Benguela: Predicting a Large Marine Ecosystem, Volume 14 (Large Marine Ecosystems)

Benguela: Predicting a Large Marine Ecosystem Large Marine Ecosystems – Volume 14 Series Editor: Kenneth Sherman Dir...

Author: Vere Shannon | Gotthilf Hempel | Coleen Moloney | John D. Woods | Paola Malanotte-Rizzoli

37 downloads 799 Views 17MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Benguela: Predicting a Large Marine Ecosystem

Large Marine Ecosystems – Volume 14 Series Editor:

Kenneth Sherman Director, Narragansett Laboratory and Office of Marine Ecosystem Studies NOAA-NMFS, Narragansett, Rhode Island, USA and Adjunct Professor of Oceanography Graduate School of Oceanography, University of Rhode Island Narragansett, Rhode Island, USA

On the cover The main cover picture illustrating the complexity of the Benguela Current Large Marine Ecosystem (BCLME) and adjacent regions is an AQUA MODIS level three, 4 km resolution, chlorophyll image for the week 2-10 February 2004, obtained from the NASA Oceancolor webpage: http://oceancolor.gsfc.nasa.gov/cgi/level3.pl The top picture, with the BCLME box inset, is the global map of average primary productivity and the boundaries of the 64 Large Marine Ecosystems (LMEs) of the world, available at www.edc.uri.edu/lme. The annual productivity estimates are based on SeaWIFS data collected between September 1998 and August 1999. The color enhanced image was provided by Rutgers University.

A list of recent publications in this series appears at the end of this volume.

Benguela: Predicting a Large Marine Ecosystem Edited by Vere Shannon Honorary Professor, Department of Oceanography University of Cape Town South Africa Gotthilf Hempel Science Advisor, Senate of Bremen, Germany Emeritus Professor, Bremen and Kiel Universities Germany Paola Malanotte-Rizzoli Professor, Department of Earth, Atmospheric and Planetary Sciences Massachusetts Institute of Technology Cambridge, Massachusetts United States Coleen Moloney Senior Lecturer, Department of Zoology University of Cape Town South Africa John Woods Emeritus Professor, Department of Earth Science and Engineering Imperial College London United Kingdom

Technical editor Sara P. Adams - Large Marine Ecosystem Program - Narragansett RI - USA

Amsterdam - Boston - Heidelberg - London - New York - Oxford - Paris San Diego - San Francisco - Singapore - Sydney - Tokyo

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, UK First edition 2006 Copyright © 2006 Elsevier B.V. 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 without the prior written permission of the publisher. Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (+44) (0) 1865 843830; fax (+44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material. Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made. Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN-13: 978-0-444-52759-2 ISBN-10: 0-444-52759-1 ISBN-13: 978-0-444-52760-8 ISBN-10: 0-444-52760-5 (CD-rom) ISSN: 1570-0461 For information on all Elsevier publications visit our website at books.elsevier.com Printed and bound in The Netherlands 06 07 08 09 10 10 9 8 7 6 5 4 3 2 1

v

Series Editor’s Introduction The world’s coastal ocean waters continue to be degraded from unsustainable fishing practices, habitat degradation, eutrophication, toxic pollution, aerosol contamination, and emerging diseases. Against this background is a growing recognition among world leaders that positive actions are required on the part of governments and civil society to redress global environmental and resource degradation with actions to recover depleted fish populations, restore degraded habitats and reduce coastal pollution. No single international organization has been empowered to monitor and assess the changing states of coastal ecosystems on a global scale, and to reconcile the needs of individual nations to those of the community of nations for taking appropriate mitigation and management actions. However, the World Summit on Sustainable Development convened in Johannesburg in 2002 in recognition of the importance for coastal nations to move more expeditiously toward sustainable development and use of ocean resources, declared that countries should move to introduce ecosystem-based assessment and management practices by 2010, and by 2015, restore the world’s depleted fish stocks to maximum levels of sustainable yields. At present, 121 developing countries are moving toward these targets in joint international projects supported, in part, by financial grants by the Global Environment Facility in partnership with scientific and technical assistance from UN partner agencies (e.g. UNIDO, UNEP, UNDP, IOC, FAO), donor countries and institutions and nongovernmental organizations including the IUCN (World Conservation Union). Many of these projects are linked to ecosystem-based efforts underway in Europe and North America in a concerted effort to overcome the North-South digital divide. The volumes in the new Elsevier Science series on Large Marine Ecosystems are bringing forward the results of ecosystem-based studies for marine scientists, educators, students and resource managers. The volumes are focused on LMEs and their productivity, fish and fisheries, pollution and ecosystem health, socioeconomics and governance. This volume in the new series, “Benguela: Predicting a Large Marine Ecosystem” progresses systematically and innovatively from studies that set the present scene, to studies constituting the “cutting edge” of forecasting changing states of the Benguela Current LME (BCLME) and move ahead to a fully integrated BCLME forecasting system. The authors are forward looking and quite deliberate in tightening up the linkages between science based assessments of the changing states of the BCLME and the socioeconomic benefits to be derived by the people of Angola, Namibia, and South Africa from the application of an LME forecasting system that is readily transparent and clearly adaptable to others of the world’s 64 LMEs. The volume provides an important LME baseline from which to support the stated goals of the Global Ocean Observing System (GOOS) and the Global Environmental Observing System of Systems (GEOSS). The volume is the fourteenth in the Elsevier Science New LME Series. Recent volumes in the Series are listed at http://www.elsevier.com/wps/find/bookseriesdescription.cws_home/BS_LME/description. Kenneth Sherman, Series Editor Narragansett, Rhode Island

ThisPage pageis intentionally This Intentionally left Leftblank Blank

vii


ix

THE PUBLICATION OF THIS BOOK HAS BEEN MADE POSSIBLE THROUGH THE GENEROUS SPONSORSHIP AND SUPPORT PROVIDED BY THE ABOVE ORGANISATIONS


xi

Foreword In recent years, there has been considerable international interest in ocean monitoring and operational oceanography to enable responsible management of marine and coastal resources and support a variety of other maritime activities through timeous provision of appropriate information, including forecasts. This is reflected in the strategies of the International Waters Programme of the Global Environment Facility (GEF) through its Large Marine Ecosystem (LME) initiatives and the Global Ocean Observing System (GOOS) of the Intergovernmental Oceanographic Commission (IOC), and endeavours are at present being made to develop close links between GOOS and the LMEs. The need for accurate forecasting, contingency planning and effective reporting mechanisms to managers and the public at large has more recently been highlighted in the wake of the devastating tsunami in SE Asia in 2005, hurricane Katrina in the USA and the broader global impacts of El Niño events and climate change. In developed countries and regions, significant advances have been made in ocean monitoring and observing systems and new generations of metocean buoys and satellite technology have been put in place which allow real-time monitoring and modeling of the processes taking place in the marine environment. This is not the case in many developing parts of the world, Africa in particular. The IOC, through its GOOS-Africa Programme and the network of African LME Programmes, is attempting to address this by developing affordable, implementable and sustainable ocean observing systems to service the needs of African countries, regions and the continent as a whole. Of major concern for the Benguela region and a key goal of the Strategic Action Plan (SAP) of the BCLME Programme has been the establishment of a viable and costeffective forecasting system that can provide resource and environmental managers and indeed the public at large in the region with early warning of catastrophic events. Benguela Niños, low oxygen anomalies and extensive harmful algal blooms periodically occur in the marine and coastal environment of the BCLME with devastating consequences on the living marine resources. This is particularly true for the northern Benguela. In November 2004, a four day International Workshop on Forecasting and Data Assimilation in the Benguela and Comparable Systems was held in Cape Town to address the key policy actions of forecasting environmental variability and its impacts on the BCLME. The information, knowledge, wisdom and advice resulting from this workshop are captured in this definitive peer-reviewed book entitled, Benguela – Predicting a Large Marine Ecosystem. This book will be a significant contribution to the BCLME Programme and its sustainable management objectives, and a blueprint for application

Foreword

xii

in other LMEs around the world and for fast-tracking the objectives of several international science organisations. The BCLME Programme is highly appreciative of the efforts of all the contributors to this volume and, in particular, I wish to thank Professors Vere Shannon, Gotthilf Hempel, Paola Malanotte-Rizzoli, Coleen Maloney and John Woods for so enthusiastically editing this volume and putting together the accompanying CD-ROM. Special thanks are due to Dr. Sara Adams, LME Program at the Northeast Fisheries Science Center’s Narragansett Laboratory in Rhode Island, USA, for her technical editing skills and preparing the book in camera ready format for publication. Finally, we of the Benguela LME community would like to dedicate this volume, the 14th in Elsevier’s Large Marine Ecosystems Series to Dr Kenneth Sherman, the founding father of the “Large Marine Ecosystems ” global movement which is now widely recognised as one of the most effective and practical strategies for operationalising the ecosystem approach to management of marine resources. His tireless efforts, support and guidance over the years in collaboration with Al Duda of the Global Environment Facility (GEF) have ensured the successful implementation of the BCLME Programme and of LME initiatives in other parts of the world. The publication of this volume was made possible through support of the BCLME Programme, the National Oceanic and Atmospheric Administration (NOAA) and the University of Cape Town. Michael John O’Toole Chief Technical Advisor BCLME Programme

xiii

Contributors Justin Ahanhanzo GOOS Africa Intergovernmental Oceanographic Commission IOC of UNESCO Paris, FRANCE Jürgen Alheit Baltic Sea Research Institute Institut für Ostseeforschung Warnemünde GERMANY Miriam Andrioli Maritime Division Forecasting Department Servicio Meteorologico Nacional 1002 Buenos Aires ARGENTINA Hernam Arango Institute of Marine and Coastal Sciences Rutgers University State University of New Jersey New Brunswick, New Jersey USA Claire Attwood Media Liaison Muizenberg 7945 Cape Town SOUTH AFRICA Awie Badenhorst SA Pelagic Fishing Industry Association SA Inshore Cape Town 8000 SOUTH AFRICA Geoff W. Bailey Department of Environmental Affairs and Tourism Marine and Coastal Management Roggebaai 8012 Cape Town, SOUTH AFRICA

Ray G. Barlow Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town, SOUTH AFRICA Chris Bartholomae Ministry of Fisheries and Marine Resources National Marine Research and Information Centre Swakopmund NAMIBIA Eric D. Barton Spanish Council for Scientific Research Instituto de Investigaciones Marinas Vigo, SPAIN Stewart Bernard Department of Oceanography University of Cape Town Rondebosch 7701 Cape Town SOUTH AFRICA Geoff Brundrit Department of Oceanography University of Cape Town Rondebosch 7701 Cape Town SOUTH AFRICA Deidre Byrne School of Marine Sciences University of Maine Orono, Maine USA Rudi Cloete Ministry of Fisheries and Marine Resources National Marine Research and Information Centre Swakopmund, NAMIBIA

xiv Andy C. Cockcroft Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town, SOUTH AFRICA R. J. M. Crawford Marine and Coastal Management (MCM) Department of Environmental Affairs and Tourism (DEAT) Roggebay 8012 Cape Town SOUTH AFRICA Philippe Cury Institut Recherche Développement (IRD) Centre de Recherche Halieutique Méditerranéenne et Tropicale Paris, Sète Cedex 10 FRANCE Antonio da Silva Instituto de Investigação Pesqueira Ministerio das Pescas Luanda, ANGOLA Hervé Demarcq Institut Recherche Développement (IRD) Centre de Recherche Halieutique Méditerranéenne et Tropicale Paris, Sète Cedex 10 FRANCE Chris M. Duncombe Rae Department of Environmental Affairs and Tourism Marine and Coastal Management (MCM) Roggebaai 8012 Cape Town SOUTH AFRICA Alex Fawcett Department of Oceanography University of Cape Town Rondebosch 7701 SOUTH AFRICA Katje Fennel Institute of Marine and Coastal Sciences Rutgers University New Brunswick, New Jersey USA

Contributors Wolfgang Fennel Baltic Sea Institute Institut für Ostseeforschung Warnemünde GERMANY Quilanda Fidel Instituto de Investigação Marinha Ministerio das Pescas Luanda ANGOLA John G. Field Zoology Department, Marine Biology Research Institute University of Cape Town Rondebosch 7701 Cape Town, SOUTH AFRICA Jim Fitzpatrick HydroQual Inc. MacArthur Blvd. Mahwah, New Jersey USA Pierre Florenchie Department of Oceanography University of Cape Town Rondebosch 7701 Cape Town, SOUTH AFRICA Regina Folorunsho Nigerian Institute for Oceanography and Marine Research Victoria Island Lagos, NIGERIA Peter J. S. Franks Scripps Institution of Oceanography University of California San Diego, California USA Pierre Fréon Institut Recherche Développement (IRD) Halieutique Méditerranéenne et Tropicale Paris, Cedex 10 FRANCE Ema Gomes Ministry of Petroleum of Angola 1279 C Luanda ANGOLA

xv

Contributors Paul Goodman Department of Geosciences University of Arizona, Tucson, AZ USA

Astrid Jarre Danish Institute for Fisheries Research North Sea Centre Hirtshals DENMARK

Leticia Greyling National Ports Authority of South Africa (NPA) Braamfontein, Jhb, 2017 SOUTH AFRICA

Ashley Johnson Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town SOUTH AFRICA

Marten Grundlingh Council for Scientific and Industrial Research – CSIR Environmentek 7599 Stellenbosch SOUTH AFRICA Johannes Guddal Norwegian Meteorological Institute (DNMI Region W)) Bergen, NORWAY Dale B. Haidvogel Institute of Marine and Coastal Sciences Rutgers University New Brunswick, New Jersey USA Jenny A. Huggett Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town SOUTH AFRICA Ian T. Hunter South African Weather Service South African Weather Bureau Pretoria 0001 SOUTH AFRICA Larry Hutchings Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town SOUTH AFRICA David W. Japp Capricorn Fisheries and Monitoring Waterfront 8002 SOUTH AFRICA

Anél Kemp Council for Scientific and Industrial Research CSIR Environmentek 7599 Stellenbosch SOUTH AFRICA Souad Kifani Institut National de Recherche Halieutique Casablanca MOROCCO Vamara Koné Institut de Recherche pour le Développement (IRD) Centre de Recherches Halieutiques Mediterranéenne et Tropicale (CRH) 34203 Sète Cedex, FRANCE Anja Kreiner Ministry of Fisheries and Marine Resources National Marine Research and Information Centre Swakopmund NAMIBIA Raphael M. Kudela Ocean Sciences Department University of California Santa Cruz, California USA Uli Lass Institut für Ostseeforschung Institute for Baltic Sea Research Warnemünde, Rostock GERMANY

Contributors

xvi Deon Louw Ministry of Fisheries and Marine Resources National Marine Research and Information Centre Swakopmund NAMIBIA

Pedro M. S. Monteiro Coast Programme Council for Scientific and Industrial Research – CSIR Environmentek 7599 Stellenbosch SOUTH AFRICA

Lima Maartens De Beers Marine Namibia Windhoek NAMIBIA

Pat D. Morant Coast Programme Council for Scientific and Industrial Research – CSIR Environmentek 7599 Stellenbosch SOUTH AFRICA

Eric Machu Institut de Recherche pour le Développement (IRD) Centre de Recherches Halieutiques Mediterranéenne et Tropicale 34203 Sète Cedex, FRANCE Paola Malanotte-Rizzoli Department of Earth, Atmospheric and Planetary Science Massachusetts Institute of Technology (MIT) Cambridge, Massachusetts USA Thomas Malone Horn Point Laboratory Center for Environmental Science University of Maryland Cambridge, Maryland, USA Patrick Marchesiello IRD Institut Recherche Développement Halieutique Méditerranéenne et Tropicale Paris, Cedex 10 FRANCE Yukio Masumoto Department of Earth and Planetary Science Graduate School of Science The University of Tokyo JAPAN Coleen L. Moloney Zoology Department University of Cape Town Rondebosch 7701 SOUTH AFRICA

Kathie R. Peard Ministry for Fisheries and Marine Resources Lüderitz Marine Research Lüderitz NAMIBIA Patrick Penven Institut de Recherche pour le Développement (IRD) Centre de Recherches Halieutiques Mediterranéenne et Tropicale (CRH) 34203 Sète Cedex, FRANCE Pavitray Pillay SANCOR Secretariat Foundation for Research Development SOUTH AFRICA Grant C. Pitcher Research Aquarium, Sea Point Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town SOUTH AFRICA Christopher J. C. Reason Department of Oceanography University of Cape Town Rondebosch 7701 Cape Town SOUTH AFRICA Mathieu Rouault Department of Oceanography University of Cape Town Rondebosch 7701 Cape Town SOUTH AFRICA

