988, 1992). However, fish yields from Asian reservoirs comprise just 0.5 million mt of this 5.5 million mt (De Silva, 1988). DeSilva (1988) estimated fish production from Asian reservoirs at only 20 kg/ha/year, with a wide variability in production that was

Aquaculture Ecosystems for Resettlement from Hydropower Dams

287

Dam and reservoir Earth crustal system Earthquakes

Atmospheric system Local weather and climates

Hydrological system Groundwater Sedimentation Erosion Limnological conditions Water balance Water quality Terrestrial biological system Cultivable land Cattle and other livestock Big game and other useful species Rare or endangered species Nature preserves and game parks Forest harvest and clearance

Aquatic biological system Substrate for aquatic production Useful aquatic plants Nuisance aquatic plants Useful aquatic animals Rare endangered species Intentionally introduced species

Human system Resettlement Public health Fishing Hunting Agriculture

Livestock production Manufacturing Service industries Commerce Tourism and recreation

Urbanism Transport Education Historical continuity Equity

Social justice Income Employment Nutrition

Fig. 10.1 Social and environmental impacts of hydropower development on a tropical watershed and landscape ecosystem (Costa-Pierce, 1997).

not always related to the size of the reservoir. Costa-Pierce & Soemarwoto (1987) calculated that an average percentage increase in reservoir area in 15 Asian nations from 1987 to 2000 would be 511%, ranging from 50% (Singapore) to 9900% (Laos). By 2000 the collective water surface in reservoirs (20.3 million ha) exceeded the surface area of Asia's natural waters (18.5 million ha). Clearly, if the huge expanse of underutilized water areas locked behind Asia's dams could be utilized for increased fish production, thousands of tons of new aquatic protein could enter Asian markets. Production of new aquatic protein is especially urgent in Asia, a region where fish is the most important source of animal protein. In addition, there is a need to create thousands of new rural jobs as a result of population growth and to find innovative ways to stem the rapid rural to urban migration. It is argued that expansion of aquatic food production in Asian reservoirs could assist in mitigating Asia's growing food and population crises (Brown, 1997). The planned development, enhancement, and management of capture (`fishing') and culture fisheries (`aquaculture') enterprises in new reservoirs as alternative livelihoods for the people displaced by dam construction has received little or no

288

Ecological Aquaculture

attention. The policies for the technical, social, environmental and economic incorporation of fisheries into the planning for dams or project site works have not been done. Fisheries has not been incorporated into the social and environmental compensation packages of hydropower dam projects (Cernea, 1997). Hydropower projects are usually controlled by dam engineers and government officials more concerned with moving people away from the new water bodies, rather than promoting new forms of intimate contact with them (such as the Indonesian transmigration projects; Fearnside, 1997). An alternative view is to develop policies and planning structures to resettle the displaced people locally and encourage development and evolution of new, community based, social±cultural interactions with the new aquatic ecosystems in order to rehabilitate the damaged social and environmental situations. The evolution of sustainable, ecological aquaculture holds this kind of rehabilitation potential. Most of the world's current population growth is occurring in nations where there is little potential for increasing the area of arable land under cultivation (Engleman & LeRoy, 1995; Brown, 1997; Pimental et al., 1997). States Engleman (1995), `The food that will be required to feed a world population of 8 billion or more in the next century will have to come almost entirely from today's farmland.' The rapid rate of urbanization in most Asian nations is causing severe population pressure on existing rural and peri-urban agricultural ecosystems. Java is one of the world's most densely populated areas of the world, and the island is losing agricultural lands to urbanization at an alarming rate. Between 1983 and 1992, new housing starts in just three cities in West Java (Bekasi, Bogor, Tangerang) ate up 61 000 ha of croplands (Firman & Dharmapatni, 1994). The USDA estimated that urban expansion claimed 20 000 ha on Java in 1994 alone (Thompson, 1995). Rapid urbanization has increased the need to preserve and intensify agricultural production on the remaining agricultural areas, and to find innovative ways to conduct intensification sustainably without further environmental and social damage. Densely populated Java and southern China have been two ancient centers of farmer innovations having numerous sustainable examples of productive, ecologically sophisticated backyard and small farm ecosystems that merge agriculture and aquaculture in unique ways (Ruddle & Zhong, 1988; De la Cruz et al., 1992; Koesoemadinata & Costa-Pierce, 1992). However, because of urbanization and associated population pressures, the era of consistent increases in the numbers of these sophisticated traditional aquaculture agroecosystems in Asia may be coming to a close, principally due to their land-intensive nature and their proximity to the main areas of urban sprawl. In the peri-urban fringes of many large Asian megacities (Jakarta, Bangkok, Manila, etc.) there is an alarming loss of these traditional agroecosystems and the indigenous knowledge systems that have developed over hundreds of years to manage them. In contrast, there are vast areas of new inland waters `locked up' in hydropower and irrigation reservoirs in the region. The surface waters of these hydropower reservoirs are almost completely vacant of any significant productive enterprise, other than being used for water storage, and subsistence level fishing that provides little other than part-time incomes (Munro et al., 1990). Nearly all Asian reservoirs outside

Aquaculture Ecosystems for Resettlement from Hydropower Dams

289

of China and Thailand have little or no water-based aquaculture systems such as fish cages, and have underdeveloped capture fisheries management programs (De Silva, 1988, 1992; Costa-Pierce, 1997). Development of aquaculture in and around Asia's hydropower and irrigation reservoirs to enhance fish production, and as a management tool to enhance capture fisheries, may be one means to provide thousands of new jobs in rural areas. Fisheries development may also be the only means left for creating new sources of freshwater aquatic protein for many densely populated Asian nations (De Silva, 1988; Costa-Pierce, 1997). Indeed, hydropower and irrigation reservoirs may be Asia's final `aquatic frontier'.

The case study of resettlement aquaculture in Indonesia A cooperative Institute of Ecology, Padjadjaran University (IOE, Bandung, Indonesia) and International Center for Living Aquatic Resources Management (ICLARM, Manila, Philippines) project in the highlands of West Java, Indonesia, was initiated to investigate the feasibility of developing cage aquaculture for local population resettlement from hydropower dam construction. Indonesia was chosen for the project because of the need to evolve alternative resettlement schemes to the transmigration program, which had caused some social and environmental problems and was costly (Fearnside, 1997). The use of floating cage aquaculture (FCA) to resettle 3000 of the 40 000 displaced families locally was proposed since FCA was deemed compatible with engineering forecasts of dam operations and drawdowns (Costa-Pierce & Soemarwoto, 1990). The purpose of this paper is to describe the initial institutional framework, planning, and policy efforts, and extension and training programs conducted during the project; and to discuss the reasons for project success, failures, and long-term constraints to environmental and social sustainability in light of the possible transfer of this model to other (especially Asian) nations with similar aquatic resources, and social and economic development scenarios. Details of the project's technical, aquaculture, fisheries, and economic aspects have been reported elsewhere extensively (Costa-Pierce & Soemarwoto, 1990; Sutandar et al., 1990; Costa-Pierce, 1997), and will be mentioned herein only for their relevance to the discussion of social, institutional and environmental sustainability and transfer of technologies.

Institutional framework and planning process Funding for the resettlement aquaculture efforts was obtained from the World Bank as a small portion of a larger loan package to the Indonesian State Electric Company (PLN) for dam construction at Cirata. The aquaculture development portion of the World Bank loan was divided into local (Rp 917 441 530) and foreign currency (US$416 690) components and administered by IOE (local) and ICLARM (foreign) agencies for project implementation (at project initiation in January 1986, 1 US$ = 1130 Indonesian Rp (IDR); in September 1986 the IDR was devalued to 1640).

290

Ecological Aquaculture

The primary objectives of the project were to provide: (1)

(2)

Technical support: applied ecological, capture fisheries, aquaculture research, extension, training, and other support services to the West Java Provincial Fisheries Agency (Bandung, Indonesia) in order to facilitate rapid resettlement of 3000 families in either full- or part-time jobs in reservoir aquaculture, capture fisheries, or related support industries in the two hydropower reservoir areas: the Saguling reservoir (1500 families) and Cirata (1500 families) reservoir areas. Planning: a comprehensive reservoir fisheries and aquaculture development and management plan for population resettlement.

Despite the relatively large information base available in freshwater aquaculture development and capture fisheries management in Asia, there were no previous experiences anywhere involving the resettlement of such a large number of persons through a planned development program of reservoir cage aquaculture, and its allied land-based aquaculture support systems (such as hatcheries and nurseries). For this reason, The World Bank suggested that IOE associate with an international fisheries research organization (ICLARM) to assist with technical advice in fisheries and project implementation. A new institutional framework had to be formulated for project implementation because loan documents were inadequate in defining an overall administrative structure for the project and in delegating individual institutional responsibilities. The IOE was chosen as the lead agency to coordinate a multidisciplinary applied research and training program that involved agricultural economists, sociologists, demographers, anthropologists, agronomists, forestry, animal husbandry, capture fisheries, and aquaculture professionals. To assist IOE's applied research program, ICLARM provided the services of a full-time resident consultant fisheries scientist. ICLARM also provided specific short-term scientific expertise in capture fisheries, aquaculture, and fisheries marketing and economics. The Indonesian government's West Java Provincial Fisheries Agency created a special reservoir fisheries technical implementation unit that established offices at both the Saguling and Cirata dam sites to lead the local extension and training efforts. IOE also hired fisheries extension personnel on three-year contracts to facilitate more rapid transfer of applied research results to villagers and to coordinate extension activities with the government's technical implementation unit. The Indonesian State Electric Company (PLN) provided funding, institutional coordination, and unlimited access to the reservoir areas for research and development activities, since, after compensation monies were paid, the reservoirs and their drawdown areas were PLN's `private property'. Upon delegation of responsibilities, a planning process was undertaken by IOE and ICLARM to apportion the large workload into teams in order to implement the diverse number of applied research activities to be conducted with the displaced people (Table 10.1). Geographic areas where applied research was to be conducted were chosen in new or expanded villages having the largest numbers of displaced residents (taken from the electric company's compensation reports). Applied research was done in a participatory manner with villagers in capture

Aquaculture Ecosystems for Resettlement from Hydropower Dams

291

fisheries, aquaculture, drawdown agriculture, sociology, fisheries economics and marketing. A number of different types of low cost, water-based, cage aquaculture systems were developed and tested (Costa-Pierce & Hadikusumah, 1990, 1995). Fish cage capital costs ranged from Rp 491 000 (US$299) for cages having recycled oil drums and wood frames for flotation, to Rp 177 500±274 500 ($108±$167) for cages having bamboo or banana logs for flotation. Low cost `mini' cages (17.3 m3) were also developed that cost between $20 and $70 to construct (Costa-Pierce & Hadikusumah, 1990). In addition, a suite of low-cost land-based hatchery and nursery aquaculture systems were developed. In contrast to the report of Zerner (1992), aquaculture systems developed were simple and not elaborate and were well within the financial means of the majority of the displaced people (Costa-Pierce & Hadikusumah, 1990; Rusydi & Lampe, 1990). Technological interventions in land-based aquaculture were based upon existing models of traditional knowledge systems in integrated agriculture±aquaculture farming ecosystems present for over a hundred years in West Java (Djajadiredja et al., 1980; Thornburn, 1982). Cage aquaculture technologies were based upon previous Indonesian and overseas experiences (Gonzales, 1984; Rifai, 1985; Pullin, 1986; Hai & Zweig, 1987; Costa-Pierce et al., 1988, 1989a, b, c, d, e). Development of the landbased and water-based aquaculture systems was initiated in an attempt to integrate the new, water-based reservoir cage systems into an existing social ecological system of traditional pond hatchery and rice±fish nursery systems, thereby creating as many new jobs as possible by using aquaculture's `multiplier effects' (Costa-Pierce, 1992, 1997; Koesoemadinata & Costa-Pierce, 1992).

Farming systems extension and training programs Two extension approaches were used to disseminate to displaced persons aquaculture technologies. For new `water farmers' totally unfamiliar with traditional aquaculture technologies in West Java, adoption/diffusion techniques (Pollnac, 1982) were chosen. For farmers who knew about traditional cage aquaculture, or land-based aquaculture techniques, a farming systems research and extension approach was used (Chambers et al., 1989). This division of extension efforts was not a hard and fast categorization, however. It was left up to farmers themselves to choose which training sessions they wanted to attend. Many farmers attended sessions using both extension techniques.

Traditional training methods In 1982, three years before the first reservoir was to be flooded (Saguling was flooded in 1985±86), formal classroom and hands-on training in cage aquaculture was conducted with a small group of selected residents (mostly village leaders) that were to be displaced. Hands-on practical training in cage aquaculture was conducted in a small, shallow lake in Bogor (Lake Lido), and in an existing, downstream reservoir (Jatiluhur, Indonesia). In addition, a special `fisheries dike area' was created by the State Electric

292

Step 2

Step 1 Synthesis of project agreements and goals

Preliminary surveys and data gatherings in three sub-program areas of the project:

Formal and informal meetings with cooperating agencies

Aquaculture development:

Capture fisheries development and management:

Sociocultural, economics and marketing:

Assignment of project staff into three subprograms: . aquaculture development . capture fisheries development and management

. information on existing traditional and modern systems and knowledge base; production; feeds and feeding practices; systems design, economics, organization and

. broad environmental surveys of fisheries ecosystems and changes occurring in them, port/ harbors; industrial developments; valueadded products; postharvest traditional

. data on population sizes; income; fish consumption patterns; forms/sizes of fish eaten; cultural reactions to fish products; land use patterns; agricultural interfaces;

Step 3

Step 4

Summary report on findings in each subprogram area in chapters comprising draft implementation plans for: . aquaculture development; . capture fisheries development and management; . sociocultural, economics and marketing.

Complete synthesis of three sub-program plans for a single implementation plan through development Implementation plan is subject of seminar of invited experts and is revised according to changes and recommendations Final implementation plan for fisheries and aquaculture development as a means of large-scale resettlement

Ecological Aquaculture

Table 10.1 Planning process used to complete an implementation plan for applied research in integrated reservoir fisheries and aquaculture development

. sociocultural, economics and marketing

techniques; . main hydromorphometric, biological, limnological, climatological factors; . information on past/ present fishing efforts; species; yields; gear used; catch per unit efforts; labor; routines; transportation costs/ patterns; . fishermen's impression of new environments, and effects on new ecosystems from demographic, local, pollution, market, political changes.

.

. .

.

land tenure; rapid rural appraisals; adaptations to new reservoirs; special environment/people interactions; resource use patterns; exploitation/mining activities; constraints and limiting factors to resource use and conservation; availability of funds; knowledge of loans, credit; use of banks; familiarity with the concepts of loans/ credit; analyses of fish markets; market structures; distribution and transportation systems; marketing costs; middlemen; market supply statistics from all administrative levels; market price fluctuations with peak supply; market capability projections.

Aquaculture Ecosystems for Resettlement from Hydropower Dams

interactions; constraints; past performance; species; rationales for use; labor use; success/failures; . secondary data from national, provincial, regional, city, village fisheries institutions, universities, private operators, regarding specific systems likely be of use in reservoirs; . interviews with farmers, middlemen, trips to markets, infrastructure, daily/weekly routines; problems with materials; labor; environment; technology; neighbor relations.

293

294

Ecological Aquaculture

Company in the Saguling Reservoir for cage aquaculture experiments before the flooding of the main part of the reservoir (Rifai, 1985). Aquaculture training was conducted for 24 displaced persons by the Research Institute for Freshwater Fisheries, Bogor and the Department of Fisheries, Padjadjaran University in this diked area. Two displaced farmers from this latter group subsequently received loans from the Bank of Indonesia to develop cage aquaculture businesses after initial experimental results showed the potentials for high yields and profits (Rifai, 1985). The two pioneer farmers successfully paid back loans from the Bank of Indonesia by just the second year of operation. Their positive results attracted a great deal of community interest. At the same time, the State Electric Company facilitated a highly publicized visit by the Governor of West Java to these two farmers. These preinundation events created a widespread awareness of the potential of cage aquaculture among Saguling's displaced people, and the two successful farmers later became kader-kader petani ikan (fish farmer facilitators). Further experimental work from 1986 to 1990 developed a `basket' of low-cost technologies in cage aquaculture, small-scale (`backyard') fish hatcheries, waterbased hatcheries, and integrated rice±fish nursery systems appropriate to the rural expertise, availability of capital and construction materials, and management complexity of displaced farmers (Costa-Pierce & Hadikusumah, 1990, 1995; Costa-Pierce, 1992, 1997). All research employed displaced villagers (the beneficiaries for whom the technologies were intended) who worked with scientists from the outset. Simultaneously the West Java Provincial Fisheries Agency and IOE/ICLARM collaborated to offer a number of short courses of 1 week to 3 months duration for villagers at offices of village headmen. These courses were held in over 20 districts in the Cirata region and were attended by over 500 persons. Courses covered operations of numerous types of land- and water-based aquaculture systems having high and low capital costs (e.g. intensive raceway systems, hatcheries, small and large cages, pen systems, rice±fish culture), plus instruction to villagers on how to formulate fish feeds, and process and market fish (Table 10.2). A Saguling Fish Farmers' Association (SFFA) was formed in late 1985, and by end 1989 had over 700 members. Leadership of the SFFA was by the two pioneers who successfully repaid their loans (subsequently, these two and their families became the most powerful members of the fish farming community). The SFFA was formed by the Technical Implementation Unit of the West Java Provincial Fisheries Agency who also assisted farmers with obtaining bank loans and marketing fish. In Cirata, a government Village Cooperative Unit (Korporasi Unit Desa, or KUD) took the lead in cage aquaculture development with assistance from the Government's Technical Implementation Unit. In 1989, the cooperative obtained a government loan package in excess of Rp100 million to develop cage aquaculture in Cirata.

Farmer participatory training methods A farming systems research and extension approach was chosen since it was known that West Java had a unique cultural heritage and a large bank of traditional

Aquaculture Ecosystems for Resettlement from Hydropower Dams

295

Table 10.2 Aquaculture systems subject areas covered in formal aquaculture training courses offered by the aquaculture resettlement project in impacted villages Hatcheries (1) Site selection (2) Designs and construction (3) Brood stock selection, management, feeds/ feeding (4) Spawning, egg collectors (5) Nursery pond management (6) Water quality and natural feeds management (7) Harvesting, economics, marketing, security

Rice±fish culture (1) Pond and trench design (2) System types and sizes (3) Water quality and management (4) Feeding and fertilization (5) Integration with rice schedules (6) Integrated pest management (7) Harvesting methods (8) Economics, marketing and security

Small bamboo and floating net cages (1) Site selection (2) Designs and construction (3) Aquaculture species production (4) Aquaculture management: seed management, feeds, feeding (5) Diseases (6) Harvesting, economics, marketing

Running water systems (1) Site selection (2) Design and construction (3) Water quality and management (4) Aquaculture species production (5) Aquaculture management (6) Harvesting, economics, marketing, security

Pen systems (1) Community organization, group process training (2) Site selection (3) Design and construction (4) Seeding and fertilization (5) Compost as feeds and fertilizers (6) Harvesting methods (7) Marketing and security

Fish feeds processing (1) Feeds available, their composition and costs (2) Raw materials available (3) Low-cost processing methods (4) Economics and marketing

knowledge in many areas of land- and water-based aquaculture and integrated farming systems (for examples see Djajadiredja et al., 1980; Thornburn, 1982; Little & Muir, 1987; Costa-Pierce & Effendi, 1988). Many traditional aquaculture systems and much of the existing farmers' knowledge could be used directly, or modified and adapted for use. It was decided that all adaptive research should employ the villagers from the outset, and that the villagers would be involved throughout the success or failures of the aquaculture research, development, and adaptation processes. Using this approach, farmer recipients were made active, valued participants in both the process and evaluation of the suitability of chosen technologies for their needs. There was a high level of indigenous aquaculture farming knowledge in the rural society where the development project was undertaken. Surveys before the project began documented that a wide diversity of aquaculture systems already existed; that farmers in the surrounding region had an impressive management capability; and that farmers were already doing detailed practical experiments (Koesoemadinata & Costa-Pierce, 1992). In the design of the efforts project scientists realized that it was best if the scientists recognized these farmers and their indigenous knowledge for the value of their innovations, since this approach would speed the choice of more promising and more relevant research topics of direct value to the intended beneficiaries. Lightfoot (1987) has also pointed to the unique value of indigenous

296

Ecological Aquaculture

research by farmers to setting the research agenda of both on- and off-station research workers. Three cage aquaculture research stations were constructed to test a variety of lowcost technologies (`the basket') under three different but prevailing limnological and social/cultural conditions in Saguling in the years just after reservoir filling (the `onstation' experimentation phase).

Establishment of community integrated aquaculture schools While development of the aquaculture resettlement option was the main concern of the project, it was clear from the outset that a more holistic approach to applied research, extension and environmental restoration would be necessary. To accomplish this, community schools (called `Pusat Penelitian Sistem Terintegrasi Tanaman, Ternak dan Ikan'; or Research Centers for the Integrated System of Plants, Farm Animals, and Fish) were created in three villages surrounding the Saguling reservoir with the largest percentage of displaced residents. The project rented village houses for a 3-year period in two villages in Saguling's northern region (Cangkorah and Cipondoh villages) and one in the southern region (Awilarangan village), in areas having excellent technical capabilities for cage aquaculture, and having boat and road access to markets. Another house was rented in Mande village to coordinate applied research and community training activities in Cirata. Village schools had a permanent IOE/ICLARM staff member stationed at the houses who coordinated all applied research projects, hands-on training, and community relationships. Villagers were employed to carry out all aquaculture construction, labor and routine tasks at the schools. Village schools were the center of all collaborative research activities with villagers and between the various outside institutions and other outside villagers visiting the schools. Village schools were visited regularly on overnight stays by IOE, ICLARM, Provincial and Technical Unit personnel to mentor progress, to discuss results internally, and to meet and discuss progress with villagers. Village schools were replete with displays, photographs and extension aquaculture books written in both the local and national languages (Costa-Pierce et al., 1989a, b, c, d, e). Programs at the village schools not only focused on aquaculture but also had working demonstrations in animal husbandry, composting, soil conservation, capture fisheries, fish feed formulation, and fish processing technologies. The village schools also promoted an environmental rehabilitation system which took a system ecology approach to small farm development. Many ecological principles were intuitively familiar to the sophisticated rural farmers, but scientific staff also introduced them to a wide range of new technologies (cage aquaculture, landbased aquaculture systems such as hatcheries and new rice±fish systems), rabbit husbandry, earthworm culture and composting, insect culture, fish and animal processing and marketing, and agroforestry and erosion control (Maskana et al., 1990).

Aquaculture Ecosystems for Resettlement from Hydropower Dams

297

It was estimated from records kept at the schools that over 4000 villagers visited the four village schools from 1986 to 1990.

Farmer-to-farmer visits Once the Saguling cage culture industry began its remarkable expansion, the task of attracting new entrants was of little concern for the project. Indeed control of the number of cages became an issue as early as 1989, since cages became concentrated in one of the southern sectors of Saguling reservoir (Bongas region). By the end of 1989, over 80% of the cage aquaculturists in Saguling were concentrated in this region (Rusydi & Lampe, 1990). The Bongas region had an excellent technical capability for cage aquaculture (a long, deep, sheltered bay with good flushing), a good economic infrastructure, and had excellent market access to fish fingerling and feed suppliers. Given the obvious success of development of cage aquaculture in Saguling, by mid1988, project extension efforts shifted to the new Cirata reservoir. Since, in familiar cultural settings, diffusion of innovations can occur as rapidly with informal farmer contacts as in formal courses (Kang & Song, 1984), a `hands-off extension approach' was used to develop aquaculture in Cirata. Simple farmer-tofarmer visits were sponsored for displaced residents from the new Cirata reservoir. In structured visits, prospective water farmers from the Cirata reservoir region were sponsored to visit the `aquaculturally developed' regions of the Saguling reservoir. Extension personnel were present to answer questions, to distribute free aquaculture workbooks in the local language, to provide `social lubrication', and to take care of personal needs. These visits were a tremendous success. By end 1989, 94 cages (40 families) were operating in Cirata with no formal coursework or extension programs having been conducted. By end 1992, fish production in Cirata was estimated at 3880 mt (Table 10.3). Table 10.3 Fish production in the Saguling and Cirata reservoirs from 1985 to 1996. Saguling was filled in 1985±86 and Cirata in 1986±87. NA = not available

Year

Saguling cages (numbers)

Saguling production (t)

1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997

0 NA NA 765 1 250 1 365 1 724 1 846 2 219 4 250 4 425 NA 8 199

0 522 1 443 2 043 2 204 2 490 2 926 4 053 6 666 6 039 4 420 4 405 NA

Cirata cages (numbers) 0 0 0

NA 94 361 1 043 1 723 2 091 3 860 7 485 NA 25 558

Cirata production (t)

Total production (t)

0 0 0 26 458 798 2 241 3 880 6 556 11 383 14 644 20 091 NA

0 522 1 443 2 069 2 662 3 288 5 167 7 933 13 222 17 877 19 064 24 496 NA

298

Ecological Aquaculture

Information resources It was found that Saguling's cage fish farmers were well educated, with 94% having completed elementary school (Rusydi & Lampe, 1990). Almost all of the people could read extension workbooks if they were in the local Sundanese language. Far fewer villagers could read extension materials in the national language (Bahasa Indonesia), and almost no one could read English. As a result, simple `comic book' type workbooks on floating net cage, pen, small-scale hatchery, and small cages were published in the local Sundanese dialect and distributed widely (Costa-Pierce et al., 1989a, b, c, d, e). Workbooks were made available free at all IOE/ICLARM community schools, and at the offices of the Government's Technical Implementation Unit at the reservoir dam sites. Books were used widely by extension officers and trainers in formal courses in villages. Workbooks were also available free to all members of the Saguling Fish Farmers' Association and Village Cooperative Units in Cirata. During visits to cage culture operators in Saguling and Cirata in 1989±1990, it was frequently observed that these workbooks were among the only reading materials available in village residences. Children seemed to particularly value the `comic book nature' of the materials; to the point that cage culture `toys' appeared in one village.

Study tours to nations with relevant experiences Many aquaculture technologies successful in one developing nation can be transferred to other nations with similar development circumstances after adaptive research is undertaken (World Bank, 1982). However, West Java has a unique aquaculture history, a wealth of experience, and capable fisheries institutions and scientists involved in aquaculture of common carp and other species (Costa-Pierce & Hadikusumah, 1990). However, it was noted that: (1) (2)

Saguling and Cirata were very eutrophic reservoirs with a large potential for `no feed', or extensive cage aquaculture, and much of the specific technology to diversify the reservoir cage culture industry and assist the poorest of displaced residents (e.g. by evolving a low cost or extensive cage aquaculture, particularly for Nile tilapia (Oreochromis niloticus) and Chinese silver carp (Hypophthalmichthys molitrix) and bighead carp (Aristichthys nobilis) was lacking.

It was felt by project scientists that transfer of modern methods and management practices in extensive aquaculture to the Indonesian reservoirs would not be a prolonged proposition or an expensive exercise. The Philippines has a wide diversity of successful tilapia cage and pen aquaculture, and extensive cage aquaculture for the Chinese carps has been successful in lakes in China, Nepal and Singapore (Chookajorn, 1982; Guerrero, 1982; Gonzales, 1984; Pullin, 1986; Hai & Zweig, 1987).