Contributors Jean-Paul Roux Ministry of Fisheries and Marine Resources Lüderitz Marine Research Lüderitz, NAMIBIA Claude Roy Institut Recherche Développement (IRD) Centre IRD de Bretagne BP 70 29280 Plouzané FRANCE Hidehary Sasaki Earth Simulator Center JAMSTEC Yokohama JAPAN Lynne J. Shannon Department of Environmental Affairs and Tourism Marine and Coastal Management (MCM) Rogge Bay 8012 Cape Town SOUTH AFRICA L. Vere Shannon Department of Oceanography University of Cape Town Rondebosch 7701 Cape Town SOUTH AFRICA Kenneth Sherman NOAA, National Marine Fisheries Service Northeast Fisheries Science Center Narragansett Laboratory Narragansett, Rhode Island USA Frank A. Shillington Department of Oceanography University of Capetown Rondebosch 7701 Cape Town SOUTH AFRICA Geoff G. Smith Coast Programme Council for Scientific and Industrial Research CSIR Environmentek 7599 Stellenbosch SOUTH AFRICA

xvii Neville Sweijd BENEFIT Secretariat c/o Ministry of Fisheries and Marine Resources National Marine Research and Information Centre Swakopmund, NAMIBIA Tomoki Tozuka COE Research Associate Depatrment of Earth and Planetary Science Graduate School of Science University of Tokyo JAPAN Roy C. van Ballegooyen Hydroodynamics Coast Programme and Marine and Estuarine Water Quality Council for Scientific and Industrial Research CSIR Environmentek 7599 Stellenbosch SOUTH AFRICA Carl D. van der Lingen Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town SOUTH AFRICA Anja K. van der Plas Ministry of Fisheries and Marine Resources National Marine Research and Information Centre Swakopmund NAMIBIA Jennifer Veitch Department of Oceanography University of Cape Town Rondebosch 7701 Cape Town SOUTH AFRICA F. Vaz-Velho Instituto de Investigação Marinha Ministerio das Pescas Luanda ANGOLA

xviii Hans M. Verheye Department of Environmental Affairs and Tourism Marine and Coastal Management Rogge Bay 8012 Cape Town SOUTH AFRICA C. K. Wainman Institute for Maritime Technology Simon’s Town 7995 SOUTH AFRICA Scarla J. Weeks University of Queensland Centre for Marine Studies Brisbane AUSTRALIA

Contributors John Wilkin Institute of Marine and Coastal Sciences Rutgers University State University of New Jersey New Brunswick, New Jersey USA John D. Woods Complex System Modelling Department of Earth Science and Engineering Royal School of Mines Imperial College London London SW7 2AZ UK Toshio Yamagata Department of Earth and Planetary Science Graduate School of Science University of Tokyo JAPAN

xix

Contents Series Editor’s introduction Ministers’ page: Towards forecasting a changing ocean: An African Perspective (Salomão José Xirimbimbi, Minister of Fisheries, Angola; Abraham Iyambo, Minister of Fisheries and Marine Resources, Namibia; Marthinus van Schalkwyk, Minister of Environmental Affairs and Tourism, South Africa) Sponsorship page Foreword by Michael John O’Toole List of contributors

PART I: BY WAY OF INTRODUCTION 1. A plan comes together Vere Shannon

v vii

ix xi xiii

1 3

2. Forecasting within the context of Large Marine Ecosystem Programs Kenneth Sherman

11

3. The Global Ocean Observing System for Africa (GOOS Africa): Monitoring and Predicting in Large Marine Ecosystems Justin Ahanhanzo

35

PART II: SETTING THE SCENE Data, time series and models: What we think we know about variability in the Benguela and comparable systems.

47

4. Large scale physical variability of the Benguela Current Large Marine Ecosystem (BCLME) Frank A. Shillington, Chris J. C. Reason, Chris M. Duncombe Rae, Pierre Florenchie and Patrick Penven

49

5. Low oxygen water (LOW) variability in the Benguela system: Key processes and forcing scales relevant to forecasting Pedro M. S. Monteiro and Anja K. van der Plas

71

6. Variability of plankton with reference to fish variability in the Benguela Current Large Marine Ecosystem – An overview Larry Hutchings, Hans M. Verheye, Jenny A. Hugget, Hervé Demarcq, Rudi Cloete, Ray G. Barlow, Deon Louw and Antonio da Silva

91

xx

Contents

7. The variability and potential for prediction of harmful algal blooms in the southern Benguela ecosystem Grant C. Pitcher and Scarla J. Weeks

125

8. Resource and ecosystem variability, including regime shifts, in the Benguela Current system Carl D. van der Lingen, Lynne J. Shannon, Philippe Cury, Anja Kreiner, Coleen L. Moloney, Jean-Paul Roux and F. Vaz-Velho

147

9. Modelling, forecasting and scenarios in comparable upwelling ecosystems --California, Canary, Humboldt Pierre Fréon, Jürgen Alheit, Eric D. Barton, Souad Kifani and Patrick Marchesiello

185

PART III: HOPES, DREAMS AND REALITY Forecasting in the Benguela: Our collective wisdom

221

10. Influences of large scale climate modes and Agulhas system variability on the BCLME region 223 Chris J. C. Reason, Pierre Florenchie, Mathieu Rouault and Jennifer Veitch 11. Developing a basis for detecting and predicting long-term ecosystem 239 changes Astrid Jarre, Coleen L. Moloney, Lynne J. Shannon, Pierre Fréon, Carl. D. van der Lingen, Hans M. Verheye, Larry Hutchings, Jean-Paul Roux and Philippe Cury 12. The requirements for forecasting harmful algal blooms in the Benguela Stewart Bernard, Raphael M. Kudela, P. J. S. Franks, Wolfgang Fennel, Anél Kemp, A. Fawcett and Grant C. Pitcher

273

13. Low oxygen water (LOW) forcing scales amenable to forecasting in the Benguela ecosystem Pedro M. S. Monteiro, Anja K. van der Plas, Geoff W. Bailey, Paola Malanotte-Rizzoli, Chris M. Duncombe Rae, Deidre Byrnes, Grant Pitcher, Pierre Florenchie, Patrick Penven, Jim Fitzpatrick and H. Uli Lass

295

14. Forecasting shelf processes of relevance to living marine resources in the BCLME Carl D. van der Lingen, Pierre Fréon, Larry Hutchings, Claude Roy, Geoff W. Bailey, Chris Bartholomae, Andy C. Cockcroft, John. G. Field, Kathie R. Peard and Anja K. van der Plas

309

Contents

xxi

15. Environmental data requirements of maritime operations in the Benguela coastal ocean Marten.L. Gründlingh, Pat D. Morant, Roy C. van Ballegooyen, Awie Badenhorst, Ema Gomes, Leticia Greyling, Johannes Guddal, Ian T. Hunter, David W. Japp, Lima Maartens, Kathie R. Peard, Geoff G. Smith and C. K. Wainman

PART IV:

THE WAY AHEAD

349

373

16. Towards a future integrated forecast system Geoff Brundrit, Chris Bartholomae, Quilanda Fidel, Ashley Johnson and Johannes Guddal

375

17. Forecasting a large marine ecosystem John Woods

387

INDEX

401


xxiii

CD-ROM Contents Benguela: Predicting a Large Marine Ecosystem INSTRUCTIONS TO THE USER The main menu (FOLDERS A through F) allows the user to view documents on the CD, browse the BCLME Forecasting Workshop Webpage, install required software and browse the contents of the CD. Adobe Reader 7.0 is required to view the PDF documents directly from the menu system and the installation software is located in Folder F of the main menu. Although animations can be viewed directly from the CD menu, links to the QuickTime and IrfanView websites are also included in Folder F. Both Packages can be freely downloaded from the given websites and allow the user to have greater control when viewing mov, flic or gif animations. All the animations close automatically after running once, or can be closed with the escape <esc> key. The menu system is navigated by moving the mouse cursor over the folder icons. Menu titles will pop up and the folder letter will be highlighted as the mouse cursor moves over the folder. Click on the highlighted folder letter to access information on the CD. The CD menu system was developed for viewing in Windows XP and Windows 2000. INTRODUCTION PART 1 – (FOLDER A) provides comprehensive details about the International Workshop on Forecasting and Data Assimilation in the Benguela and Comparable systems, held in Cape Town, South Africa in November 2004, its planning, a persoal perspective about the Workshop in the form of a concluding summary by John Woods, and closing remarks by Gotthilf Hempel, the ‘Grandfather’ of the BCLME and BENEFIT Programmes. A: Workshop website B. Aspects of BCLME variability amenable to forecasting C. Presentation of workshop summary D. Closing remarks PART 2 – (FOLDER B) – Observations of model outputs - comprises outputs from observational work and models, including several animations which highlight the spatial and temporal variability in the Benguela. A. High resolution ocean general circulation model (OFES), provided by Tomoki Tozuka, Hidehary Sasaki, Yukio Masumoto and Toshio Yamagata.

xxiv

CD-ROM Contents

B. Forecasting anomalous climatic events in the tropical Atlantic sector using the NLOM Prediction System, provided by Pierre Florenchie. C. ROMS modelled surface chlorophyll a, provided by Eric Machu and Vamara Koné. This is a contribution of the IDYLE and ECO-UP Programmes of the IRD and of EUR-OCEANS. D. Monthly climatology of 18 Years of NOAA SST and 5 Years of SeaWiFS chlorophyll a pigment at 4 km resolution for the BCLME, provided by Hervé Demarcq. This is a contribution of the IDYLE and ECO-UP Programmes of the IRD and of EUR-OCEANS. E. Eight-day composites of ROMS modelled SST for the period 1992 – 2000 using realistic wind forcing. This is a contribution of the IDYLE and ECO-UP Programmes of the IRD and of EUR-OCEANS. F. Five-day SST composites of ROMS modelled for ten years using climatological wind forcing, provided by Patrick Penven. G. Five-day composites of AVISO altimetry around southern Africa, provided by Patrick Penven. H. Example of expert system model: Predicting anchovy recruitment PART 3 – (FOLDER C) – Supplementary material presented at the Benguela Forecast Workshop - contains some selected contributions presented at the Benguela Forecast Workshop, providing additional insights and inputs relevant to measuring, modelling and predicting. A. Requirements and needs of a viable observing and forecasting system in Angola, provided by Quilanda Fidel B. Developing operational forecasting capabilities for coastal GOOS, provided by Thomas Malone, Dir. Ocean, US Office, USA. C. Multi-scale modelling studies on the Northeast U.S. continental shelves, provided by Dale B. Haidvogel, John Wilkin, Katje Fennel, Hernam Arango and Paul Goodman D. Mechanisms and tools in oceanographic capacity building. Provided by Miriam Andrioli, Regina Folorunsho, Geoff Brundrit and Johannes Guddal. E. Ecosystem modelling approaches for South African Fisheries Management, provided by Lynne J. Shannon, Coleen L. Moloney, Carl D. van der Lingen, R.J.M. Crawford, Pierre Fréon F. Offshore oil and gas industry: Marine environment needs, provided by Pat D. Morant PART 4 – (FOLDER D) Information about programmes, network and data centre in the BCLME region - describes four major southern African regional initiatives, including the BCLME and BENEFIT Programmes. A. The BCLME Programme, prepared by Claire Attwood B. Benguela Environment Fisheries Interaction and Training (BENEFIT) Programme, prepared by Neville Sweijd C. SADCO: South African Data Centre for Oceanography, prepared by Marten Grundlingh D. South African Network for Coastal and Oceanic Research (SANCOR), prepared by Pavitray Pillay

CD-ROM Contents

xxv

PART 5 – (FOLDER E) - List of websites that authors and editors believe will be useful to readers. PART 6 - (FOLDER F) – Software installation and browse CD A. Launch Adobe Reader 7.0 installer B. Launch QuickTime Website C. Launch IrfanView Website D. Browse CD Contents


Part I: By Way of Introduction

This Page is Intentionally Left Blank

Large Marine Ecosystems, Vol. 14 V. Shannon, G. Hempel, P. Malanotte-Rizzoli, C. Moloney and J. Woods (Editors) © 2006 Elsevier B.V. All rights reserved.

3

1 A Plan Comes Together Vere Shannon

UNIQUE ENVIRONMENT The Benguela Current Large Marine Ecosystem (BCLME) is situated along the coast of south-western Africa, stretching from east of the Cape of Good Hope in the south equatorwards to the Angola (Cabinda) Front, near the northern border of Angola (Figure 1-1). It encompasses one of the four major coastal upwelling ecosystems of the world which lie at the eastern boundaries of the oceans. Like the Humboldt, California and Canary systems, the Benguela is an important centre of marine food production. The BCLME’s distinctive bathymetry, hydrography, chemistry and trophodynamics combine to make it one of the most productive ocean areas in the world. This high level of primary productivity of the BCLME supports an important global reservoir of biodiversity and biomass of zooplankton, fish, sea birds and marine mammals. Near-shore and off-shore sediments hold rich deposits of precious minerals (particularly diamonds), as well as oil and gas reserves. The natural beauty of the coastal regions, many of which are still pristine by global standards, have also enabled the development of significant tourism along parts of the coast. Pollution, poorly planned coastal developments, population pressure and near-shore activities such as mining are, however, resulting in rapid degradation of some vulnerable coastal habitats. The main area of coastal upwelling extends from the Angola-Benguela Front north of the Angola/Namibia border, southwards around the Cape of Good Hope, and intermittently as far east as Port Elizabeth (Figure 1-1). The upwelling system as we know it is about 2 million years old, and much of the adjacent land area is arid, e.g. the Namib Desert in Namibia. The principal upwelling centre near Lüderitz in southern Namibia, is one of the most concentrated and intense found in any upwelling regime (e.g. Shannon 1985). What also makes the Benguela upwelling system somewhat unique in the global context is that it is bounded at both northern and southern ends by warm water systems, viz. the tropical/equatorial Eastern Atlantic and the Indian Ocean’s Agulhas Current respectively. Sharp horizontal gradients (fronts) exist at the boundaries of the upwelling system, but these display substantial variability in time and in space – at times pulsating in phase and at others not.

4

Figure 1-1 Currents and boundaries of the Benguela Current Large Marine Ecosystem

V. Shannon

A plan comes together

5

Interaction between the BCLME and adjacent ocean systems occurs over thousands of kilometres. For example, much of the BCLME, in particular off Namibia, is naturally hypoxic –even anoxic – at depth partly as a consequence of subsurface flow southwards from the tropical Atlantic. (This hypoxia is compounded by depletion of oxygen from more localised biological decay processes.) The response of the South Atlantic to ENSO has been documented by Colberg et al. (2004) while links between the Benguela and processes in the North Atlantic may also exist. Moreover the southern Benguela lies at a major choke point in the “Global Climate Conveyor Belt” whereby on timescales of decades to centuries warm upper layer waters move from the Pacific via the Indian Ocean into the North Atlantic. (The South Atlantic is the only ocean in which there is a net transport of heat towards the equator.) As a consequence, not only is the Benguela at a critical location in terms of the global climate system, but it is also potentially extremely vulnerable to climate change and climate variability. TEN YEARS OF CLOSE REGIONAL COLLABORATION In mid-1995, recognising the need for a more holistic approach to the study and ultimately the sustainable management of the living resources of the Benguela region, the Namibian Ministry of Fisheries and Marine Resources hosted a Workshop/Seminar on Fisheries Resource Dynamics in the Benguela Current Ecosystem in the coastal town of Swakopmund. At this seminal meeting the seed was sown for two regional cooperative initiatives, BENEFIT and the BCLME Programmes. Country driven jointly by Angola, Namibia and South Africa, and with strong international encouragement and support, particularly from Norway, Germany and France, these Programmes have been instrumental in building goodwill, trust and close cooperation at all levels. Launched in 1997, BENEFIT (BENguela-Environment-FisheriesInteraction & Training) has as its overall goal the development of enhanced science capability required for the optimal and sustainable utilization of living resources of the Benguela by (a) improving knowledge and understanding of the dynamics of important commercial stocks, their environment and linkages between the environmental processes and the stock dynamics, and (b) building appropriate human and material capacity for marine science and technology in the countries bordering the Benguela ecosystem. Following its conception in mid-1995, an embryonic plan for the BCLME Programme was formulated by Kenneth Sherman, Les Clark, Michael O’Toole and Vere Shannon later that year. With enthusiastic country support and incremental funding from the Global Environment Facility (GEF), a comprehensive programme was developed over the next four years (see Anon 1999). Following the approval of the BCLME Project Document early in 2002, funds were released by the UNDP/GEF to enable the full implementation of the BCLME Programme, including reduction of uncertainty and improvement of predictability, and the Programme commenced in April 2002. Whereas the focus of BENEFIT is on science and science capacity building, the BCLME Programme is a broad-based multi-sectoral initiative aimed at sustainable and integrated management of the Benguela Current ecosystem as a whole, having as its developmental goal “to sustain the ecological integrity of the BCLME through integrated transboundary ecosystem management.”