Aquaculture Ecosystems for Resettlement from Hydropower Dams

299

It was decided to arrange two study tours in 1987±88 for selected scientists from the Indonesian State Electric Company, IOE, and the Technical Implementation Unit to transfer technology rapidly from Asian nations with relevant experiences in low-cost cage aquaculture to Indonesia (Costa-Pierce et al., 1988; 1989e). This transfer of technology by sponsoring overseas study tours was very successful. By the end of 1989, 26 small-scale cage hatcheries for an Indonesian variety of red tilapia (hybrids of Oreochromis spp.) were being operated by resource poor farmers in Saguling. Two tilapia grow-out operations started in 1989±1990. In addition, in 1990, a number of cage operators started doing polycultures of common carp and red tilapia in cages. Active research in tilapia aquaculture was started by the Provincial Fisheries Agency, at the local university (Department of Fisheries, Padjadjaran University), and at IOE.

Short-term analysis of successes and failures Aquaculture as a means of sustainable rural development in developing nations has been sharply criticized by a number of authors (Zerner, 1992; Folke et al., 1994; Costa-Pierce, 1996; Kautsky et al., 1997); and, in general, remains an elusive goal. The main problems have been insufficient attention to realistic economic appraisals, a lack of social concerns, especially social impacts on equity, a lack of expertise, and a lack of appropriate technologies incorporating traditional knowledge (Pollnac & Sihombing, 1996). The reservoir aquaculture resettlement project was successful in a technical sense in developing numerous new land- and water-based aquaculture systems and associated aquaculture support industries in villages surrounding the two new hydropower reservoirs (IOE & ICLARM, 1989; Costa-Pierce & Soemarwoto, 1990; Sutandar et al., 1990; Costa-Pierce, 1997). The project trained a documented 2081 persons and recorded over 4000 visits to village schools; by 1992, an estimated 7527 persons were directly or indirectly involved in fish production; and at the end of 1996 total annual fish production from cages from the two reservoirs was an amazing 24 496 mt. It is important to reflect a little on the magnitude of this production. The reported range of capture fisheries production in Southeast Asian reservoirs is just 5±675 kg/ ha/year (De Silva, 1988). In 1996, cage aquaculture in the Saguling and Cirata reservoirs produced 2130 kg/ha/year (24 496/11 500 ha). And expansion of production is possible with existing technology and better siting. Costa-Pierce & Hadikusumah (1990) demonstrated that each cage could produce fish at 3 mt/year if adequate supplies of fingerlings were available. Applied across the 16 400 cages (at carrying capacity) this would yield 49 200 mt of fish and generate an estimated US$49.2 million per year ($1 per kg) at capacity. And the production potential of simple cage systems doesn't stop there. Costa-Pierce (1997) noted the proliferation of a new type of `condominium' cage aquaculture systems in the Bongas area of the Saguling reservoir that had a production potential of 10 mt per cage per year! Clearly, the cage aquaculture systems in the Saguling and Cirata reservoirs present a exciting new model of

300

Ecological Aquaculture

large-scale protein food production for a protein-hungry Asia, that could, if sustainable, represent a new, globally important food resource ecosystem. The most important factors contributing to the initial technical success of the development efforts in the first seven years were: (1)

(2)

(3)

(4)

(5)

Presence of a defined, educated target group. Lists of names with addresses of displaced persons from reservoir inundation were obtained from the electric company along with how much compensation money these people obtained. While in many cases the electric company lists were found to be outdated or wrong, the fact that some information did exist helped to identify: (a) the exact geographic scope of the project (new and old villages with largest numbers of displaced families), (b) families who had backyard fish ponds before the reservoir, and therefore had a traditional knowledge of aquaculture ecosystems. Ready availability of investment capital. Lack of the ready availability of startup capital often constrains aquaculture development among the rural poor (the target group of many development assistance projects). All villagers in Saguling obtained compensation money from the electric company; 92% received less than Rp 6 million, and 8% over Rp 6 million (Suwartapradja & Achmad, 1990). Having a large amount of cash available allowed immediate investment in new aquaculture businesses. However, for the poorest residents, compensation monies were not enough to replace homes and lands lost due to increased land prices, speculation, and inflation (Cernea, 1997). Lack of alternative employment opportunities in both rural agricultural and human ecosystems. Rural population densities increased 2±3 times due to the reservoirs, from a range of 237±1691 persons/km2 before the reservoir to 476± 4292 persons/km2 after (Suwartapradja & Achmad, 1990). These are among the highest `rural' population densities anywhere in the world. According to Collier et al. (1977), by the late 1970s the rice agroecosystem in Java could not absorb more rural labor. They predicted massive migration to coastal Javanese cities if a solution was not found (and one wasn't: Jakarta boomed from 11.9 to 17.1 million in the 1980s). Local traditional knowledge of aquaculture and cage culture. Dahlman & Westphal (1981) describe the success of development assistance in terms of the `technological mastery' of a system, which they define as `the autonomous ability to identify, select, and generate technological improvements and changes'. Rapid adoption of the fish cage systems in the reservoirs was influenced by the inherent innovativeness of farmers in West Java. Farmers operating existing agro- and aquaecosystems in the province had an impressive amount of indigenous knowledge and vibrant on-farm `trial and error research' systems that were characterized by a great deal of individual innovation. Flingel (1984) also viewed the adoption of change as being directly influenced by the basal level of innovativeness present in a society. Large market demands and relatively high prices for freshwater fish. Ikan mas (common carp) are the most preferred fish in West Java. Common carp play a

Aquaculture Ecosystems for Resettlement from Hydropower Dams

(6) (7)

(8)

(9)

301

unique cultural role as food eaten at important celebrations (weddings, graduations, family reunions, etc.) among the dominant Sundanese culture in the province. The price for the fish is therefore higher than that for other fish such as tilapia. Kusnadi & Lampe (1990) observed that price fluctuations for common carp in Jakarta were small even though strong seasonal fluctuations in fish supplies occurred, indicating that a large demand exists for the fish. Given the increasing population density and increasing incomes of Jakarta residents (who eat about 14 kg of fish per capita per year), market demands for freshwater fish were large and unsaturated. Ready access to large urban markets on paved roads. Suitable environment. Saguling had many deep, sheltered bays very suitable for cage aquaculture (IOE & ICLARM, 1989; Soemarwoto et al., 1990; CostaPierce, 1997). Institutional and international cooperation. Although difficult to coordinate, and necessitating a larger than anticipated administrative load, the cooperation and technical assistance of international (ICLARM), government, electric company, university ecology, and local fisheries organizations was critical. Having an international reservoir fisheries expert on site for three years was also beneficial. Accessibility of rural extension services. Rusydi & Lampe (1990) report that 90% of Saguling's cage culturists with a single fish cage participated in extension and training programs, and that 44% of all cage farmers got information from training or extension programs.

Problems and lack of sustainability of the aquaculture resettlement option (1994±present) The main problems experienced in the course of this project were: (1)

Resources for whom? The problem of equity. When the cage aquaculture industry began to take off, rich people from the urban centers of Bandung/ Jakarta began to enter the industry. By the end of 1996 these `outsiders' owned 52% of the cages in Cirata (Table 10.4), and had acquired nearly complete control over the marketing sectors. The West Java Provincial Fisheries Agency made government regulations on the cage industry specifying that only displaced persons could get permits to use the reservoir waters for aquaculture and fisheries. These regulations were later codified into provincial laws, stipulating that only the displaced people were allowed to operate cages and that the number of cages could not exceed four cages per family. The permitting process was to be controlled by the Fisheries Department. However, by 1994 outsiders had found that the `back door route' was an easy way to obtain access, and they paid off government agents, village leaders, and other local people to get permits. Outsiders undermined the ownership of the new aquatic resources with their financial abilities to employ the displaced

302

Ecological Aquaculture

Table 10.4 Numbers of resettled or outsider families operating fish cages in the Cirata reservoir at the end of 1996. (Data from the West Java Provincial Fisheries Service, 1996.) 2±4 cages

5±8 cages

9±12 cages >13 cages

No.

Village

R

N

R

N

R

N

R

N

1 2 3 4 5 6 7 8 9 10

Cipicung Tegal datar Patok beusi Ciputri Janggari Nyalempet Neundeut Kebon coklat Bongas Calincing

38 67 11 19 22 30 8 8 21 25

18 22 4 14 9 31 27 12 17 22

50 112 15 24 33 61 17 24 23 48

30 41 29 32 23 83 46 32 27 43

45 87 10 10 10 48 6 14 20 37

26 42 34 30 16 78 21 30 18 35

29 82 10 14 9 56 9 12 9 28

Total families

249

176

407

386

287

330

258

Total of families R

N

44 42 34 31 11 121 17 23 12 44

162 348 46 67 74 195 40 58 73 138

118 147 101 107 59 313 111 97 74 144

379

1 201

1 271

Total cages 3 105 5 190 1 682 1 653 1 054 5 946 1 272 1 546 1 227 2 883 25 558

Notes: R (resettled) = families that lost land flooded by reservoir impoundment; N (newcomers) = immigrants from outside reservoir shore line.

(2)

(3)

(4)

persons as managers or laborers in the cage industry in return for `shadow ownership'. The aquaculture permit was in the name of a displaced person but the owner was an absentee `waterlord' in the city. Consolidation of the aquaculture industry into the hands of the rich puts into question any long-term social benefits of the project (Zerner, 1992). Multidisciplinary nature of the efforts. While commendable, such multidisciplinary environmental efforts require more administration than traditional, disciplinary ones. Such extra administrative efforts must be funded. In this case, they were not. It was difficult to coordinate all the professionals needed for project implementation. Many persons had difficulty seeing beyond the narrow bounds of their professional training. It was felt, however, that the developmental situation would have been even worse if the project had been led by a conventional fisheries or aquaculture research organization, rather than by an ecology institution. Ecologists, in general, have a more `generalist' training, and overall were more sensitive to interactions and interfaces. Vague nature of institutional agreements and responsibilities. The institutional framework for project implementation between electric company, university, fisheries, and regional and village political institutions had to be created by the project, necessitating the above-mentioned larger than anticipated administrative load. In addition, each institution occasionally (and repeatedly) had their own interpretation of what the project agreement actually said. And in some cases, these separate institutions actually carried out their interpretations of the agreements without communicating with others, causing duplication and disagreements. Self-pollution. IOE & ICLARM (1989) calculated that the aquaculture carrying capacity of the two reservoirs was 16 400 cages (5800 in Saguling and 10 600 in Cirata). Depending on the availability of adequate numbers of fingerlings, each

Aquaculture Ecosystems for Resettlement from Hydropower Dams

303

cage could produce, conservatively, 1±3 mt fish/year, or a final carrying capacity of 16 400±49 200 mt of fish per year (Costa-Pierce & Hadikusumah, 1990). IOE & ICLARM (1989) also included a dispersal of cages throughout most of the suitable sites for cages (deep, sheltered, and well-flushed bays) (Figure 10.2).

Fig. 10.2 Aquaculture development plan for the Cirata reservoir (Effendi, 1988). The plan apportions water-based aquaculture systems (floating cages, net pens) and land-based aquaculture systems (hatcheries, rice-field fish nurseries and intensive running water systems) in and around the 6200 ha reservoir. Effendi (1988) forecast aquaculture production from floating cages in Cirata to be 1597 mt by 1992. Fish aquaculture production in Cirata actually grew much more rapidly ± annual fish production reached 3880 mt by 1992 (Table 10.3). Administratively, the reservoir was located in two regions (Bandung and Cianjur; kab. = kabupaten) and four districts (kec. = kecamatan) of kabupaten Cianjur.

304

(5)

Ecological Aquaculture

This areal distribution of cages did not occur. Farmers crowded into a very few select bays of the Saguling reservoir (southern Bongas region) and Cirata reservoir (Janggari region) due to better availability of the village and economic infrastructures, and better access to large fish markets in Sukabumi, Bandung, and Jakarta. This crowding led to `self-pollution' by the cages due to waste feed and nitrogen discharges from the cages (and the wastes from an increasing number of residents on the surface of the water) (Soemarwoto et al., 1990). The result has been development of nuisance algal blooms and more frequent oxygen depletions, leading to large fish kills. Industrial pollution. In the 1990s Indonesia experienced high rates of industrial growth, especially in the manufacturing sector. In the Saguling±Cirata reservoir region, there has been an especially rapid growth in the textile industry. Some of these textile mills discharged untreated wastes into the Saguling reservoir, and were cited by residents as causing fish kills. Industrial wastes discharged to the reservoirs threaten the very basis of the entire cage aquaculture developments and are a principal concern to the public and product safety now and into the future.

The lack of social sustainability: aquaculture's role in equity and the rural poor Aquaculture development has often promised the rural poor in developing countries increased access to rural jobs, and better incomes in an environmentally sensitive, non-polluting business. In this case study, however, even strong government laws and other regulations controlling access, the number of cages (4 per family), and a permitting system reserved only for displaced persons could not stop the new aquaculture industry from being usurped and consolidated into the hands of the urban rich. In addition, planning could not stop the haphazard crowding of cages into a very few areas of the reservoirs where the wastes caused self-pollution and economic losses. Benefits of the new reservoir aquaculture enterprises in Indonesian reservoirs have accrued increasingly to those with adequate capital and power who are not displaced persons (Table 10.4). The poorest of the displaced residents have been resigned to laboring in the industry, rather than controlling it. The institutional will to enforce control over access has evaporated when cage numbers exceeded the guidelines and problems appeared. The fisheries department is the authority responsible for the cages, the water resources directorate owns the water, and the State Electric Company the dam and the reservoir bottom, but none of these agencies is willing to restrict new cage development or enforce their own laws (T. Walton, personal communication, 1997). In addition, there is a new problem because, in the drive for privatization, the reservoirs are now operated not by the State Electric Company (PLN), but by wholly owned subsidiary companies of PLN. However, there have been a number of positive, `trickle-down' type of benefits for the rural poor. It has been estimated that the new aquaculture industry created many new jobs in a rural area of severe underemployment (22 new types of rural jobs have

Aquaculture Ecosystems for Resettlement from Hydropower Dams

305

been documented; Costa-Pierce, 1997). These new jobs were higher paying (it has been estimated that cage aquaculture workers earned about Rp 56 000 per month more than rice field workers in the same area); and the new work was less rigorous than hard labor in rice paddies (Costa-Pierce, 1997). In 1992, it was estimated that 7527 persons were employed either full- or part-time in the reservoir aquaculture industry in the two reservoirs: 1162 directly and 6365 indirectly (Costa-Pierce, 1997). More difficult to discern is whether or not any of the trickle-down benefits were due to these new and high paying aquaculture jobs or were simply due to the increased economic opportunities that have been available to most of Java in the 1990s.

The lack of environmental sustainability: how to ensure sustainability of cage aquaculture production in tropical reservoirs The West Java Provincial Fisheries Agency led the drafting of two 5-year plans for reservoir aquaculture development in Saguling (1985±89) and Cirata (1988±92) (Effendi, 1985, 1988). Plans called for a stepwise development of cage aquaculture with 3195 mt of fish to be produced from cages in Saguling and Cirata in equal amounts by 1992 (6390 mt total), along with development of other land- and waterbased aquaculture support systems (Table 10.5). Cage aquaculture developed much more rapidly and in a much more haphazard fashion. Instead of 6390 mt by the end of 1992, fish production was estimated at 7933 mt (Table 10.3). The IOE & ICLARM (1989) reservoir fisheries and aquaculture development plan estimated a carrying capacity of 5800 cages for Saguling and 10 600 for Cirata, or 16 400 total fish cages distributed throughout the reservoirs (Fig. 10.2). There are an estimated 25 558 cages in Cirata in 1996 (Table 10.4), and these are crowded into very few areas. As a result, more frequent oxygen depletions have occurred since 1993, leading to large fish kills in Saguling (> 500 mt in 1994±95). There is a significant role for an ecosystems approach to developing aquaculture, by designing non-polluting aquaculture ecosystems and closing biogeochemical nutrient cycles as an alternative approach to the `feedlot' aquaculture scenarios (Folke & Kautsky, 1992; Costa-Pierce, 1996). High nutrient inputs enter the reservoirs from surrounding urban areas making a high biomass of plankton available year round. Plankton could be cropped by fish in cages in `extensive', or no feed, cage culture of the tilapias and Chinese carps (Hai & Zweig, 1987; Costa-Pierce, 1997). In addition, `condominium' cages having one insert cage suspended above another with fish that are fed, and an outer cage having a crop of fish that is unfed, could increase fish production from the same area of water surface, increase feed efficiency, and decrease self- and external environmental pollution (CostaPierce, 1997). It is recommended that an expansion of `no feed' cage aquaculture systems for Chinese carps and tilapias be developed and that more emphasis be placed on development of land-based hatcheries and rice±fish nursery systems. Emphasizing an ecosystems approach and developing ecosystems technologies would also better concentrate aquaculture's local economic development and multiplier effects.

306

Activity

Unit

1985

1986

1987

Floating net cages Pens Hatcheries

7 6 7 6 2.5 m 1 ha 1500 m2

75 35 52

230 20 2

340 15 2

Activity

Unit

1988

1989

1990

Floating net cages Pens Hatcheries Running water Rice±fish

7 6 7 6 2.5 m 1 ha 1500 m2 100 m2 1 ha

75 27 20 5 115

230 24 18 4 15

340 15 16 2 10

Year

Year

1988

1989

Total

1985

1986

Production (kg) 1987 1988

340 15 2

355 15 2

1 340 100 60

337.5 63 24 128

1 035 36 928

1 530 27 928

1991

1992

Total

1988

1989

Production (kg) 1990 1991

340 15 14 2 10

355 15 12 2 10

1 340 96 80 15 160

337.5 156.6 9 280 25 40.3

1 035 1 392 8 352 20 53

1 530 87 7 424 10 3.5

1 530 27 928

1 530 87 6 496 10 3.5

1989

Total

1 597.5 27 928

6 030 180 27 840

1992

Total

1 597.5 87 5 568 10 3.5

6 030 556.8 37 120 75 56.1

Ecological Aquaculture

Table 10.5 Five-year step-by-step development plan for the development of aquaculture systems for resettlement in the Saguling and Cirata reservoirs (Effendi, 1985, 1988)

Aquaculture Ecosystems for Resettlement from Hydropower Dams

307

Due to uncontrolled population growth and increased demands for water and power, there has been a boom in hydropower construction since the 1950s, especially in Asia. New methods are needed to manage reservoirs for sustainable food production and restore the livelihoods of displaced peoples. In these studies, an interdisciplinary, `social ecological' approach was used to develop aquaculture as a means of local population resettlement and income restoration. Village participatory research methods were used to develop a `basket of low cost aquaculture systems', and encourage development of intimate societal interactions with the new land± water aquatic ecosystems in villages most impacted by hydropower reservoirs, as opposed to moving people far away from these new environments. There are numerous situations in the developing countries of Asia where this paradigm may be of interest. Aquaculture's role in sustainable rural development must be determined by the agenda of the intended beneficiaries. Sophisticated traditional methods of aquatic ecosystems management exist in numerous developing nations, especially in Asia. By involving the target group from the outset to develop a step-by-step technology and social adaptation program, we have evolved a set of appropriate technologies based on traditional aquaculture ecosystems in the Saguling±Cirata reservoir region that present a exciting new model of large-scale protein food production for proteinhungry Asia. It is argued that social and environmental restoration of damaged watersheds from dam construction can only be accomplished through such active involvement of displaced people who have an investment in rehabilitation. In this case an integrated program using the principles of farming systems research and extension methods allowed the necessary flexibility of choices to be made by the people themselves on the best component technologies they could capitalize on and manage. Reservoirs offer unique opportunities to educate people about their new environments and for formulating innovative, flexible, and evolutionary ownership patterns and agreements between electric companies, research institutions, nongovernmental organizations, and communities to meet a common set of restoration goals. Few reported aquaculture development experiences in any country have met with such remarkable success in as short a period of time as the reservoir cage aquaculture developments in Saguling and Cirata in Indonesia. But success of the fish cage aquaculture industry cannot only be measured by tons of fish. These aquaculture systems are fragile, and presently are unsustainable both environmentally and socially. This study shows clearly that sustainability of aquaculture requires government support in the form of technical extension inputs, strict enforcement of its own regulations on access permits, systems numbers, and pollutant discharges, and clear institutional commitments to equity. If a means could be found to ensure the more equitable distribution of long-term benefits to the target group, this notion of developing floating cage aquaculture in artificial reservoirs as a new source of aquatic protein could, in the future, represent a new, globally important food resource. The development scenario reviewed here is characterized by rapid, dynamic and constant change. For this reason, it is recommended that comprehensive, long-term studies of the fish cages and the people be accomplished. These people have made a

308

Ecological Aquaculture

huge social leap from being principally highland rice farmers to fish farmers, transiting in a few short years from being land farmers to water farmers. They are a remarkable group of people, a unique living laboratory of new indigenous knowledge. These people could teach us a great deal about the global and localized factors contributing to success, and provide more accurate forecasts in order for scientists and policy-makers to better judge the applicability of the aquaculture resettlement option to other developing countries.

Conclusions From 1985 to 1988 the Saguling and Cirata hydropower reservoirs in the densely populated highlands of West Java, Indonesia, displaced over 40 000 families. As part of a comprehensive resettlement plan, an attempt to resettle 3000 families in waterbased, floating fish cage aquaculture and land-based aquaculture support was initiated. Over a 4-year period, aquaculture research, demonstration, extension, and training programs were conducted and study tours to other Asian nations with relevant experiences arranged. By the end of 1992, water- and land-based aquaculture ecosystems in and around the Saguling and Cirata reservoirs employed 7527 persons. At the end of 1996, total fish production was 24 496 metric tons (approximately 95% common carp, Cyprinus carpio and 5% hybrid red tilapia, Oreochromis spp.), an amount equal to about 20% of the fish entering the Bandung district (estimated population 3 million persons). Total 1996 gross revenue from fish was over US$24 million, over twice the estimated annual revenue (US$10.4 million adjusted for inflation to 1996) from the 5783 ha of ricelands lost to the reservoirs. However, from 1994 to the present, aquaculture developments in the Saguling and Cirata reservoirs are neither environmentally nor socially sustainable. The benefits of floating cage aquaculture, which were guaranteed originally to the displaced people by provincial legislation and were designed to give them exclusive ownership over the production and marketing sectors of the industry, were usurped by the politically powerful, and consolidated into the hands of the urban rich from Bandung and Jakarta. Survey of fish cage ownership in 1996 in Cirata showed only 48% of cage owners were resettled persons (1201 out of 2480 total aquaculture families). Most of the displaced people were involved as employees of absentee owners. Guidelines set on the numbers of cages (10 600 in Cirata and 5800 for Saguling) to protect the reservoir environments were not enforced, causing environmental degradation and self-pollution (at the end of 1996, Cirata had 25 558 cages and Saguling 8199 (1997 survey)). Fish cages were developed haphazardly in very few areas of the reservoirs where market access was good, rather than where the environments were suitable, degrading water quality. As a result of overcrowding and water-column turnovers, there were numerous fish kills in the upstream Saguling reservoir (aquaculture production dropped from 6666 mt in 1993 to 4405 mt in 1996). Numerous Saguling farmers moved downstream to the Cirata and Jatiluhur reservoirs, where they crowded cages in the waters beyond sustainable levels (Cirata's fish production

Aquaculture Ecosystems for Resettlement from Hydropower Dams

309

increased from 6556 mt in 1993 to 24 496 mt in 1996). Thus, while the reservoir cage aquaculture developments were successful from a fish production viewpoint, aquaculture has not been sustainable socially or environmentally over the long term. Cage aquaculture in reservoirs can be an important new means of large-scale population resettlement from hydropower dam construction and protein production in tropical developing countries only if: (1) (2) (3) (4) (5)

adequate government planning for fisheries is included before dam construction (too often fisheries are viewed as another `simple engineering problem'); adequate financial compensation for lost assets is given; there is rigid enforcement of institutional regulations guaranteeing the long-term benefits of the new lakes for the exclusive use of the displaced people; there is enforcement of regulations on cage numbers to prevent environmental degradation; and adequate government subsidies are provided for aquaculture job creation, training, long-term extension support, and active monitoring.

References Alam, M., Mirza, M. & Maughan, O. (1995) Constraints and opportunities in planning for the use of natural resources in developing countries: example of a hydropower project. Environmental Conservation, 22, 352±358. Brown, L. (1997) Facing the prospect of food scarcity. In: State of the World (ed. L. Brown), pp. 23±41. Worldwatch Institute, Washington, DC. Cernea, M. (1988) Involuntary Resettlement in Development Projects. The World Bank, Washington, DC. Cernea, M. (1997) The risks and reconstruction model for resettling displaced populations. World Development, 25, 156±158. Chambers, R., Pacey, A. & Thrupp, L. (1989) Farmers First: Farmer Innovation and Agricultural Research. Intermediate Technology Publications, London. Chookajorn, T. (1982) Cage fish culture in reservoirs in Singapore. Thai Fisheries Gazette, 35, 661±665. Collier, W.L., Hadikoesworo, H. & Saropie, S. (1977) Income, Employment, and Food Systems in Javanese Coastal Villages. Center for International Studies, Ohio University, Athens, Ohio. Costa-Pierce, B.A. (1992) Rice±fish systems as intensive nurseries. In: Rice±Fish Research and Development in Asia (eds C. De la Cruz, C. Lightfoot, B. Costa-Pierce, V. Carangal & M. Bimbao), pp. 117±130. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A. (1996) Environmental impacts of nutrients discharged from aquaculture: towards the evolution of sustainable, ecological aquaculture systems. In: Aquaculture and Water Resource Management (eds D.J. Baird, M.C.M. Beveridge, L.A. Kelly, & J.F. Muir), pp. 81±113. Blackwell Science, Oxford. Costa-Pierce, B.A. (1997) From Farmers to Fishes: Developing Reservoir Aquaculture for People Displaced by Dams. The World Bank, Washington, DC.

310

Ecological Aquaculture

Costa-Pierce, B.A. & Soemarwoto, O. (1987) Proliferation of Asian reservoirs: the need for integrated management. Naga, The ICLARM Quarterly, 10, 9±10. Costa-Pierce, B.A. & Effendi, P. (1988) Sewage fish cages of Kota Cianjur. Naga, The ICLARM Quarterly, 11, 7±9. Costa-Pierce, B.A., Zainal, S. & Effendi, P. (1988) ICLARM and south±south technology transfer: Philippine aquaculture technology and Indonesia. Part 1. Naga, The ICLARM Quarterly, 11, 10±11. Costa-Pierce, B.A., Rusydi, K., Safari, A. & Wira Atmadja, G. (1989a) Small Scale Hatchery for Common Carp. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A., Rusydi, K., Safari, A. & Wira Atmadja, G. (1989b) Culture of Common Carp in Floating Cages. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A., Rusydi, K., Safari, A. & Wira Atmadja, G. (1989c) Growing Fish in Pen Systems. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A., Wira Atmadja, G., Rusydi, K. & Safari, A. (1989d) Growing Fish in Cages. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A., Bimbao, M.A., Zainal, S. & Effendi, P. (1989e) ICLARM and south±south technology transfer: Philippine aquaculture technology and Indonesia. Part 2. Naga, The ICLARM Quarterly, 12, 14±16. Costa-Pierce, B.A. & Soemarwoto, O. (1990) Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A. & Hadikusumah, H. (1990) Research on cage aquaculture systems in the Saguling Reservoir, West Java, Indonesia. In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia (eds B.A. Costa-Pierce & O. Soemarwoto), pp. 112± 217. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A. & Hadikusumah, H. (1995) Production management of double-net tilapia Oreochromis spp. hatcheries in a eutrophic tropical reservoir. Journal of the World Aquaculture Society, 26, 453±459. Dahlman, C.J. & Westphal, L.E. (1981) The meaning of technological mastery in relation to the transfer of technology. Annals of the American Academy of Political Science, 458, 12±25. De la Cruz, C., Lightfoot, C., Costa-Pierce, B.A., Carangal, V. & Bimbao, M. (eds) (1992) Rice±Fish Research and Development in Asia. International Center for Living Aquatic Resources Management, Manila, Philippines. De Silva, S.S. (1988) The reservoir fishery of Asia. In: Reservoir Fishery Management and Development in Asia (ed. S.S De Silva), pp. 19±28. International Development Research Centre, Ottawa, Canada. De Silva, S.S. (1992) Reservoir Fisheries of Asia. International Development Research Centre, Ottawa, Canada. Djajadiredja, R., Jangkaru, Z. & Junus, M. (1980) Freshwater aquaculture in Indonesia, with special reference to small scale agriculture±aquaculture farming systems in West Java. In: Integrated Agriculture±Aquaculture Farming Systems (eds R. Pullin & Z. Shehadeh), pp. 143± 166. International Center for Living Aquatic Resources Management, Manila, Philippines.