6

V. Shannon

Key to the implementation of the BCLME Programme was the endorsement by the Governments of Angola, Namibia and South Africa in 1999/2000 of six main policy actions. These policy actions recognise the need to address both the impacts of anthropogenic factors and natural processes occurring in the ecosystem, including the highly variable nature of the environment of the BCLME. Fundamental to the above is knowledge and understanding of how the ecosystem, and components thereof, will respond to human actions (e.g. fishing, pollution, habitat alteration) and to natural environmental events and change (e.g. Benguela Niños, hypoxia/anoxia, harmful algal blooms i.e. HABs,). Important outputs of the GEF intervention are inter alia: • enhancement of sustainable management and utilisation of transboundary marine resources • assessment of environmental variability and its ecosystem impacts, and improvement of predictability for enhancing management of living marine resources • maintenance of ecosystem health and biodiversity and management of pollution to safeguard fisheries and other resources • increasing donor participation and co-financing throughout life of Programme and beyond Clearly there is a need to improve predictability of the natural and anthropogenic regimes, i.e. forecasting changes, major perturbations and ecosystem responses and impacts. OBSERVING AND PREDICTING IN THE BCLME WITHIN THE INTERNATIONAL CONTEXT There is considerable international interest in regional ocean forecasting as evident from the strategies of the International Association for the Physical Sciences of the Ocean (IAPSO) and the Global Ocean Observing System (GOOS) of the Intergovernmental Oceanographic Commission (IOC). In this respect it should be noted that the IAPSO strategy makes specific reference to promoting the creation of real-time forecasting strategies in developing countries, capitalising on expertise developed in the USA and Europe, and implemented through co-sponsorship of focussed workshops in targeted regions. Permanent, continuously operating real-time regional ocean prediction systems are increasingly required to support a variety of critical activities in the coastal environment, including navigation, fisheries and marine operations, response to oil and hazardous material spills, search and rescue, and prediction of harmful algal blooms and other ecosystem or water quality phenomena. The implementation of such systems in turn requires advanced technologies in sensors and observing systems, and numerical models and data assimilation, as well as the infrastructures necessary to jointly use them. Coastal ocean observation networks are now being constructed at numerous locations, and the USA and European networks can be prototypes for more extensive systems. Enabling technologies that make this possible include the rapid advances in sensor and platform technologies, multiple real-


7

time communication systems for transmitting the data and the emergence of a universal method for the distribution of results via the World Wide Web. Future sensors that will expand observing capabilities include new ocean colour satellites, altimeters, HF radars and autonomous vehicles. Particularly important are the efforts of the Global Ocean Observing System to develop an observational network for the global ocean and that also meets the requirements for regional ocean observations and forecasting (c.f. GOOS-Africa, Abidjan Convention etc). Concurrently, hydrodynamic and ecological models for the regional systems have been developed and are beginning to show considerable skills. The crucial step allowing for real-time regional forecasting is the development, started in the late 80s, of oceanographic data assimilation, which is now becoming a reality. International thinking as reflected above is very much in keeping with the strategy and workplan of the BCLME Programme, where a key policy action is the assessment of environmental variability, ecosystem impacts and the improvement of predictability. Two cornerstones are the development of an environmental early warning system and the improvement of predictability of extreme events in the BCLME. To give effect to this, an Environmental Variability Advisory Group (EVAG) and associated Activity Centre (EVAC) were established in 2002, the terms of reference for the requisite projects were developed, contracts were awarded, and work on these is now progressing well. This builds on, and is being integrated with, ongoing modelling activities in the Benguela and comparable systems, which are being undertaken in partnership with overseas scientists and institutions – for example the French Institute for Research and Development (IRD). FAST-TRACKING THE DEVELOPMENT OF A REGIONAL OBSERVING SYSTEM AND PREDICTIVE CAPABILITY At its Strategy Task Group meeting in November 2002, IAPSO identified the Benguela region as a promising candidate site for development of a real-time forecasting capability. In view of the of the long history of ocean science in southern Africa, the coming into being of the BCLME Programme as well as international interest and regional needs, IAPSO decided that the most beneficial approach would be to host an international workshop in the region in partnership with other organisations, and involving specialists in the field of predicting and data assimilation. Accordingly, the concept was explored further in consultation with various regional and other international bodies and in November 2004 the International Workshop on Forecasting and Data Assimilation in the Benguela and Comparable Systems was held in Cape Town. Sponsored by the BCLME Programme, IAPSO and IOC-GOOS, together with seven other international, regional and national organisations, the Workshop addressed a key policy action of the BCLME, viz. the assessment of environmental variability, ecosystem impacts and the improvement of predictability. The Workshop was planned, developed and structured by an international Scientific Programme Committee and implemented by a regional Local Organising Committee. Participation in the four-day meeting was strictly by invitation, and over 100 leading international and regional experts, including Kenneth Sherman, the mastermind behind the global

8

V. Shannon

LME initiative and Gotthilf Hempel, the “Grandfather” of both BENEFIT and the BCLME Programmes attended the Workshop. An overarching objective of the Workshop was to contribute to BCLME management by improving assessment of variability and developing an effective and affordable forecasting capability for the region. In order to address this the Workshop had to assess what was known about variability in the BCLME and ascertain which aspects are amenable to forecasting of value, to review present status and recent advances in forecasting and data assimilation in the BCLME, to review advances in forecasting in comparable ecosystems (e.g. Humboldt, Canary, California), to specify minimum data, modelling and human capacity requirements for an early warning system and a blueprint for implementation, to transfer expertise and technology from leading overseas individuals and institutions to the BCLME region and promote collaboration, partnerships and networking, and to help improve numerical literacy skills of regional marine scientists and decision makers and generally build human capacity. The Workshop addressed a broad range of subjects (ocean and atmosphere physics, chemistry and biology – including ecosystem and resource dynamics) of importance for the development of a predictive capability for the greater Benguela Current region and comparable systems. Topics of relevance to forecasting on time scales ranging from hours to months, and even years and decades, were inter alia: • wind forcing on various scales • variability of the Angola (Cabinda) Front, Angola Current and Angola Dome • variability of the Angola-Benguela Front, Benguela Niños and other events of tropical origin • alongshore and cross-shelf process associated with the principal upwelling cell (Lüderitz cell) • intrusions of the Agulhas Current and Sub-Antarctic water into the Benguela • wind and waves and impacts on marine structures and maritime operations • hypoxia/anoxia and its impacts on fish stocks • harmful algal blooms and their predictability • advection and dispersal of pollutants, sulphur “eruptions” • modelling food chain dynamics, including regime shifts • environmental impacts on fish resources and environmental constraints on the distribution of fish vis-à-vis modelling and forecasting • impacts of environmental variability on the ecosystem on inter-annual and decadal time scales • developments in observing and forecasting in comparable systems (Humboldt Current, Canary Current etc) • models and data requirements • ocean observing system appropriate for the BCLME region and key elements of an early warning system for the BCLME. In order to make optimal use of time and the available knowledge and expertise, a programme was devised (somewhat similar to the successful “Dahlem” model)


9

whereby definitive overviews of the state of knowledge and understanding of the BCLME variability were presented in plenary on the first day, followed by eight parallel specialist discussion, review and planning sessions on the remaining three days during which regular report-backs were made in plenary in order to inform and promote integration. Full particulars about the Workshop, its sponsors, the scientific programme, the overviews, specialist sessions and the organisers and the participants are provided in the accompanying CD-ROM to this book. ABOUT THIS BOOK This book draws on material presented at the Benguela Forecast Workshop and the specialist discussions, but it is not the proceedings per se. The idea for the book came from a meeting in 2003 among representatives of the BCLME Programme, GOOSAfrica and IAPSO, and was seen as a combination of an overview of the state of knowledge of the variability in the Benguela and the collective wisdom of experts about the predictability of the system. This concept was explored further with other interested international and regional organisations and the “father” of the global LMEs, Kenneth Sherman, for possible inclusion in the ongoing LME series published by Elsevier. The book is in four parts. Part I introduces the topic of prediction within the context of the international Large Marine Ecosystem initiative and GOOS-Africa. Part II sets the scene through a suite of five definitive overviews of aspects of Benguela variability and an overview of variability and change in comparable ecosystems such as the Guinea Current, Humboldt and California systems. Part III – titled “Hopes, Dreams and Reality” gets to the heart of the subject of forecasting and data assimilation, and captures the collective thinking on Benguela predictability. Part IV gives pointers to what is needed, what is possible and what should be done, and concludes with a vision in the form of an essay on modelling and forecasting in the BCLME and other eastern boundary upwelling systems. Complementary material, including model outputs, animations illustrating variability, graphical displays as well as some selected contributed papers, is included in the CD-ROM which accompanies this book. The editors and authors hope that the reader will find the book both a useful reference work and a source of inspiration as to how best to address the complex issue of predicting in the coastal oceans of developing countries and regions. We also hope that the book will provide a blueprint for managers and scientists for converting hopes and dreams into action in the BCLME and elsewhere. If it does, then the plan whose foundation was laid in Swakopmund, Namibia in 1995 and which has gathered momentum over the past decade through regional collaboration and international support, really will have “come together.”

10

V. Shannon

ACKNOWLEDGEMENT The introductory paragraphs of this chapter are based on previously published material which was prepared for the UNDP/GEF during the development of the BCLME Programme, e.g. Anon (1999), and which is now in the public domain. REFERENCES Anon. 1999. Benguela Current Large Marine Ecosystem Programme: Transboundary Diagnostic Analysis. UNDP/GEF/UNOPS. 51p. Colberg, F., C. J. C. Reason and K. Rogers. 2004. South Atlantic response to ENSO and induced climate variability in an OGCM. J. Geophys. Res. 100:15835-15847. Shannon, L. V. 1985. The Benguela Ecosystem Part I. Evolution of the Benguela, physical features and processes. In Barnes, M., ed. Oceanogr. Mar. Biol. Ann. Rev. 23:105-182.


11

2 Forecasting Within the Context of Large Marine Ecosystems Programs Kenneth Sherman LME DEFINITION: DELINEATION AND MAJOR STRESSORS Large marine ecosystems are natural regions of ocean space encompassing coastal waters from estuaries to the seaward boundary of continental shelves and the outer margins of coastal currents. They are relatively large regions of 200,000 km2 or greater, the natural boundaries of which are based on four ecological criteria: bathymetry, hydrography, productivity, and trophically related populations. The concept that critical processes controlling the structure and function of biological communities can best be addressed on a regional basis (Ricklefs 1987) has been applied to the ocean by using large marine ecosystems as the distinct units for marine resources assessment, monitoring, and management. In turn, the concept of assessment, monitoring, and management of renewable resources from an LME perspective has been the topic of a series of ongoing national and international studies, symposia case studies and workshops initiated since 1984; in each instance, the geographic extent of the LME has been defined on the basis of bathymetry, hydrography, productivity, and trophodynamics. A list of peer reviewed published volumes of LME case studies is given in Table 2-1. The marine areas of the world most stressed from habitat degradation, pollution, and overexploitation of resources are the coastal ecosystems. Ninety percent of the usable annual global biomass yield of fish and other living marine resources is produced in 64 LMEs (Figure 2-1) identified within, and in some cases extending beyond, the boundaries of the EEZs of coastal states (Sherman 1994; Garibaldi and Limongelli 2003). Levels of primary production are persistently higher around the margins of the ocean basins than in the open-ocean pelagic areas (Figure 2-2). High population density characterizes these coastal ocean areas and contributes to the pollution that has its greatest impact on natural productivity cycles through eutrophication from high levels of nitrogen and phosphorus effluent from estuaries or air-born sources. The presence of toxins, harmful algal blooms, and loss of wetland nursery areas to coastal development are ecosystem-level problems that also need to be addressed.

12

K. Sherman

Table 2-1. Published Studies and volumes on LMEs

Author(s)

LME

7

Okemwa

East China Sea

5

Dwividi

Yellow Sea

7

Hazizi

East Bering Sea

1

West Greenland Shelf

8 3 5 10 2 4 5

Incze & Schumacher Livingston et al. Hovgård & Buch Blindheim & Skjoldal Rice Skjoldal & Rey Borisov Skjoldal

Kuroshio Current Sea of Japan

10

Dalpadado et al.

12 3 5

Matishov Ellertsen et al. Blindheim & Skjoldal Daan Reid McGlade Hempel

LME Somali Coastal Current Bay of Bengal

Barents Sea

Norwegian Shelf

Vol.

North Sea

1 9 10 12

Iceland Shelf

10

Faroe Plateau

10

Astthorsson, Vilhjálmsson Gaard et al.

1 3 5

Scully et al. Hempel Scully et al.

1 4 5

MacCall Mullin Bottom

Antarctic

California Current

Pacific American Coastal Humboldt Current Gulf of Thailand South China Sea Indonesian Sea Northeast Australian Shelf

12 8

Lluch-Belda et al. Bakun et al.

5 12 5 11

Bernal Wolff et al. Piyakarnchana Pauly & Chuenpagdee Christensen Zijlstra, Baars Bradbury & Mundy

5 3 2 5 8, 12

Kelleher Brodie

Oyashio Current Okhotsk Sea Gulf of Mexico

Southeast U.S. Shelf Northeast U.S. Shelf

Scotian Shelf Caribbean Sea Patagonian Shelf South Brazil Shelf East Brazil Shelf North Brazil Shelf Baltic Sea Celtic-Biscay Shelf Iberian Coastal Mediterranean Sea Canary Current Guinea Current

Benguela Current Black Sea

Vol. 8

Author(s) Chen & Shen

2, 5, 12 2

Tang Terazaki

8

Terazaki

2 5 2

Minoda Kusnetsov et al. Richards & McGowan

4 9 9 4

Brown et al. Shipp Gracia & Vasquez Baden Yoder

1

Sissenwine

4 6 10, 12 8 3 5 12

Falkowski Anthony Sherman

12

Ekau & Knoppers

12

Ekau & Knoppers

1 12 10

Kullenberg Jansson Lavin

2 10 5

Perez-Gandaras Wyatt & Porteiro Caddy

5 12

Bas Roy & Cury

5 11 11 11

Binet & Marchal Koranteng & McGlade Mensah & Quaatey Lovell & McGlade

11 11 2

Cury & Roy Koranteng Crawford et al.

12 5 12

Shannon & O’Toole Caddy Daskalov

Zwanenburg et al. Richards & Bohnsack Bakun Ekau & Knoppers

Forecasting within the LME programs context

Volume No. 1 2

3 4 5 6 7 8 9 10 11 12

13

Volume description 1986. Variability and Management of Large Marine Ecosystems. Sherman and Alexander, eds. AAAS Symposium 99. Westview Press, Boulder, CO. 319p 1989. Biomass Yields and Geography of Large Marine Ecosystems. Sherman and Alexander, eds. AAAS Symposium 111. Westview Press, Boulder, CO. 493p 1990. Large Marine Ecosystems: Patterns, Processes, and Yields. Sherman, Alexander and Gold, eds. AAAS Symposium. AAAS, Washington, DC. 242p 1991. Food Chains, Yields, Models, and Management of Large Marine Ecosystems. Sherman, Alexander and Gold, eds. AAAS Symposium. Westview Press, Boulder, CO..320p 1992. Large Marine Ecosystems: Stress, Mitigation and Sustainability. Sherman, Alexander and Gold, eds. AAAS Press, Washington, DC. 376 p. 1996. The Northeast Shelf Ecosystem: Assessment, Sustainability and Management. Sherman, Jaworski and Smayda, eds. Blackwell Science, Cambridge, MA. 564p 1998. Large Marine Ecosystems of the Indian Ocean: Assessment, Sustainability and Management. Sherman, Okemwa and Ntiba, eds. Blackwell Science, Malden, MA. 394p 1999. Large Marie Ecosystems of the Pacific Rim: Assessment, Sustainability and Management. Sherman and Tang, eds. Blackwell Science, Malden, MA. 455p 1999. The Gulf of Mexico Large Marine Ecosystem: Assessment, Sustainability and Management. Kumpf, Steidinger and Sherman, eds. Blackwell Science, Malden, MA. 736p 2002. Large Marine Ecosystems of the North Atlantic: Changing States and Sustainability. Skjoldal and Sherman, eds. Elsevier Science, New York. and Amsterdam.449p 2002. Gulf of Guinea Large Marine Ecosystem: Environmental Forcing and Sustainable Development of Marine Resources. McGlade, Cury, Koranteng, Hardman-Mountford, eds. Elsevier Science, Amsterdam and New York. 392p 2003. Large Marine Ecosystems of the World: Trends in Exploitation, Protection and Research. Hempel and Sherman, eds. Elsevier Science, New York and Amsterdam. 423p

Figure 2-1. Map showing 64 large marine ecosystems and linked watersheds

14

K. Sherman

Figure 2-2. Global map of average primary productivity and the boundaries of the 64 Large Marine Ecosystems (LMEs) of the world, available at www.lme.noaa.gov. The annual productivity estimates are based on SeaWiFS satellite data collected between September 1998 and August 1999, and the model developed by M. Behrenfeld and P.G. Falkowski (Limnol.Oceangr. 42(1): 1997, 1-20). The colorenhanced image provided by Rutgers University) depicts a shaded gradient of primary productivity from a high of 450 gCm2yr-1 in red to less than 45 gCm2yr-1 in purple.