Aquaculture Ecosystems for Resettlement from Hydropower Dams

311

Effendi, P. (1985) Fisheries Development Program in Saguling during the National Five Year Plan, 1985±89. West Java Provincial Fisheries Agency, Bandung, Indonesia. Effendi, P. (1988) Fisheries Development Program in Cirata during the National Five-Year Plan, 1988±92. West Java Provincial Fisheries Agency, Bandung, Indonesia. Engleman, R. (1995) Feeding tomorrow's people from today's land. Environmental Conservation, 22, 97±98. Engleman, R. & LeRoy, P. (1995) Conserving Land: Population and Sustainable Food Security. Population Action International, Washington, DC. Fearnside, P. (1997) Transmigration in Indonesia: lessons from its environmental and social impacts. Environmental Management, 21, 553±570. Firman, T. & Dharmapatni, I.A.I. (1994) The challenges to sustainable development in the Jakarta metropolitan region. Habitat International, 18, 79±91. Flingel, F.C. (1984) Extension communication and adoption process. In: Agricultural Extension: A Reference Manual (ed. B. E. Swanson), pp. 77±88. Food and Agriculture Organization (FAO), Rome. Folke, C. & Kautsky, N. (1992) Aquaculture with its environment: prospects for sustainability. Ocean and Coastal Management, 52, 457±472. Folke, C., Kautsky, N. & Troell, M. (1994) The costs of eutrophication from salmon farming: Implications for policy. Journal of Environmental Management, 40, 173±182. Gardner, G. & Perry, J. (1995) Big-dam construction is on the rise. Worldwatch, 8, 36±37. Gonzales, E.R. (1984) Small scale tilapia cage technology adopted in fishing villages in Laguna Lake, Philippines. Aquaculture, 41, 161±169. Guerrero, R.D. III (1982) Development, prospects and problems of the tilapia cage culture industry in the Philippines. Aquaculture, 27, 313±315. Hai, L.Y. & Zweig, R.D. (1987) Cage culture in Kunming: an effective means of resource recovery. Aquaculture Magazine, 13, 28±31. Hunter, J.M., Rey, L. & Scott, D. (1983) Man-made lakes ± man-made diseases. World Health Forum, 4, 177±182. IOE (Institute of Ecology) & ICLARM (International Center for Living Aquatic Resources Management) (1989) Final Report. Development of Aquaculture and Fisheries Activities for Resettlement of Families from the Saguling and Cirata Reservoirs. Vol. 1. Executive Summary and Management Plan. Padjadaran University, Institute of Ecology, Bandung, Indonesia, and International Center for Living Aquatic Resources Management, Manila, Philippines. Kang, J.T. & Song, H.K. (1984) Individual and group extension teaching methods. In: Agricultural Extension: a Reference Manual (ed. B.E. Swanson), pp. 130±143. Food and Agriculture Organization, Rome. Kautsky, N., Berg, H., Folke, C., Larsson, J. & Troell, M. (1997) Ecological footprint for assessment of resource use and development limitations in shrimp and tilapia aquaculture. Aquaculture Research, 28, 753±766. Koesoemadinata, S. & Costa-Pierce, B.A. (1992) Status of rice±fish culture in Indonesia, In: Rice±Fish Research and Development in Asia (eds C.R. de la Cruz, C. Lightfoot, B.A. CostaPierce, V.R. Carangall & M.P. Bimbao), pp. 45±62. International Center for Living Aquatic Resources Management, Manila, Philippines. Kusnadi, E.W. & Lampe, H. (1990) Marketing of common carp from the Saguling reservoir. In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia. (eds B.A.

312

Ecological Aquaculture

Costa-Pierce & O. Soemarwoto), pp. 240±256. International Center for Living Aquatic Resources Management, Manila, Philippines. Lelek, A. (1984) Ecological aspects of the damming of a river in Borneo. Applied Geography and Development, 24, 59±73. Lightfoot, C. (1987) Indigenous research and on-farm trials. Agricultural Administration and Extension, 24, 79±89. Little, D. & Muir, J. (1987) A Guide to Integrated Warmwater Aquaculture. Institute of Aquaculture, University of Stirling, Stirling. Maskana, S., Astuti, S., Sudiarto, B. & Soemarwoto, O. (1990) Integrated system for environmental management in the Saguling-Cirata reservoir region. In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia (eds B.A. Costa-Pierce & O. Soemarwoto), pp. 370±378. International Center for Living Aquatic Resources Management, Manila, Philippines. McCully, P. (1996) Silenced Rivers. The Ecology and Politics of Large Dams. Zed Books, London. Munro, J., Iskandar, A. & Costa-Pierce, B. (1990) Fisheries of the Saguling reservoir and a preliminary appraisal of management options. In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia (eds B.A. Costa-Pierce & O. Soemarwoto), pp. 285±328. International Center for Living Aquatic Resources Management, Manila, Philippines. Petr, T. (1978) Tropical man-made lakes ± their ecological impact. Archives fuÈr Hydrobiologie, 81, 368±385. Pimental, D., Houser, J., Preiss, E., White, O., Fang, H., Mesnick, L., Barsky, T., Tariche, S., Schreck, J. & Alpert, S. (1997) Water resources: agriculture, the environment, and society. BioScience, 47, 97±106. Pollnac, R.B. (1982) Sociocultural aspects of implementing aquaculture systems in marine fishing communities. In: Aquaculture Development in Less Developed Countries (eds L.J. Smith & S. Peterson), pp. 31±52. Westview Press, Boulder, Colorado. Pollnac, R. & Sihombing, S. (1996) Cages, controversies and conflict: carp culture in Lake Toba, Indonesia. In: Aquaculture Development. Social Dimensions of an Emerging Industry (eds C. Bailey, S. Jentoft & P. Sinclair), pp. 230±256. Westview Press, Boulder, Colorado. Postel, S., Daily, G. & Ehrlich, P. (1996) Human appropriation of renewable fresh water. Science, 271, 785±788. Pullin, R.S.V. (1986) Aquaculture development in Nepal ± pointers for success. Naga, The ICLARM Quarterly, 9, 9±10. Rifai, S. (1985) Experimental Project in Floating Net Cages in the Saguling Reservoir. Padjadjaran University, Bandung, Indonesia. Ruddle, K. & Zhong, G. (1988) Integrated Agriculture±Aquaculture in South China. Oxford University Press, Oxford, England. Rusydi, K. & Lampe, H. (1990) Economics of floating net cage common carp culture in the Saguling Reservoir, West Java, Indonesia. In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia (eds B.A. Costa-Pierce & O. Soemarwoto), pp. 218±239. International Center for Living Aquatic Resources Management, Manila, Philippines. Soemarwoto, O., Roem, C., Herawti, T. & Costa-Pierce, B.A. (1990) Water quality suitability

Aquaculture Ecosystems for Resettlement from Hydropower Dams

313

of the Saguling and Cirata reservoirs for floating net cage aquaculture, In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia (eds B. Costa-Pierce & O. Soemarwoto), pp. 18±111. International Center for Living Aquatic Resources Management, Manila, Philippines. Sutandar, Z., Costa-Pierce, B.A., Iskandar, A., Rusydi, K. & Hadikusumah, H. (1990) The aquaculture resettlement option in the Saguling reservoir, Indonesia: its contribution to an environmentally-oriented hydropower project. In: The Second Asian Fisheries Forum (eds R. Hirano & I. Hanyu), pp. 253±258. The Asian Fisheries Society, Manila, Philippines. Suwartapradja, O.S. & Achmad, H. (1990) Population density and general socioeconomic conditions around the Saguling and Cirata reservoirs. In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia (eds B. Costa-Pierce & O. Soemarwoto), pp. 7±17. International Center for Living Aquatic Resources Management, Manila, Philippines. Thompson, S. (1995). USDA, Foreign Agriculture Service, Grain: World Markets and Trade. United States Department of Agriculture, Washington, DC. Thornburn, C. (1982) Teknologi Kampungan: A Collection of Indigenous Indonesian Technologies. Volunteers in Asia, Stamford, Connecticut. World Bank (1982) Fishery Sector Policy Paper. The World Bank, Washington, DC. Zerner, C. (1992) Development of small-scale freshwater cage culture fishery in reservoirs in Java: legal, environmental, and socioeconomic issues. In: Contributions to Fishery Development Policy in Indonesia (eds R.B. Pollnac, C. Bailey & A. Poernomo), pp. 1001±1128. Central Research Institute for Fisheries, Jakarta, Indonesia.

Chapter 11

The Role of Aquaculture in the Restoration of Coastal Fisheries Mark A. Drawbridge Hubbs-SeaWorld Research Institute Introduction When aquaculture is used as a vehicle to help restore fisheries, it is referred to as `sea ranching' or `stock enhancement'. These terms are often used interchangeably, especially as they pertain to marine programs, but they have been appropriately and separately defined (Bannister, 1991). Sea ranching involves marking and releasing organisms so they can later be identified and harvested by the releasing organization. Salmon are often ranched in this fashion, while the ranching of branded cattle offers a good land-based analogy. Unlike sea ranching programs, stock enhancement is typically initiated and implemented for the public good ± no single user group is rewarded (Bannister, 1991). The commonality between sea ranching and stock enhancement is that organisms are released into an ecosystem from an external source. It should be noted that the broader definition of stock enhancement involves not only culture-based activities, but also habitat protection and modification. Specific examples are the creation of protected areas (e.g. refugia), artificial habitats (e.g. reefs), and nursery areas. These categories of enhancement are not necessarily directed toward any one species or age class, as is often the case for stocking programs. Interest in establishing stocking programs in the marine environment has developed from a variety of circumstances. The most compelling of these has been the plight of fisheries resources during the past century. According to recent statistics, among 200 marine fisheries for which data are available, approximately 60% are either fully or over-exploited (FAO, 1997). Since 1985, worldwide fish landings have stabilized at approximately 80 million metric tons (mt) with 75% of that total being food fish (New, 1997). In 1995, aquaculture production of fish and shellfish added another 21 million mt to this total. Unlike capture fisheries, which showed no increased yield, aquaculture production almost doubled between 1990 and 1995. New (1997) estimated that fishery production needs to reach 52 million mt by 2025, taking into consideration the current level of seafood consumption (approximately 13.5 kg/ caput/yr in 1995) and increasing human populations. Since production by capture fisheries is not expected to increase significantly, if at all, aquaculture must play a large role in meeting this deficit. Because the goal of stock enhancement is to use

Role of Aquaculture in the Restoration of Coastal Fisheries

315

cultured organisms to complement fishery production, it represents a bridge between capture and culture-based fisheries that will likely have some impact on the future availability of seafood. Like commercial fishermen, recreational fishers are often powerful stakeholders in fishery conservation programs. The balance of political power between recreational and commercial fishers often dictates the focus of stock enhancement programs. In the United States, the total economic impact of recreational fishing is estimated to be $108 billion ± 23% of that being attributed to saltwater fishing activity (Maharaj & Carpenter, 1996). Bartley (1999) reported that among stock enhancement programs that were not classified as experimental, approximately 12% were directed toward recreational species, 60% toward commercial species, and the remaining 28% toward species that were both recreationally and commercially valuable.

Stock enhancement ± past and present Historical perspective Surprisingly, concerns about depletion of wild fishery resources, disputes between recreational and commercial fishers, and plans to establish hatcheries for stock enhancement were reported as early as 1870 in the United States (Bowen, 1970). During this period marine hatcheries were also established in Canada, France, Great Britain, New Zealand, and Norway. Stocking programs of this era involved stripspawning of adults from the wild and releasing fertilized eggs or newly hatched larvae into coastal waters. In addition to a number of anadromous species, early stocking programs focused on cod (Gadus morhua), haddock (Melanogrammus aeglefinus), pollack (Pollachius virens), winter flounder (Pseudopleuronectes americanus), herring (Clupea harengus), tautog (Tautog onitis) and Atlantic mackerel (Scomber scombrus) (Shelbourne, 1964; Richards & Edwards, 1985). Although biologists released millions of eggs and fry, they found no conclusive evidence to indicate that target populations were being enhanced as a result. Even without validation of success, stocking programs continued into the early 1950s. At this time, funding was eliminated and programs shut down under mounting criticism regarding the lack of evidence of efficacy. The failure of these early stocking efforts was due to two primary factors. First, technological limitations prevented culturists from growing large numbers of fish to metamorphosis, so that only eggs and larvae were released. Post-release mortality was likely very high, either from predation or from starvation. Secondly, the efficacy of the stocking efforts was not rigorously assessed. To a certain extent, this failure was again a technological limitation. To accurately evaluate the success of releases, it was necessary to mark hatchery fish so that they could be distinguished from wild fish. At that time techniques were not readily available to mark or otherwise identify released eggs and larvae. Led by Japanese scientists working with red sea bream (Pagrus major), culture of marine fishes progressed quickly from 1965 onwards. By the 1980s, British and French scientists were successful in establishing the techniques required for

316

Ecological Aquaculture

large-scale production of sea bass (Dicentrarchus labrax), gilthead sea bream (Sparus aurata), and turbot (Scophthalmus maximus) (Sorgeloos et al., 1993). With the scientific achievements in marine fish culture, particularly in larval nutrition, came a revived interest in marine stock enhancement. This interest was bolstered by the fact that the depletion of marine resources recognized in the previous century had not been corrected by traditional management strategies. Quite the contrary, technological advances in the commercial fishing industry increased harvests, and development of coastal areas severely impacted critical nursery and spawning habitats.

Modern techniques While modern science has overcome the primary technological hurdles experienced by early investigators, the potential pitfalls associated with the improper application of stock enhancement are still present and new concerns have emerged. The notion that `more is better' or `quantity over quality' is still a political reality in some areas. Some programs continue to fail in evaluating the effectiveness of hatchery releases adequately, even though numerous techniques for marking fish and other marine organisms are now available (Parker et al., 1990). A more stringent regulatory environment has impeded program development in some areas, including hatchery production. In cases where enhancement hatcheries are capable of the mass production of juveniles, new concerns have emerged about the potential genetic and ecological impacts of releasing so many hatchery-produced individuals. Notwithstanding these concerns, the revival of marine stock enhancement is moving quickly, as scientists and managers are eager to test its potential and evaluate the true concerns regarding its application. From the early controversy, and recent pilot-scale feasibility studies, guidelines for establishing a responsible, comprehensive approach to marine stock enhancement are being developed. The guidelines outlined by Blankenship & Leber (1995) include: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

prioritizing and selecting species for enhancement; developing a comprehensive species management plan; defining qualitative measures for success; applying genetic resource management; applying disease and health management; forming stock enhancement objectives and strategies using all available ecological, biological, and life-history information; identifying hatchery fish and assessing the effects of stocking; using an empirical process to define optimal release strategies; identifying economic and policy objectives; and using adaptive management.

Because of their relevance to this discussion, many of these points are covered in this chapter.

Role of Aquaculture in the Restoration of Coastal Fisheries

317

Program planning and review Prior to stocking, careful consideration and evaluation should be given as to how and why the organism will be released, the potential impacts of stocking, the goals for the stocking program, and how success will be measured. A step-wise strategy for developing and implementing a stocking program is illustrated in Fig. 11.1. The stock enhancement plan should be integrated within a more general species management plan that includes other resource management tools, as well as contingency procedures for modifying the plan as new data become available. Another important component to consider early in the planning phase is a review process that will allow an unbiased evaluation of the performance of each component of the species management plan, including releases of juveniles.

Selection of species Developing and evaluating a list of candidate species is one of the first steps in the planning process. To do this, a thorough review of the history and current status of local stocks should be conducted. The species on this list are then ranked using criteria that are developed a priori. The process of prioritizing or ranking potential species for stock enhancement is often based on a combination of economic, sociopolitical and biological factors. Given the uncertain success associated with new enhancement programs for marine species, and the costs associated with both culture operations and post-release assessment, it is not surprising that economic considerations (i.e. the economic value of the species) weigh heavily in this process. Similarly from a biological standpoint, species with a known culture potential may be favored over species that have not been cultured in order to reduce the risk of failure. Sociopolitical factors become important when multiple user groups (e.g. recreational and commercial fishers) advocate enhancement of different depleted resources. Blankenship & Leber (1995) describe a step-wise process of workshops, surveys and interviews that is designed to reduce conflict frequently associated with the species selection process. They recommend using a numerical index derived from pre-established criteria to reduce bias, as well as using a trained facilitator.

Ecological criteria for success As part of the planning and species-selection process, it is important to understand the reason for the decline in the resource, because this will have a direct impact on the success of the stocking program. Similarly, while difficult to measure, an understanding of the carrying capacity of the system is necessary in order to optimize enhancement programs. Munro & Bell (1997) point out that two basic ecological conditions must exist in order to successfully enhance fish populations through stocking. First, following natural recruitment, sufficient habitat of appropriate quality must be available to accommodate stocked juveniles. Secondly, prey abundance associated with these habitats must be sufficient to support the growth and survival of the stocked fish.

318 Ecological Aquaculture

Fig. 11.1 Suggested strategy for planning, implementing and evaluating a stocking program in order to minimize risks and maximize benefits (adapted from Cowx, 1994, and Howell, 1998).

Role of Aquaculture in the Restoration of Coastal Fisheries

319

It is often reported that habitat and trophic resources of marine ecosystems are rarely fully exploited. This statement is supported by historical patterns of recruitment where peak years (well above the average recruitment) still result in successful colonization (Munro & Bell 1997). If the species to be enhanced has been overexploited, as is often the case, then it would seem intuitive that habitat and trophic resources once occupied by wild fish would be available to support stocked fish. However, it may also be that other species in the community that were not similarly exploited now utilize these resources, making it less likely that stocking will be effective. This phenomenon of alternating species abundance has been well documented between sardines and anchovies (Lluch Belda et al., 1992). Certainly if critical habitats have been reduced or significantly altered, it may not be possible to reach historical production levels again. The challenge for stock enhancement programs is to try to understand the complexities of carrying capacity and make the best use of underutilized spatial or trophic resources. As hatchery technology improves, and greater numbers of juveniles are available for release, the potential for overstocking or saturating an environment will increase. In cases where the carrying capacity is reached because of stocking, cultured animals may displace wild ones or fail to survive (Hume & Parkinson, 1987). Densitydependent factors such as reduced growth or condition have also been reported as the carrying capacity is approached (Peterman, 1991).

Establishing objectives and measures of success The objectives and measures of success for stocking should be clearly defined before fish are released. The objectives for stocking typically fall into one of four categories: stocking to restore, mitigate, augment, or create new fisheries (Cowx, 1994; Fig. 11.1). In restoration-oriented stocking programs, the first step is to identify the factor responsible for limited population growth, and then correct it. The bottleneck may be related to anthropogenic impacts such as poor water quality or restricted migratory access. Once the bottleneck is eliminated, stocking is used to rebuild the population but is not necessary over the long term (see Richards & Rago, 1999, for case study). Mitigation programs involve a known anthropogenic impact that limits productivity but is not corrected because of sociopolitical factors. The negative impact of hydroelectric dams on salmon populations is a classic example where mitigation hatcheries are utilized. In this case, stocking is required as long as the bottleneck exists. Stocking programs with an enhancement or augmentation objective are the most common, especially in the marine environment. In this case, the goal is to enhance or supplement an existing stock for which numbers are below the carrying capacity of the system. Unlike restoration programs, in this case the limitation to population growth may not be clearly identified. The use of stocking to create new fisheries was widely practiced in the past, particularly in freshwater systems, but is becoming less common today because of the potential for negative ecological impacts. The indicators used to measure the success of a stocking program will vary according to the specific objectives described above. Regardless of the objective, three

320

Ecological Aquaculture

pieces of information are required to fully evaluate the success of enhancement. This information includes: (1) the cost to produce fish to release size; (2) the number of stocked fish surviving to harvest; and (3) the value of the harvested fish. During the planning phase, the specific indicators and techniques used to quantitatively assess success should be identified and stated precisely for each objective. For example, one objective might be to attain a specific percentage increase in landings within a specified period of time (Blankenship & Leber, 1995).

Pre-release evaluations and feasibility assessment The targeted population and its native habitat must be well understood before initiating a culture program and releasing fish. Knowledge of historical patterns of distribution (habitat use) and abundance for different life stages, as well as interactions with predator, prey and competitor species, is necessary in the planning process. This information will help the evaluation of feasibility, program implementation (e.g. broodstock collection and management, designing release strategies), and impact assessment (e.g. positive or negative interaction with native species).

Genetic considerations The potential impact cultured fish impose on the genetic integrity of native populations is one of the greatest concerns associated with stocking programs, especially those programs involving small populations (see Bert et al., Chapter 3). Concerns related to genetic quality assurance in stocking programs have been reviewed and evaluated by a number of investigators (Brannon, 1993; Joerstad et al., 1994; Naevdal, 1994; Bartley et al., 1995; Shaklee & Bentzen, 1998; Tringali & Bert, 1998). The rapid growth and improvement of genetic identification techniques is resulting in a corresponding increase in this body of literature. Because the importance of genetic quality assurance is now clear, guidelines for maintaining genetic integrity among enhanced populations have been developed (Kapuscinski & Jacobson, 1987; FAO, 1992). A good understanding of the genetic population structure for each species is important because the seed for enhancement programs typically comes from captive adults. Broodfish should be collected from within the population that is targeted for enhancement. This approach facilitates local adaptation and survival of the progeny and promotes genetic integrity among interbreeding wild and hatchery-reared fish (Brannon, 1993; Conover, 1998; Utter, 1998). In cases where the targeted population no longer exists, geneticists recommend using adults from as similar an environment as possible (Travis et al., 1998). The genetic population structure of marine animals tends to be more homogenous than freshwater or anadromous species. The lack of genetic differentiation is primarily attributed to patterns of egg and larval dispersal. Marine organisms often have relatively long planktonic stages of development. Ocean currents and a lack of physical barriers facilitate the dispersal of these planktonic stages in the ocean (Bartley et al., 1995; Conover, 1998). However, clinal differences in gene frequency

Role of Aquaculture in the Restoration of Coastal Fisheries

321

have been observed in some marine species, making it necessary to evaluate each target species independently (King et al., 1995; Tringali & Bert, 1998). Responsible stocking practices require a thorough understanding of the genetic characteristics of the fish being released. The size of the breeding population must be evaluated on a species-by-species basis, and the relative contribution of each potential parent must be assessed. Allendorf & Ryman (1987) recommend a minimum broodstock population of 100 males and 100 females when information regarding genetic diversity and spawning habits is limited but adult fish are plentiful. Geneticists also recommend that culturists encourage single pair matings among their broodstock and that eggs be collected equally from all females in the population. Using empirical data from both wild and hatchery-reared white seabass, Bartley et al. (1995) recommended a minimum broodstock population size of 150 adults. These investigators implemented a broodstock management plan that included 200 adult fish and provisions for rotating males among maturation pools and integrating new stock into the captive-breeding program (Figs 11.2, 11.3). A similar broodstock management plan was established in Texas for red drum (McEachron et al., 1995).

Bioeconomic modeling A bioeconomic model is a useful tool for evaluating enhancement programs before, during and after implementation of stocking. As the name suggests, this type of model

Fig. 11.2 Schedule of broodstock rotation and replacement as part of an overall broodstock management plan. Cylinders represent mass spawn pools maintained on controlled temperature and day length cycles with 25 males and 25 females (adapted from Bartley et al., 1995).

322

Ecological Aquaculture

Fig. 11.3 Researchers work on a sedated seabass broodfish collected from the wild. Prior to being added to the breeding stock, each fish is biopsied to determine its sex, weighed and measured to assess growth, passive induced transponder tagged for physical identification, and fin clipped for genetic analyses.

combines biological criteria (growth, survival, etc.) with economic ones (cost to culture, harvest product value, etc.). Before stocking is initiated, a bioeconomic model can help evaluate the likelihood of success in economic terms. Once an enhancement program is initiated, bioeconomic models can be used as a simulation tool to identify specific areas within the stocking program where improvements will result in a higher rate of return. Bioeconomic models can be used to set goals (i.e. specific targets for cost to benefit) and should be refined as new data is available. Outputs from the model in the form of cost-to-benefit ratios represent an excellent measure of the program success (Sproul & Tominaga, 1992; Ungson et al., 1993; Hilborn, 1998). It should be emphasized that the output from any model is only as good as the information entered into it. Care must be taken to select a model that is sufficiently detailed and that represents the components of the stocking program accurately. A model that does not have these features can produce very misleading results.

Pilot-scale studies to optimize post-release survival Marking techniques A thorough review of marking techniques used to evaluate stocking success is beyond the scope of this work. However, it is important to note that a wide variety of tools have been developed in recent years that allow researchers to mark individuals within the full range of life history stages for almost any species (Parker et al., 1990). Access to and implementation of these tools is critical to understanding and evaluating stocking efforts, and to assure optimum success of the releases. The choice of marking technique is dictated by the age, size and species involved, as well as the specific objectives of the stocking program. Chemical markers, including oxytetracycline and calcein, can be used effectively to mark animals as early as the egg and larval stages (Wilson et al., 1987; Monaghan, 1993; Brooks et al., 1994; Secor et al., 1995; Mohler, 1997). The primary advantages to this technique are its ability to mark small life stages, and its ease of application as a bath or in the feed. Genetic

Role of Aquaculture in the Restoration of Coastal Fisheries

323

marking techniques are also being developed for stocking programs (Murphy et al., 1983; Joerstad et al., 1991; Gaffney et al., 1996; Kristiansen et al., 1997; Wilson et al., 1997). The primary benefit of this technique is that it allows assessment of the reproductive contribution of stocked fish, as genetic markers are passed down to subsequent generations. As illustrated in Fig. 11.4, a variety of physical tags and marking techniques have been developed for larger fish, including those that are applied internally such as coded wire, passive induced transponder, or visible implants; or externally such as fin clipping, anchor-type, or branding (Bergman et al., 1992; Guy et al., 1996). The specific objectives of the study will dictate which type of tag is most appropriate. Tags differ in the degree of invasiveness to the organism and the longevity as a mark. Consideration must also be given to the techniques and costs associated with implanting and later identifying the mark, the amount of information the tags encode, and any bias associated with results.

Hatchery techniques ± fitness considerations Successful, responsible stocking practices require that the organisms being released are of the highest quality possible. Individuals whose behavior or health is compromised will likely starve or be eaten, or may transmit diseases to wild fish. Therefore, hatchery managers must consider factors that will affect post-release performance, and develop techniques to minimize their impact. Culture techniques used in hatcheries that produce animals for market sale may not be adequate for those that will be released into the ocean.

Fish health and nutrition Good husbandry practices that result in vigorous, healthy animals are required to produce maximum post-release survival of stocked fish and to minimize potential impacts to native species. The goal of any stocking program should be to release organisms that are as similar to wild conspecifics as possible. The similarities should include physical, physiological, and biochemical characteristics, and parasite load. This is obviously a difficult goal and although deviations from `normal' are commonly reported in the literature, the implications of these differences are not always clear. Physical differences range from subtle deviations in meristics, morphometrics and color patterns, to the grossly abnormal (Fig. 11.5). The latter have obvious relationships to decreased fitness (Balbontin et al., 1973; Blaxter, 1976; Fukuhara, 1990; Stoner & Davis, 1994; Ellis et al., 1997a). Physical malformations are most often attributed to improper nutrition or sub-optimal holding conditions (e.g. stressful lighting regimes, high stocking densities, poor water quality). Reduced fitness of stocked fish has also been attributed to low stress tolerance caused by improper nutrition and chemical imbalances (Howell, 1994; Olla et al., 1994, 1998; Wallin & Van den Avyle, 1995). Reduced tolerance to stress is particularly important at the time of release when stress levels are high due to handling and transport conditions.

324 Ecological Aquaculture

Fig. 11.4 Different tags used historically on fish and their attachment sites. Approximate minimum tag sizes are given for scaling (adapted from Guy et al., 1996, with permission, from the original image by Wydoski & Emery, 1983).

Role of Aquaculture in the Restoration of Coastal Fisheries

Fig. 11.5 Two species of organisms being evaluated for enhancement, demonstrating clear differences between hatchery and wild individuals. (a) Cultured conch on the left has poorly developed shell compared with wild counterpart on the right (photo with permission from A.W. Stoner, National Marine Fisheries Service). Note streamer tags used for marking. (b) Cultured California halibut on the left has natural pigmentation compared with abnormal pigmentation on the right. In both cases, animals released with these abnormalities would have reduced predator avoidance capacities and therefore limited stocking effectiveness.

325

(a)

(b)

A number of cases have been reported in which diseases associated with hatcheryreared fish spread to native fish, sometimes with devastating results (Heggberget et al., 1993; Hindar & Jonsson, 1995). Transfer of disease can occur through the release or escape of unhealthy hatchery stock, or from cage and farm effluents. Conversely, the transfer of disease from wild to hatchery-reared organisms is also possible, especially in cases where animals are transferred from a relatively sterile, hatchery environment into the natural environment. In this case, the danger may not be immediate for the wild population, but the stocked or caged animals may act as a reservoir for the proliferation of the particular pathogen (Cowx, 1994). Disease outbreaks occur when typically stable but delicate relationships among the host animal, pathogen and the environment become imbalanced. For example, poor water quality (environment) may lead to stressed animals (host) and increased susceptibility and proliferation of disease (pathogens). Within this triangle, pathogenic organisms can often be isolated and identified, and therefore they receive a great deal of attention. However, a thorough understanding of the interrelationships between the host and its environment is generally lacking, especially for wild populations (Hedrick, 1998). Diseases occur in natural populations of aquatic organisms, but are

326

Ecological Aquaculture

poorly documented. A better understanding of disease processes is required to effectively and responsibly manage disease interactions within culture-based fisheries (Coutant, 1998). High quality seed animals can be produced by applying sound genetic principles, using modest stocking densities, and providing a high-quality diet. It may also be advantageous to modify larval rearing techniques in order to cull out genetically inferior groups of individuals early in the culture process, similar to what would occur naturally in the wild. Promoting survival among only the hardiest larvae would likely reduce the incidence of disease outbreaks in later life-history stages and allow maximum post-release survival among stocked fish. These modifications might include reducing the use of antibiotics in larval rearing systems (Coutant, 1998), modifying filtration systems so the water is not completely sterile, and purposefully exposing larvae to natural bacteria or `probiotics' that are carefully selected (Kennedy et al., 1998). To minimize the risk of releasing inferior or compromised organisms, a certified professional should inspect each batch of fish prior to transfer or release.

Fish behavior, conditioning and performance Behavioral differences between hatchery-reared and wild organisms have been reviewed by a number of investigators (Blaxter, 1976; Howell, 1994; Olla et al., 1994, 1998). Variations in behavior related to feeding (Ersbak & Hasse, 1983; Iglesias & Rodriguez-Ojea, 1994; Steingrund & Fernoe, 1997), migration (Jonsson et al., 1991), schooling, and predator avoidance (Schiel & Welden, 1987; Stoner & Davis, 1994; Ellis et al., 1997b) have been reported. Because behavioral processes are complex, the effects that behavioral differences impose on the fitness of stocked organisms are not always clear. For example, differences in feeding may involve the mode of prey capture (Steingrund & Fernoe, 1997), the diurnal pattern of feeding (Howell, 1994), or the type and quantity of prey captured (Nordeide & Salvanes, 1991; Kristiansen & Svasand, 1992). Deficits in any one of these facets of behavior will negatively impact the fitness of stocked fish by either increasing the expenditure or decreasing the intake of energy, or by increasing the risk of predation due to poor foraging skills. Because behavioral deficits are often a by-product of traditional culture practices, there is a need to better evaluate the impacts and, if possible, mitigate them. The performance of stocked animals may be improved by using more extensive rearing techniques and by conditioning animals prior to release. For example, Olla et al. (1994) demonstrated that fish exposed to predators in the laboratory (experienced) survived better upon release than fish with no prior exposure (naõÈ ve). While many predator avoidance and feeding behavioral characteristics are innate, a controlled learning period often is required to successfully implement them (Ellis et al., 1997b; Steingrund & Fernoe, 1997). There is a growing body of evidence to suggest that culturing organisms under more natural conditions for at least a short period prior to release will have a significant, positive impact on post-release survival and the ability of cultured fish to integrate with wild conspecifics. More `natural' culture conditions include lower stocking densities, more variable water quality, and exposure to natural substrates or

Role of Aquaculture in the Restoration of Coastal Fisheries

327

structure, prey and predators. The combination of these modifications will increase feeding success, and reduce stress and predation (Howell, 1994; Olla et al., 1994; Maynard et al., 1995; Tanaka et al., 1998; Berejikian et al., 2000). In California, white seabass destined for stocking are acclimated to natural conditions for 5±6 months prior to release by growing them in net cages in protected embayments (Kent & Drawbridge, 1999; Fig. 11.6).

Fig. 11.6 Cage system used to grow-out and acclimate white seabass prior to release. Cage is located in Marina del Rey, California, and is owned and operated by volunteers from a local fishing club.

Release strategies ± biological and ecological considerations After the culture protocols are developed, strategies for releasing the animals must be evaluated. While the principal goal of the culture program is to produce animals of the highest quality, release programs are designed to achieve the highest rate of recruitment possible. The size and numbers of organisms to release, and the season and habitat in which to release them are the primary factors that can be manipulated to optimize recruitment success. It is important to note that there may be strong interactive effects among these components, so release strategies must be developed accordingly. Similarly, interspecific differences require that each species be evaluated separately.

Fish size at release The most appropriate or `optimum' size to release fish should be empirically defined prior to full implementation of any stocking program. Leber (1995) defined the `optimum release size' as the size of fish at which stocking results in the greatest rate of return or recruitment. Intuitively one would think that the larger the organism is at release, the greater its chance of survival. In fact, this has been demonstrated empirically for a number of different species but may not always be the case when interactive effects of release season or habitat are considered (Tsukamoto et al., 1989; Ray et al., 1994; Yamashita et al., 1994; Leber, 1995; Willis et al., 1995). In addition to the size-specific aspects of growth on survival, the optimum release size is also commonly evaluated in relation to the economic considerations associated with

328

Ecological Aquaculture

raising organisms to a larger size. That is, the benefits of releasing fish at a larger size (i.e. increased post-release survival) are typically weighed against the added costs of the extended growout (Kent et al., 1995). In addition to the optimum release size, Leber et al. (1995) also define a `critical release size' as `the size at release below which the probability of survival to reproductive size approaches zero'. In that study involving striped mullet, the critical size was 70 mm total length (TL) for summer releases. While this is only one example, it is not difficult to understand why early marine stocking programs, releasing eggs and larvae, were likely to be very ineffective.

Season of release The season or timing of release is another important factor to consider when refining release strategies. Stocking of organisms is possible year-round for many species, even at the same target size, because spawning can be induced artificially and out of phase with wild populations. Similarly, growth rates in the hatchery environment can be manipulated to bring the development of organisms into or out of phase with wild counterparts. Stocking success is often dictated not only by the size of the animal at release, but also by the timing of the release. In fact, these two parameters are usually highly correlated (Hume & Parkinson, 1988; Willis et al., 1995; Leber et al., 1997, 1998). Return rates for organisms released during different times of year can vary widely. For example, Willis et al. (1995) reported a rate of return greater than 800% for red drum stocked in phase with wild fish compared with similar sized individuals stocked six months out of phase. Leber et al. (1997) demonstrated that the timing of releases can have a dramatic effect on recapture rates among groups of fish released at different sizes. While few small mullet (<70 mm TL) survived summer releases, the rate of recapture following springtime releases was significant. As an explanation for these results, the authors provided evidence that successful recruitment of stocked fish was associated with their ability to mix with similar sized wild fish.

Release habitat In order to identify appropriate habitats for releases, one should determine the habitats used by wild conspecifics of similar life stage to those that will be stocked. Since organisms are typically stocked as juveniles, the most appropriate release areas are often nursery areas (Fig. 11.7). Munro & Bell (1997) emphasized that subtle differences often exist among microhabitats, so field trials should be conducted to empirically define the suitability of each habitat. Even known historical nursery grounds should be evaluated due to the dynamics of these systems and changes caused naturally or by human activity. Thorough habitat assessments are especially important for organisms that are not highly mobile such as mollusks (Stoner, 1994). Once the suitability of different release habitats is understood and this understanding is applied, the results can be dramatic. For example, Leber et al. (1996) reported a 600% increase in recapture rate of striped mullet over previous trials when

Role of Aquaculture in the Restoration of Coastal Fisheries

329

Fig. 11.7 Researchers and volunteers count and then release batches of tagged white seabass into an embayment in southern California.

releases were focused in areas with freshwater influx from streams. When fish were not stocked into a nursery area, return rates were as low as 0%.

Stocking density Stocking at an appropriate density for a given habitat is an important part of optimizing the efficiency of any stocking program. This requires a good understanding of the carrying capacity of the area to be enhanced. Stocking at too high a density (i.e. above the carrying capacity) will increase competitive interactions with conspecifics and other species, resulting in decreased growth and survival, and displacement from the nursery area. Although understocking an area does not maximize stocking efforts, it is a good approach when carrying capacity is not well understood because the risk of negative ecological impacts is reduced. A good understanding of historical recruitment patterns, including year-to-year fluctuations, will facilitate the development of initial stocking density estimates. Traditional yield-per-recruit models can be adapted for enhancement programs to help predict the contribution of hatchery releases at different levels of stocking (Polovina, 1990, 1991). Modeling simulations can then be supported by release experiments designed specifically to test densityrelated assumptions (Leber et al., 1995).

Program implementation and monitoring The stocking program can be fully implemented once the release strategies have been optimized. Hatchery production can be scaled to meet targeted release size and

330

Ecological Aquaculture

stocking density. Operational considerations such as broodstock collection and management, and pre-release conditioning can also be accommodated. Releases of hatchery fish should be monitored and evaluated for the life of the program (Fig. 11.8). Continuous monitoring will allow the program to react to natural and anthropogenic factors such as fluctuating environmental conditions and changing management strategies that may affect target population levels. Ongoing monitoring programs are useful for identifying any unanticipated ecological or genetic impacts that may occur so that adjustments can be made as quickly as possible.

Fig. 11.8 A researcher scans white seabass for the presence of coded wire tags at a local fish market. Tissue samples and otoliths are also collected at this time for genetic analyses and aging studies.

An evaluation of modern marine stock enhancement programs Are modern-day marine stocking programs successful? As expected, the reviews are mixed. It often depends on whom you ask and what your definition of success is. In many cases it is too early to draw conclusions because of the complexity and longterm nature of the evaluation process. In other cases, appropriate evaluation tools are not in place to allow for unbiased assessment of success. As more programs employ a scientific approach, it is becoming easier to evaluate performance ± the performance often being measured in economic terms. In an external review of eight marine stocking programs (including three for salmonids), Hilborn (1998) reported only one program (Japanese chum salmon) that could clearly be described as economically successful. Stocking programs for pink salmon in the US, chinook and coho salmon in the US and Canada, lobster in the UK and France, and cod in Norway were not economically viable according to this review. In another recent review of eight marine species stocked in Japan, three were reported to economically increase net fishery production (Kitada, 1999). Enhancement was successful for chum salmon (in agreement with Hilborn, 1998), scallop, and red sea bream. Enhancement of flounder appears to have great potential and is economically successful in some areas (Kitada et al., 1992; Masuda & Tsukamoto, 1998). Stocking of Kuruma prawns, swimming crabs, abalone and sea urchins was reported to be uneconomical at this time. Kitada (1999) acknowledged that there were wide

Role of Aquaculture in the Restoration of Coastal Fisheries

331

variations in stocking effectiveness between prefectures for all species except chum salmon and scallops. It is clear from these reviews that aquaculture can play a significant role in the restoration and sustained viability of coastal fisheries. Responsible stocking practices supported by an experimental approach are helping to critically evaluate and improve the performance of marine enhancement programs worldwide.

References Allendorf, F.W. & Ryman, N. (1987) Genetic management of hatchery stocks. In: Population Genetics and Fishery Management (eds N. Ryman & F. Utter), pp. 141±159. University of Washington Press, Seattle. Balbontin, F., De Dilva, S.S. & Ehrlich, K.F. (1973) A comparative study of anatomical and chemical characteristics of reared and wild herring. Aquaculture, 2, 217±240. Bannister, R.C.A. (1991) Stock enhancement (workshop report). In: The Ecology and Management Aspects of Extensive Mariculture (ed. S.J. Lockwood), pp. 191±192. ICES, Nantes, 20±23 June 1989. Bartley, D.M. (1999) Marine ranching: a global perspective. In: Stock Enhancement and Sea Ranching (eds B.R. Howell, E. Moksness & T. Svasand), pp. 79±90. Fishing News Books, Oxford. Bartley, D.M., Kent, D.B. & Drawbridge, M.A. (1995) Conservation of genetic diversity in a white seabass hatchery enhancement program in southern California. American Fisheries Society Symposium, 15, 249±258. Berejikian, B.A., Tezak, E.P., Flagg, T.A., LaRae, A.L., Kummerow, E. & Mahnken, C.V.W. (2000) Social dominance, growth, and habitat use of age-0 steelhead (Oncorhynchus mykiss) grown in enriched and conventional hatchery rearing environments. Canadian Journal of Fisheries and Aquatic Sciences, 57(3), 628±636. Bergman, P.K., Haw, F., Blankenship, L.H. & Buckley, R.M. (1992) Perspectives on design, use, and misuse of fish tags. Fisheries, 17, 20±25. Blankenship, H.L. & Leber, K.M. (1995) A responsible approach to marine stock enhancement. American Fisheries Society Symposium, 15, 167±175. Blaxter, J.H.S. (1976) Reared and wild fish ± how do they compare? In: Tenth European Symposium on Marine Biology, Ostend, September 17±23, 1975 (eds G. Persoone & E. Jaspers), pp. 11±26. Bowen, J.T. (1970) A history of fish culture as related to the development of fishery programs. American Fisheries Society Symposium, 7, 71±94. Brannon, E.L. (1993) The perpetual oversight of hatchery programs. Fisheries Research, 18, 19±27. Brooks, R.C., Heidinger, R.C. & Kohler, C.C. (1994) Mass-marking otoliths of larval and juvenile walleyes by immersion in oxytetracycline, calcein, or calcein blue. North American Journal of Fisheries Management, 14, 143±150. Conover, D.O. (1998) Local adaptation in marine fishes: evidence and implications for stock enhancement. Bulletin of Marine Science, 62, 477±494. Coutant, C. C. (1998) What is `normative' for fish pathogens? A perspective on the controversy

332

Ecological Aquaculture

over interactions between wild and cultured fish. Journal of Aquatic Animal Health, 10, 101±106. Cowx, I.G. (1994) Stocking strategies. Fisheries Management and Ecology, 1, 15±30. Ellis, T., Howell, B.R. & Hayes, J. (1997a) Morphological differences between wild and hatchery-reared turbot. Journal of Fish Biology, 50, 1124±1128. Ellis, T., Howell, B.R. & Hughes, R.N. (1997b) The cryptic responses of hatchery-reared sole to a natural sand substratum. Journal of Fish Biology, 51, 389±401. Ersbak, K. & Hasse, B.L. (1983) Nutritional deprivation after stocking as a possible mechanism leading to mortality in stream-stocked brook trout. North American Journal of Fisheries Management, 3, 142±151. FAO (1992) Report of the expert consultation on utilization and conservation of aquatic genetic resources. FAO Fisheries Report 491, Food and Agriculture Organization of the United Nations, Rome. FAO (1997) Review of the state of world fishery resources: marine fisheries. FAO Fisheries Circular. 920, Food and Agriculture Organization of the United Nations, Rome. Fukuhara, O. (1990) Importance of qualitative evaluation of hatchery-bred fish for aquaculture. In: Proceedings of the Fifteenth U.S.±Japan Meeting on Aquaculture: Marine Farming and Enhancement (ed. A.K. Sparks), pp. 117±121. NOAA Technical Report NMFS 85, Kyoto, Japan, 1986. Gaffney, P.M., Rubin, V.P., Hedgecock, D., Powers, D.A., Morris, G. & Hereford, L. (1996) Genetic effects of artificial propagation: signals from wild and hatchery populations of red abalone in California. Aquaculture, 143, 257±266. Guy, C.S., Blankenship, L.H. & Nielsen, L.A. (1996) Tagging and marking. In: Fisheries Techniques, 2nd edition (eds B.R. Murphy & D. Willis), pp. 353±383. American Fisheries Society, Bethesda, Maryland. Hedrick, R.P. (1998) Relationships of the host, pathogen, and environment: implications for diseases of cultured and wild fish populations. Journal of Aquatic Animal Health, 10, 107±111. Heggberget, T.G., Johnsen, B.O., Hindar, K. et al. (1993) Interactions between wild and cultured Atlantic salmon: a review of the Norwegian experience. Fisheries Research, 18, 123±146. Hilborn, R. (1998) The economic performance of marine stock enhancement projects. Bulletin of Marine Science, 62, 661±674. Hindar, K. & Jonsson, B. (1995) Impacts of aquaculture and hatcheries on wild fish. In: Protection of Aquatic Biodiversity. Proceedings of the World Fisheries Congress (eds D.P. Philipp, J.M. Epifanio, J.E. Marsden & J.E. Claussen), pp. 70±87. Science Publishers, Inc. Lebanon, USA. Howell, B. (1998) Marine fisheries. In: Stocking and Introduction of Fish. (ed. I. G. Cowx), pp. 289±293. Fishing News Books, Oxford. Howell, B.R. (1994) Fitness of hatchery-reared fish for survival in the sea. Aquaculture and Fisheries Management, 25, 3±17. Hume, J.M.B. & Parkinson, E.A. (1987) Effect of stocking density on the survival, growth, and dispersal of steelhead trout fry (Salmo gairdneri). Canadian Journal of Fisheries and Aquatic Sciences, 44, 271±281. Hume, J.M.B. & Parkinson, E.A. (1988) Effects of size at and time of release on the survival

Role of Aquaculture in the Restoration of Coastal Fisheries

333

and growth of steelhead fry stocked in streams. North American Journal of Fisheries Management, 8, 50±57. Iglesias, J. & Rodriguez-Ojea, G. (1994) Fitness of hatchery-reared turbot, Scophthalmus maximus L., for survival in the sea: first year results on feeding, growth and distribution. Aquaculture and Fisheries Management, 25(Supplement 1), 179±188. Joerstad, K.E., Skaala, O. & Dahle, G. (1991) The development of biochemical and visible genetic markers and their potential use in evaluating the interaction between cultured and wild fish populations. In: The Ecology and Management Aspects of Extensive Mariculture (ed. S. J. Lockwood), pp. 200±205. ICES, Nantes, June 20±23, 1989. Joerstad, K.E., Paulsen, O.I., Naevdal, G. & Thorkildsen, S. (1994) Genetic studies of cod, Gadus morhua L., in Masfjorden, western Norway: comparisons between the local stock and released, artificially reared cod. Aquaculture and Fisheries Management, 25, 77±91. Jonsson, B., Jonsson, N. & Hansen, L.P. (1991) Differences in life history and migratory behaviour between wild and hatchery-reared Atlantic salmon in nature. Aquaculture, 98, 69±78. Kapuscinski, A.R. & Jacobson, L.D. (1987) Genetic Guidelines for Fisheries Management. Sea Grant Research Report 17, University of Minnesota, Duluth. Kennedy, S.B., Tucker, J.W., Neidig, C.L. et al. (1998) Bacterial management strategies for stock enhancement of warmwater marine fish: a case study with common snook (Centropomus undecimalis). Bulletin of Marine Science, 62, 573±588. Kent, D.B., Drawbridge, M.A. & Ford, R.F. (1995) Accomplishments and roadblocks of a marine stock enhancement program for white seabass in California. American Fisheries Society Symposium, 15, 492±498. Kent, D.B. & Drawbridge, M.A. (1999) Developing a marine ranching programme: A multidisciplinary approach. In: Marine Ranching: Global Perspectives with Emphasis on the Japanese Experience, pp. 66±78. FAO Fisheries Circular 943. King, T.L., Ward, R., Blandon, I.R., Colura, R.L. & Gold, J.R. (1995) Using genetics in the design of red drum and spotted seatrout stocking programs in Texas: a review. American Fisheries Society Symposium, 15, 499±502. Kitada, S. (1999) Effectiveness of Japan's stock enhancement programs: current perspectives. In: Stock Enhancement and Sea Ranching (eds B.R. Howell, E. Moksness & T. Svasand), pp. 103±131. Fishing News Books, Oxford. Kitada, S., Taga, Y. & Kishino, H. (1992) Effectiveness of a stock enhancement program evaluated by a two stage sampling survey of commercial landings. Canadian Journal of Fisheries and Aquatic Sciences, 49, 1573±1582. Kristiansen, T.S. & Svasand, T. (1992) Comparative analysis of stomach contents of cultured and wild cod, Gadus morhua L. Aquaculture and Fisheries Management, 23, 661±668. Kristiansen, T.S., Jorstad, K.E., Ottera, H., Paulsen, O.I. & Svasand, T. (1997) Estimates of larval survival of cod by releases of genetically marked yolk-sac larvae. Journal of Fish Biology, 51(Supplement A), 264±283. Leber, K.M. (1995) Significance of fish size-at-release on enhancement of striped mullet fisheries in Hawaii. Journal of the World Aquaculture Society, 26, 143±153. Leber, K.M., Brennan, N.P. & Arce, S.M. (1995) Marine enhancement with striped mullet: Are hatchery releases replenishing or displacing wild stocks? American Fisheries Society Symposium, 15, 376±390.

334

Ecological Aquaculture

Leber, K.M., Arce, S.M., Sterritt, D.A. & Brennan, N.P. (1996) Marine stock-enhancement potential in nursery habitats of striped mullet, Mugil cephalus, in Hawaii. Fishery Bulletin, 94(3), 452±471. Leber, K.M., Blankenship, H.L., Arce, S.M. & Brennan, N.P. (1997) Influence of release season on size-dependent survival of cultured striped mullet, Mugil cephalus, in a Hawaii estuary. Fishery Bulletin, 95, 267±279. Leber, K.M., Brennan, N.P. & Arce, S.M. (1998) Recruitment patterns of cultured juvenile Pacific threadfin, Polydactylus sexfilis (Polynemidae), released along sandy marine shores in Hawaii. Bulletin of Marine Science, 62, 389±408. Lluch Belda, D., Schwartzlose, R.A., Serra, R., Parrish, R., Kawasaki, T., Hedgecock, D. et al. (1992) Sardine and anchovy regime fluctuations of abundance in four regions of the world oceans: A workshop report. Fisheries Oceanography, 1, 339±347. Maharaj, V. & Carpenter, J.E. (1996) The 1996 Economic Impact of Sportfishing in the United States. American Sportfishing Association, Alexandria, VA. Masuda, R. & Tsukamoto, K. (1998) Stock enhancement in Japan: review and perspective. Bulletin of Marine Science, 62, 337±358. Maynard, D.J., Flagg, T.A. & Mahnken, C.V.M. (1995) A review of seminatural culture strategies for enhancing the postrelease survival of anadromous salmonids. American Fisheries Society Symposium, 15, 307±316. McEachron, L.W., McCarty, E. & Vega, R. (1995) Beneficial uses of marine fish hatcheries: enhancement of red drum in Texas coastal waters. American Fisheries Society Symposium, 15, 161±166. Mohler, J.W. (1997) Immersion of larval Atlantic salmon in calcein solutions to induce a nonlethally detectable mark. North American Journal of Fisheries Management, 17, 751±756. Monaghan, J.P., Jr (1993) Comparison of calcein and tetracycline as chemical markers in summer flounder. Transactions of the American Fisheries Society, 122, 298±301. Munro, J.L. & Bell, J.D. (1997) Enhancement of marine fisheries resources. Reviews in Fisheries Science, 5, 185±222. Murphy, B.R., Nielsen, L.A. & Turner, B.J. (1983) Use of genetic tags to evaluate stocking success for reservoir walleyes. Transactions of the American Fisheries Society, 112, 457±463. Naevdal, G. (1994) Genetic aspects in connection with sea ranching of marine fish species. Aquaculture and Fisheries Management, 25, 93±100. New, M.B. (1997) Aquaculture and the capture fisheries ± balancing the scales. World Aquaculture, 11±30. Nordeide, J.T. & Salvanes, A.G.V. (1991) Observations on reared newly released and wild cod (Gadus morhua L.) and their potential predators. In: The Ecology and Management Aspects of Extensive Mariculture (ed. S.J. Lockwood), pp. 139±146. ICES, Nantes, June 20±23, 1989. Olla, B.L., Davis, M.W. & Ryer, C.H. (1994) Behavioural deficits in hatchery-reared fish: Potential effects on survival following release. Aquaculture and Fisheries Management, 25, 19±34. Olla, B.L., Davis, M.W. & Ryer, C.H. (1998) Understanding how the hatchery environment represses or promotes the development of behavioral survival skills. Bulletin of Marine Science, 62, 531±550. Parker, N.C., Giorgi, A., Heidinger, E.R.C., Jester, D.B., Prince, E.D. & Winans, G.A. (1990) Fish Marking Techniques. American Fisheries Society Symposium 7.