Efforts are underway to meet the challenges of forecasting changing biotic and abiotic conditions within the boundaries of LMEs (USEO 2004; USOAP 2004; UN General Assembly 2001). Given the multi-sectoral and multi-disciplinary demand for timeseries data, consideration should be given to the use of standard and inter-calibrated protocols for measuring changing ecological states of the watersheds, bays, estuaries, and coastal water of LMEs. Long-term historical time series data on living marine resources (some up to 40-yr), coupled with measured or inferred long-term pollutant loading histories, have proven useful for relating the results of intensive monitoring to the quantification of ‘cause and effect’ mechanisms affecting the changing ecological states of LMEs. Temporal and spatial scales influencing biological production and changing ecological states in marine ecosystems have been the topic of a number of theoretical and empirical studies. The selection of scale in any study is related to the processes under investigation. An excellent treatment of this topic can be found in Steele (1988) (see also Denman et al. 1989). Steele indicates that in relation to general ecology of the sea, the best known work in marine population dynamics includes studies by Schaefer (1954), and Beverton and Holt (1957), following the earlier pioneering approach of Lindemann (1942). However, as noted by Steele (1988), this array of models is unsuitable for consideration of temporal or spatial variability in the ocean. A heuristic projection was produced by Steele (1988) to illustrate scales and ecosystem indicators of importance in monitoring pelagic components of the ecosystem including phytoplankton, zooplankton, fish, frontal processes, and short-term but large-area


15

episodic effects (Figure 2-3). Advances in technology allow for cost effective methods for measuring the changing states of LMEs using suites of indicators including those depicted in Figure 2-3, supplemented with other modular suites of indicators.

Figure 2-3. A simple set of scale relations for the pelagic food web. (P) Phytoplankton, (Z) zooplankton, (F) fish, (MM) marine mammals, (B) birds. Two physical processes are indicated by (X) Predictable fronts with small cross-front dimensions, and (Y) weather events occurring over relatively large scales. (Adapted from Steele 1988)

LME INDICATOR MODULES A five-module indicator approach to the assessment and management of LMEs has been proven to be useful in ecosystem-based projects in the United States and elsewhere. The modules provide time-series data to support forecasting efforts. They are customized to fit the situation within the context of a transboundary diagnostic analysis (TDA) process and a strategic action plan (SAP) development process for the groups of nations or states sharing an LME. These processes are critical for integrating science into management in a practical way and establishing appropriate governance regimes. The five modules consist of 3 that are science-based indicators focused on: productivity, fish/fisheries, pollution/ecosystem health; the other two, socio-economics and governance, are focused on economic benefits to be derived from a more sustainable resource base and implementing governance mechanisms for providing stakeholders and stewardship interests with legal and administrative support for ecosystem-based management practices. The first four modules support the TDA process while the governance module is associated with periodic updating of the Strategic Action Plan. Adaptive management regimes are encouraged through periodic assessment processes (TDA updates) and updating of SAPs as gaps are filled (Wang 2004; Duda and Sherman 2002).

16

K. Sherman

Productivity Module Indicators Primary productivity can be related to the carrying capacity of an ecosystem for supporting fish resources (Pauly and Christensen 1995). Measurements of ecosystem productivity can be useful indicators of the growing problem of coastal eutrophication. In several LMEs, excessive nutrient loadings of coastal waters have been related to algal blooms implicated in mass mortalities of living resources, emergence of pathogens (e.g., cholera, vibrios, red tides, and paralytic shellfish toxins), and explosive growth of non-indigenous species (Epstein 1993). The ecosystem parameters measured and used as indicators of changing conditions in the productivity module are hydrography, nutrients, primary production, zooplankton biomass and species composition (Edwards et al. 2000a, 2000b). Plankton inhabiting LMEs have been measured over decadal time scales by deploying continuous plankton recorder systems monthly across ecosystems from commercial vessels of opportunity as well as from fixed stations. Advanced plankton recorders can be fitted with sensors for temperature, salinity, chlorophyll, nitrate/nitrite, petroleum, hydrocarbons, light, bioluminescence, and primary productivity, providing the means for in situ monitoring and for calibrating satellite-derived oceanographic data. Properly calibrated satellite data can provide information on such ecosystem aspects as physical state (i.e. surface temperature), nutrient characteristics, primary productivity and chlorophyll concentration (Berman and Sherman 2001; Aiken et al. 1999). Fish and Fisheries Module Indicators Changes in biodiversity and species dominance within fish communities of LMEs have resulted from excessive exploitation, naturally occurring environmental shifts due to climate change and coastal pollution. Changes in biodiversity and species dominance in a fish community can rise up the food web to apex predators and cascade down the food web to plankton components of the ecosystem (Frank et al. 2005; Choi et al. 2004; Pauly and Christensen 1995). The Fish and Fisheries Module includes both fisheries-independent bottom-trawl surveys and pelagic-species acoustic surveys to obtain time-series information on changes in fish biodiversity, population dynamics, and abundance levels. Standardized sampling procedures, when employed from small calibrated trawlers, can provide important information on changes in fish populations (NOAA 1993; NEFSC 1999, 2002) Sherman et al. 2002, 2003). Commercial fish catch provides biological samples for stock identification, stomach content analyses, age-growth relationships, fecundity, as well as data for preparing stock assessments and for clarifying and quantifying multispecies trophic relationships and pathological conditions. The survey vessels can also be used as platforms for obtaining water, sediment, and benthic samples for monitoring harmful algal blooms, diseases, anoxia, and changes in benthic communities.


17

Pollution and Ecosystem Health Module Indicators In several LMEs, pollution and eutrophication have been important driving forces of change in biomass yields. Assessing the changing status of pollution and health of an entire LME is scientifically challenging. Ecosystem health is a concept of wide interest for which a single precise scientific definition is difficult. The health paradigm is based on multiple-state comparisons of ecosystem resilience and stability, and is an evolving concept that has been the subject of a number of meetings (NOAA 1993). To be healthy and sustainable, an ecosystem must maintain its metabolic activity level and its internal structure and organization, and must resist external stress over time and space scales relevant to the ecosystem (Costanza 1992). The Pollution and Ecosystem Health Module measures pollution effects on the ecosystem through the bivalve monitoring strategy of the U.S. Environmental Protection Agency’s (EPA’s) Mussel-Watch Program, through the pathobiological examination of fish; through the estuarine and nearshore monitoring of contaminants and contaminant effects in the water column, the substrate, and in selected groups of organisms through similar efforts. Where possible, bioaccumulation and trophic transfer of contaminants are assessed, and critical life history stages and selected food web organisms are examined for indicators of exposure to, and effects from, contaminants. Effects of impaired reproductive capacity, organ disease, and impaired growth from contaminants are measured. Assessments are made of contaminant impacts at both species and population levels. Implementation of protocols to assess the frequency and effect of harmful algal blooms, emergent diseases, and multiple marine ecological disturbances (Sherman 2000) are included in the pollution module. In the United States, the EPA has developed a suite of 5 coastal condition indicators: water quality index, sediment quality index, benthic index, coastal habitat index, and fish tissue contaminants index. The 2004 report, “National Coastal condition Report II,” includes results from EPA’s analyses of coastal condition indicators and NOAA’s fish stock assessments by LMEs aligned with EPA’s National Coastal Assessment (NCA) regions (USEPA 2001, 2004). Socioeconomic Module Indicators This module emphasizes the practical application of scientific findings to managing LMEs and the explicit integration of social and economic indicators and analyses with all other scientific assessments to assure that prospective management measures are cost-effective. Economists and policy analysts work closely with ecologists and other scientists to identify and evaluate management options that are both scientifically credible and economically practical with regard to the use of ecosystem goods and services. In order to respond adaptively to enhanced scientific information, socioeconomic considerations must be closely integrated with science. This component of the LME approach to marine resources management has recently been described as the human dimensions of LMEs. A framework has been developed by the Department of Natural Resource Economics at the University of Rhode Island for monitoring and assessment

18

K. Sherman

of the human dimensions of LMEs and for incorporating socioeconomic considerations into an adaptive management approach for LMEs (Sutinen et al. 2000). One of the more critical considerations, a method for economic valuations of LME goods and services, has been developed using framework matrices for ecological states and economic consequences of change (Hoagland et al. 2005). Governance Module Indicators The Governance Module is evolving, based on demonstration projects now underway in several ecosystems, such that ecosystems will be managed more holistically than in the past. In LME assessment and management projects supported by the Global Environment Facility for the Yellow Sea, the Guinea Current, and the Benguela Current LMEs, agreements have been reached among the environmental ministers of the countries bordering these LMEs to enter into joint resource assessment and management activities as part of building institutions. One of the major goals of the Benguela Current LME (BCLME) Programme is to establish a Benguela Current Commission which will enable Angola, Namibia and South Africa to engage constructively and peacefully in resolving the transboundary fisheries and environmental issues that threaten the integrity of the BCLME. A preliminary study has found that the establishment of a Benguela Current Commission (BCC) can be justified on several grounds. These include the need for an appropriate institution to implement an ecosystem-based management approach in the BCLME and the need to fulfill the international obligations and undertakings of the three countries of the Benguela. Other motives for the establishment of a regional commission include the need to develop a better understanding of the BCLME, to improve the management of human impacts on the BCLME, to facilitate regional capacity building and to increase the benefits derived from transboundary management and harvesting of fish stocks. A phased approach towards establishing a Benguela Current Commission has been recommended. The first priority would be to draft the necessary agreement between the three countries of the Benguela region. Thereafter, working groups and joint management committees could be brought into operation to address the most pressing transboundary concerns. An Interim Benguela Current Commission (IBCC) is seen as a preliminary step towards a permanent Commission. It would provide the three countries with an opportunity to test and strengthen the institutional structures that will be required for a permanent Commission. It is envisaged that the BCLME Programme’s existing structures would support the IBCC until new structures are made operational. Elsewhere, the Great Barrier Reef LME and the Antarctic LME are also being managed from an ecosystem perspective, the latter under the Commission for the Conservation of Antarctic Marine Living Resources. Governance profiles of LMEs are being explored to determine their utility in promoting long-term sustainability of ecosystem resources (Juda and Hennessey 2001). In each of the LMEs, governance jurisdiction can be scaled to ensure conformance with existing legislated mandates and authorities.


19

APPLICATION OF INDICATOR MODULES TO LME MANAGEMENT SUPPORTED BY THE GLOBAL ENVIRONMENT FACILITY (GEF) Indicator data derived from spatial and temporal applications of the five modules is being applied by a growing number of nations in the assessment and management of LMEs with the financial assistance of the Global Environment Facility (GEF). Among the stressors affecting the sustainability of LMEs is the growing problem of coastal eutrophication. And the depletion of fish and fishery resources and biomass yields. Continued over-fishing in the face of scientific warnings, fishing down food webs, destruction of habitat, and accelerated nutrient loading, especially nitrogen export, have resulted in significant degradation to coastal and marine ecosystems of both rich and poor nations. Fragmentation among institutions, international agencies, and disciplines, lack of cooperation among nations sharing marine ecosystems, and weak national policies, legislation, and enforcement all contribute to the need for a new imperative for adopting ecosystem-based approaches to managing human activities in these systems in order to avoid serious social ad economic disruption. Following a three-year pilot phase (1991-1994), the Global Environment Facility (GEF) was formally launched to forge cooperation and finance actions in the con text of sustainable development—actions that address critical threats to the global environment from biodiversity loss, climate change, degradation of international waters, ozone depletion, and persistent organic pollutants. GEF-LME projects are implemented by UNDP, UNEP, and the World Bank. Expanded opportunities exist for participation by other agencies (GEF 1995, 2004). At present, 121 countries are in the planning and/or implementation phase of GEF/LME projects supported by $650 million in funding support for introducing an ecosystem-based approach to the assessment and management of marine resources and their environments (Table 2-2). SCIENCE-BASED ASSESSMENTS OF LME BIOMASS YIELDS The growing awareness that biomass yields are being influenced by multiple driving forces has broadened monitoring strategies to encompass food chain dynamics and the effects of environmental perturbations and pollution on living marine resources from an ecosystem perspective. To assist stewardship agencies in implementing ecosystembased assessment and management practices, Transboundary Diagnostic Analyses (TDAs) are being focused on the root causes of trends in LME biomass yields. In addition, information on principal driving forces of biomass yields from 29 LME case studies by marine resource experts has been analyzed. A list of the principal investigators, constituting the expert-systems analyses, appearing in 12 peer-reviewed and published LME volumes, is given in Table 2-1. The biomass yields in Table 2-3 are based on the mid-point value, 1995, of LME yields compiled by FAO for 19901999 (Garibaldi and Limongelli 2003). Biomass yield data for four LMEs not included in the FAO report were taken from published LME case studies. Based on expert systems analyses, principal and secondary driving forces were assigned to each LME using four categories (climate, fisheries, eutrophication, and inconclusive) and listed in descending order of catch level, as seen in Table 2-3.

20

K. Sherman

Table 2-2. 121* Countries Participating in GEF/Large Marine Ecosystem Projects

_______________________________________________________________ Approved GEF Projects

LME Gulf of Guinea (6)………………………… Yellow Sea (2)…………………………..... Patagonia Shelf/Maritime Front (2)………….. Baltic (9)………………………………………

Benguela Current (3)…………………………. South China Sea (7)…………………………… Black Sea (6)………………………………….. Mediterranean (19)……………………………. Red Sea (7)……………………………………. Western Pacific Warm Water Pool-SIDSa (13)…

Countries Benin, Cameroon, Côte d’Ivoire, Ghana, Nigeria, Togo China, Korea Argentina, Uruguay Denmark, Estonia, Finland, Germany, Latvia, Lithuania, Poland, Russia, Sweden Angola, Namibia, South Africa Cambodia, China*, Indonesia, Malaysia, Philippines, Thailand, Vietnam Bulgaria, Georgia, Romania, Russia*, Turkey, Ukraine Albania, Algeria, Bosnia-Herzegovina, Croatia, Egypt, France, Greece, Israel, Italy, Lebanon, Libya, Morocco, Slovenia, Spain, Syria, Tunisia, Turkey*, Yugoslavia, Portugal Djibouti, Egypt* Jordan, Saudi Arabia, Somalia, Sudan, Yemen Cook Islands, Micronesia, Fiji, Kiribati, Marshall Islands, Nauru, Niue, Papua New Guinea, Samoa, Solomon Islands, Tonga, Tuvalu, Vanuatu

Total number of countries: 70* GEF Projects in the Preparation Stage

Canary Current (7)............................................. Bay of Bengal (8)…………………………….. Humboldt Current (2)………………………… Guinea Current (16)…………………………..

Gulf of Mexico (3)……………………………. Agulhus/Somali Currents (8)…………………. Caribbean LME (23)…………………………..

Cape Verde, Gambia, Guinea, Guinea-Bissau, Mauritania, Morocco*, Senegal Bangladesh, India, Indonesia*, Malaysia*, Maldives, Myanmar, Sri Lanka, Thailand* Chile, Peru Angola*, Benin*, Cameroon*, Congo, Democratic Republic of the Congo, Côte d’Ivoire*, Gabon, Ghana*, Equatorial Guinea, Guinea*, Guinea-Bissau*, Liberia, Nigeria*, São Tomé and Principe, Sierra Leone, Togo* Cuba, Mexico, United States Comoros, Kenya, Madagascar, Mauritius, Mozambique, Seychelles, South Africa*, Tanzania Antigua and Barbuda, The Bahamas, Barbados, Belize, Colombia, Costa Rica, Cuba*, Grenada, Dominica, Dominican Republic, Guatemala, Haiti, Honduras, Jamaica, Mexico*, Nicaragua, Panama, Puerto Ricob, Saint Kitts and Nevis, Saint Lucia, Saint Vincent and the Grenadines, Trinidad and Tobago, Venezuela Total number of countries: 63*

*Adjusted for multiple listings Provisionally classified as Insular Pacific Provinces in the global hierarchy of LMEs and Pacific Biomes (Watson et al. 2003). b A self-governing commonwealth in union with the United States a

____________________________________________________________________


21

Table 2-3. PRIMARY AND SECONDARY DRIVING FORCES OF LME BIOMASS YIELDS1 Based on published expert assessments in LME volumes listed in Table 1 LEVEL PRIMARY SECONDARY MMT

LME

Humboldt Current

climate

fishing

16.0

South China Sea

fishing

climate

10.0

East China Sea North Sea Eastern Bering Sea

fishing fishing -

climate climate -

3.8 3.5 2.1

Bay of Bengal

fishing

climate

2.0

Okhotsk Sea2

climate

fishing

2.0

Canary Current

climate

fishing

1.8

Norwegian Shelf

climate

fishing

1.5

Iceland Shelf

climate

fishing

1.3

Benguela Current

climate

fishing

1.2

Gulf of Thailand Mediterranean Sea of Japan3

fishing fishing climate

climate eutrophication fishing

1.1 1.1 1.0

Gulf of Mexico

fishing

climate

0.9

Guinea Current

climate

fishing

0.9

Baltic Sea

fishing

eutrophication

0.8

California Current

climate

fishing

0.7

U.S. Northeast Shelf

fishing

climate

0.7

Scotian Shelf

fishing

climate

0.7

Black Sea

eutrophication

fishing

0.5

EXPERT ASSESSMENTS Alheit and Bernal Wolff et al. Pauly and Christensen

5 12 5

Chen and Shen McGlade Schumacher et al. Dwividi Hazizi Kusnetsov et al. Roy and Cury Bas Ellertsen et al. Blindheim and Skjoldal Astthorsson and Vilhjálmsson Crawford et al. Shannon and O’Toole Pauly and Chuenpagdee Caddy Terazaki Richards and McGowan

8 10 12 5 7 5 12 5 3 5 10

Brown et al. Shipp Binet and Marchal Koranteng and McGlade Kullenberg Jansson MacCall Lluch-Belda et al. Sissenwine

Barents Sea

climate

fishing

0.5

Caribbean Sea Iberian Coastal Newfoundland-Labrador

fishing climate fishing

climate fishing climate

0.4 0.3 0.2

Yellow Sea4

fishing

climate

0.2

Great Barrier Reef

fishing

climate

0.1

West Greenland Shelf

climate

fishing

0.1

Faroe Plateau

climate

fishing

0.1

VOL REF

2 12 12 5 8 2 4 9 5 11 1 12 1 12 1

Murawski Sherman et al.