Role of Aquaculture in the Restoration of Coastal Fisheries

335

Peterman, D. (1991) Density-dependent marine processes in North Pacific salmonids: lessons for experimental design of large-scale manipulations of fish stocks. ICES Marine Science Symposium, 192, 69±77. Polovina, J.J. (1990) Application of yield-per-recruit models to fishery enhancement through juvenile releases. In: Marine Farming and Enhancement: Proceedings of the Fifteenth US± Japan Meeting on Aquaculture (ed. A.K. Sparks), pp. 29±33. NOAA, Kyoto, Japan, October 22±23, 1986. Polovina, J.J. (1991) Evaluation of hatchery releases of juveniles to enhance rockfish stocks, with application to Pacific ocean perch Sebastes alutus. US Fishery Bulletin, 89, 129±136. Ray, M., Stoner, A.W. & O'Connell, S.M. (1994) Size-specific predation of juvenile queen conch Strombus gigas: Implications for stock enhancement. Aquaculture, 128, 1±2. Richards, R.A. & Rago, P.J. (1999) A case history of effective fishery management: Chesapeake Bay striped bass. North American Journal of Fisheries Management, 19, 356±375. Richards, W.T. & Edwards, R.E. (1985) Stocking to restore or enhance marine fisheries. In: The Role of Fish Culture in Fisheries Management (ed. R. H. Stroud), pp. 75±80. American Fisheries Society Symposium, Lake Ozark, MO, March 31±April 3, 1985. Schiel, D.R. & Welden, B.C. (1987) Responses to predators of cultured and wild red abalone, Haliotis rufescens, in laboratory experiments. Aquaculture, 60, 173±188. Secor, D.H., Houde, E.D. & Monteleone, D.M. (1995) A mark-release experiment on larval striped bass Morone saxatilis in a Chesapeake Bay tributary. ICES Journal of Marine Science, 52, 87±101. Shaklee, J.B. & Bentzen, P. (1998) Genetic identification of stocks of marine fish and shellfish. Bulletin of Marine Science, 62, 589±622. Shelbourne, J.E. (1964) The artificial propagation of marine fish. Advances in Marine Biology, 2, 1±83. Sorgeloos, P., Dehasque, M., Dhert, P. & Lavens, P. (1993) Review of some aspects of marine fish larviculture. In: Mass Rearing of Juvenile Fish (eds K. Pittman, R.S. Batty & J. Verreth), pp. 138±142. ICES Marine Science Symposia Vol. 201, Bergen, Norway, June 21±23, 1993. Sproul, J.T. & Tominaga, O. (1992) An economic review of the Japanese flounder stock enhancement project in Ishikari Bay, Hokkaido. Bulletin of Marine Science, 50, 75±88. Steingrund, P. & Fernoe, A. (1997) Feeding behaviour of reared and wild cod and the effect of learning: two strategies of feeding on the two-spotted goby. Journal of Fish Biology, 51, 334±348. Stoner, A.W. (1994) Significance of habitat and stock pre-testing for enhancement of natural fisheries: experimental analyses with queen conch Strombus gigas. Journal of the World Aquaculture Society, 25, 155±165. Stoner, A.W. & Davis, M. (1994) Experimental outplanting of juvenile queen conch, Strombus gigas: comparison of wild and hatchery-reared stocks. Fishery Bulletin, 92, 390±411. Tanaka, M., Seikai, T., Yamamoto, E. & Furuta, S. (1998) Significance of larval and juvenile ecophysiology for stock enhancement of the Japanese flounder, Paralichthys olivaceus. Bulletin of Marine Science, 62, 551±572. Travis, J., Coleman, F., Grimes, C., Conover, D., Bert, T. & Tringali, M. (1998) Critically assessing stock enhancement: An introduction to the Mote Symposium. Bulletin of Marine Science, 62, 305±312. Tringali, M.D. & Bert, T.M. (1998) Risk to genetic effective population size should be an

336

Ecological Aquaculture

important consideration in fish stock-enhancement programs. Bulletin of Marine Science, 62, 641±660. Tsukamoto, K., Kuwada, H., Hirokawa, J. et al. (1989) Size-dependent mortality of red sea bream, Pagrus major, juveniles released with fluorescent otolith-tags in News Bay, Japan. Journal of Fish Biology, 35 (Supplement A), 59±69. Ungson, J.R., Matsuda, Y., Hirata, H. & Shiihara, H. (1993) An economic assessment of the production and release of marine fish fingerlings for sea ranching. Aquaculture, 118, 3±4. Utter, F. (1998) Genetic problems of hatchery-reared progeny released into the wild, and how to deal with them. Bulletin of Marine Science, 62, 623±640. Wallin, J.E. & Van den Avyle, M.J. (1995) Interactive effects of stocking-site salinity and handling stress on short-term survival of striped bass stocked in the Savannah River, Georgia±South Carolina. American Fisheries Society Symposium, 15, 15. Willis, S.A., Falls, W.W., Dennis, C.W., Roberts, D.E. & Whitchurch, P.G. (1995) Assessment of season of release and size at release on recapture rates of hatchery-reared red drum. American Fisheries Society Symposium, 15, 354±365. Wilson, C.A., Beckman, D.W. & Dean, J.M. (1987) Calcein as a fluorescent marker of otoliths of larval and juvenile fish. Transactions of the American Fisheries Society, 116, 668±670. Wilson, R.R., Jr, Donaldson, K.A., Frischer, M.E. & Young, T.B. (1997) Mitochondrial DNA control region of common snook and its prospect for use as a genetic tag. Transactions of the American Fisheries Society, 126, 594±606. Wydoski, R. & Emery, L. (1983) Tagging and marking. In: Fisheries Techniques (eds L.A. Nielsen & D.L. Johnson), pp. 215±237. American Fisheries Society, Bethesda, Maryland. Yamashita, Y., Nagahora, S., Yamada, H. & Kitagawa, D. (1994) Effects of release size on survival and growth of Japanese flounder Paralichthys olivaceus in coastal waters off Iwate Prefecture, northeastern Japan. Marine Ecological Progress Series, 105, 269±276.

Part 4

Conclusion

Chapter 12

Ecology as the Paradigm for the Future of Aquaculture Barry A. Costa-Pierce Rhode Island Sea Grant College Program `We suspect that living in true harmony with the natural world, in a manner sustainable over the long run, is something no modern human society has yet learned how to do. The survival of the natural world, however, and likely our survival as a species, depends on our learning to do this. It will be a unique experience in human history.' (Bock & Bock, 2000)

Introduction: the global protein ± ecological crisis The world is in the midst of the `greatest human migration of all time' (Rees & Wackernagel, 1996). Millions of people are moving from rural, inland areas to coastal cities. By 2015 the United Nations (1994) estimates there will be 27 megacities of 10 million or more people ± 23 of them in economically developing countries ± and most of these will be located on the coast. By 2025, the world's urban population will increase to 5.1 billion people, equivalent to the entire human population on Earth in 1930 (United Nations, 1994). To sustain these people the United Nations World Commission on Environment and Development projects a five- to tenfold expansion in industrial activities (WCED, 1987). Most of the ocean's biodiversity and productivity is concentrated near the coasts ± estimated at 80% of all living marine species. Soule (1991) predicts that if present trends continue about one-third of the species alive in 1900 will be gone by 2100. The huge human demand on the Earth's rich marine resources could lead to massive losses of biodiversity ± and the complete dismantling of coastal ecosystems ± if we defer the necessary planning needed to ensure the sustainability of both nature and millions of coastal peoples. Meeting basic human needs for protein foods in the future will be a difficult challenge. Approximately 1.3 billion people live on less than a dollar a day ± the cost of a half a pint of beer ± and half of the world's population lives on less than 2 dollars a day (Watson, 1999). Since 1950 there has been a 100% increase in the per capita demand for fish, a 40% increase for grain, and 33% for wood. FAO (2000) predicts world fish consumption to increase from 16 kg (1997) to 19±20 kg by 2030, raising total human use of aquatic foods to 150±160 million metric tons (mt). Capture

340

Ecological Aquaculture

fisheries can provide no more than 100 million mt, so the bulk of the increase will need to come from aquaculture. Most of the world's population is concentrated in Asia. As incomes increase in Asia, global demands for natural resources and protein foods increase dramatically. Meat demand is forecast to increase by 160% by 2020 in the economically developing countries. China's pork consumption increased 14% in 1995 alone (Brown et al., 1998). More comprehensive planning for these future needs is needed; but our world continues in the expansionist mode of a never-ending, bountiful Earth. We are cutting down forests to increase the area needed for plant and animal agriculture; increasing the intensification of agriculture to irrigate more land, and increasing fertilizer use; and pursuing almost every fish in the ocean. Fishery products are the world's most important source of animal protein, especially for the poor (FAO, 2000). Fish contribute more animal protein for human consumption than beef and poultry combined. The coastal economies of all nations are dependent on fishery products. An estimated 15 million people are employed directly in fishing, and millions more work in marketing, processing, and other associated manufacturing activities (FAO, 2000). Due to the exploding world population, however, demands for fishery products are greater than supplies, resulting in unemployment, price fluctuations, and grotesque market and regulatory inefficiencies. About two-thirds of the major marine fisheries are fully exploited, overexploited, or depleted (FAO, 2000). Over 25% of the world's fisheries catch are discarded (Alverson et al., 1994). Global fisheries lose an estimated $50 million per year, which are made up by ineffective government subsidies (Christy, 1997). Some fish species are in such high demand that they are being sold at outrageous prices. In the Tokyo fish market, a single bluefin tuna has sold for $173 000 (Jim McVey, personal communication). Native, wild-caught fish are being replaced in traditional markets by exotic, cultured fish with new `market names', e.g. Jamaican cultured red tilapia (Oreochromis spp.) becomes `Caribbean red snapper'. These examples are part of a growing trend ± one that will only increase in the future ± of shortages in wild fishery resources; price increases; and of replacement of wild fish by cultured fish in generic `white fish markets'. Unless we change our current practices there will have to be per capita declines in the use of fish, forest, crop and irrigated lands (Watson, 1999). We cannot catch more fish from the sea. But the world can turn to farming the waters ± not just hunting them ± and rapidly accelerate the `blue revolution'. This blue revolution cannot be a modern clone of the `green revolution' since what is required is an `evolution', not a `revolution'. This aquaculture evolution will be a modern, twentyfirst century, knowledge-based process to pioneer the development of sustainable, ecologically integrated aquaculture systems that have positive impacts on both natural and social ecosystems. What this book has described is the history, practices and pedagogy of `ecological aquaculture' necessary to develop integrated aquaculture ecosystems that enhance natural aquatic ecosystems, ecosystem services, and the social fabric of coastal societies. Scientists and policy-makers have a vital role in directing the evolution of the `blue revolution' to meet the projected protein needs in the twenty-first century.

Ecology as the Paradigm for the Future of Aquaculture

341

Ecological aquaculture, a `new' field of applied environmental scholarship (actually an `ancient' field, see Beveridge & Little (Chapter 1) and Costa-Pierce (Chapter 2), this volume) needs to emerge throughout the world at the major land and sea universities to assist aquaculture's rapid transition to social and environmental sustainability, and to integrate aquaculture into the mainstream planning for sustainable fisheries and coastal zone management. The blue revolution will go `belly up' unless it embraces a sustainable pedagogy having environmentally and socially sensitive codes of conduct that both industry and communities can accept.

The social ecology of aquaculture `Perhaps the greatest single role an ecological ethics can play is a discriminating one ± to help us distinguish which of our actions serve the thrust of natural evolution and which of them impede it. That human interests of one kind or another may be involved in these actions is not always relevant to the ethical judgments we are likely to make. What really counts are the ethical guidelines that determine our judgment.' (Bookchin, 1982) All of the world's marine and terrestrial environmental challenges for the twenty-first century are colored by the explosive growth of the world's population. We live on a `human-dominated planet' (Lubchenco, 1998) where there are no wild areas on Earth yet untouched by humans (with the possible exception of the deep, mid-oceanic realm). The world's nature `reserves' and `wilderness' areas will lose their ecological integrity unless these ecosystems are managed by humans! As a result, for the first time in the history of our species the destiny of the Earth's natural ecosystems is in the hands of our scientific, political, and social institutions. Brown et al. (2000) state, `To turn around current degradation trends, humans need to reinvent the completely ecologically dysfunctional political and economic systems of our times.' Rees (2000) states that humans need to `turn from managing the environment to managing themselves.' Ensuring that accelerated aquaculture developments be done in an ecological manner is much more than a simple technological exercise ± it is an exercise in multidisciplinary, multi-institutional environmental scholarship. Millions of people whose lives depend upon harvesting marine resources from fishing and farming require that we devise a planned system that includes them, and ensures their future. Behavioral changes will be required that can be accomplished through social investments, strategic subsidies, and market mechanisms that facilitate change in consumer behaviors. Jamieson (1996) believes the most effective strategy for sustainability is not technological, but solutions `located in their source: humans, their behavior, and their institutions'. In this regard, development of ecological aquaculture is essentially a `conscious exercise in social engineering' (Rees, 2000). Due to the urgent nature of the ocean food crisis, the viability of the everyday incomes and protein foods of millions of coastal and rural people are threatened. The

342

Ecological Aquaculture

success of aquaculture is dependent not only on its technical needs for hatcheries to produce seed, and feed mills to produce feeds, but also on markets, equipment, and the capacities and capabilities of the entire seafood infrastructure. To date, macroeconomic factors have been the main controllers of aquaculture developments, with environmental and social costs externalized (Born et al., 1994). The future challenge for planners ± who clearly need to accelerate aquaculture development ± is to plan for new production ± not only technically, but also as community development ± and consider the social ecology (Bookchin, 1982) of aquaculture developments. Proper planning for ecological aquaculture internalizes all of nature's and society's costs as part of an entire regional development activity, or `aquaculture production network' that connects aquatic seed and feed production centers and markets in order to maximize local economic multiplier effects (Fig. 12.1). Hatchery systems

Nursery systems

Ponds

Ricefields

Ponds

Grow-out systems

Markets

Ponds

Local

Ricefields Export

Tanks

Cages

Fig. 12.1 An aquaculture production network connects the resident hatchery, nursery and grow-out systems with feed sources and markets in a regional planning approach for community development in order to maximize local economic impacts.

If aquaculture is planned as grow-out operations only ± and using a feedlot concept ± then the benefits to communities are small. However, if aquaculture is planned as community-based development of a highly integrated, local operation, then employment opportunities and the potential for positive community impacts increase dramatically. Bailey (1988) states that, `Aquaculture must be understood as a human enterprise designed to meet human needs, including the need for economically viable communities, especially in rural areas where most aquaculture production occurs.' Aquaculture can play an important economic role by creating new economic niches ± generating employment in areas where there are few alternative job choices ± and providing local sources of high quality food, and opportunities for attractive investments for local entrepreneurs to invest in the local economy, thereby increasing local control over economic development (Bailey, 1997). There are many cases where aquaculture expansion has fueled the hope of fragile coastal and inland rural communities that have undergone unprecedented changes in their traditional ways of life. Aquaculture has provided significant multiplier effects

Ecology as the Paradigm for the Future of Aquaculture

343

on the local economies increasing both direct and indirect personal spending in these coastal communities (Ruddle & Zhong, 1988; Costa-Pierce, 1997a, b; Edwards & Demaine, 1997; Brummett & Costa-Pierce, Chapter 7, this volume; Rakocy, Chapter 9, this volume). And yet, communities in many parts of the world actively oppose aquacultural development because such development is perceived as a threat to local social and/or ecological systems (Bailey et al., 1996). If intensive aquaculture operations ± no matter how advanced technically ± have no community roots, and feeds, seeds, supplies, equipment and human expertise are procured from great distances from the sites of production, there will be community opposition. In these cases it is easy to see why some community members view aquaculture development as `all we get is your pollution'. Planning for aquaculture development as community development and environmental enhancement must thereby encompass regional planning processes to accommodate aquaculture's vital support industries (inputs), and the use of aquaculture resources and wastes in agriculture or in environmental enhancement projects (outputs). Regional planning for ecological aquaculture developments will have much higher positive impacts on jobs and the environment, and will eventually dissolve the opposition from communities who will see the newcomers as one of their own.

Defining ecological aquaculture I define `ecological aquaculture' as an alternative model of aquaculture research and development that brings the technical aspects of ecological principles and ecosystems thinking to aquaculture, and incorporates ± at the outset ± principles of natural and social ecology, planning for community development, and concerns for the wider social, economic, and environmental contexts of aquaculture (Kautsky & Folke, 1991; Folke & Kautsky, 1992; Grove & Edwards, 1993; Costa-Pierce, 1997a, b; Edwards & Demaine, 1997; Edwards, 1998). Ecological aquaculture research is oriented to the design, development, and monitoring of aquatic farming systems that preserve and enhance the form and functions of the natural and social environments in which they are situated. Aquaculture depends upon inputs from various food, processing, transportation and other industries, and can produce valuable, uncontaminated wastewaters and fish processing wastes, all of which can be a vital part of an ecological system that can be planned and organized for community-based aquatic foods production ± and natural ecosystem rehabilitation, reclamation and enhancement ± not degradation. Ecological aquaculture takes a global view, integrating ecological science and sharing technological information in a sophisticated, knowledge-based manner, promoting innovation and efficiency in the global marketplace and incorporating social and environmental costs ± not externalizing them. Systems ecology approaches are used to develop aquaculture production networks for the target species in a highly diversified, segmented manner, with numerous interconnections supplying inputs and outputs using local resources and recycled wastes and materials, planning for maximal job creation, and closing leaky loops of

344

Ecological Aquaculture

energy and materials that can potentially degrade natural ecosystems. Ecological aquaculture treats and recycles its own wastes rather than relying upon public subsidies, and integrates people with technologies in new synergies to create new employment and biotechnical advances with a global view, integrating ecological science and sharing technological information with innovation in the global marketplace. The six characteristics of ecological aquaculture are: l

l

l

l

l

l

Preservation of the form and functions of natural ecosystems. Sites do not disrupt or displace valuable natural ecosystems; but if localized displacement/degradation does occur, active research and development programs for ecosystem rehabilitation and enhancement are initiated and sustained. It practices trophic level efficiency as the world's most efficient protein producer, relying on plant, waste animal or seafood processing wastes, with fish meal used in the production process not as the major protein or energy source but to solve issues of diet palatability only. It practices nutrient management by not discharging any nutrient or chemical pollution, and does not use chemicals or antibiotics harmful to human or ecosystem health in the production processes. It uses native species/strains and does not contribute to `biological' pollution; but if exotic species/strains are used, complete escapement control and recovery procedures are in place, and active research and development programs provide complete documentation and public information. It is integrated with communities to maximize job creation and training for displaced `sea workers', and is a good community citizen; exporting to earn profits, but also marketing products locally to contribute to community development. It is a global partner, producing information for the world, avoiding the proprietary.

Need to incorporate ecological aquaculture into the planning for sustainable fisheries In many nations, planning for aquaculture developments is not incorporated into the overall planning framework for sustainable fisheries and coastal zone management. For example, in the USA, the Magnuson±Stevens Fishery Conservation and Management Act (Magnuson Act), passed in 1976, says nothing about aquaculture. In addition, although aquaculture expands the production of commercially valuable species, it depends upon natural ecosystems and ecosystem services. Although capture fisheries and aquaculture operations are researched, planned, and managed as if they were independent entities, they both share common concerns about genetic diversity in hatchery-raised fish, feeds, and the sustainability of fish meal/oil fisheries and industries. Recently, there have been questions as to whether aquaculture contributes to the depletion of world fisheries (Naylor et al., 1998, 2000). This `aquaculture paradox' recognizes the dependence of both wild

Ecology as the Paradigm for the Future of Aquaculture

345

and farmed fish stocks on many of the same marine and agricultural resources ± from food to habitats. Fisheries science needs to incorporate aquaculture into the longer-term outlook for managing the fisheries of the future. Analyses of the trends in aquaculture and capture fisheries are required along with in-depth examinations of interdependencies. As a example of what is required, the Lochaber Management Group was formed recently in Scotland as a pioneering effort between fishing, aquaculture, non-governmental, and local government groups to `ensure the maintenance of a healthy stock of wild fish while promoting a sustainable aquaculture industry' (FFI, 2001). Vital to establishing a framework for incorporating aquaculture into the planning for comprehensive fisheries and coastal zone management is the need to: (1)

Recognize the vital contribution of culture fisheries (aquaculture) and enhanced fisheries (ranching) to global fisheries production. FAO (2000) reports that aquaculture is the fastest growing form of global food production, accounting for over 25% of total world fish production; and aquaculture will provide most of the growth in world fisheries production over the next 50 years. There are intimate ± albeit unplanned ± connections between capture fisheries, enhanced fisheries (`ranching'), and culture fisheries (`aquaculture') (Fig. 12.2).

Capture fisheries 'hunting'

Culture fisheries 'aquaculture' Broodstock selection 'Open' aquaculture production network

Hatchery

Fry

Nursery

Fingerlings

Grow-out

Adults

Broodstock selection Enhance fisheries 'ranching' 'stock enhancement'

'Closed' aquaculture production network

Fig. 12.2 Dynamic interactions of the subfields of modern fisheries. A `closed' aquaculture production network refers to the completion of the entire life cycle of a plant or animal in captivity, whereas an `open' aquaculture production network still relies upon broodstock captured from natural sources (e.g. much of Chinese pond culture of carps still relies upon wild caught broodstock and fingerlings taken from rivers).

346

(2) (3)

Ecological Aquaculture

Properly value natural ecological resources and services (Folke et al., 1998; Naylor et al., 2000). Use market and tax incentives to enhance aquaculture's efficiencies; to eliminate wastes; and to improve aquaculture's economic and social returns in order to protect ecosystems and ecosystem services. For example, one radical method to control overexploitation is to decrease demands for seafood in the rich countries where demands for aquatic proteins are no longer controlled by price, e.g. the rich countries can pay for what they want, regardless of the price. Shortages of supplies will teach consumers in the rich countries to change their behaviors.

Choosing the most efficient protein-production system for a crowded planet Aquaculture is the fastest growing sector of the world food economy, growing at 11% per year in the past 10 years, from 13 to 31 million mt (FAO, 2000; Worldwatch Institute, 2001). At this rate of growth, by 2010, aquaculture will overtake cattle ranching (Worldwatch Institute, 2001). Aquaculture is most developed in Asia ± especially China ± but aquaculture is also growing rapidly in India, Vietnam, Bangladesh, Indonesia, and Thailand. Major increases in aquaculture production are also projected for numerous Latin and South American nations (Fitzsimmons, 2000). Aquaculture is growing more slowly in Europe and North America due to user conflicts, high production costs, labor and market constraints, and regulatory and environmental issues. Of the rich nations, Japan is the largest and most diverse producer (0.8 million mt of scallops, oysters, yellowtail tuna, etc.). The US production is mostly catfish (0.45 million mt). Canada, Norway, and the UK produce mostly salmon. Since 1990 there has been little growth in either free-range beef production or ocean capture fisheries; and both are reaching their productive limits. Between 1950 and 1990 beef production ± four-fifths of it from range lands ± nearly tripled from 19 to 53 million mt (Worldwatch Institute, 2001). Fish catches grew from 19 to 86 million mt during this period but have not grown since. A world that is reaching resource limits on oceanic fisheries and range lands while adding 80 million people each year needs efficient new sources of animal protein such as aquaculture. The great environmental and social issues of the twenty-first century are all interconnected. Evolving ecological aquaculture will require the use of modern ecological and social science and participatory technology development methods (CostaPierce, this volume). Most scientists and policy-makers concerned about the future of aquaculture deal with natural and human resource management issues in isolation. Rapid evolution of ecological aquaculture will require a much tighter coupling of science, policy, and management. Aquaculture has intimate connections not only with capture fisheries, but also with agriculture, markets, and marine policy and regulatory environments. Accelerated production of aquatic proteins cannot be evaluated in a vacuum separate from other

Ecology as the Paradigm for the Future of Aquaculture

347

types of land-based animal agriculture or ocean-based capture fisheries since these food systems use similar (and sometimes competing) inputs and outputs, face similar policy and regulatory environments, and have to deal with common consumers and decision-makers. More holistic planning perspectives are needed to ensure the survival of traditional coastal fishing and aquaculture communities, and to link aquaculture science, industry, and society in order to design effective policies, practices and technologies to address the many challenges ahead. In the last decade there have been concerns raised about aquaculture developments from the very communities that should be its natural supporters ± coastal communities and environmentalists (Goldburg & Triplett, 1997; Scherr, 1997; Table 12.1). There is a reason for much concern, since the rapid changes occurring in the coastal zones of the world threaten the traditional livelihoods of many fishing/aquaculture communities, and could threaten ecosystems and ecosystem services (Naylor et al., 1998, 2000). `Trophy' homes beside wildlife refuges, and lawsuits against the `look' (and smell) of traditional fishing villages are occurring more frequently. However, Schrecker (2000) states that the position of `let them work in ecotourism' used by affluent urban environmentalists is unmatched by a commitment to alternative economic development strategies that would generate comparable sources of employment at comparable or higher wages. He states that `Such claims of moral authority are tenuous even as applied to environmental politics in rich countries, and even more tenuous in the global context.' Accelerated aquaculture developments need to be ± in both perception and reality ± an ecologically and socially appropriate means of sustaining coastal communities, not `industrial' developments which threaten the sustainability of marine ecosystems and coastal communities. In this regard, ecological aquaculture can be an important means to rescue traditional coastal fishing communities from becoming `museum curios' in low paying, poorly planned, ecotourism schemes. Constraints to accelerating aquaculture development can be overcome by accelerating applied research in ecological aquaculture, which will require scientists to embrace a pedagogy that makes seamless connections between research, extension/

Table 12.1 Non-governmental environmental organizations concerned about unsustainable aquaculture Organizations

Web sites

Greenpeace

http://www.greenpeace.org

International Shrimp Action Network; Shrimp Sentinel Online

http://www.earthsummit.org/shrimp/index.html

Mangrove Action Project

http://www.earthisland.org/map/index.html

Environmental Defense

http://www.edf.org

World Wildlife Fund

http://www.wwf.org

Natural Resources Defense Council

http://www.nrdc.org

Sierra Club of Canada

http://www.sierraclub.ca

348

Ecological Aquaculture

outreach and communications to `backstop' (and mold) public policy. Principal constraints are: (1)

(2)

(3)

(4)

The accelerated use of fish meals/oils in aquaculture feeds; resulting in aquaculture being protein intensive to the point of being a net consumer ± rather than a producer ± of animal protein (Schroeder, 1980; Table 12.2); plus aquaculture not adding to ± but subtracting from ± the world's protein supplies (`fishing down and farming up marine food webs'; Naylor et al., 2000; Pauly et al., 2000). Concerns about intensive aquaculture operations being feedlots (Wohlfarth & Schroeder, 1979) that are energy intensive (Weatherly & Cogger, 1977), producing nutrient pollution loads comparable to human sewage (Bergheim & Sivertsen, 1981; Bergheim et al., 1982), and leading to accelerated eutrophication, harmful algal blooms, and unacceptable modification of benthic ecosystems. Habitat destruction ± especially mangrove destruction and water diversions ± that disrupt nearshore and riverine ecosystems (Macintosh & Phillips, 1992; Pullin et al., 1993; Dierberg & Kiattisimku, 1996; Goldburg & Triplett, 1997). Chemical use and nutrient discharges to the environment (Beveridge et al., 1991; Pullin et al., 1993; Folke et al., 1994; Costa-Pierce, 1997a).

Table 12.2 The aquaculture paradox: `The pond consumes fish' (modified from Schroeder, 1980) Process

Conversions

Results

Food conversion ratio (FCR)

1.0 kg dry food/1.0 kg wet fish

Typical FCR in salmon aquaculture (Fig. 12.3)

Fish meal content of feed

35% fish meal (Fig. 12.4)

0.35 kg dry fish meal/1.0 kg wet weight of `cultured' fish

Wild fish weight consumed in production

Fish meal is 25% dry matter (Jobling, 1993)

1.4 kg wet `wild' fish/1.0 kg wet weight of `cultured' fish

Aquaculture systems of greatest concern are intensive shrimp ponds in lowland tropical estuaries, and fish pens in the nearshore environment growing finfish (especially salmon) (Naylor et al., 2000). Concerns about shellfish, grazing finfish, and marine plant agronomy (seaweed aquaculture) are much less. Indeed, shellfish aquaculture has been shown to provide a vital ecosystem service ± nutrient removal (Haamer, 1996). A recent review of aquaculture's effect on world fish supplies by Naylor et al. (2000) has neglected to complete comparative production and energy efficiencies of aquaculture versus other wild-capture and terrestrial animal protein-production alternatives. Only by comparing efficiencies of terrestrial and aquatic protein production systems can scientists, policy-makers and the public be able to address in a more rigorous manner the research, policy and regulatory needs for ecological aquaculture. Comparisons of energy and production efficiencies of aquaculture versus an array

Ecology as the Paradigm for the Future of Aquaculture

349

of fisheries and terrestrial agriculture systems confirm that aquaculture ± even as it is presently practiced ± is an efficient mass producer of animal proteins for a crowded coastal planet. Goodland & Pimental (2000) state, `In the worldwide effort to increase food production, aquaculture merits more attention than raising grain-fed cattle.' (Table 12.3). Production efficiencies of edible mass for aquaculture range from 2.5 to 4.5 kg dry feed/kg edible mass compared with 3.0 to 17.4 for conventional terrestrial animal production systems (Table 12.4). Beef cattle require over 10 kg of feed to add 1 kg of edible weight, whereas catfish can add 1 kg of edible weight with less than 3 kg of feed. Wild capture fisheries are inefficient from both trophic and energy efficiency perspectives in comparison with aquaculture. Coastal and oceanic ecosystems have efficiencies of 10 to 15%, and mean trophic levels of 3.0 to 5.0 (Ryther, 1969), whereas mean trophic levels in aquaculture systems range from 2.3 to 3.3, with highest aquaculture trophic levels in North America and Europe (Pullin et al., 2001). Pullin et al. (2001) state that `. . . most ocean fish consumed by humans have trophic levels ranging from 3.0 to 4.5 (Pauly et al., 1998); i.e. 0 to 1.5 levels above that of lions'. To produce 1 kcal of catfish protein about 34 kcal of fossil fuel energy is required ± Table 12.3 Environmental sustainability and food chain ranking (from Goodland, 1997) Worst Most impact Least efficient Least healthy To be taxed highest

;

Best Least impact Most efficient Healthiest Zero tax

Homeotherms (warm-blooded) 1. Mammals: swine/cattle/goats/sheep/rodents/lagomorphs/camelids/ deer (eggs/cheese/milk/butter) 2. Birds: chickens/geese/ducks/pigeons/turkeys Poikilotherms (cold-blooded) 3. 4. 5. 6.