6 10

Zwanenburg et al. Zwanenburg Caddy Daskalov Skjoldal and Rey

10 12 5 12 2

Borisov Blindheim and Skjoldal Matishov et al.

4 5 12

Richards and Bohnsack Wyatt and Perez-Gandaras Rice et al. Tang Tang Brodie Hovgard and Buch Pederson and Rice Gaard et al.

3 2 10 2 12 12 3 10 10

22

K. Sherman

Notes to Table 2-3 1

Annual biomass yield levels based on 1990-1999 mid-decadal data (1995) from FAO 2003 Okhotsk Sea LME data from Kusnetsov et al. 1993 based on mid-decadal (1972) data on fishing yields from 1962 to 1982 1 Biomass yield data from Terazaki (1999) based on mid-decadal data (1985) from Sea of Japan 1980-1990. 1 Biomass yield data from Tang (2003) based on mid-decadal data for demersal species for the Yellow Sea, 1952 to 1992. 1

It would appear that the management regime for nearly half of this yield from the 29 case study LMEs (27.0 mmt) will need to focus efforts for forecasting biomass yields on the climate signal, whereas the management regime for slightly less of the biomass yield (24.8 mmt) will need to focus primarily on catch control and secondarily on the climate signal, to recover depleted fish stocks and achieve maximum sustainable yield levels. The influence of climate forcing in biomass yields for the California Current LME has been analyzed and illustrated by Lluch-Belda et al. (2003) for anchovy and sardine catches. They also provide a time-series indicator of regime shift (RIS) for sardine and anchovy populations off the coast of Japan, and for the California Current, Benguela Current and Humboldt Current LMEs (Figure 2-4).

Figure 2-4. Historic sardine (upper) and anchovy (middle) catches (1920-2000) from the California Current LME, and the Regime Indicator Series (RIS lower). Catch data were obtained from Schwarzlose et al. (1999). RIS is a composite series reflecting synchronous variability of sardine and anchovy populations of the Japan, California, Benguela and Humboldt currents. Modified from Lluch-Cota, D.B. et al. (1997). (Lluch-Belda et al. 2003)


23

Evidence of climate forcing for the Humboldt Current LME has been given by Wolff et al. (2003) and for the Iceland Shelf LME by Astthorsson and Vilhjálmsson (2002). For the Benguela Current LME, the effects of climate forcing appear to be mediated through productivity, pollution-policy and fisheries interactions (Shannon et al. 2004). In contrast, the argument for urgent reduction in fishing effort is supported by the data in Sherman et al. (2003) for the US Northeast Shelf LME, and for the Gulf of Thailand based on the expert analysis of Pauly and Chuenpagdee (2003) (Figure 2-8). Of the 29 LME case studies, 13 were assigned to climate forcing as the principal driver of change in biomass yield, 14 were assigned to fisheries as principal driver, one was assigned to eutrophication, and the remaining one was deemed inconclusive. In all but the Mediterranean Sea LME, where climate forcing was the principal driver of changing biomass yield, fisheries forcing was the secondary driver. In the case of the Mediterranean Sea LME, the secondary driver was eutrophication (Caddy(1993). The contribution of the 29 LMEs to the annual global biomass yields amounts to 54.4 million metric tons (mmt) or 64% based on the average annual yield from 1995-1999 of 85 mmt (Garibaldi and Limongelli 2003). The observation that excessive fishing effort can alter the structure of the ecosystem, resulting in a shift from relatively highpriced, large-sized, long-lived demersal species, down the food chain toward lowervalued, smaller, short-lived pelagic species (Pauly and Christensen 1995), is supported by the LME data on species biomass yields.

Figure 2-5. A conceptual model of how climatic conditions in the Icelandic Shelf LME may affect production at lower trophic levels and eventually the yield from the Icelandic cod stock in Astthorsson and Vilhjálmsson (2002).

Evidence from the East China Sea, Yellow Sea, and Gulf of Thailand suggests that these three LMEs are approaching a critical state of change, wherein recovery to a

24

K. Sherman

previous ratio of demersal to pelagic species may become problematic. In all three cases, the fisheries are now being directed toward fish protein being provided by catches of smaller-sized species of low value (Chen and Shen 1999; Pauly and Chuenpagdee 2003; Tang 2003). The species change in biomass yields of the Yellow Sea, as shown in Tang (2003), represents an extreme case wherein the annual demersal species biomass yield was reduced from 200,000 mt in 1995 to less than 25,000 mt through 1980. The fisheries then targeted the pelagic anchovy and, between 1990 and 1995, landings of anchovy reached an historic high of 500,000 mt.

Figure 2-6. Major features of the Gulf of Thailand LME fisheries and trophic level. (A) Catches, by major species groups (excluding tuna and other large pelagics). Note stagnation and decline of demersal catches, following their rapid increase in the 1960s and 1970s. Also note increasing contribution of small and medium pelagics, and overall decline in the 1990s. (B) Trophic level (TL) trends in the catch of research trawlers (reflecting relative abundances in the ecosystems), and in the total landings (both series excluding large pelagics). Lower TL in 1977 to 1997 series is due to inclusion of small pelagics and other low-TL organisms caught by gears other than trawl. From Pauly and Chuenpagdee (2003).

RECOVERING FISHERIES BIOMASS The GEF/LME projects presently funded or in the pipeline for funding in Africa, Asia, Latin America and eastern Europe, represent a growing network of marine scientists, marine managers and ministerial leaders who are engaged in pursuing the ecosystem and fishery recovery goals. The significant annual global biomass yields of marine fisheries from ecosystems in the GEF-LME Network of almost half the world marine landings provides a firm basis for moving toward the 2004 Johannesburg World Summit on Sustainable Development (WSSD) goal for introducing an ecosystembased assessment and management approach to global fisheries by 2010 (Table 2-4).


25

Table 2-4. LMEs With Relatively Stable Biomass Yields, 1990-1999

LME

LEVEL MMT

Arabian Sea

2.2

Bay of Bengal

2.3

Mediterranean

1.1

Indonesian Seas

1.6

Sulu-Celebes

0.8

North Brazil

0.1 TOTAL: 8.1

Percentage of Global Marine Biomass Yield: 9.5%

There is an international instrument supported by most coastal nations that could have immediate applicability to reaching Summit fishery goals. The FAO Code of Conduct for Responsible Fishery Practice of 2002 argues for moving forward with a “precautionary approach” to fisheries sustainability given a situation wherein available information can be used to recommend a more conservative approach to fish and fisheries total allowable catch levels (TAC) than has been the general practice over the past several decades (Holling 1973, 1986, 1993). Based on the decadal profile of LME biomass yields from 1990 to 1999 (Garibaldi and Limongelli 2003), it appears that the yields of total biomass and the biomass of 11 species groups of 6 LMEs have been relatively stable or have shown marginal increases over the decade (Sherman 2005). The yield of marine biomass for these 6 LMEs was 8.1 mmt, or 9.5 percent of the global marine fisheries yield in 1999 (Table 2-4). The countries bordering these six LMEs—Arabian Sea, Bay of Bengal, Indonesian Sea, Northeast Brazil Shelf, Mediterranean Sea and the Sulu-Celebes Sea—are among the world’s most populous, representing approximately one-quarter of the total human population. These LME border countries are increasingly dependent on marine fisheries for food security and for national and international trade. Given the risks of “fishing-down-the-food-web,” and in the absence of fishing-effort data, it would appear opportune for the stewardship agencies responsible for the fisheries of the bordering countries to consider options for mandating precautionary total allowable catch levels during a period prior to reductions in catch. Evidence for species recovery following a significant reduction in fishing effort through mandated actions is encouraging. Following management actions to reduce fishing effort, the robust condition of the U.S. Northeast Shelf ecosystem with regard to the average annual level of primary productivity (350 gCm2y), stable annual average levels of zooplankton (33 cc100m3), and a relatively stable oceanographic regime (Sherman et al. 2002), contributed to: (1) a relatively rapid recovery of depleted herring and mackerel stocks, with the cessation of foreign fisheries in the mid-1970s; and (2) initiation of the recovery of depleted yellowtail flounder and haddock stocks at the Georges Bank sub-area of the ecosystem following a mandated 1994 reduction in fishing effort (Figure 2-7).

26

K. Sherman

Figure 2-7. Increase in biomass of the Northeast Shelf LME, Georges Bank sub-area yellowtail flounder and haddock follows reduction in fishing effort (exploitation rate): http://www.nefsc.noaa.gov/nefsc/publications/crd/crd0216/

Three LMEs remain at high risk for fisheries biomass recovery expressed as a pre1960s ratio of demersal to pelagic species—the Gulf of Thailand, East China Sea and Yellow Sea. However, mitigation actions have been initiated by the People’s Republic of China toward recovery by mandating 60 to 90 day closures to fishing in the Yellow Sea and East China Sea during summer months (Tang 2003). The country-driven planning and implementation documents supporting the ecosystem approach to LME assessment and management practices can be found at www.iwlearn.org.


27

NITROGEN OVER-ENRICHMENT OF LMEs Nitrogen over-enrichment has been reported as a coastal problem for two decades, from the southeast coast of the US as described by Duda (1982) twenty years ago, to the Baltic and other systems (Helsinki 2001). More recent estimates of nitrogen export to LMEs from linked freshwater basins are summarized in Jaworski 1999. These recent human-induced increases in nitrogen flux range from 4-8 in the US from the Gulf of Mexico to the New England coast while no increase was documented in areas with little agricultural or population sources in Canada (Howarth et al. 2000). In European LMEs, recent nitrogen flux increases of from 3-fold in Spain to 4-fold in the Baltic and 11-fold in the Rhine basin draining to the North Sea LME have been recorded (Howarth et al. 2000). Duda and El-Ashry (2000) described the origin of this disruption of the nitrogen cycle from the “Green Revolution” of the 1970s as the world community converted wetlands to agriculture, utilized more chemical inputs, and expanded irrigation to feed the world. As noted by Duda (1982) for the Southeast estuaries of the US, and by Rabalais (1999) for the Gulf of Mexico, much of the large increase in nitrogen export to LMEs is from agricultural inputs, both from the increased delivery of fertilizer nitrogen as wetlands were converted to agriculture and from concentrations of livestock as shown in Duda and Finan (1983) for eastern North Carolina, where the increase in nitrogen export over the forested situation ranged from 20- to 500-fold in the late 1970s. Industrialized livestock production in the last two decades increases the flux, the eutrophication, and the oxygen depletion even more as reported by the NRC (2000). The latest GESAMP Assessment (2001) also identified sewage as a significant contributor to the eutrophication from drainages from large cities; atmospheric deposition from automobiles/agricultural activities also contributes, depending on proximity to sources.

DIN Export by Rivers for World Regions 1990 and 2050 BAU Scenario 16 14

Tg N y-1

12 10 2050

8 6

1990

4 2 0

North South Africa Am erica Am erica

Europe

NE Eastern Asia Asia

Southern Asia

Figure 2-8. Model-predicted nitrogen [dissolved inorganic N in Tg (or 1012g) Ny-1] export by rivers to coastal systems in 1990 and in 2050 (based on a business-as-usual [BAU] scenario). Figure modified from Kroeze and Seitzinger (1998).

28

K. Sherman

The excessive levels of nitrogen contributing to coastal eutrophication constitute a new global environment problem that is cross-media in nature. Excessive nitrogen loadings have been identified as problems in the following LMEs that are receiving GEF assistance: Baltic Sea, Black Sea, Adriatic portion of the Mediterranean, Yellow Sea, South China Sea, Bay of Bengal, Gulf of Mexico, and Plata Maritime Front/Patagonia Shelf. In fact, preliminary global estimates of nitrogen export from freshwater basins to coastal waters were assembled by Seitzinger and Kroeze (1998) as part of a contribution to better understanding LMEs (2-8)). These preliminary estimates of global freshwater basin nitrogen export are alarming for the future sustainability of LMEs. Given the expected future increases in population and fertilizer use, LMEs may be, without significant N mitigation efforts, subjected to a future of increasing harmful algal bloom events, reduced fisheries, and hypoxia that further degrades marine biomass and biological diversity. LME MODELING AND DRIVING FORCES OF CHANGE Empirical and theoretical aspects of yield models for large marine ecosystems have been reviewed by several ecologists. According to Beddington (1986,1995; Daan (1986), Levin (1990) and Mangel (1991), published dynamic models of marine ecosystems offer little guidance on the detailed behavior of communities. However, these authors concur on the need for covering the common ground between observation and theory by implementing monitoring efforts on the large spatial and long temporal scales (decadal) of key components of the LMEs. The sequence for improving the understanding of the possible mechanisms underlying observed patterns in LMEs is described by Levin (1990) as: 1) examination of statistical analyses of observed distributional patterns of physical and biological variables, 2) construction of competing models of variability and patchiness based on statistical analyses and natural scales of variability of critical processes, 3) evaluation of competing models through experimental and theoretical studies of component systems, and 4) integration of validated component models to provide predictive models for population dynamics and redistribution. The approach suggested by Levin (1990) is consistent with the observation by Mangel (1991) that empirical support for the currently used models of LMEs is relatively weak, and that a new generation of models is needed that serves to enhance the linkage between theory and empirical results. Three models of ecosystem structure and function are being applied to LMEs with financial assistance from the GEF through one mid-sized proposal, “Promoting Ecosystem-based Approaches to Fisheries Conservation” (www.gefweb.org). (1) Estimates of carrying capacity using ECOPATH-ECOSIM food web approaches for the world’s 64 LMEs are being prepared in a collaboration between scientists of the University of British Columbia and marine specialists from developing countries. (2) A 24-month training project is being implemented by scientists from Rutgers University in collaboration with IOC/UNESCO to estimate expected nitrogen loadings for each LME over the next 50 years. (3) Scientists from Princeton University, under


29

the direction of Simon Levin, and from the University of California at Berkeley, under the direction of Thomas “Zack” Powell, are modeling particle spectra pattern formation in LMEs, on the assumption that they represent complex adaptive systems in which ecological systems interact with socioeconomic ones. The goal is to describe and understand patterns of particle sizes emerging from multiple activation and inhibition processes that operate on diverse scales of space, time and organization. This work is informed by earlier studies that suggest, for example, the emergence of smaller size spectra of organisms following ecosystem-wide perturbations within LMEs (Chave and Levin 2003; Levin 2003; Cavendar-Bares et al. 2001; Gin Guo and Cheong 1998, Sheldon and Parsons 1967; Duplisea and Castonguay 2006). Additionally, the American Fisheries Society and the World Council of World Fisheries Societies are collaborating to create an electronic network that will expedite information access and communication among marine specialists participating in GEFsupported LME projects. There is a growing awareness among marine scientists, geographers, economists, government representatives, and lawyers of the utility of a more holistic ecosystem approach to resource management (Byrne 1986; Christy 1986; Alexander 1989; Belsky 1989; Crawford et al. 1989; Morgan 1989; Prescott 1989). On a global scale, the loss of sustained biomass yields from LMEs from mismanagement and overexploitation has not been fully investigated, but is likely very large. Effective management strategies for LMEs will be contingent on identification of major driving forces causing large-scale changes in biomass yields. Management of species responding to strong environmental signals will be enhanced by improving the understanding of the physical factors forcing biological change, thereby enhancing forecasts of El Niño-type events. In other LMEs, where the prime driving force is overfishing, options can be explored for reductions of fishing effort and implementing adaptive management strategies (Collie 1991). Further, remedial actions are required to ensure that the nitrogen overloading and the pollution of the coastal zone of LMEs is reduced and does not become a principal driving force in any LME. Recent reports explore the application of ecosystem-based research and modeling that is focused on management outcomes (Browman and Stergiou 2004) and on macroecology (Belgrano 2004; Hoagland et al. 2005; Edwards 2005; Grigalunas et al. 2005). Considerable effort has been focused on studies of modeling to improve forecasts of changing conditions in LMEs. For the Benguela Current LME, Shannon et al (2004) provide an excellent volume of ecosystem modeling studies pertinent to the Benguela Current LME. Based on 24 year time-series of fisheries and environmental data, Shannon et al. (2004) in their chapter on dynamic modeling concluded that fishing effects on fish stocks of the Benguela Current LME were less important than physical forcing on many of the important groups of fishes studied. Cury et al. (2000) consider the effects of physical forcing as triggering a critical predator-prey response between small pelagic fish and their zooplankton food base as an important consequence of forcing that is affecting biomass yields of the small pelagic fishes of the Benguela LME. The subsequent chapters in the present volume provide new insights on the components of physical forcing that could serve to improve forecasts of biomass yields as well as sustainable levels of yields for the Benguela Current LME.

30

K. Sherman

REFERENCES Aiken, J., R. Pollard, R. Williams, G. Griffiths, I. Bellan. 1999. Measurements of the upper ocean structure using towed profiling systems. In Sherman, K. and Q. Tang, eds. Large Marine Ecosystems of the Pacific Rim: Assessment, Sustainability, and Management. Malden: Blackwell Science, Inc. 346-362. Alexander, L.M. 1989. Large marine ecosystems as global management units. In Sherman, K., L.M. Alexander eds. Biomass Yields and Geography of Large Marine Ecosystems. American Association for the Advancement of Science (AAAS) Selected Symp. 111. Westview Press, Inc., Boulder, Colorado. 339-344. Astthorsson, O.S. and H. Vilhjálmsson. 2002. Iceland Shelf large marine ecosystem. In Sherman, K. and H.R. Skjoldal, eds. Large Marine Ecosystems of the North Atlantic: Changing States and Sustainability. Elsevier Science. Netherlands, London, New York, Tokyo. 219-244. 449p. Beddington, J.R. 1986. Shifts in resource populations in large marine ecosystems. In Sherman, K. and L.M. Alexander, eds. Variability and Management of Large Marine Ecosystems. AAAS Selected Symposium 99. Westview Press. Boulder, Colorado. 9-18. 319p. Beddington, J.R. 1995. The primary requirements. Nature 374:213-214. Behrenfeld, M. and P.G. Falkowski. 1997. Photosynthetic rates derived from satellite-based chlorophyll concentration. Limnol. Oceangr. 42(1): 1-20. Belgrano, A. coord. 2004. Theme Section: Emergent properties of complex marine systems: A macroecological perspective. Marine Ecology Progress Series 273:227-302. Belsky, M.H. 1989. The ecosystem model mandate for a comprehensive United States ocean policy and Law of the Sea. San Diego L. Rev. 26(3): 417-495. Berman, M.S. and K. Sherman. 2001. A towed body sampler for monitoring marine ecosystems. Sea Technology 42(9): 48-52. Beverton, R.J.H. and S.J. Holt. 1957. On the dynamics of exploited fish populations. Fish. Invest. Minist. Agric. Fish Food. (G.B.) Ser.II 19:1-533. Browman, H.I. and K.I. Stergiou, coord. 2004. Theme Section: Perspectives on ecosystem-based approaches to the management of marine resources. Marine Ecology Progress Series 274:269-298. Byrne, J. 1986. Large marine ecosystems and the future of ocean studies. In Sherman, K., L.M. Alexander, eds. Variability and Management of Large Marine Ecosystems. AAAS Selected Symp. 99. Westview Press, Inc. Boulder, Colorado. 299-308. Caddy, J.F. 1993. Contrast between recent fishery trends and evidence for nutrient enrichment in two large marine ecosystems: The Mediterranean and the Black Seas. In Sherman, K., L.M. Alexander and B.D. Gold, eds. Large Marine Ecosystems: Stress, Mitigation and Sustainability. AAAS Press. Washington, D.C. 376p. 137-147. Cavender-Bares, K.K, A. Rinaldo and S.W. Chisholm. 2001. Microbial size spectra from natural and nutrient enriched ecosystems. Limnology and Oceanography 46(4): 778-789. Chave, J. and S.A. Levin. 2003. Scale and scaling in ecological and economic systems. In Dasgupta, P. and K.-G. Möler, eds. The Economics of Non-convex Ecosystems, Special Issue, Environmental & Resource Economics 26:527-557. Chen, Ya-Qu and Xin-Qiang Shen. 1999. Changes in the biomass of the East China Sea ecosystem. In Sherman, K. and Q. Tang, eds. Large Marine Ecosystems of the Pacific Rim: Assessment, Sustainability and Management. Blackwell Science. Malden, Massachusetts. 221-239. 465p. Choi, J.S., K.T. Frank, W.C. Leggett, and K. Drinkwater. 2004. Transition to an alternate state in a continental shelf ecosystem. Can. J. Fish. Aquat. Sci. 61:505-510. Christy, F.T. 1986. Can large marine ecosystems be managed for optimum yields? In Sherman, K., and L.M. Alexander, eds. Variability and Management of Large Marine Ecosystems. AAAS Selected Symp. 99. Westview Press, Inc., Boulder, Colorado. 319p. 263-267. Collie, J.S. 1991. Adaptive strategies for management of fisheries resources in large marine ecosystems. In Sherman, K., L.M. Alexander, B.D. Gold, eds. Food Chains, Yields, Models, and Management of Large Marine Ecosystems. Westview Press, Inc. Boulder, Colorado. 320p. 225-242.


31

Costanza, R. 1992. Toward an operational definition of ecosystem health. In Costanza, R., B.G. Norton, B.D. Haskell, eds. Ecosystem Health: New Goals for Environmental Management. Island Press, Washington DC. 239-256. Crawford, R.J.M., L.V. Shannon, P.A. Shelton. 1989. Characteristics and management of the Benguela as a large marine ecosystem. In Sherman, K., L.M. Alexander, eds. Biomass Yields and Geography of Large Marine Ecosystems. AAAS Selected Symp. 111. Westview Press, Inc., Boulder, Colorado. 493p. 169-219. Cury, P., A. Bakun, R. Crawford, A. Teichmann, R. Quiñones, L.J. Shannon and H.M. Verheye. 2000. Small pelagics in upwelling systems: Patterns of interaction and structural changes in “wasp-waist” ecosystems. ICES J. mar. Sci. 57:603-618. Daan, N. 1986. Results of recent time-series observations for monitoring trends in large marine ecosystems with a focus on the North Sea. In Sherman, K., L.M. Alexander, eds. Variability and Management of Large Marine Ecosystems. AAAS Selected Symp. 99, Westview Press, Inc., Boulder, Colorado. 319p. 145-174. Denman, K.L., H.J. Freeland and D.L. Mackas. 1989. Comparisons of time scales for biomass transfer up the marine food web and coastal transport processes. In Beamish, R.J. and G.A. McFarlane, eds. Effects of ocean variability on recruitment and an evaluation of parameters used in stock assessment models. Can. Spec. Publ. Fish. Aquat. Sci. 108. Dept. of Fisheries and Oceans, Ottawa. 379p. Duda, A.M. 1982. Mulicipal point sources and agricultural non-point source contributions to coastal eutrophication. Water Resources Bulletin 18(3): 397-407. Duda, A.M. and M.T. El-Ashry. 2000. Addressing the global water and environmental crises through integrated approaches to the management of land, water, and ecological resources. Water International 25:115-126. Duda, A.M. and D.S. Finan. 1983. Influence of livestock on nonpoint source nutrient levels of streams. Transactions of American Society of Agricultural Engineers 26(6): 1710-1726. Duda, A. and K. Sherman. 2002. A new imperative for improving management of large marine ecosystems. Ocean & Coastal Management 45(2002):797-833. Duplisea, D.E. and M. Castonguay. 2006. Comparison and utility of different size-based metrics of fish communities for detecting fishery impacts. Can. J. Fish. Aquat. Sci. 63:810-820. Edwards, S. 2005. Ownership of multi-attribute fishery resources in large marine ecosystems. In: Hennessey, T. and J. Suitinen, eds. Sustaining Large Marine Ecosystems: The Human Dimension. Elsevier. 368p. 137-154. Edwards, C.A., T.M. Powell, and H.P. Batchelder. 2000a. The stability of an NPZ model subject to realistic levels of vertical mixing. J. Mar. Res. 58:37-60. Edwards, C.A., H.P. Batchelder and T.M. Powell. 2000b. Modeling microzooplankton and mesozooplankton dynamics within a coastal upwelling system. J. Plankton Res.22:1619-1648. Epstein, P.R. 1993. Algal blooms and public health. World Resource Review 5(2): 190-206. FAO (The U.N. Food and Agriculture Organization). 2000. The State of the World Fisheries, Aquaculture. Rome. FAO. 142p. FAO Code of Conduct for Responsible Fisheries. 2002. www.fao.org/FI/agreem/codecond/ficonde.asp Frank, K.T., B. Petrie, J.S. Choi, W.C. Leggett. 2005. Trophic cascades in a formerly cod-dominated ecosystem. Science 308:1621-1623. Garibaldi, L. and L. Limongelli. 2003. Trends in oceanic captures and clustering of large marine ecosystems: Two studies based on the FAO capture database, as reported to the FAO by official national sources. FAO Fisheries Technical paper 435. Food and Agriculture Organization of the United Nations. Rome. 71p. GEF. 2004. Promoting Ecosystem-based Approaches to Fisheries Conservation. http://www.GEFweb.org,/ mid-sized proposals, March 23, 2004. GEF (Global Environment Facility). 1995. GEF Operational Strategy. Washington, DC: Global Environment Facility. GESAMP (Group of Experts on the Scientific Aspects of Marine Pollution). 1990. The state of the marine environment. UNEP Regional Seas Reports and Studies No. 115. Nairobi. Gin, K.Y.H., J. Guo, H-F Cheong. 1998. A size-based ecosystem model for pelagic waters. Ecological Modelling 112;53-72. Grigalunas, T.A., J.J. Opaluch, J. Diamantides and D-S Woo. 2005. Eutrophication in the Northeast Shelf large marine ecosystem: Linking hydrodynamic and economic models for benefit estimation. In

32

K. Sherman

Hennessey T., and J. Sutinen, eds. Sustaining Large Marine Ecosystems: The Human Dimension. Elsevier, Netherlands. 368p. 229-248. Helsinki Commission. 2001. Environment of the Baltic Sea Area 1994-1998. Baltic Sea Environment Proceedings No. 82A, Helsinki. 23p. Hoagland, P., D. Jin, E. Thunberg, and S. Steinback. 2005. Economic activity associated with the Northeast Shelf large marine ecosystem: Application of an input-output approach. In Hennessey, T. and J.Sutinen, eds. Sustaining Large Marine Ecosystems: The Human Dimension. Elsevier, Netherlands. 371p. 159-181. Holling, C.S. 1993. Investing in research for sustainability. Ecol. Applic. 3:552-555. Holling, C.S. 1986. The resilience of terrestrial ecosystems local surprise and global change. In Clark, W.C., and R.E. Munn, eds. Sustainable Development of the Biosphere. Cambridge Univ. Press, London. 292-317. Holling, C.S. 1973. Resilience and Stability of Ecological Systems. Institute of Resource Ecology. Univ. of British Columbia, Vancouver. Howarth, R., D. Anderson, J. Cloern, C. Elfring, C. Hopkinson, B. Lapointe, T. Malone, N. Marcus, K. McGlathery, A. Sharpley, D. Walker. 2000. Nutrient Pollution of coastal rivers, bays, and seas. ESA Issues in Ecology 7:1-15. Jaworski, N.A. 1999. Comparison of nutrient loadings and fluxes into the US Northeast Shelf LME with the Gulf of Mexico and other LMEs. In Kumpf, H., K. Steidinger, K. Sherman, eds. The Gulf of Mexico Large Marine Ecosystem: Assessment, Sustainability, and Management. Blackwell Science. Malden, Massachusetts. 704p. 360-371. Juda, L. and T. Hennessey. 2001. Governance profiles and the management of the uses of large marine ecosystems. Ocean Development and International Law. 32:41-67. Kroeze, C. and S.P. Seitzinger. 1998. Nitrogen inputs to rivers estuaries and continental shelves and related nitrous oxide emissions in 1990 and 2050: a global model. Nutrient Cycling in Agroecosystems 52:195-212. Kuznetsov, V.V., V.P. Shuntov, L.A. Borets. 1993. Food chains, physical dynamics, perturbations, and biomass yields of the Sea of Okhotsk. In Sherman, K., L.M. Alexander, and B.D. Gold, eds. Large Marine Ecosystems: Stress, Mitigation, and Sustainability. AAAS Press. 376p. 69-78. Levin, S.A. 1990. Physical and biological scales, and modeling of predator-prey interactions in large marine ecosystems. In Sherman, K., L.M. Alexander, B.D. Gold, eds. Large Marine Ecosystems: Patterns, Processes, and Yields. AAAS Press, Washington, DC. 242p. 179-187. Levin, S.A. 1993. Approaches to forecasting biomass yields in large marine ecosystems. In Sherman, K., L.M. Alexander, B.D. Gold, eds. Large Marine Ecosystems: Stress, Mitigation and Sustainability. AAAS Press, Washington, DC. 376p.36-39. Levin, S.A. 2003. Complex adaptive systems: Exploring the known, the unknown and the unknowable. Bulletin of the American Mathematical 40:3-19. Lindemann, R.L. 1942. The trophic dynamic aspect of ecology. Ecology 23:399-418. Lluch-Belda, D., D.B. Lluch-Cota and S.E. Lluch-Cota. 2003. Figure 9, p.212, Chapter 9, Interannual variability impacts on the California Current large marine ecosystem. In Hempel, G. and K. Sherman, eds. Large Marine Ecosystems of the World: Trends in Exploitation, Protection and Research. Elsevier Science, Amsterdam, Netherlands. 423p. Mangel, M. 1991. Empirical and theoretical aspects of fisheries yield models for large marine ecosystems. In Sherman, K., L.M. Alexander and B.D. Gold, eds. Food Chains, Yields, Models, and Management of Large Marine Ecosystems. Westview Press, Boulder, Colorado. 243-261. Morgan, J.R. 1989. Large marine ecosystems in the Pacific Ocean. In Sherman, K., L.M. Alexander, eds. Biomass Yields and Geography of Large Marine Ecosystems. AAAS Selected Symp. 111. Westview Press, Inc. Boulder, Colorado. 493p. 377-394. NEFSC. 2002. Assessment of 20 Northeast Groundfish Stocks through 2001: A Report of the Groundfish Assessment Review Meeting (GARM), Northeast Fisheries Science Center, Woods Hole, Massachusetts, October 8-11, 2002. This report is available at http://www.nefsc.noaa.gov/nefsc/publications/crd/crd0216/ NEFSC. 1999. Atlantic Herring. In: Report of the 27th Northeast Regional Stock Assessment Workshop (27th SAAW). Stock Assessment Review Committee (SARC) Consensus Summary of Assessments. Woods Hole Laboratory Reverence Document No. 98-15. NOAA (National Oceanic and Atmospheric Administration). 1993. Emerging theoretical basis for monitoring the changing states (Health) of large marine ecosystems. Summary report of two