Other vertebrates: fish/reptiles/amphibians Invertebrates: crustaceans/insects/annelids/mollusks Saprophytes: fungi/yeasts/other microbes Autotrophs: legumes/grains/vegetables/starch crops/fruits/nuts/algae

Table 12.4 Production efficiencies of edible proteins from aquaculture compared with terrestrial animal agriculture systems (modified from Phillips et al., 1978) System

Tilapia Catfish Freshwater prawns Marine shrimp Milk Eggs Broiler chickens Swine Rabbits Beef Lamb

Food conversion ratio (kg dry food/kg wet weight gain) (standard deviations)

Per cent edible portion

Production efficiency (kg dry food/kg edible wet mass)

1.5 (0.2) 1.5 (0.2) 2.0 (0.2) 2.5 (0.5) 3.0 (0.0) 2.8 (0.2) 2.0 (0.2) 2.5 (0.5) 3.0 (0.5) 5.9 (0.5) 4.0 (0.5)

0.60 0.60 0.45 0.56 1.00 0.90 0.59 0.45 0.47 0.49 0.23

2.5 2.5 4.4 4.5 3.0 3.1 3.1 5.6 6.4 10.2 17.4

350

Ecological Aquaculture

Table 12.5 Production efficiencies of fossil fuel energy required to produce edible proteins from aquaculture compared with capture fisheries and terrestrial animal agriculture systems (Rawitscher & Mayer, 1977; Pimental & Pimental, 1979; Brown & Lugo, 1981; Folke & Kautsky, 1992; Goodland & Pimental, 2000) Systems

Capture fisheries Cod fisheries Atlantic salmon fisheries Pacific salmon fisheries King salmon fisheries Lobster fisheries Shrimp fisheries Enhanced fisheries (ranching) Atlantic salmon ranching Agriculture systems Vegetable row crops Rangeland beef Broiler chickens Swine Feedlot beef Aquaculture systems Seaweed culture Mussel longline culture Rainbow trout cages Catfish Atlantic salmon cages

kcal fossil fuel energy input/kcal protein output

20 29 18±30 40 192 198 7±33 2±4 10 22 35 20±78 1 10 24 34 50

lobster and shrimp capture fisheries use more than 5 times this amount of energy. Energy costs for even the most intensive salmon cages are less than lobster and shrimp fishing, and are comparable to beef production in feedlots (Table 12.5). Concerns over aquaculture's competitive use of industrial fish meals and oils (Wilks, 1995; Naylor et al., 1998, 2000) must be weighed carefully on the basis of comprehensive scientific analyses that include analyses of the: (1) (2) (3)

sustainability of fish meal/oil fisheries; current and projected future uses of protein meals/oils by society's terrestrial and aquatic animal protein production systems; and impacts of aquaculture science on the market demands for feed proteins.

Fisheries for fish meal/oil species are well known and strictly regulated (Table 12.6). The average annual tonnage of fish reduced to fish meal and oil from 1994 to 1999 was 29.1 million mt (FAO, 2000). Average annual fish oil production from 1976 to 1997 has been 6.7 million mt (FAO, 2000). It is common knowledge that there is little room to increase these finite supplies; these stocks are classified by FAO (2000) as `fully fished'. Nearly all fish meal and oil species are small, low quality, low cost,

Ecology as the Paradigm for the Future of Aquaculture

351

Table 12.6 Regulatory mechanisms on industrial fish meal fisheries (from Barlow, 2000) Meal/oil species

Total fleet effort regulations

Total catch limits

Closed seasons

Closed areas

Minimum landing sizes

Minimum mesh sizes

Anchovy Sardines Herring Capelin Jack mackerel Horse mackerel Sand eels Sprats Pout Blue whiting

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

Yes Yes Yes Yes Yes Yes Yes Yes No No

Yes Yes Yes No Yes Yes No Yes Yes No

Yes Yes Yes Yes Yes Yes No No No No

Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes

pelagic species with short lifespans and small maximum sizes, principally stocks of anchoveta, Chilean jack mackerel, capelin, herring, sand eels, pilchards, menhaden and sardines. Fish meal production has remained stable over the past 10 years (Barlow, 2000) ± except for El NinÄo years (Table 12.7). A large El NinÄo in 1998, for example, decreased fish oil production to 4.8 million mt but this recovered to 6.6 million mt in 1999, close to the 6.5 million mt annual average from 1976 to 1999 (FAO, 2000). Aquaculture currently uses about 35% of the world's fish meal supplies, increasing its market share from just 10% in 1988 (FAO, 2000), but over 60% of the world's fish meal is consumed by poultry and swine. Table 12.7 Percentage changes in industrial fish meal catches due to El NinÄo events (from Barlow & Pike, 1997) Years

Intensity of event

% Change

1951 1953 1957/58 1965 1969 1972/73 1976 1982/83 1987 1991/92 1997/98

Moderate Weak Strong Moderate Weak Strong Moderate Strong Moderate Moderate Strong

Insignificant Insignificant Insignificant 718.3 712.4 755.5 727.2 756.1 722.7 723.8 726.1

In highly regulated fisheries such as fish meal and fish oil fisheries, which have three major market sectors (swine, poultry, aquaculture, and a smaller amount to ruminants), supplies, demands, and prices fluctuate with economic conditions of highly interconnected markets for many feed proteins and oils. Markets for fish meals and oils are affected not only by demands from various aquatic and terrestrial protein production enterprises, but also by meal/oil production from terrestrial plant and animal sources (animal and bone meals, soybean meals and oils, corn and wheat

352

Ecological Aquaculture

gluten meals, canola, linseed, and cottonseed oils, etc.). Market demands and interrelated economic factors determine the route and magnitudes of product flows; not fishery supplies or fishery management issues. On January 1, 2001, the European Union (EU) banned the use of processed meat and bone meals in animal feeds and fish meal in ruminant feeds in efforts to control `mad cow disease'. As a result, fish consumption in the UK in 2000±2001 has soared. The EU's decision removed 2.5 million tons of feed proteins from the market, affecting prices and demands for other feed proteins such as soybean meals (Starkey, 2001). At the same time, Chilean production of fish meal (mostly sardines, anchovies, mackerels, herring and sand eels) dropped 22% from levels exported in 2000 due to climatic factors, but in 2001 has recovered dramatically. Naylor et al. (2000) contend that increased aquaculture use of fish meals/oils will decrease fish supplies for human needs and threaten fragile marine ecosystems from overfishing. They acknowledge the intimate connections of industrial meal/oil fisheries to many market sectors; but they neglect to conduct analyses of economics, market factors, and the sustainability of industrial fish meal/oil fisheries. Chamberlin & Barlow (2000) stated, in a response to Naylor et al. (2000), that, owing to the primacy of market forces as opposed to fisheries management ones, `if fishmeal were completely eliminated from aquaculture feeds, it would continue to be produced for land animals.' In addition, not all protein and oil in aquaculture feeds comes from fish meal and fish oils; and diet compositions change for different stages of the life cycle of cultured organisms, e.g. protein percentages are different (higher) for juveniles, and lower for adult, grow-out stages. Animal and bone meals, soybean meals and oils, corn and wheat gluten meals, canola, linseed, and cottonseed oils, among other alternative protein meals, can replace between one-third and one-half of the fish meal used in feeds for carnivorous species such as salmon, trout, sea bass and sea bream (Arndt et al., 1999). All of the fish meal and oil requirements can be met by non-fish meal sources for herbivorous and omnivorous fish such as catfish, tilapia, and carp. Current protein requirements for catfish are met by fish meal at 12±15% of the diet, but this can be replaced completely by soybean meal and corn (Robinson et al., 2000). Studies have shown that complete replacement of fish meal can be obtained even for carnivorous species such as salmon and seabass (Elangovan & Shim, 2000). Palatability issues with soybean-based diets can be corrected by adding other protein sources such as blood meal, corn gluten meal, blue mussel meat, and krill (Arndt et al., 1999; Kikuchi, 1999). These rapid advances in the science of aquaculture nutrition have led to dramatic declines in the percent fish meal used in salmon feeds from 1972 to 2000 (Fig. 12.4). Continuation of these trends will mean that fish meal use per metric ton of aquaculture production will continue to decline. Barlow (2000) forecasts aquaculture production to triple by 2010 with only a 24% increase in fish meal use in aquaculture. Naylor et al. (2000) have given proper attention to the need to decrease aquaculture's dependence on industrial fish meal/oil fisheries, and to improve the sustainability of aquaculture practices. But Naylor et al. (2000) did not accurately

Ecology as the Paradigm for the Future of Aquaculture

353

FCR 1972–2000

2.5

FCR

2.0

Fig. 12.3 Reduction in food conversion ratio (FCR) for salmon aquaculture from 1972 to 2000 (Sveinson & Engelstad, 2001).

1.5 1.0 0.5 0

1972

1975

1980

1985

1989

1995

2000

Year

100

% Fishmeal in fish feed 1972–2000

%

75

Fig. 12.4 Reduction in the percentage incorporation of fishmeal in diets for salmon aquaculture from 1972 to 2000 (Sveinson & Engelstad, 2001).

50 25 0

1972

1985

2000

Year

characterize the rapid advances in aquaculture research and development towards sustainability, choosing instead to skew their data towards the high end of the ranges of their data ± rather than the means or modes ± and develop arguments towards worst case scenarios that support their advocacy positions. For example, Naylor et al. (2000, see their Table 2) estimated that 3.16 kg of wild fish from fish meal fisheries were required to produce 1.0 kg of salmon, using a water content of fish of 80%, a fish meal content of diets at 45%, and a food conversion ratio (FCR) of 1.5. These figures do not apply to the modern salmon industry ± nor the future one ± as they would be uneconomic. For salmon grown profitably using best management practices, FCRs of 1:1 (Fig. 12.3) and diets of 35% fish meal protein are used (Fig. 12.4), resulting in a conversion of 1.40 kg of wild fish to produce 1.0 kg of salmon (Table 12.2). Many salmon farms use even lower fish meal incorporation rates of 25%, which would imply a conversion of 1.00 kg of wild fish to produce 1.0 kg of salmon (FFI, 2001).

Aquaculture pollution and habitat losses Pollution is defined as `an undesirable change in the physical, chemical or biological characteristics of air, water or land that may or will be harmful to humans and other living organisms' (Odum, 1975). Large-scale corporate agriculture is the largest source of pollution in the rich countries. Dent & Anderson (1971) stated that agriculture in most developed countries has `ignored the larger ecological framework in which farming is conducted and as a result agricultural production has often exploited the

354

Ecological Aquaculture

natural environment'. Overall, pollution from aquaculture is not widespread since aquaculture is a very rare occupation throughout the world. Edwards (1993) states that even though Asia accounts for 85% of total world aquaculture production: `Aquaculture is far less widespread in Asia than is widely supposed. Perhaps less than 1% of farmers are involved in aquaculture in the region.' However, the degraded state of most aquatic ecosystems, combined with public concerns about adding any `new' sources of aquatic pollution to already overburdened ecosystems, will require aquaculture to develop ecosystems approaches and sustainable operating procedures ± and to articulate a sustainable, ecological pedagogy. In the twenty-first century, aquaculture developers will need to spend as much time on technological advances coming to the field as they do in designing ecological approaches to aquaculture development that clearly exhibit stewardship of the environment. Folke & Kautsky (1991) state that `one must expand the boundaries and one's actions far beyond the cultivation site, and realize that there is an unavoidable complementarity between the life-support environment and aquaculture production'. Clear, unambiguous linkages between aquaculture and the environment must be created and fostered; and the complementary roles of aquaculture in contributing to environmental sustainability, rehabilitation and enhancement must be developed and clearly articulated to a highly concerned, increasingly educated and involved public (Costa-Pierce, 1997a). The area impacted by a species ± based on resource consumption and waste discharge from a specie's activities ± is defined as the `ecological footprint' (Rees & Wackernagel, 1994). An ecological footprint extends the impact humans have on nature to an area several times greater than the land directly inhabited. For example, salmon farming in Sweden was estimated to have an ecological footprint 40 000± 50 000 times greater than the area of the cages, whereas mussel aquaculture had a footprint 20 times greater than the mussel production area (Folke et al., 1998). However, Black et al. (1997) and Roth et al. (2000) point out that ecological footprint analyses do not provide dependable analyses for management purposes ± failing to analyze how aquaculture reduces the benefits obtained for society. These authors challenge the application by Folke et al. (1998) of marginal values from the present scale of production to increased levels in the future. Poorly sited and managed marine cage aquaculture operations have caused environmental impacts, but assessments of impacts have too often been based upon outdated literature, scientific misinterpretation, and advocacy (Costa-Pierce, 1997a; Boyd & Schmittou, 1999). Studies of benthic impacts of cage aquaculture have shown ecological impacts to be localized and reversible by fallowing (Stewart, 1998; Table 12.8). The Net Pen Advisory Work Group of the Washington Department of Ecology (WDOE), after extensive studies, also found that benthic impacts of salmon farming in Puget Sound were limited to within 30 m of the net pen perimeter, and that impacts were reversible by fallowing. Based upon their data WDOE has decided to manage salmon pens by allowing a sediment impact zone within a 30 m `footprint' of the cages. Outside of this perimeter, water quality and benthic `performance standards' would have to be met (Rensel, 2001). Additional research needs to be conducted in this area since very few data exist to date on the long-term assimilative capacity of benthic communities in different climatic regions. For example, Angel et al. (1992)

Ecology as the Paradigm for the Future of Aquaculture

355

Table 12.8 Reported impacts on benthic ecosystems from cage aquaculture (modified from Costa-Pierce, 1997a) Studies

Benthic impacts

Mattsson & Linden (1983)

Species composition changed up to 20 m away from mussel farm

Brown et al. (1987)

Species composition changed up to 15 m away from cage edge

Gowen et al. (1988)

Species composition changed up to 30±40 m away from cages

Lumb (1989)

Impacts restricted to within 50 m of cage edges and dependent on seabed type

Ritz et al. (1989)

Macrofaunal community under the farm adopted an undisturbed condition 7 weeks post-harvest of farm stock

Kupka-Hansen et al. (1991)

Species composition changed up to 25 m away from cages

Weston (1990)

Farm effects on sediment chemistry evident up to 45 m from the farm; species composition changed at least to 150 m away from cages

Johannessen et al. (1994)

No influence of fish farming could be detected 250 m away from cages

Krost et al. (1994)

Affected area extended 3±5 m from the fish farm margin

Wu et al. (1994)

Impacted area extended to 1000 m with industry using trash fish as feed and poor water flushing exists

McGhie et al. (2000)

Farm wastes largely restricted to area beneath sea cages; most of the sediment organic input from feces; and 12-month fallowing period sufficient to return site to pre-farm oxic conditions

Morrisey et al. (2000)

Large temporal and spatial variabilities depending on water velocities; recovery times estimated at between 3 and 12 years

Dominguez et al. (2001)

No effect on physical and chemical sediment characteristics due to fish farm operation in high average water current velocity (6 cm/s) site

Findlay et al. (1995)

Little or no carbon flux 10 m from the pen; aquaculture impacts comparable to seasonal trends and storm-related resuspension events; effects are spatially limited

Wu (1995)

Very high pollution levels from marine fish cages due to feeding unpelleted, ground-up trash fish; but impacts confined to within 1 km of cages

found that organic matter decomposition in sediments under fish cages in the Gulf of Aqaba may be three to four times greater than in temperate waters. Coastal aquaculture development (fish and shrimp ponds) has caused a reduction in mangrove areas in Asia and the Americas. The extent of destruction due to aquaculture is, however, debatable; and is likely overstated (Boyd & Schmittou, 1999). Primavera (1991) estimates that Philippine mangroves declined from 400 000± 500 000 ha in the 1920s to 140 000 ha in the 1990s, with 60% of the destruction due to coastal aquaculture of milkfish and shrimp ponds. In contrast, Larsson et al. (1994) reported that there was no evidence that mangroves have been cut down for aquaculture development in Columbia. Binh et al. (1997) reported that from 1983 to 1991 one district of Vietnam (the Ngoc Hien district) lost 48% of its mangroves to shrimp pond development. There are hopeful signs that governments and shrimp farmers are realizing that it is not in their best interests to destroy mangroves, and are legislating

356

Ecological Aquaculture

against locating aquaculture ponds in mangrove areas. Ochoa (1997) describes one effort between a non-governmental organization, a local community, and a shrimp aquaculture company who negotiated a deal with the following conditions: (1) (2) (3) (4) (5)

all shrimp ponds were to be located 50 m behind the mangroves, no alteration of mangrove cover would be allowed, no alteration of natural water flows by dams, walls, or diversions was permitted, traditional uses and access to mangrove areas would be guaranteed to the local peoples, ecotourism activities and collaborative research were to be encouraged.

Despite these concerns, the current body of scientific literature has numerous examples portraying aquaculture facilities as preserving and enhancing the form and function of the natural environment, serving as vital habitats for enhanced wild fish populations ± both juvenile and adult forms ± and providing extensive structure to support productive biofouling communities (Brzeski & Newkirk, 1997; Costa-Pierce & Bridger, 2001).

Examples of ecological aquaculture systems Aquaculture is not a uniform `industry' or a standard set of practices easy to classify ± or label ± and regulate. There are a wide diversity of systems and species which can be classified in many different ways (Table 12.9). Given this diversity, the possibilities for engineering ecological aquaculture systems are numerous (Mitsch & Jorgensen, 1989). For example, the integration of aquaculture, agriculture and animal husbandry on small farms in Asia creates definable aquaculture ecosystem types (Fig. 12.5). These aquaculture ecosystems closely resemble natural ecosystems having their own structure, closely coupled nutrient recycling pathways, and ecological management strategies (Dalsgaard & Oficial, 1997; Edwards, 1998). Such integrated systems provide the following advantages: synergy, complementarity, adaptability: `polyvalent' technologies are `marketdriven'; l drought-proofing: efficient use in agriculture of warm, fertile irrigation waters from aquaculture (Costa-Pierce, 1987, 1998); l waste-treatment capability: ponds are `sunlit rumens' (Schroeder, 1980) processing low quality agricultural by-products into high quality aquatic proteins; and l restoration capability: conversion of marginal lands to prime agricultural lands by managing long-term rotations and natural/social cycles between agriculture and aquaculture systems (Costa-Pierce, 1985). l

There is an extensive literature, and a growing body of research, on models of sustainable, ecological aquaculture systems for the world to adopt (Tables 12.10, 12.11). In these farming ecosystems, addition of aquaculture into the mix of farm

Ecology as the Paradigm for the Future of Aquaculture

Table 12.9 Classifications of aquaculture systems (from Costa-Pierce, 1997a) Types

Kinds and levels

Stocking, management and economic intensity levels

Intensive Semi-intensive Extensive

Water salinities

Freshwater Brackishwater Seawater

Water flow characteristics

Running water (lotic) Standing water with flushing Standing water (lentic)

Amount of on-site waste treatment and recirculation

Open, no recirculation Semi-closed, partial recirculation Closed, full recirculation

Environmental location

Indoor Outdoor±natural Outdoor±artificial

Feed qualities

Complete Supplemental Natural

Feeding stategies

Continuous Scheduled Natural

Species' stocking strategies

Monoculture Janitorial polyculture

Species' temperature tolerances

Eurythermal Stenothermal cold water warm water

Species' salinity tolerances

Euryhaline Stenohaline

Species' natural food habits

Carnivorous Omnivorous Herbivorous Opportunistic

Fry sources

Hatcheries Wild capture of broodstock Natural

Level of systems integration

Stand alone Integrated

Unit types

Raceways Tanks, cages (floating, fixed) Net pens (fixed) Rafts (ropes, nets) Ponds

Marketing channels

Human food (local, export) Sport, recreation, tourism

357

358

Ecological Aquaculture

AQUACULTURE

Crop-aquaculture systems

Animal aquaculture systems Crop-animal aquaculture systems

AGRICULTURE

'Traditional' agriculture systems

ANIMAL HUSBANDRY

Fig. 12.5 Aquaculture ecosystems combining aquaculture, agriculture and animal husbandry have definable structure, nutrient cycling, and management protocols that parallel those of natural, land±water interactive ecosystems. In these farming ecosystems, addition of aquaculture into the mix of farm enterprises increases efficiencies of resource flows, profitability, and overall sustainability (Lightfoot et al., 1993; Dalsgaard & Oficial, 1997).

enterprises can greatly increase the efficiencies of bioresource flows and profitability (Ruddle & Zhong, 1988; Costa-Pierce, 1992; Zweig, 1992; Lightfoot et al., 1993).

The need for rapid progress towards sustainable, ecological aquaculture systems The Brundtland Commission (WCED, 1987) has defined sustainable development as `the ability to meet the needs of the present without compromising the ability of the future generations to meet their own needs.' Caldwell (1990) believes we are entering one of history's great transitional eras ± the sustainability transition ± that will be played out on a global scale. In this regard, there is an urgent need for aquaculture to complete the sustainability transition by adopting sustainability criteria and developing the pedagogy of ecological aquaculture in a bold and direct manner (Fig. 12.6). Edwards (1993) states that the `problem for developing countries is essentially how to stimulate agricultural (and aquaculture) productivity and profitability without further environmental degradation, in contrast to the need to reduce the level of intensification to a sustainable level in the developed world.' As an infant enterprise the

Ecology as the Paradigm for the Future of Aquaculture

359

Table 12.10 Examples of ecologically integrated mariculture systems System description

References

Integration of mussel and fish cage aquaculture

Folke & Kautsky (1989, 1992); Bodvin et al. (1996); Kautsky et al. (1997); Troell et al. (1997, 1999a)

Integration of shrimp, oysters and Gracilaria edulis

Jones et al. (2001)

Cultivation of Porphyra with salmon cage aquaculture

Chopin et al. (1999)

Devoting 30±50% of shrimp pond area to mangroves in an integrated shrimp±mangrove ecosystem gave highest annual economic returns

Binh et al. (1997)

4-ha modular shrimp aquaculture ecosystem can produce 40 t whole shrimp, 7 t mullet and 0.5 million oysters/year

Sandifer & Hopkins (1996)

Grey mullet confined in bottom cages underneath commercial sea bream cages reduced organic matter accumulations and oxidized bottom sediments

Angel et al. (1992)

Integrated shrimp/oyster ecosystem produces greater economic yields and increased ecological efficiencies

Wang (1990)

Integration of sea bream culture with seaweed (Ulva) improved water quality of discharged water and improved water conservation and economic yields

Jimenez del Rio et al. (1996); Neori et al. (1996, 2000)

Integration of salmonids and Gracilaria seaweeds improves water quality in the systems and effluents, with increased economic yield

Buschmann et al. (1994, 1996); Troell et al. (1999b)

Land-based integrated culture of fish, bivalves (Crassostrea gigas, Tapes semidecussatus, Haliotis tuberculata) and Ulva lactuca

Shpigel et al. (1993); Neori et al. (1996); Shpigel & Neori (1996)

Integrated kelp (Laminaria saccharina, Nereocystis luetkeana) and salmon cage aquaculture

Petrell & Alie (1996)

Production of 7 t/ha/crop of shrimp with no water exchange or filtration

Hopkins et al. (1995)

Integration of shrimp and Ulva pertusa

Danakusumah et al. (1991)

world over, aquaculture can ill afford to recreate the sorry history of commercial agriculture where huge toxic, nutrient and chemical loads were (and still are) washed down a primitive path of the `solution to pollution is dilution'. Aquaculture should be pro-active, promoting and developing itself as the world's most ecologically integrated industry, and adopting a new strategy ± that of a community-based, sustainable, ecological aquaculture industry that produces ecologically and socially certified produce ± adopting input management strategies (Odum, 1989) and `codes of practice'. The Food and Agriculture Organization of the United Nations (FAO) Code of Conduct for Responsible Fisheries (FAO, 1995) contains a key recommendation that: `States should produce and regularly update aquaculture development strategies and

360

Ecological Aquaculture

Table 12.11 Examples of ecologically integrated freshwater aquaculture systems Systems descriptions

References

Chinese dike-pond aquaculture ecosystem integrates aquaculture, plant and animal agriculture, silviculture and sericulture, sustainably producing 20±40 t/ha/y

Li (1987); Korn (1996); Ruddle & Zhong (1988)

Trout farms produce 8 million kg of settleable solids. 2±3 kg of fish manure equaled 1 kg N in cornfields

Yarris (1981)

Tilapia aquaculture integrated into irrigation schemes in the US southwest

Olsen & Fitzsimmons (1994)

`Janitorial' polyculture of carps in prawn (Macrobrachium rosenbergii) and fish ponds

Costa-Pierce et al. (1985)

Biculture cage ecosystems (one species being grown in a culture system receiving formulated feeds placed above or at higher elevation from a second system below it holding species that are unfed)

Costa-Pierce & Hadikusumah (1990)

Reuse of saline aquaculture effluents to irrigate halophytes suitable for forages

Brown & Glenn (1999); Brown et al. (1999)

Effluents from channel catfish ponds coupled with bulrush, cutgrass and maidencane aquatic wetlands achieved excellent removal of all nutrients, BODs and solids

Schwartz & Boyd (1995)

Polyculture of carps, mullet and prawns, with effluents irrigating a mixed tropical orchard

Costa-Pierce (1987)

Mixed intensive tilapia culture in tanks integrated with hydroponics producing commercially viable fish and plant yields and using sludges for pastures

Rakocy & Hargreaves, 1993; Rakocy, this volume

plans, as required, to ensure that aquaculture development is ecologically sustainable and to allow the rational use of resources shared by aquaculture and other activities.' Additional key strategies have been the development of the Holmenkollen Guidelines for Sustainable Aquaculture (Moe & Svennevig, 1998); the Bangkok Declaration and Strategy (NACA/FAO, 2000); and the Global Aquaculture Alliance's Responsible Aquaculture Program that has produced criteria for environmentally certified cultured shrimp. But codes of conduct and guidelines for certifying sustainability are too complex (Bertollo, 1998). Pullin et al. (2001) have made a leap forward by suggesting a simple set of easily quantifiable indicators for sustainability in aquaculture: Biological: domestication, trophic level, nutrient/energy conversion, Ecological: footprint, emissions, escapees, l Intersectoral: water sharing, diversity, cycling, stability, capacity. l l

The need for science in the development of ecological aquaculture Aquaculture is an ancient practice but until recently has developed mainly on a trial and error basis. Carps still dominate production (FAO, 2000). Of the 25 000 fish

FanLi – 1960s

Industrial Monoculture Linear Luxury proteins Shareholders control Ex. Ecuador shrimp

'Sustainability transition' 21st Century

Fig. 12.6 Adoption of the pedagogy, principles, and practices of ecological aquaculture is vital to complete the `sustainability transition' and to direct the `evolution of the blue revolution'.