33

workshops: 23 April 1992, National Marine Fisheries Service, Narragansett, Rhode Island, and 11-12 July 1992, Cornell University, Ithaca, New York. NOAA Technical Memorandum NMFS-F/NEC100. NRC (National Research Council). 2000. Clean Copastal Waters: Understanding and reducing the effects of nutrient pollution. National Academy Press, Washington, DC. Pauly, D., V. Christensen. 1995. Primary production required to sustain global fisheries. Nature 374:255257. Pauly, D. and R. Chuenpagdee. 2003. Development of fisheries in the Gulf of Thailand large marine ecosystem: Analysis of an Unplanned experiment. In Hempel, G. and K. Sherman, eds. Large Marine Ecosystems of the World: Trends in Exploitation, Protection and Research. Elsevier Science. Amsterdam, Netherlands. 423p. 337-354. Prescott, J.R.V. 1989. The political division of large marine ecosystems in the Atlantic Ocean and some associated seas. In Sherman, K. and L.M. Alexander, eds. Biomass Yields and Geography of Large Marine Ecosystems. AAAS Selected Symp. 111. Westview Press, Inc. Boulder, Colorado. 395-442. Rabalais, N.N., R.E. Turner, W.J. Wiseman Jr. 1999. Hypoxia in the Northern Gulf of Mexico: Linkages with the Mississippi River. In Kumpf, H., K. Steidinger, K. Sherman, eds. The Gulf of Mexico Large Marine Ecosystem: Assessment, Sustainability, and Management. Malden, MA. Blackwell Science, Inc. 704p. 297-322. Ricklefs, R.E. 1987. Community diversity: Relative roles of local and regional processes. Science 235(4785):161-171. Schaefer, M.B. 1954. Some aspects of the dynamics of populations important to the management of the commercial marine fisheries. Bull. Inter-Am. Trop. Tuna Comm. 1:27-56. Schwartzlose, R.A., J. Alheit, T. Baumgartner, R. Cloete, R.J.M. Crawford, W.J. Fletcher, Y. Green Ruiz, E. Hagen, T. Kawasaki, D. Lluch-Belda, S.E. Lluch-Cotta, A.D. MacCall, Y. Matsuura, M.O. Nevárez-Martínez, R.H. Parrish, C. Roy, R. Serra, K.V. Shust, N.M. Ward and J.Z. Zuzunaga. 1999. Worldwide large-scale fluctuations of sardine and anchovy populations. S. Afr. J. Mar. Sci. 21:289347. Seitzinger, S.P. and C. Kroeze. 1998. Global distribution of nitrous oxide production and N inputs to freshwater and coastal marine ecosystems. Global Biogeochemical Cycles 12:93-113. Shannon, L.J., K.L. Cochrane and S.C. Pillar, eds. 2004. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26. Republic of South Africa Department of Evironmental Affairs and Tourism: Marine and Coastal Management. Creda Communications, Cape Town, South Africa. 328p. Shannon, L.J., V. Christensen and C.J. Walters. 2004. Modelling stock dynamics in the southern Benguela Ecosystem for the period 1978-2002. In Shannon, L.J., K.L. Cochrane and S.C. Pillar, eds. Ecosystem approaches to fisheries in the southern Benguela. African Journal of Marine Science 26. 328p. 179195. Sheldon, R.W. and T.R. Parsons. 1967. A continuous size-spectrum for particulate matter in the sea. J. Rish. Res. Bd. Canada 235:909-915. Sherman, B. 2000. Marine ecosystem health as an expression of morbidity, mortality, and disease events. Marine Pollution Bulletin 41(1-6): 232-54. Sherman, K. 2005. The Large Marine Ecosystem Approach for assessment and management of ocean coastal waters. In Hennessey, T. and J. Sutinen, eds. Sustaining Large Marine Ecosystems: The Human Dimension. Elsevier. Amsterdam, Netherlands. 368p. 3-16. Sherman, K. 1994. Sustainability, biomass yields, and health of coastal ecosystems: An ecological perspective. Marine Ecology Progress Series 112:277-301. Sherman, K., J. O’Reilly and J. Kane. 2003. Assessment and sustainability of the U.S. Northeast Shelf Ecosystem. In Hempel, G. and K. Sherman. Large Marine Ecosystems of the World: Trends in Exploitation, Protection, and Research. Elsevier B.V., Amsterdam, Netherlands. 423p. 93-120. Sherman, K., J. Kane, S. Murawski, W. Overholtz and A. Solow. 2002. The U.S. Northeast Shelf large marine ecosystem: Zooplankton trends in fish biomass recovery. In Sherman, K. and H.R. Skjoldal, eds. Large Marine Ecosystems of the North Atlantic: Changing States and Sustainability. 449p. 195215. Steele, J.H. 1988. Scale selection for biodynamic theories. In Rothschild, B.J., ed. Toward a Theory on Biological-physical Interactions in the World Ocean. NATO ASI Series C: Mathematical and Physical Sciences, Vol 239. Kluwer Academic Publishers, Dordrecht. 513-526.

34

K. Sherman

Sutinen, J. ed. 2000. A Framework for Monitoring and Assessing Socioeconomics and Governance of Large Marine Ecosystems. NOAA Technical Memorandum NMFS-NE-158. 32p. Tang, Q. 2003. Figure 10, p.137 from Chapter 6, The Yellow Sea and mitigation action. In Hempel, G. and K. Sherman, eds. Large Marine Ecosystems of the World: Trends in Exploitation, Protection and Research. Elsevier Science. Amsterdam, Netherlands. 423p. Terazaki, M. 1999. The Sea of Japan large marine ecosystem. In Sherman, K. and Q. Tang, eds. Large Marine Ecosystems of the Pacific Rim: Assessment, Sustainability, and Management. Blackwell Science. 465p. 199-220. USEPA. 2004. National Coastal Condition Report. EPA-620-R-03/002 Washington, DC. http://www.epa.gov/owow/oceans/nccr/2005/nccr2-factsheet.html. USEPA. 2001. National Coastal Condition Report. EPA-620/R-01/005 Washington, DC. 204p. USEO. 2004. Executive Order 121704. Committee on Ocean Policy. http://www.whitehouse.gov/news/releases/2004. USOAP. 2004. U.S. Ocean Action Plan, Office of the President of the United States. 17 December 2004. http://ocean.ceq.gov/actionplan.pdf United Nations General Assembly. 2001. Report on the work of the United Nations Open-ended Informal Consultative Process established by the General Assembly in its resolution 54/33 in order to facilitate the annual review by the Assembly of developments in ocean affairs at its second meeting. Report A/56/121, 22 June, New York. 62p. Wang, H. 2004. An evaluation of the modular approach to the assessment and management of large marine ecosystems. Ocean Development & International Law 35:267-286. Watson, R., D. Pauly, V. Christensen, R. Froese, A. Longhurst, T. Platt, S. Sathyendranath, K. Sherman, J. O’Reilly, and P. Celone. 2003. Mapping fisheries onto marine ecosystems for regional, oceanic and global integrations. In Hempel, G., and K. Sherman. Large Marine Ecosystems of the World: Trends in Exploitation, Protection, and Research. Elsevier. 423p. Wolff, M., C. Wosnitza-Mendo and J. Mendo. 2003. The Humboldt Current LME. In: Hempel, G. and K. Sherman, eds. Large Marine Ecosystems of the World: Trends in Exploitation, Protection and Research. Elsevier Science, Amsterdam, Netherlands. 423p. 279-309.


35

3 The Global Ocean Observing System for Africa (GOOS-Africa): Monitoring and Predicting in Large Marine Ecosystems Justin Ahanhanzo

INTRODUCTION The United Nations Conference on Environment and Development (UNCED) 1992 called for the creation of a global system of ocean observations to enable effective and sustainable management and development of seas and oceans, and prediction of future change. It is worth noting from a historical perspective that the Second World Climate Conference in 1990 also called for the establishment of such a system to provide the oceanographic data needed for the Global Climate Observing System (GCOS), which was proposed shortly afterward. Consequently, in response to these needs, the Twenty Fifth Session of the Executive Council of the Intergovernmental Oceanographic Commission of UNESCO formally initiated the Global Ocean Observing System (GOOS) in 1992, to which later the World Meteorological Organisation (WMO), the United Nations Environment Programme (UNEP) and the International Council of Scientific Unions (ICSU) committed themselves. GOOS is: (i) a sustained, coordinated international system for gathering data about the oceans and seas of the Earth, (ii) a system for processing such data, with other relevant data from other domains, to enable the generation of beneficial analytical and prognostic environmental information services, and (iii) the research and development on which such services depend for their improvement. Ultimately GOOS provides scientific monitoring and predicting for the implementation of the United Nations Framework Convention on Climate and the United Nations Convention on Biodiversity. THE LARGE MARINE ECOSYSTEM (LME) CONCEPT AND STRATEGY The coastal areas and the margins of the Pacific, Indian and Atlantic Oceans have been delineated into 64 Large Marine Ecosystems (LMEs), (Sherman and Alexander 1986; Sherman 1994; Sherman 1999). The LME boundaries were determined on the basis of four ecological criteria: (i) bathymetry, (ii) hydrography, (iii) productivity, and (iv) trophically dependant populations. Assessments of changing states of LMEs provide science-based information for the management of LME goods and services (Duda and Sherman 2002). Monitoring, assessment and management of the LMEs require long term and sustained ocean observations, measurements, data collection, processing, analysis, interpretation, and provision of products and services that may serve as

36

J. Ahanhanzo

decision making tools for governments, managers, planners and stakeholders. Marine data and information are crucial for monitoring, predicting and managing the LMEs and forecasting environmental change. The Global Ocean Observing System is conceived to provide these requirements on behalf of LMEs and other user communities. On the African coasts, there are six major LMEs including the Canary Current LME, Guinea Current LME and the Benguela Current LME on the Atlantic coast and the Red Sea LME, Somali Current and Agulhas Current LMEs in the Indian Ocean (Table 3-1) and (Figure 3-1). Research and investigations are ongoing to explore the criteria for the possibility of establishing a Mascarene LME.

Table 3-1. African LMEs and participating countries AFRICAN LMEs

Participating Countries

Agulhas Current

Comoros, Madagascar, Mozambique, South Africa

Benguela Current

Angola, Namibia, South Africa

Canary Current

Cape Verde, Gambia, Guinea, Guinea Bissau, Mauritania, Morocco, Senegal

Guinea Current

Angola, Benin, Cameroon, Congo, Democratic Republic of Congo, Equatorial Guinea, Gabon, Ghana, Guinea, Guinea Bissau, Ivory Coast, Liberia, Nigeria, Sao Tome and Principe, Sierra Leone, Togo

Red Sea

Djibouti, Egypt, Jordan, Saudi Arabia, Somalia, Sudan, Yemen

Somali Current

Kenya, Tanzania, Seychelles, (Somali not yet involved due to political issues)

THE RISE OF THE GLOBAL OCEAN OBSERVING SYSTEM IN AFRICA (GOOS-AFRICA) Based on several decades of experience, studies and sound expertise in oceanography and marine sciences, African institutions, marine scientists and stakeholders came to the conclusion that there is a need for an integrated approach to operational oceanography in support of the activities in LMEs and for other stakeholders. The initial GOOS-Africa information document was presented to the African Forum at the GCLME Seminar and Workshop in Abidjan, Côte d’Ivoire in early 1998. A core network of African scientists was established in 1998 to take forward the development of the Global Ocean Observing System for Africa. Upon the request of the Government of Mozambique, in the framework of the Pan-African Conference on

GOOS-Africa

37

Figure 3-1. Chlorophyll a biomass in African LMEs determined with ENVISAT/MERIS in June 2003 (courtesy of Coast Watch/ACRI)

Sustainable Integrated Coastal Management (PACSICOM), and with the support of African countries and institutions, UNESCO and its IOC, the first GOOS-Africa workshop was organized in Maputo in Mozambique, 18-20 July 1998. This workshop entitled, “Data for Sustainable Integrated Coastal Management, Global Ocean Observing System for Sustainable Integrated Coastal Management in Africa,” laid the foundations for the rise of the Global Ocean Observing System in Africa (GOOSAfrica). The programme addresses the fact that Africa can be impacted by extreme events such as El Niño, and La Niña, which affect rainfall and crops, as well as by floods, drought and tropical cyclones that have various causes in the atmosphere and ocean. In addition, the recent Indian Ocean tsunami also affected African countries on the East coast of the continent. A multidisciplinary approach to collecting observations, needed to forecast and interpret such events, becomes imperative because of the linkages between ocean, earth and meteorological processes and climate change: GOOS-AFRICA recognises that what happens at the coasts is commonly a complex function of earth, ocean and atmosphere on regional and global scales. Ocean processes can affect economic values of African investments in: (i) offshore and coastal oil and gas (ii) shipping and trade; (iii) offshore and coastal mining; (iv) coastal and offshore fisheries; (v) integrated coastal zone management; (vi) monitoring and predicting in large marine ecosystems; (vii) seaside tourism; (viii) public safety/health and protection of properties; (ix) early warning systems. The GOOS-Africa mandate includes a provision of a common platform of coastal and ocean services for monitoring and predicting dynamics of the large marine ecosystems through (i) assessing, (ii) hindcasting, predicting and forecasting, and (iii) establishing early warning systems. These will provide information on potential floods, sea level rise, regime shifts and their impacts on ecosystems and on the people who depend on them.

38

J. Ahanhanzo

GOOS-AFRICA STRATEGIC PARTNERSHIPS The GOOS-Africa Network recognized that a grassroots approach, with national and regional leadership and ownership is the key for building the long term institutional and scientific capacity required to get the desired results. In addition, positive synergy with reliable partners, in particular the regional financing institutions and industries, is needed to help to develop financial models enabling businesses and Governments to share the financial burden of the Regional Ocean Observing and Forecasting Systems for Africa (ROOFS-Africa) that will be the mechanism for implementing GOOS in Africa. THE AFRICAN LMES ARE CORE AND VITAL STRATEGIC PARTNERS FOR GOOS-AFRICA. GOOS-Africa is conceived to provide long term, sustained and systematic observations

both from in situ collecting devices and satellite remote sensing, combined with modelling, data assimilation and forecasting to support the monitoring, prediction and management of the resources of African LMEs (Figure 3-1) and (Table 3-1). Strategic partnerships have been established with relevant ongoing African and overseas programmes and specialized institutions. These programmes and partners include all ongoing African LMEs, the fisheries programmes, the UNESCO crosscutting project on the Applications of Remote Sensing for Integrated Management of Ecosystems and Water Resources in Africa, the African Monsoon Multidisciplinary Analysis (AMMA), the Ocean Data and Information Network for Africa (ODINAFRICA), the European Commission funded project for the GOOS Regional Alliances Network Development, the African Centre of the Meteorological Applications for Development (ACMAD); African Association of Remote Sensing of the Environment, (AARSE); National Oceanographic Center of Southampton, UK Meteorological Office, the European Space Agency (ESA), the US- National Oceanic and Atmospheric Administration (NOAA), the French Institute of Research for Development (IRD); the United Nations Agencies including the United Nations Environment Programme (UNEP); United Nations Industrial Development Organization (UNIDO), United Nations Development Programme (UNDP), United Nations Office of Outer Space Affairs (UNOOSA), and other multilateral and bilateral partners. In May 2005, GOOS-AFRICA was invited to sit as an observer and partner in the Met Ocean Committee of the International Association of Oil and Gas (OGP), given the considerable interest and the potential for mutual benefit between the GOOS-AFRICA Networks and the OGP members. GOOS-AFRICA CONTRIBUTION TO INTEGRATED MONITORING AND PREDICTING OF LARGE MARINE ECOSYSTEMS Based on a multidisciplinary and integrated strategy of five complementary work packages or modules, GOOS-Africa’s observing and forecasting system (ROOFSAfrica) will provide the observations and forecasting underpinning for ecosystembased management and rational use and exploitation of marine resources towards

GOOS-Africa

39

sustainable development of marine environment. The modules are listed in Table 3-2. To ensure flexibility, these modules are carried out in a complementary way. Their value will be maximised when all are integrated. Each module comprises a rationale, key objectives, key tasks, and a list of outputs, results and deliverables.

Table 3-2. Modular approach and strategy of GOOS-Africa

Modules

Title

M1

The African network of in situ ocean observing and monitoring systems including sea level records for monitoring coastal zones and impacts of global change in Africa

M2 M3

Remote sensing and satellite applications to marine and coastal environment Modelling, hindcasts/forecasts and data assimilation based on in situ and satellite data Effective involvement of different stakeholders at different stages of project implementation, and development of end-user interactive communication and information delivery system Industry and business partnerships towards reinforcing a Regional Ocean Observing and Forecasting System for Africa (ROOFS-Africa)

M4 M5

Module 1: The African network of in situ ocean observing and monitoring systems including sea level records In-situ ocean measurements are key elements of any coastal ocean observation system for Africa. These measurements must be made for the long term so as to provide the underpinning for accurate understanding and forecasting of water levels and water quality that are essential for a wide variety of uses and users including port management, shipping, fisheries, tourism, offshore and coastal installations and coastal erosion. This basic information is essential for the warnings and mitigation of natural disasters and the management of marine resources and offshore operations, including fish stocks and marine pollution. Measuring sea level and other ocean parameters provide a vital component of oceanographic observation programmes needed for immediate operational requirements of ships, navigation, and storm surge forecasting, for long-term monitoring and prediction of global sea level changes due to climate variations. In particular, tidal information is needed for addressing the following: (i) coastal erosion; (ii) flooding; (iii) salt water intrusion; (iv) sea-level topographic map production with satellite calibration; (v) assessment of ecosystem health; (vi) marine navigation and transportation; (vii) oil exploration and exploitation activities; (viii) marine pollution and oil spill mitigation; (ix) early warning systems.