Ecology as the Paradigm for the Future of Aquaculture

Community Polyculture Ecosystems Recycling Human needs Government control Ex. China carp

Community-based Industrial ecology Ecological engineering Zero discharge Product diversity • organic • health conscious • socially and environmentally responsible

361

362

Ecological Aquaculture

species only a few are harvested for direct human consumption. Aquaculture science is still poorly developed (and funded). There are very few centers of excellence in aquaculture, and no ecological aquaculture experiment stations. Clearly, aquaculture is the `poor cousin' of agriculture and capture fisheries. Most scientific inquiry in aquaculture is `discipline-oriented', i.e. organized in a traditional, compartmentalized manner in major academic centers. Multidisciplinary aquaculture scientists ± whom David Orr (1999) has called `specialists in whole things' ± are rare. Aquaculture urgently needs a true interdisciplinary environmental scholarship to evolve ± much like the field of oceanography. To develop aquaculture more rapidly, aquaculture science needs to break through disciplinary bounds and tie together real-world knowledge and academic disciplines to create a new, knowledgebased infrastructure and support system to develop ecological aquaculture. Lubchenco (1998) calls for a `new social contract for science' which would `facilitate the investigation of complex, interdisciplinary problems that span multiple spatial and temporal scales; to encourage interagency and international cooperation on societal problems; and to construct more effective bridges between policy, management, and science, as well as between the public and private sectors. Most of our efforts to address economic and social problems are as yet devoid of ecological knowledge.'

Certifying sustainable, ecological aquaculture systems Ecolabelling has been defined as the `voluntary granting of labels by a private or public body in order to inform consumers and thereby promote consumer products which are determined to be environmentally more friendly than other functionally and competitively similar products' (Salzman, 1991). Most ecolabelling guidelines follow the International Organization for Standardization (ISO) criteria which prescribe that `the development of environmental labels and declarations shall take into consideration all relevant aspects of the life cycle of the product' (ISO, 1998). Certification of plant and animal protein products as being produced or harvested in a sustainable manner is gathering much attention from both producers and consumers. In the US there are over 40 state and independent organic certification agencies/firms; with more throughout the world, led by the UK Soil Association, the International Federation of Organic Agriculture Movements (IFOAM), Sweden's KRAV, Germany's `Naturland', and FAO's Codex Alimentarius Commission. Wessells et al. (1999) found, however, that successful ecolabelling programs must accelerate consumer education programs so that consumers become more aware of differences in species, geographic regions, and certifying agencies. In Scotland production of organic salmon is growing rapidly. Total organic production is estimated at 750±850 mt, produced from five farms in Shetland and in Orkney. Aquaculture farms receive a premium price of between £4.00±5.00/kg, as consumers are increasingly willing to pay extra for the organic trademark. Conventionally farmed salmon presently (2001) cost just over £2.00/kg (C. Ritch, Balta

Ecology as the Paradigm for the Future of Aquaculture

363

Island Seafare Ltd, personal communication). Organic certification is obtained by the UK Soil Association, a licensing body for all organic products. Some of the regulations on fish farms are: l l

l l l l

density of fish in organic salmon farming has to be less than 10 kg/m3 (traditional salmon farming has 20 kg/m3); salmon feed has to come from certified feed manufacturers which supply a specified organic salmon feed that has no genetically modified ingredients, no artificial colors, no fish meal from industrial fishing (fish meal produced from fish processing wastes is used), and all cereals used have to come from organic farming; fish should be fed by hand; no carbon dioxide is allowed in slaughtering; no anti-fouling chemicals are allowed in treating nets; no polystyrene boxes are allowed to market fish (waxed cardboard or reusable plastic boxes are used).

In addition, Trouw Aquaculture have achieved official certification of registration from the UK Soil Association for a new line of organic trout and salmon feeds marketed under Trouw's new `green product' lines. Certification entails complete auditing of the ingredients and the manufacturing process. Feeds will use the Soil Association's certification mark on the organic diets. Organic oysters are being produced in New Zealand. Oyster farmers obtain certification from a private company known as Bio-Gro New Zealand that analyzes the complete farming ecosystem from both external impacts and internal farming practices. External environmental impacts on the oyster aquaculture operation are due mainly to poor farm siting where adjacent or upstream `non-organic' farms or residential developments contribute nutrient and chemical pollution. Bio-Gro (http:// www.biogro.co.nz) analyzes internal farming practices to certify that there is no use of prohibited chemicals; that the business is energy efficient; and that recycling is a part of the business. Two salmon producers in Ireland have been accredited by an independent certification body, the Irish Food Quality Certification (IFQC), under the internationally recognized EN45011 standard. Salmon is the first animal product in Ireland to achieve the EN45011 standard. IFQC audits salmon farms seeking accreditation on a yearly basis while processors have bi-annual inspections. Accredited products carry an IFQC quality logo and are able to be easily traced. The Global Aquaculture Alliance (GAA) Responsible Aquaculture Program intends to produce environmentally certified shrimp according to the following criteria (George Chamberlin, personal communication): Voluntary: farms are under no pressure to participate; Transparent: compliance is based on published, quantitative standards; l Practical: standards are challenging, but not utopian; l Verifiable: farms are certified by an accredited third party; l l

364

l l

Ecological Aquaculture

Labeled: certified product is eligible for the `Responsible Aquaculture' label; Identity preserved: identity of labeled products is preserved through chain-ofcustody certification.

GAA Verification will be completed by third-party certifiers who are accredited by an international certifying group using ISO 65 Guidelines. Inspectors will be required to pass a GAA training course demonstrating their familiarity with GAA quantitative standards and compliance plans. Ecolabelling of seafoods for consumers has been led by the efforts of the Audubon Society (http://www.audubon.org/campaign/lo/ow/index.hml) and the Marine Stewardship Council (MSC) (http://www.msc.org). The MSC has been certifying products from both capture fisheries and aquaculture under their sustainability ecolabel. Anyone wanting to attach MSC's ecolabel to fish products must first obtain a chain of custody certificate from the MSC which can cost a minimum of a few thousand dollars for a seafood processor. In addition, the company must enter into a licensing agreement with the MSC before it can use MSC's ecolabel.

Conclusion: ecological aquaculture as the paradigm for the blue revolution In the twenty-first century, aquaculture developers will need to spend as much time on the technological advances coming to the field as they do in designing ecological approaches to aquaculture development that clearly exhibit stewardship of the environment. For aquaculture development to proceed to the point where it will be recognized worldwide as the most efficient contributor to new protein production, clear, unambiguous linkages between aquaculture and the environment must be created and fostered and the complementary roles of aquaculture in contributing to environmental sustainability, rehabilitation and enhancement must be developed and clearly articulated to a highly concerned, increasingly educated and involved public. The damaged and degraded state of most aquatic ecosystems worldwide ± combined with public concerns about adding any `new' sources of pollution to already overburdened natural ecosystems ± will require aquaculture to develop ecosystems approaches and sustainable operating procedures, and to articulate a sustainable, ecological aquaculture pedagogy. Clearly, most of the public will not tolerate the addition of any new sources of pollution or the further degradation of the natural environment which is perceived to come at the expense of the degradation of the quality of life. Connections between the disruption of our environment and human health are being made by an increasingly skeptical public determined to fight `the experts'. In many cases the simple implication of the presence of a chemical implies a hazard and a threat to human health. In order to change the public perception of aquaculture as `outsiders' or `industrial polluters', the aquaculture world community needs to focus its attention on a new paradigm, ecological aquaculture, in order to evolve an `aquaculture revolution' that is technically sophisticated, knowledge-based, and ecologically and

Ecology as the Paradigm for the Future of Aquaculture

365

socially responsible. Aquaculture in the twenty-first century must plan to become part of a community and a region, and have a wider plan for community development that works with policy-makers to: create a diversity of unprocessed and value-added products, and to provide local market access to needed inputs; recycle wastes; and to plan for job creation and environmental enhancement on local and regional scales. Ecological aquaculture brings modern sustainability, ecological methods and systems thinking to aquaculture, incorporating social, economic, and planning considerations for its wider social and environmental contexts in fisheries and coastal zone management. Ecological aquaculture will create new opportunities for a more diverse group of professionals and entrepreneurs to get involved in aquaculture since new advances will be needed not only in treatment technologies, production management and feed technologies; but also in energy technologies, information management, public information and outreach, community facilitation and networking.

References Alverson, D., Freeberg, M., Murawski, A. & Pope, J. (1994) A Global Assessment of Fisheries Bycatch and Discards. FAO Technical Paper 339. FAO, Rome. Angel, D., Krost, R., Zuber, D., Mozes, N. & Neori, A. (1992) The turnover of organic matter in hypertrophic sediments below a floating fish farm in the oligotrophic Gulf of Eliat. Bamidgeh, 44, 143±144. Arndt, R., Hardy, R., Sugiura, S. & Dong, F. (1999) Effects of heat treatment and substitution level on palatability and nutritional value of soy defatted flour in feeds for coho salmon, Oncorhynchus kisutch. Aquaculture, 180, 129±145. Bailey, C., Jentoft, S. & Sinclair, P. (1996) Aquacultural Development: Social Dimensions of an Emerging Industry. Westview Press, Boulder, CO. Bailey, C. (1988) The social consequences of tropical shrimp mariculture development. Ocean & Shoreline Management, 11, 31±44. Bailey, C. (1997) Aquaculture and basic human needs. World Aquaculture, 28, 28±31. Barlow, S. (2000) Fishmeal and fish oil: sustainable ingredients for aquafeeds. The Advocate, April, pp. 85±88. Barlow, S. & Pike, I. (1997) International Fish Meal and Oil Manufacturers Association (IFOMA). Fish Farming International, 24, 6±8. Bergheim, A. & Sivertsen, A. (1981) Oxygen consuming properties of effluents from fish farms. Aquaculture, 22, 185±187. Bergheim, A., Sivertsen, A. & Selmer-Olsen, A. (1982) Estimated pollution loadings from Norwegian fish farms. I. Investigations 1978±1979. Aquaculture, 28, 347±361. Bertollo, P. (1998) Assessing ecosystem health in governed landscapes: a framework for developing core indicators. Ecosystem Health, 4, 32±51. Beveridge, M., Phillips, M. & Clarke, R. (1991) A quantitative assessment of wastes from aquatic animal production. In: Aquaculture and Water Quality (eds D. Brune & J. Tomasso), pp. 506±533. Advances in World Aquaculture, Vol. 3. World Aquaculture Society, Baton Rouge, LA.

366

Ecological Aquaculture

Binh, C.T., Phillips, M.J. & Demaine, H. (1997) Integrated shrimp±mangrove farming systems in the Mekong Delta of Vietnam. Aquaculture Research, 28, 599±610. Black, E., Gowen, R., Rosenthal, H., Roth, E., Stechy, D. & Taylor, F.J.R. (1997) The costs of eutrophication from salmon farming: implications for policy ± a comment. Journal of Environmental Management, 50, 105±109. Bock, C. & Bock, J. (2000) The View From Bald Hill: Thirty Years in an Arizona Grassland. University of California Press, Berkeley, CA. Bodvin, T., Indergaard, M., Norgaard, E., Jensen, A. & Skaar, A. (1996) Clean technology in aquaculture ± a production without waste products? Hydrobiologia, 326/327, 83±86. Bookchin, M. (1982) The Ecology of Freedom. Cheshire Books, Palo Alto, CA. Born, A.F., Verdegem, M. & Huisman, E. (1994) Macro-economic factors influencing world aquaculture production. Aquaculture and Fisheries Management, 25, 519±536. Boyd, C.E. & Schmittou, H.R. (1999) Achievement of sustainable aquaculture through environmental management. Aquaculture Economics and Management, 3, 59±69. Brown, S. & Lugo, A. (1981) Management and Status of US Commercial Marine Fisheries. Council on Environmental Quality, Washington, DC. Brown, J.R., Gowen, R. & McLusky, D. (1987) The effect of salmon farming on the benthos of a Scottish sea loch. Journal of Experimental Marine Biology and Ecology, 109, 39±51. Brown, L., Renner, M. & Flavin, C. (1998) Vital Signs 1998: The Environmental Trends That Are Shaping Our Future. Worldwatch Institute, Washington, DC. Brown, J. & Glenn, E. (1999) Reuse of highly saline aquaculture effluent to irrigate a potential forage halophyte, Suaeda esteroa. Aquacultural Engineering, 20, 91±111. Brown, J., Glenn, E., Fitzsimmons, K. & Smith, S. (1999) Halophytes for the treatment of saline aquaculture effluent. Aquaculture, 175, 255±268. Brown, D., Manno, M., Westra, L., Pimental, D. & Crabbe, P. (2000) Implementing global ecological integrity: a synthesis. In: Ecological Integrity (eds D. Pimental, L. Westra & R. Noss), pp. 385±405. Island Press, Washington, DC. Brzeski, V. & Newkirk, G. (1997) Integrated coastal food production systems ± a review of current literature. Ocean & Coastal Management, 34, 55±71. Buschmann, A., Mora, O., Gomez, P. et al. (1994) Gracilaria chilensis outdoor tank cultivation in Chile: use of land-based salmon culture effluents. Aquacultural Engineering, 13, 283±300. Buschmann, A., Troell, M., Kautsky, N. & Kautsky, L. (1996) Integrated tank culture of salmonids and Gracilaria chilensis (Gracilariales, Rhodophyta). Hydrobiologia, 326/327, 75±82. Caldwell, L. (1990) Between Two Worlds: Science, The Environmental Movement and Policy Choice. Cambridge University Press, Cambridge, UK. Chamberlin, G. & Barlow, S. (2000) A balanced assessment of aquaculture. The Advocate, August 2000, p. 7. Chopin, T., Yarish, C., Wilkes, R., Belyea, E., Lu, S. & Mathieson, A. (1999) Developing Porphyra/salmon integrated aquaculture for bioremediation and diversification of the aquaculture industry. Journal of Applied Phycology, 11, 463±472. Christy, F. (1997) Economic waste in fisheries: impediments to change and conditions for improvement. In: Global Trends: Fisheries Management (eds E. Pikitch, D. Huppert & M. Sissenwine), pp. 28±39. American Fisheries Society Symposium 20. AFS, Bethesda, MD. Costa-Pierce, B.A. (1985) Aquaculture in ancient Hawaii. BioScience, 37, 320±331.

Ecology as the Paradigm for the Future of Aquaculture

367

Costa-Pierce, B.A. (1987) Initial feasibility of integrated aquaculture on lava soils in Hawaii. Aquacultural Engineering, 6, 171±182. Costa-Pierce, B.A. (1992) Rice±fish systems as intensive nurseries. In: Rice±Fish Research and Development in Asia (eds C.R. de la Cruz, C. Lightfoot, B.A. Costa-Pierce, V.R. Carangall & M.P. Bimbao), pp. 117±130. International Center for Living Aquatic Resources Management, Manila, Philippines. Costa-Pierce, B.A. (1997a) Environmental impacts of nutrients discharged from aquaculture: towards the evolution of sustainable, ecological aquaculture systems. In: Aquaculture and Water Resource Management (eds D.J. Baird, M.C.M. Beveridge, L.A. Kelly & J.F. Muir), pp. 81±113. Blackwell Science, Oxford. Costa-Pierce, B.A. (1997b) From Farmers to Fishers: Developing Reservoir Aquaculture for People Displaced by Dams. World Bank Technical Paper No. 369 (Fisheries Series), World Bank, Washington, DC. Costa-Pierce, B.A. (1998) Preliminary investigation of an integrated aquaculture±wetland ecosystem using tertiary-treated municipal wastewater in Los Angeles County, California. Ecological Engineering, 10, 341±354. Costa-Pierce, B.A., Malecha, S. & Laws, E. (1985) Effects of polyculture and manure fertilization on water quality and heterotrophic productivity in Macrobrachium rosenbergii ponds. Transactions of the American Fisheries Society, 114, 896±906. Costa-Pierce, B.A. & Hadikusumah, H. (1990) Research on cage aquaculture systems in the Saguling Reservoir, West Java, Indonesia. In: Reservoir Fisheries and Aquaculture Development for Resettlement in Indonesia (eds B.A. Costa-Pierce & O. Soemarwoto), pp. 112±217. International Center for Living Aquatic Resources Management (ICLARM), Manila, Philippines. Costa-Pierce, B.A. & Bridger, C.J. (in press) The role of marine aquaculture facilities as habitats and ecosystems. In: Responsible Marine Aquaculture (eds R. Stickney & J. McVey). CAB International, London. Dalsgaard, J. & Oficial, R. (1997) A quantitative approach for assessing the productive performance and ecological contributions of smallholder farms. Agricultural Systems, 55, 503±533. Danakusumah, E., Kadowaki, S. & Hirata, H. (1991) Effects of coexisting Ulva pertusa on the production of Kruma prawn. Nippon Suisan Gakkaishi, 578, 1597. Dent, J.B. & Anderson, J. (1971) Systems, management and agriculture. In: Systems Analysis in Agricultural Management (eds J. Dent & J. Anderson), pp. 3±14. Wiley & Sons, New York. Dierberg, F. & Kiattisimku, W. (1996) Issues, impacts, and implications of shrimp aquaculture in Thailand. Environmental Management, 20, 649±666. Dominguez, L.M., Calero, G., Martin, J. & Robaina, L. (2001) A comparative study of sediments under a marine cage farm at Gran Canaria Island (Spain): preliminary results. Aquaculture, 192, 225±231. Edwards, P. (1993) Environmental issues in integrated agriculture±aquaculture and wastewater-fed fish culture systems. In: Environment and Aquaculture in Developing Countries (eds R. Pullin, H. Rosenthal & J. Maclean), pp. 139±170. ICLARM, Manila, Philippines. Edwards, P. (1998) A systems approach for the promotion of integrated aquaculture. Aquaculture and Economics Management, 2, 1±12. Edwards, P. & Demaine, H. (1997) Rural Aquaculture: Overview and Framework for Country

368

Ecological Aquaculture

Reviews. Regional Office for Asia and the Pacific, Food and Agriculture Organization of the United Nations (FAO), Bangkok, Thailand. Elangovan, A. & Shim, K. (2000) The influence of replacing fish meal partially in the diet with soybean meal on growth and body composition of juvenile tin foil barb (Barbodes altus). Aquaculture, 189, 133±144. FAO (Food and Agriculture Organization of the United Nations) (1995) Code of Conduct for Responsible Fisheries. FAO, Rome. FAO (Food and Agriculture Organization of the United Nations) (2000) State of the World's Fisheries and Aquaculture. FAO, Rome. FFI (Fish Farming International) (2001) Wild and farmed go hand in hand! Fish Farming International, March 2001, p. 5. Findlay, R., Watling, L. & Mayer, L. (1995) Environmental impact of salmon netpen culture on marine benthic communities in Maine: a case study. Estuaries, 18, 145±179. Fitzsimmons, K. (2000) Future trends of tilapia aquaculture in the Americas. In: Tilapia Aquaculture in the Americas, Vol. 2 (eds B.A. Costa-Pierce & J.E. Rakocy), pp. 252±264. World Aquaculture Society, Baton Rouge, LA. Folke, C. & N. Kautsky (1989) The role of ecosystems for a sustainable development of aquaculture. Ambio, 18, 234±243. Folke, C. & Kautsky, N. (1991) Ecological economic principles for aquaculture development. In: Nutritional Strategies and Aquaculture Wastes (eds C.B. Cowey & C.Y. Cho), pp. 207± 222. University of Guelph, Ontario. Folke, C. & Kautsky, N. (1992) Aquaculture with its environment: prospects for sustainability. Ocean and Coastal Management, 17, 5±24. Folke, C., Kautsky, N. & Troell, M. (1994) The cost of eutrophication from salmon farming: implications for policy. Journal of Environmental Management, 40, 173±182. Folke, C., Kautsky, N., Berg, H., Jansson, A. & Troell, M. (1998) The ecological footprint concept for sustainable seafood production: a review. Ecological Applications, 8, S63±S71. Goldburg, R. & Triplett, S. (1997) Murky Waters: Environmental Effects of Aquaculture in the United States. Environmental Defense Fund, Washington, DC. Goodland, R. (1997) Agricultural sustainability: diet matters. Ecological Economics, 23, 189±200. Goodland, R. & Pimental, D. (2000) Environmental sustainability and integrity in the agriculture sector. In: Ecological Integrity (eds D. Pimental, L. Westra & R. Noss), pp. 121±38. Island Press, Washington, DC. Gowen, R., Brown, J., Bradbury, N. & McLusky, D. (1988) Investigations into Benthic Enrichment, Hypernutrification and Eutrophication Associated with Mariculture in Scottish Coastal Waters (1984±1988). Department of Biology, University of Stirling, Scotland. Grove, T. & Edwards, P. (1993) Do we need a new developmental paradigm? Agriculture, Ecosystems and Environment, 46, 135±145. Haamer, J. (1996) Improving water quality in a eutrophied fjord system with mussel farming. Ambio, 25, 356±362. Hopkins, J., Browdy, C., Hamilton, R. & Heffernan, J. (1995) The effect of low-rate sand filtration and modified feed management on effluent quality, pond water quality and production of intensive shrimp ponds. Estuaries, 18, 116±123. ISO (International Organization for Standardization) (1998) Environmental labels and declarations: general principles. Principle 5. ISO 14020. Geneva.

Ecology as the Paradigm for the Future of Aquaculture

369

Jamieson, D. (1996) Ethics and intentional climate change. Climatic Change, 33, 323±336. Jimenez del Rio, M., Ramazanov, Z. & Garcia-Reina, G. (1996) Ulva rigida (Ulvales, Chlorophyta) tank culture as biofilters for dissolved inorganic nitrogen from fishpond effluents. Hydrobiologia, 326/327, 61±66. Jobling, M. (1993) Nutrition, diet formulation and feeding practices. In: Salmon Aquaculture (eds K. Heen, R. Monahan & F. Utter), pp. 83±126. Fishing News Books, Oxford. Johannessen, P.J., Botnen, H. & Tvedten, O. (1994) Macrobenthos: before, during and after a fish farm. Aquaculture and Fisheries Management, 25, 55±66. Jones, A., Dennison, W. & Preston, N. (2001) Integrated treatment of shrimp effluent by sedimentation, oyster filtration and macroalgal absorption: a laboratory scale study. Aquaculture, 193, 155±178. Kautsky, N. & Folke, C. (1991) Integrating open system aquaculture: ecological engineering for increased production and environmental improvement through nutrient recycling. In: Ecological Engineering for Wastewater Treatment (eds C. Etnier & B. Guterstam), pp. 320± 324. Lewis Publishing, Chelsea, Michigan. Kautsky, N., Troell, M. & Folke, C. (1997) Ecological engineering for increased production and environmental improvement in open sea aquaculture. In: Ecological Engineering for Wastewater Treatment (eds C. Etnier & B. Guterstam), pp. 325±355. Lewis Publishing, Chelsea, Michigan. Kikuchi, K. (1999) Use of defatted soybean meal as a protein source in the diet of Japanese flounder (Paralichthys olivaceus). Aquaculture, 179, 3±11. Korn, M. (1996) The dike-pond concept: sustainable agriculture and nutrient recycling in China. Ambio, 25, 6±13. Krost, P., Chrzan, T., Schomann, H. & Rosenthal, H. (1994) Effects of a floating fish farm in Kiel Fjord on the sediment. Journal of Applied Ichthyology, 10, 353±361. Kupka-Hansen, P., Pittman, K. & Ervik, A. (1991) Organic waste from marine fish farms ± effects on the seabed. In: Marine Aquaculture and the Environment (ed. Kupka-Hansen, P.), pp. 105±119. Nordic Council of Ministers, Copenhagen, Denmark. Larsson, J., Folke, C. & Kautsky, N. (1994) Ecological limitations and appropriation of ecosystem support by shrimp farming in Columbia. Environmental Management, 18, 663±676. Li, S. (1987) Energy structure and efficiency of a typical Chinese integrated fish farm. Aquaculture, 65, 105±118. Lightfoot, C., Bimbao, M., Dalsgaard, P. & Pullin, R. (1993) Aquaculture and sustainability through integrated resource management. Outlook in Agriculture, 22, 143±150. Lubchenco, J. (1998) Entering the century of the environment: a new social contract for science. Science, 279, 491±497. Lumb, C.M. (1989) Self-pollution by Scottish salmon farms? Marine Pollution Bulletin, 20, 375±379. Macintosh, D. & Phillips, M. (1992) Environmental considerations in shrimp farming. Infofish International, 6, 38±92. Mattsson, J. & Linden, O. (1983) Benthic macrofauna succession under mussels, Mytilus edulis L. (Bivalvia), cultured on hanging long-lines. Sarsia, 68, 97±102. McGhie, T.K., Crawford, C., Mitchell, I. & O'Brien, D. (2000) The degradation of fish-cage waste in sediments during fallowing. Aquaculture, 187, 351±366. Mitsch, W. & Jorgensen, S. (1989) Ecological Engineering: An Introduction to Ecotechnology. Wiley, New York.

370

Ecological Aquaculture

Moe, J. & Svennevig, N. (1998) Holmenkollen Guidelines for Sustainable Aquaculture. Norwegian Academy of Technological Sciences, Trondheim, Norway. Morrisey, D.J., Gibbs, M., Pickmere, S. & Cole, R. (2000) Predicting impacts and recovery of marine-farm sites in Stewart Island, New Zealand, from the Findlay-Watling model. Aquaculture, 185, 257±271. NACA/FAO (Network of Aquaculture Centres in Asia/Food and Agriculture Organization of the United Nations) (2000) Aquaculture Development Beyond 2000: The Bangkok Declaration and Strategy. NACA, Bangkok, and FAO, Rome. Naylor, R.L., Goldberg, R.J., Mooney, H. et al. (1998) Nature's subsidies to shrimp and salmon farming. Science, 282, 883±884. Naylor, R.L., Goldberg, R.G., Primavera, J.H. et al. (2000) Effect of aquaculture on world fish supplies. Nature, 405, 1017±1024. Neori, A., Krom, M., Ellner, S. et al. (1996) Seaweed biofilters as regulators of water quality in integrated fish±seaweed culture units. Aquaculture, 141, 183±199. Neori, A., Shpigel, M. & Ben-Ezra, D. (2000) A sustainable integrated system for culture of fish, seaweed and abalone. Aquaculture, 186, 279±291. Ochoa, E. (1997) Majagual: the tallest mangroves in the world. Intercoast. Special Mangrove Edition No. 1, March, 17. Odum, E.P. (1975) Ecology. Holt-Saunders, New York. Odum, E.P. (1989) Input management production systems. Science, 243, 177±181. Olsen, M. & Fitzsimmons, K. (1994) Integration of catfish and tilapia production with irrigation of cotton. In: World Aquaculture '94, Book of Abstracts, pp. 172. New Orleans, LA. Orr, D. (1999) Verbicide. Conservation Biology, 13, 696±699. Pauly, D. (2000) Global change, fisheries, and the integrity of marine ecosystems: the future has already begun. In: Ecological Integrity (eds D. Pimental, L. Westra & R. Noss), pp. 227±244. Island Press, Washington, DC. Pauly, D., Christensen, V., Dalsgaard, J., Froese, R. & Torres, F. Jr (1998) Fishing down marine food webs. Science, 279, 860±869. Petrell, R. & Alie, Y. (1996) Integrated cultivation of salmonids and seaweeds in open systems. Hydrobiologia, 326/327, 67±73. Phillips, J., Harlan, A., Krumhar, K. et al. (1978) Studies of Potential Biological Components of Closed Life Support Systems for Large Space Habitats: Research and Technology Development Requirements, Costs, Priorities and Terrestrial Impacts. NASA Ames Research Center, Tucson, AZ. Pimental, D. & Pimental, M. (1979) Food, Energy and Society. Wiley, New York. Primavera, J.H. (1991) Intensive prawn farming in the Philippines: ecological, social and economic implications. Ambio, 20, 28±33. Pullin, R., Rosenthal, H. & Maclean, J. (1993) Environment and Aquaculture in Developing Countries. ICLARM Conference Proceedings 31. ICLARM, Manila. Pullin, R., Froese, R. & Pauly, D. (2001) Indicators for the sustainabilty of aquaculture. In: Ecological and Genetic Implications of Aquaculture Activities (ed. T. Bert), in press. Kluwer Press. Rakocy, J. & Hargreaves, J. (1993) Integration of vegetable hydroponics with fish culture: a review. In: Techniques for Modern Aquaculture (ed. J.-K. Wang) pp. 112±136. American Society of Agriculture Engineers, St. Joseph, MI.

Ecology as the Paradigm for the Future of Aquaculture

371

Rawitscher, M. & Mayer, J. (1977) Nutritional outputs and energy inputs in seafoods. Science, 198, 261±264. Rees, W. (2000) Patch disturbance, ecofootprints, and biological integrity: revisiting the limits to growth (or why industrial society is inherently unsustainable). In: Ecological Integrity (eds D. Pimental, L. Westra & R. Noss), pp. 139±156. Island Press, Washington, DC. Rees, W. & Wackernagel, M. (1994) Ecological footprints and appropriated carrying capacity: measuring the natural capital requirements of the human economy. In: Investing in Natural Capital: The Ecological Economics Approach to Sustainability (eds A.M. Jansson, M. Hammer, C. Folke & R. Costanza), pp. 223±248. Island Press, Washington, DC. Rees, W. & Wackernagel, M. (1996) Our Ecological Footprint. New Society Publishers, Gabriola Island, BC, Canada. Rensel, J. (2001) Salmon net pens in Puget Sound. The Advocate, February 2000, pp. 66±69. Ritz, D.A., Lewis, M. & Shen, M. (1989) Response to organic enrichment of infaunal macrobenthic communities under salmonid sea cages. Marine Biology, 103, 211±214. Robinson, E., Li, M. & Manning, B. (2000) Catfish nutrition. The Advocate, April 2000, pp. 37±38. Roth, E., Rosenthal, H. & Burbridge, P. (2000) A discussion on the use of the sustainability index: `ecological footprint' for aquaculture production. Aquatic Living Resources, 13, 461±469. Ruddle, K. & Zhong, G. (1988) Integrated Agriculture±Aquaculture in South China: The DikePond System of the Zhujiang Delta. Cambridge University Press, Cambridge. Ryther, J. (1969) Photosythesis and fish production in the sea. Science, 166, 72±76. Salzman, J. (1991) Environmental labelling in OECD countries. OECD Report No. 1. Organization for Economic Cooperation and Development, Paris. Sandifer, P. & Hopkins, S. (1996) Conceptual design of a sustainable pond-based shrimp culture system. Aquacultural Engineering, 15, 41±52. Scherr, J. (1997) Shrimp Tribunal Online. http://earthsummitwatch.org/shrimp Schrecker, J. (2000) The cost of the wild: international equity and the losses from environmental conservation. In: Ecological Integrity (eds D. Pimental, L. Westra & R. Noss), pp. 301±316. Island Press, Washington, DC. Schroeder, G. (1980) Fish farming in manure loaded ponds. In: Integrated Agriculture±Aquaculture Farming Systems (eds R.S.V. Pullin & Z. Shehadeh), pp. 73±86. ICLARM, Manila, Philippines. Schwartz, M. & Boyd, C.E. (1995) Effluent quality during harvest of channel catfish from watershed ponds. Progressive Fish-Culturist, 56, 25±32. Shpigel, M., Neori, A., Popper, D. & Gordin, H. (1993) A proposed model for `environmentally clean' land-based culture of fish, bivalves and seaweeds. Aquaculture, 117, 115±128. Shpigel, M. & Neori, A. (1996) The integrated culture of seaweed, abalone, fish and clams in modular intensive land-based systems: I. Proportions of size and projected revenues. Aquacultural Engineering, 15, 313±326. Soule, M. (1991) Conservation: tactics for a constant crisis. Science, 253, 744±750. Starkey, T. (2001) Bans, quotas hit animal protein supply. The Advocate, February, p. 29. Stewart, J. (1998) Sharing the waters: an evaluation of site fallowing, year class separation and distances between sites for fish health purposes on Atlantic salmon farms. Canadian Technical Report of Fisheries and Aquatic Sciences 2218, Dartmouth, Nova Scotia, Canada.

372

Ecological Aquaculture

Sveinson, A. & Engelstad, M. (2001) Gaining consumer confidence. Fish Farming International, 28, 17. Troell, M., Halling, C., Nilsson, A., Buschmann, A., Kautsky, N. & Kautsky, L. (1997) Integrated marine cultivation of Gracilaria chilensis (Gracilariales, Rhodophyta) and salmon cages for reduced environmental impact and increased economic output. Aquaculture, 156, 45±61. Troell, M., Ronnback, P., Halling, C. & Buschmann, A. (1999a) Ecological engineering in aquaculture: use of seaweeds for removing nutrients from intensive aquaculture. Journal of Applied Phycology, 11, 89±97. Troell, M., Kautsky, N. & Folke, C. (1999b) Applicability of integrated coastal aquaculture systems. Ocean & Coastal Management, 42, 63±69. United Nations (1994) World Urbanization Prospects: The 1994 Revision. United Nations, New York. Wang, J-K. (1990) Managing shrimp pond water to reduce discharge problems. Aquaculture Engineering, 9, 61±73. Watson, R. (1999) Common themes for ecologists in global issues. Journal of Applied Ecology, 36, 1±10. WCED (World Commission on Environment and Development) (1987) Our Common Future. Oxford University Press, Oxford, UK. Weatherly, A. & Cogger, B. (1977) Fish culture: problems and prospects. Science, 197, 427±430. Wessells, C.R., Johnston, R.J. & Donath, H. (1999) Assessing consumer preferences for ecolabelled seafood: the influence of species, certifier, and household attributes. American Journal of Agriculture Economics, 5, 1084±1089. Weston, D.P. (1990) Quantitative examination of macrobenthic community changes along an organic enrichment gradient. Marine Ecology Progress Series, 61, 233±244. Wilks, A. (1995) Prawns, profit and protein. The Ecologist, 25, 120±125. Wohlfarth, G. & Schroeder, G. (1979) Use of manure in fish farming ± a review. Agricultural Wastes, 1, 279±299. Worldwatch Institute (2001) State of the World. Worldwatch Institute, Washington, DC. Wu, R.S.S. (1995) The environmental impact of marine fish culture: towards a sustainable future. Marine Pollution Bulletin, 31, 4±12. Wu, R.S.S., Lam, K., MacKay, D., Lau, T. & Yam, V. (1994) Impact of marine fish farming on water quality and bottom sediment: a case study of the sub-tropical environment. Marine Environmental Research, 38, 115±145. Yarris, L. (1981) Trout manure as fertilizer. Agriculture Research, 30, 9. Zweig, R. (1992) Ecological±economic computer simulation models for the design and analysis of integrated aquaculture systems. Journal of the Asian Farming Systems Association, 1, 337±350.

Index

Africa history of aquaculture 7, 8±9 silvofisheries 194 village-based aquaculture ecosystems 145±57 ecological footprints 151±2 economic threshold for commercial transformation 156±7 environment, economics and food security 152±3 information transfer and sustained adoption 154±5 productivity and 149, 155±6 realization of potential 153±4 research for development of 147±9 technology adoption and transmission 149±50 agriculture 3 irrigation 3, 7, 11, 103 integrated fish and field crop system for arid areas 280±83 origins of 5±6, 9, 11 pollution and 104, 353 sustainability and 105 use of tilapia as fertilizer or feed 137±8 allele-specific oligonucleotide (ASO) hybridization 53, 54 allozyme electrophoresis 51, 58 Americas history of aquaculture 7, 18±19 see also Hawaii anchovies 319 arid area aquaculture 263±84

feeding 274 future research 283±4 integrated field crop production 280±83 stocking rates 269±73 water consumption 279±80 water quality management 275±9 Asia history of aquaculture 7, 9±13, 20, 21, 24, 25±6, 104 see also individual countries assessment methods participatory 108±10 role of aquaculture in restoration of coastal fisheries 320±22 Audubon Society 364 Australia 5 behavior escape of cultured species and 81, 93 stock enhancement and 326±7 Benin 7 biochemical cycles 206±10 bream 20 Britain 17±18, 345, 362±3 brown trout, interbreeding with escaped salmon 89 Brundtland Commission 118, 358 cage aquaculture ecosystems in hydropower reservoirs 286±309 analysis of successes and failures 299±301 ensuring sustainability 305±6 equity and rural poor and 304±5

374

Index

institutional framework and planning process 289±91 problems 301±4 training programs 291±9 community integrated aquaculture schools 296±7 farmer-to-farmer visits 297 information resources 298 participatory methods 294±6 study tours 298±9 traditional methods 291±4 capture fisheries see fishing carp 360 Chinese aquaculture 11, 12±13, 20 European aquaculture 15±16, 20±21 Indonesia 300±301 mangrove forest aquaculture 175 certification of sustainable ecological aquaculture systems 362±4 chemical markers 322 China agriculture in 5, 9, 11, 104 aquaculture in 9±13, 20, 21, 24, 25, 104 silvofisheries 174±9 meat consumption 340 cloning 52 cod 315 construction of ponds 21±2, 230±33 consumption of fish 129±30 crabs mangrove forest aquaculture 203±4, 220, 227, 244 Africa 194 Indonesia 166, 169, 173 Malaysia 179±81 Philippines 187, 188±93 Vietnam 185±6 Czech Republic, aquaculture in 16 dace 20 decomposition 201±6 definitions of aquaculture 4±5 ecological/sustainable aquaculture 118±22, 343±4 demand for fish 132±4, 136±7

denaturing gradient gel electrophoresis (DGGE) technique 53, 54 diseases escape of cultured species and 83±5, 88, 93 stock enhancement and 325±6 DNA markers see genetic tags minisatellite and microsatellite 48, 49, 55±6 mitochondrial 48, 50±51 nuclear 48, 50 sequencing 47, 52±5 dry areas see arid area aquaculture ecolabeling 362±4 ecological ethics 341 ecology ecological footprints 151±2, 354 escape of cultured species and 85±6 as paradigm for future of aquaculture 339±65 aquaculture pollution and habitat losses 348, 353±6 certification of sustainable ecological aquaculture systems 362±4 choice of most efficient protein production system 346±53 defining ecological aquaculture 343±4 examples of ecological aquaculture systems 356±8 global crisis in human protein needs 339±41 incorporation into planning for sustainable fisheries 344±6 need for progress towards sustainable systems 358±60 need for science in development of ecological aquaculture 360±62 social ecology of aquaculture 341±3 see also sustainability of ecosystems economics bioeconomic modeling of restoration of coastal fisheries 321±2 demand for fish 132±4, 136±7 environment and food security and 152±3 market structure for fish 128, 137, 138

Index

prices of fish 134±5, 137 threshold for commercial transformation in village-based aquaculture ecosystems 156±7 effluent treatment 237±9 Egypt, aquaculture in ancient times 8±9 energy flow silvofisheries 195±210 decomposition 201±6 key energy transfer components 199±200 recycling of nutrients/biochemical cycles 206±10 equity, cage aquaculture ecosystems in hydropower reservoirs and 301, 304±5 escape of cultured species 61, 66 salmonid case study 77±95 effects of domestication 78±83 life history strategies of escapees 89±94 mitigation of aquaculture escapees in the wild 94±5 occurrence of salmonid aquaculture escapees in the wild 78 potential interaction of escapees with wild conspecifics 83±9 ethics, ecological 341 Europe, history of aquaculture 13±18, 20±22, 24 extensive aquaculture 4 feral cultured species see escape of cultured species Fiji case study of tilapia 125±40 consumption of fish by Fijians 129±30 demand estimates for fish 132±4, 136±7 introduction of species 126±8 market structure 128, 137, 138 preferences by ethnic groups 131±2 prices of fish 134±5, 137 recommendations 138±40 use of tilapia as fertilizer or feed 137±8 fish farming escapement see escape of cultured species methods 19±23 broodstock management and spawning 20±21

375

feeding practices 22±3 pond construction and management 21±2 see also cage aquaculture ecosystems in hydropower reservoirs fish meal and oils 350±3 fishing aquaculture distinguished from 4 incorporation of ecology into planning for sustainable fisheries 344±6 managed fisheries 6 role of aquaculture in restoration of coastal fisheries 314±31 establishing objectives and measures of success 319±20 evaluation 330±31 hatchery techniques 323±7 pilot-scale studies 322±3 pre-release evaluations and feasibility assessment 320±22 program implementation and monitoring 329±30 program planning and review 317±19 release strategies 327±9 stock enhancement 314, 315±16 food (animal), use of tilapia as animal feed 137±8 food (fish) arid area aquaculture 274 escape of cultured species and 81, 89±90, 92 fish meal and oils 350±3 food value of silvofisheries 219±27 contribution of mangroves 219±24 utilization by aquaculture species 224±7 historical feeding practices 4±5, 22±3 food (human) choice of most efficient protein production system 346±53 consumption of fish 129±30 food security 147, 152±3 global crisis in human protein needs 339±41 footprints, ecological 151±2, 354 forest aquaculture see silvofisheries France, aquaculture in 16

376

Index

gender 110 genetic tags 47±67, 322±3 meaning of 47±8 methodologies for obtaining and utilizing 52±9 types of 48±52 uses for aquaculture 59±67 future 66±7 intensive aquaculture 59±62 stock enhancement 62±6 genetics deterioration of tilapia 125 escape of cultured species and 86, 87±9 stock enhancement and 320±21 Germany, aquaculture in 16 Global Aquaculture Alliance (GAA) 363 goldfish 21 greenwater tank culture see arid area aquaculture

Europe 13±18, 20±22, 24 Hawaii 30±42 history of mariculture 13, 14, 26 Hong Kong, silvofisheries 174±9 Human Genome Project 52 Hungary, aquaculture in 16 hunting 5 hybrids, escape of cultured species and 61, 81, 86, 87±9 hydropower reservoirs resettlement by cage aquaculture ecosystems 286±309 analysis of successes and failures 299±301 ensuring sustainability 305±6 equity and rural poor and 301, 304±5 institutional framework and planning process 289±91 problems 301±4 training programs 291±9

habitats escape of cultured species and 82 loss of 348, 353±6 stock enhancement and 328±9 haddock 315 harvesting, silvofisheries 235±7 hatchery techniques 323±7 Hawaii aquaculture in 30±42 brackishwater pond ecosystems 34±5 context of Hawaiian innovations in evolution of mariculture ecosystems 41±2 freshwater pond ecosystems 34 integrated aquaculture ecosystems 32±4 nearshore fish traps 39±41 nearshore mariculture ecosystems 35±9 social ecology of ancient Hawaii and 30±32 herring 315 history of agriculture 5±6, 9, 11 history of aquaculture 3, 6±26 Africa 7, 8±9 Americas 7, 18±19 Asia 7, 9±13, 20, 21, 24, 25±6, 104

ice production 140 India aquaculture in 12, 13 silvofisheries 193±4 indigenous knowledge, importance of 103, 295±6 Indonesia 12, 126 agriculture in 103 case study of resettlement of hydropower reservoirs 289±309 analysis of successes and failures 299±301 ensuring sustainability 305±6 equity and rural poor and 301, 304±5 institutional framework and planning process 289±91 problems 301±4 training programs 291±9 mariculture 13 silvofisheries 164±74 urbanization 288 information transfer and sustained adoption 154±5 input management 106, 121 intensive aquaculture 5 uses for genetic tags 59±62

Index

International Center for Living Aquatic Resources Management (ICLARM) 145±7, 153±6 International Union for the Conservation of Nature (IUCN) 110±11 introns 49 Ireland 363 irrigation 3, 7, 11, 103 integrated fish and field crop system for arid areas 280±83 Israel, arid area aquaculture in 264, 279±80 Italy, ancient aquaculture in 13±15 jacks, mangrove forest aquaculture 169 Japan aquaculture in 12, 21, 346 fish consumption in 129 mariculture in 13 Kenya, mangrove forest aquaculture in 194 Korea, aquaculture in 12 labeling 362±4 Laos, aquaculture in 13 local adaptation 87 mackerel 315 major histocompatibility complex (MHC) 49±50 Malawi village-based aquaculture ecosystems 145±57 ecological footprints 151±2 economic threshold for commercial transformation 156±7 environment, economics and food security 152±3 information transfer and sustained adoption 154±5 productivity and 149, 155±6 realization of potential 153±4 research for development of 147±9 technology adoption and transmission 149±50 Malaysia, silvofisheries 179±81

377

mangrove forest aquaculture systems 161±247 Africa 194 brackishwater pond/mangrove rotations 194±5 energy flow 195±210 decomposition 201±6 key energy transfer components 199±200 recycling of nutrients/biochemical cycles 206±10 food value 219±27 contribution of mangroves 219±24 utilization by aquaculture species 224±7 future recommendations 239±44 Hong Kong 174±9 India 193±4 Indonesia 164±74 Malaysia 179±81 management of silvofisheries system 227±33 models of 163±4, 244±5 Philippines 186±93 pond operation 233±9 research needs 246±7 selection of appropriate mangrove species 210±19 production 216±19 survival 213±16 Sri Lanka 193 stocking and harvesting 235±7 system models 244±5 Thailand 181±2 Vietnam 182±6 maps, aqua/agroecosystem mapping 111 mariculture 162 origins of 13, 14, 26 Hawaii 30±42 Marine Stewardship Council (MSC) 364 market structure for fish 128, 137, 138 Mauritius, mariculture in 13 Mexico, aquaculture in 18±19 migration escape of cultured species and 85, 89±93 stock enhancement and 319 milkfish 13, 34, 36 mangrove forest aquaculture 166, 169, 172, 173, 213

378

Index

minisatellite and microsatellite DNA 48, 49, 55±6 mitochondrial DNA 48, 50±51 models bioeconomic modeling of restoration of coastal fisheries 321±2 pictorial 111±12 silvofisheries 163±4, 244±5 monasteries 15, 17, 18, 24 morphology escape of cultured species and 82 morphological tagging 51±2 mullet 36, 39, 169, 198, 328 New Zealand 363 Norway, salmon farming in 79±80 nuclear DNA 48, 50 organic produce 106, 362±3 origins of aquaculture see history of aquaculture ownership 4, 242±4 oysters 13, 14, 363 parasites, escape of cultured species and 84, 88, 93 participatory approaches 106±8, 147±9, 155±6 assessment and planning methods 108±10 importance of indigenous knowledge 103, 295±6 participatory technology development (PTD) 107±8, 109, 110±11, 122 aqua/agroecosystem mapping 111 pictorial modeling 111±12 problem diagramming 112±14 role of aquaculture research station 115±18 systems diagramming 114±15 training programs for resettlement of hydropower reservoirs by cage aquaculture ecosystems 294±6 perch 20 Philippines, silvofisheries 186±93 physical tagging 65±6, 323 physiology, escape of cultured species and 83

pictorial modeling 111±12 pike 19, 20 planning methods case study of resettlement of hydropower reservoirs 289±91 participatory 108±10 restoration of coastal fisheries 317±19 pollack 315 pollution 104, 303±4, 353±6 polymerase chain reaction (PCR) 53 ponds 7 in ancient times 8±9, 14±15 construction 21±2, 230±33 Hawaiian 32±41 management 21±2, 138, 149, 194±5, 227±39 in Middle Ages 16, 17±18 silvofisheries 163±4 brackishwater pond/mangrove rotations 194±5 design and construction 230±33 empang parit model in Indonesia 164±74 gei wai model in Hong Kong 174±9 integrated silvofisheries with agriculture in Philippines 186±93 mangrove/crab pen culture in Malaysia 179±81 pond operation 233±9 shrimp/mangrove model 181±6 site selection 228±30 stocking and harvesting 235±7 water management 234±5 predation escape of cultured species and 81, 82 loss of fish from 77±8 prices of fish 134±5, 137 problem diagramming 112±14 productivity 149, 155±6 proteins choice of most efficient protein production system 346±53 genetic tags and 51 global crisis in human protein needs 339±41 quantitative trait loci (QTL) 60

Index

rainbow trout, genetic tags 60 random amplified polymorphic DNA (RAPD) markers 57±8 recycling of nutrients/biochemical cycles 206±10 religion, aquaculture and 8±9, 15, 17, 18, 21 reproduction 20±21 escape of cultured species and 80±81, 82, 86 sterilization 94 research research stations 115±18 silvofisheries 246±7 village-based aquaculture ecosystems and 147±9 reservoirs see hydropower reservoirs restocking see stock enhancement restriction fragment length polymorphism (RFLP) analysis 56±7 roach 20 salmonids case study of escape of cultured species 77±95 effects of domestication 78±83 life history strategies of escapees 89±94 mitigation of aquaculture escapees in the wild 94±5 occurrence of salmonid aquaculture escapees in the wild 78 potential interaction of escapees with wild conspecifics 83±9 development of farming of 19, 23 organic produce 362±3 sea ranching 314 sardines 319 sea bass mangrove forest aquaculture 169, 173, 175 stock enhancement 316, 321, 327 sea bream mangrove forest aquaculture 175 stock enhancement 315, 316 sea ranching 314 sediments 21±2, 178 semi-intensive aquaculture 4±5 Senegal, mangrove forest aquaculture in 194

379

sequencing of DNA 47, 52±5 shrimp ecolabeling for 363±4 mangrove forest aquaculture 198, 212, 224±7, 238 Hong Kong 174±7 Indonesia 161, 166, 169, 171, 172, 173 Philippines 188 Thailand 181 Vietnam 182±6 siganids 169 silting 21±2, 178 silvofisheries 161±247 Africa 194 brackishwater pond/mangrove rotations 194±5 energy flow 195±210 decomposition 201±6 key energy transfer components 199±200 recycling of nutrients/biochemical cycles 206±10 food value 219±27 contribution of mangroves 219±24 utilization by aquaculture species 224±7 future recommendations 239±44 Hong Kong 174±9 India 193±4 Indonesia 164±74 Malaysia 179±81 management of silvofisheries system 227±33 models of 163±4, 244±5 Philippines 186±93 pond operation 233±9 research needs 246±7 selection of appropriate mangrove species 210±19 production 216±19 survival 213±16 Sri Lanka 193 stocking and harvesting 235±7 system models 244±5 Thailand 181±2 Vietnam 182±6

380

Index

single-stranded conformation polymorphism (SSCP) analysis 53±4 site selection 228±30 social ecology of aquaculture 341±3 Soil Association 363 spawning see reproduction Sri Lanka aquaculture in 126 silvofisheries 193 steelhead trout 86 life history strategies of escapes 89±94 sterilization 94 stock enhancement 77 genetic tags and 62±7 role of aquaculture in restoration of coastal fisheries 314±31 establishing objectives and measures of success 319±20 evaluation 330±31 hatchery techniques 323±7 pilot-scale studies 322±3 pre-release evaluations and feasibility assessment 320±22 program implementation and monitoring 329±30 release strategies 327±9 stock management 20±21 stocking arid area aquaculture 269±73 silvofisheries 235±7 sustainability of ecosystems 103±22 cage aquaculture ecosystems in hydropower reservoirs 286±309 analysis of successes and failures 299±301 ensuring sustainability 305±6 equity and rural poor and 301, 304±5 institutional framework and planning process 289±91 problems 301±4 training programs 291±9 case study of tilapia in Fiji 125±40 consumption of fish by Fijians 129±30 demand estimates for fish 132±4, 136±7 introduction of species 126±8 market structure 128, 137, 138

preferences by ethnic groups 131±2 prices of fish 134±5, 137 recommendations 138±40 use of tilapia as fertilizer or feed 137±8 definition of sustainable ecological aquaculture 118±22 ecology as paradigm for future of aquaculture 339±65 aquaculture pollution and habitat losses 348, 353±6 certification of sustainable ecological aquaculture systems 362±4 choice of most efficient protein production system 346±53 defining ecological aquaculture 343±4 examples of ecological aquaculture systems 356±8 global crisis in human protein needs 339±41 incorporation into planning for sustainable fisheries 344±6 need for progress towards sustainable systems 358±60 need for science in development of ecological aquaculture 360±62 social ecology of aquaculture 341±3 farmer participatory approaches 106±8 importance of indigenous knowledge 103, 295±6 mangrove forest aquaculture systems 161±247 Africa 194 brackishwater pond/mangrove rotations 194±5 energy flow 195±210 food value 219±27 future recommendations 239±44 Hong Kong 174±9 India 193±4 Indonesia 164±74 Malaysia 179±81 management of silvofisheries system 227±33 models of 163±4, 244±5 Philippines 186±93

Index

pond operation 233±9 research needs 246±7 selection of appropriate mangrove species 210±19 Sri Lanka 193 system models 244±5 Thailand 181±2 Vietnam 182±6 need for 103±6 participatory assessment and planning methods 108±10 participatory technology development (PTD) 107±8, 109, 110±11, 122 aqua/agroecosystem mapping 111 aquaculture research station and 115±18 pictorial modeling 111±12 problem diagramming 112±14 systems diagramming 114±15 system models, silvofisheries 244±5 systems diagramming 114±15 Tanzania, mangrove forest aquaculture in 194 tautog 315 taxation 243, 244 technology adoption and transmission of 149±50 participatory technology development (PTD) 107±8, 109, 110±11, 122 aqua/agroecosystem mapping 111 pictorial modeling 111±12 problem diagramming 112±14 role of aquaculture research station 115±18 systems diagramming 114±15 temperature gradient gel electrophoresis (TGGE) 54 tench 20 Thailand, silvofisheries 181±2 tilapia 8, 9 arid area aquaculture 264, 265, 267, 269±84 case study of tilapia in Fiji 125±40 consumption of fish by Fijians 129±30 demand estimates for fish 132±4, 136±7 introduction of species 126±8 market structure 128, 137, 138

381

preferences by ethnic groups 131±2 prices of fish 134±5, 137 recommendations 138±40 use of tilapia as fertilizer or feed 137±8 hydropower reservoir resettlement aquaculture and 298±9 mangrove forest aquaculture 166, 169, 172, 176, 197, 198, 239 training programs resettlement of hydropower reservoirs by cage aquaculture ecosystems 291±9 community integrated aquaculture schools 296±7 farmer-to-farmer visits 297 information resources 298 participatory methods 294±6 study tours 298±9 traditional methods 291±4 transport 140 traps 39±41 turbot 316 United Kingdom 17±18, 345, 362±3 United Nations Commission on Environment and Development 339 Conference on Environment and Development 161 Food and Agricultural Organization (FAO) 4, 145, 345, 359±60 United States of America 126, 344, 346 urbanization 288, 339 Vietnam aquaculture in 12, 13 silvofisheries 182±6 village-based aquaculture ecosystems 145±57 ecological footprints 151±2 economic threshold for commercial transformation 156±7 environment, economics and food security 152±3 information transfer and sustained adoption 154±5 productivity and 149, 155±6

382

Index

realization of potential 153±4 research for development of 147±9 technology adoption and transmission 149±50 Virgin Islands, arid area aquaculture in 263±84 water consumption in arid area aquaculture 279±80

effluent treatment 237±9 irrigation 3, 7, 11, 103 integrated fish and field crop system for arid areas 280±83 management arid area aquaculture 275±9 silvofisheries 234±5 pollution 104, 303±4, 353±6 winter flounder 315