40

J. Ahanhanzo

Module 2: Remote sensing and satellite applications to marine and coastal environment Satellite data, obtained synoptically every day from a number of different sensors together with in situ data necessary to establish sensor dependent error statistics, constitute a vital component of ROOFS-Africa. In addition, these data are required to constrain, force and initialize ocean model systems. Sea surface temperature, ocean colour (and derivative water quality parameters), surface wind speed, rain rates and distributions, solar radiation, sea surface height and state are priority data sets that should be ingested and archived in an operational manner to serve and foster the ROOFS for Africa. Synthetic aperture radar (SAR) and aircraft remote sensing measurements are also required for events (e.g., HAB, oil spill) requiring a rapid response. Remote sensing, as important as it is, will not on its own provide adequate answers to marine and coastal problems (Figure 3-2). Bencal 41 0m

Measured in-water particle size distribution data from the FRS Africana, showing the difference in size of the two algal assemblages despite their similar chlorophyll a concentrations

Bencal 37 0m Diatoms Pseudonitzschia sp. & Thalassiosira sp. dominant Chl a = 5.4 mg m-3 Deff = 12 μm

Dinoflagellates Alexandrium catenella & Prorocentrum micans dominant Chl a = 6.5 mg m-3 Deff = 24 μm

Size and Chl a data from the inverse reflectance algorithm - despite the speckling at low bio-mass, the size product demonstrates that it is possible to distinguish between a large cell size dinoflagellate community and a smaller sized diatom community

Figure 3-2. Application of the multispectral reflectance algorithm to SeaWiFS data. The SeaWiFS overpass is from the 15th of October 2003, during the BenCal bio-optical cal/val cruise in the southern Benguela. An extraordinary bloom of the toxic dinoflagellate Alexandrium catenella was reported in the Lamberts and Elands Bay vicinity several days later. The precursive expression of this bloom can be seen in the effective diameter image, which shows the presence of a large cell assemblage off Elands Bay. Note the ability of the algorithm to differentiate between water types dominated by differently sized phytoplankton at approximately equal chlorophyll concentrations (From Dr. Stewart Bernard et al., Oceanography Department University of Cape Town, South Africa. This study was funded and carried out under the framework of the UNDP/GEF BCLME project).

GOOS-Africa

41

Module 3: Modelling and forecasting based on in situ and satellite data Numerical modelling and forecasting based on in situ and satellite data constitute the basis for many services that provide forecasts or assessments useful to decision-makers working in the public or private sectors in the marine and coastal environment. The modelling and forecasting systems serve for coastal protection and better management of coastal erosion; pollution; marine transportation; fisheries; tourism; and pristine ecosystems. Module 4: Effective involvement of different stakeholders at different stages of project implementation and development of an end-user interactive communication and information delivery system This is a crosscutting package with the others. The main objective is to build up an end-user interactive information delivery system derived from active stakeholder participation in all stages of applications development and dissemination (from data gathering, to the packaging of information in forms that facilitate decision-making that will improve the life of societies). The ACMAD Communication System is the quantum leap in closing the information gap, providing missing links in the chain of development: ACMAD has developed a communication system (Radio and Internet – RANET) that involves uploading information to the AFRISTAR satellite from ‘editors’ such as ACMAD, downlink via solar powered receivers to local solar powered FM stations for rebroadcasts in local languages, with local interpretation to holders of wind-up radios in rural villages. Module 5: Industry and business partnerships towards reinforcing the Regional Ocean Observing and Forecasting System for Africa (ROOFSAfrica) The drivers consist of enhancing the transfer of meteorological and oceanographic (met ocean) information from the data providers to the users through establishing and fostering working partnerships between governments and industry. These partnerships will roadmap the role of met ocean information in the industry and business decision process and will establish the economic value of the information. There are four development and philosophical drivers for Industry and Business Partnerships: (i) The New Partnership for Africa’s Development (NEPAD) goals as exemplified in their existing projects; (ii) the WSSD goals as outlined in Africa (Johannesburg, South Africa, 2002); (iii) the Millennium Development Goals of the United Nations, particularly for the alleviation of poverty; and (iv) the Global Business Governance goals on sustainability as outlined in the “Triple Bottom Line” principles. Aligning industry and business partnerships with each of these goals can ensure the fulfilment of the sustainability mission.

42

J. Ahanhanzo

CONCLUDING REMARKS: SUCCESS STORIES

GOOS-Africa as an integral part of the African Renaissance is a viable scientific and technical framework for sustainable management of the Large Marine Ecosystems and for the protection of coastal and ocean environments. GOOS-Africa is the African contribution to the development and implementation of the Global Ocean Observing System (GOOS).

GOOS-Africa: Operational tool for Monitoring and Predicting the dynamics of African LMEs Successful arrangements concluded between GOOS-Africa and European Space Agency (ESA) enable free release of MERIS/ENVISAT data related to the African waters for the use by the members of the GOOS-Africa Network. The BCLME is the first GOOS-AFRICA partner that took advantage of these opportunities, in collaboration with the Department of Oceanography of the University of Cape Town (UCT) and the Marine and Coastal Management (MCM) of the Government of South Africa, combining remotely sensed MERIS and MODIS data with in situ measurements and observations to generate useful marine products and services for monitoring and predicting the health and status of the BCLME. Recently, in March 2005, responding to the call of African countries and Governments following the devastating Indian Ocean Tsunami, the GCLME in partnership with GOOS-Africa organized the Workshop on Coastal Dynamics in Integrated Areas Management and Early Warning Systems in Africa to identify the criteria for establishing an integrated Multi-hazards Early Warning and Mitigation System. GOOS-Africa has been called upon to contribute to the Project Development Phase of the CCLME and is participating in the Road Map Workshop that will develop the work programme for implementation. The joint leadership role of the BCLME, GCLME, GOOS–Africa, UCT and MCM to foster the development of operational oceanography in Africa shows the evidence of the positive synergy and complementarity between GOOS and the LMEs programmes, at least in the African context, towards integrated monitoring and predicting the environment and the ecosystem in LMEs. GOOS-Africa provided the observation background and substantial elements for the development of the ODINAFRICA-III project. This contribution enables installation of a number of modern tides gauges along African coasts reinforcing the networks of in situ measurements.

GOOS-Africa

43

GOOS-AFRICA FORWARD LOOK The positive synergy between GOOS-Africa and the African LMEs fosters a rapid implementation of capacity building and forecasting priorities and needs identified in the two previous chapters, 1 and 2, of this book through joint major initiatives in preparation, notably: (i)

Pan-African LMEs/GOOS-AFRICA Leadership Workshop on Operational Oceanography and Remote Sensing in Africa, Cape Town, South Africa 6-10 November 2006

(ii)

The Second Pan-African LMEs Forum, Cape Town, South Africa 13 November 2006

(iii)

The Third Forum of the GOOS Regional Alliances, Cape Town, South Africa, 14-17 November 2006-05-26

ACKNOWLEDGEMENTS The Author is grateful to Professor Vere Shannon, former Director of the Sea Fisheries Research Institute of South Africa for his continuous encouragement and patient pedagogical advice. Special thanks to Dr. Steward Bernard, Mr. Cristo Whittle (UCT) and Dr. Antoine Mangin (Coastwatch/ACRI) who provided part of the data supporting the illustrations of this Chapter. The Author thanks also Professor Geoff Brundrit, the past Chairman of GOOS-AFRICA; Dr Kwame Koranteng, Current Chairman of GOOS-AFRICA; Dr. Silvana Vallerga, the Chairperson of the Intergovernmental Committee of GOOS; Professor John Woods, Chairman of the first Experts Group on GOOS; Dr. Kenneth Sherman, US-NOAA founder of the LME Concept; Dr Brad Brown, former Director of the US-NOAA Southeast Sea Fisheries Center; Professor Gotthilf Hempel, former Director of the Institute of Tropical Ecology of Bremen; Professor Chidi Ibe, the Regional Director of the GCLME; Dr. Mike O’Toole, the Chief Technical Advisor of the BCLME; Mr. Mohammed Boulahya, first Director General, Co-Founder of ACMAD and Senior Expert Climate and Environment NEPAD Secretariat; Dr. Pierre-Philippe Mathieu, ESA; Dr. Mary Altalo, VicePresident, US-based Sciences Applications International; Dr. Colin Summerhayes, Former Director of the GOOS Project Office at IOC/UNESCO and current Executive Director, Scientific Committee on Antarctic Research (SCAR); and the Assistant Director General of IOC/UNESCO, Dr. Patricio Bernal. REFERENCES AAAS. 1994. Science in Africa: The Challenges of Capacity-Building, A forum organized by the AAAS Sub-Saharan Africa Program, American Association for the Advancement of Science, Washington, DC, May 10, 1994.

44

J. Ahanhanzo

AAAS. 1995. New Directions for Science and Technology in South Africa: Opportunities for US Collaboration, A Seminar organized by the AAAS Sub-Saharan Africa Program American Association for the Advancement of Science Washington, DC, May 2, 1995. AAAS. 1996. Utilizing Africa's Genetic Affluence through Natural Products Research and Development, A Symposium at the 1996 Annual Meeting, American Association for the Advancement of Science, Washington DC 1996. Agenda 21. 1992. The United Nations Conference on Environment and Development, UNCED Document A/CONF. 151/4 (Parts I and II). The United Nations Convention on the Law of the Sea, United Nations, New York, 1983. Ahanhanzo, J. 1995. Integrated Development of Modern Oceanography and the Management of Oceanographic Resources in the Benguela Current Region and Comparable Eastern Boundary Upwelling Ecosystems. Ahanhanzo, J. 1998. The GOOS-AFRICA Concept, in the Guinea Current Large Marine Ecosystem, UNIDO, Editor Professor Ibe Chidi. Ahanhanzo, J. 2003. The Rise of GOOS-AFRICA In the IOC Annual Report, 2003. Chu, P.C. and J.C. Gascard, editors. 1991. Deep Convection and Deep Water Formation in the Oceans, Elsevier Oceanography Series 57, Amsterdam, 382 p. CNES. 1998. French space ambitions, Paris. Dahlin, H., N.C. Flemming, K. Nittis, S.E. Petersson, editors. 2003. Building the European Capacity in Operational Oceanography, Proceedings of the Third International Conference on EuroGOOS, Elsevier Oceanography Series 69, Amsterdam. Duda, A.M. and K. Sherman. 2002. A new imperative for improving management of large marine ecosystems. Ocean & Coastal Management 45:797-833. EuroGOOS. 1997, Annual Report 1995-96, EuroGOOS Publication No. 2, Southampton Oceanography Centre, Southampton UK. Flemming, N.C., S. Vallerga, N. Pinardi, H.W.A. Behrens, G. Manzella, D. Prandle, J.H. Stel, editors. 1999. Operational Oceanography: Implementation at the European and Regional Scales, Proceedings of the Second International Conference on EuroGOOS, Elsevier Oceanography Series 66, Amsterdam. Guymer, T.H., N.C. Flemming, J. Font, P. Gaspar, J. Johannessen, G.H. van der Kolff, C. le Provost, A. Ratier and D. Williams. 2001. EuroGOOS Conference on Operational Ocean Observations from Space, EuroGOOS Publication No. 16, Southampton Oceanography Centre, Southampton. Pinardi, N. and N.C. Flemming, editors. 1998. The Mediterranean Forecasting System Science Plan, EuroGOOS Publication No. 11, Southampton Oceanography Centre, Southampton. Sherman, K. and L.M. Alexander, editors. 1986. Variability and Management of Large Marine Ecosystems. AAAS Selected Symposium 99. Westview Press, Colorado. 319p. Sherman, K. 1994. Sustainability, biomass yields, and health of coastal ecosystems: An ecological perspective. Mar Ecol Prog Ser. 112:277-301. Sherman, K. and Q. Tang, editors. 1999. Marine Ecosystems of the Pacific Rim: Assessment, Sustainability, and Management. Blackwell Science, Inc., Malden, MA. 465p. Sherman, K. and A.M. Duda. 1999. An ecosystem approach to global assessment and management of coastal waters. Mar Ecol Prog Ser.190: 271-287. UNESCO/IOC. 1992, Twenty-fifth Session of the Executive Council, Paris, 10-18 March 1992, IOC Reports of Governing and Major Subsidiary Bodies. UNESCO/IOC. 1993, IOC Committee for the Global Ocean Observing System (GOOS), First Session, Paris, 16-19 February 1993, IOC Reports of Meetings of Experts and Equivalent Bodies. UNESCO/IOC. 2005, IOC-IUCN-NOAA Consultative Meeting on Large Marine Ecosystems (LMEs), Sixth Session, 29-30 March 2004, Paris, IOC Reports of Meetings of Experts and Equivalent Bodies. UNESCO/IOC 1998. “The GOOS 1998” IOC, Paris. U.S. GOOS. 1992. First Steps Towards A U.S. GOOS, Report of a Workshop on Priorities for U.S. Contributions to a Global Ocean Observing System, Woods Hole, Massachusetts, 14-16 October 1992, Joint Oceanographic Institutions Inc., Washington. GOOS Report 62, IOC-UNESCO Publications. Woods, J.D. 1991a. Oceanography on a global scale: The new challenge Phys. Ed 26:159-163, 168. Woods, J.D. 1991b. Global Ocean Observing and Climate Forecasting Science in Parliament 48 (3):4-10 Woods, J.D. 1992. Monitoring the ocean. In Cartledge, B. ed. Monitoring the Environment, Oxford University Press

GOOS-Africa

45

Woods, J.D. 1993. The case for GOOS (The Global Ocean Observing System) Intergovernmental Oceanographic Commission IOC /INF-915, Paris Woods, J.D. Dahlin, H., L. Droppert, M. Glass, S. Vallerga,. and N.C. Flemming. 1996. The Strategy for EuroGOOS. EuroGOOS Publication N.1, Southampton Oceanography Centre, Southampton. ISBN 0904175227 Woods, J.D., H. Dahlin, L. Droppert, M. Glass, S. Vallerga, and N.C. Flemming. 1997. The EuroGOOS Plan. EuroGOOS Publication N.1, Southampton Oceanography Centre, Southampton. ISBN 0904175226.


Part II: Setting the Scene Data time series and models: What we think we know about variability in the Benguela and comparable systems.



49

4 Large Scale Physical Variability of the Benguela Current Large Marine Ecosystem (BCLME) F. A. Shillington, C. J. C. Reason, C. M. Duncombe Rae, P. Florenchie, and P. Penven

INTRODUCTION The Benguela Current Large Marine Ecosystem (BCLME) is situated off the west coast of Africa between 5-37ºS, 0-26ºE, and spans the three countries of Angola, Namibia and South Africa. It is one of the four major eastern boundary current upwelling systems of the world oceans (Hill et al. 1998), and although it has some similar characteristics to the other eastern boundary upwelling areas, a unique feature is that it is bounded on both the equatorial and poleward extremities by warm water current systems (the tropical warm Angola Current system in the north, and the Indian Ocean western boundary Agulhas Current System in the south; Shannon and Nelson, 1996; Shillington 1998; Shannon and O’Toole 2003). In the region between about 1537ºS, the surface currents are generally equatorward, with vigorous coastal upwelling cells, strong and narrow equatorward shelf edge jets (near Cape Town which is situated at 34ºS, 18ºE and off Lüderitz; 28ºS, 15ºE), and a poleward undercurrent along the shelf slope and bottom. The warm tropical Angola Current System (Ajao and Houghton 1998) generally has southward moving coastal currents which meet the Benguela Upwelling System at the Angola-Benguela Frontal Zone (ABFZ) at ~1517ºS (Shannon et al. 1987; Field and Shillington 2005; Monteiro and van der Plas, this volume; Veitch et al. 2006). The Angola Current is affected by input from the equatorial wave guide, the South Equatorial Current (SEC) and the South Equatorial Counter Current (SECC) at ~5°S (Peterson and Stramma 1991). Details of the circulation of the Angola Gyre and the nature of the Angola Dome are addressed by Monteiro and van der Plas (this volume), and by Reason et al. (this volume). At the centre of the BCLME region is an area of year-round coastal upwelling, 1530°S (Boyer et al. 2000); and a region of seasonal upwelling, 30-34°S. Coastal trapped waves have been observed to propagate polewards on the continental shelf at regular synoptic time scales (~3-10 day periods) from Walvis Bay in Namibia (20°S), and to continue around the Cape of Good Hope and up to 800 km east along the eastern coast of South Africa (Brundrit et al. 1987; Schumann and Brink 1990).

50

F.A. Shillington et al.

At the southern end the BCLME region, the Agulhas Bank (see Fig. 1-1 in Chapter 1 for the shelf topography of the BCLME region) is a very wide shelf region along the southern coast of Africa from 18-26°E, that has a highly vertically stratified water column in the west in summer, and a well mixed water column in the winter (Schumann 1998). Closer to the coast, there is summer upwelling of cool nutrient rich water at the major coastal embayments on the African coast between these longitudes. In the middle of the Agulhas Bank, there is a seasonal cold tongue apparent in surface and near-surface waters; the circulation around this feature appears to be cyclonic (Boyd and Shillington 1994). This feature is particularly visible as a ridge of elevated chlorophyll, occurring in the period from March-June (Demarcq et al 2003; HardmanMountford et al. 2003). The Agulhas Bank is very important for the spawning of pelagic fish such as anchovy and pilchard from September to March (Hutchings et al. 2002). After spawning, eggs and larvae drift northwards in the jet current past Cape Town, until juvenile fish recruitment occurs about 150 km north along the coast at ~32°S in St Helena Bay. Adult fish then make their way back to the Agulhas Bank to spawn in the following austral spring-summer (van der Lingen et al. this volume). Large-scale, multiyear climatic variations in the Benguela upwelling region have been observed from time to time and have been dubbed “Benguela Niños” as an analogue to the Pacific event (Shannon et al. 1986). The Benguela Niño, like its Pacific counterpart, has a strong effect on regional fisheries and this in turn has led to an effort to forecast these events. Benguela Niños have been observed/reported in 1934, 1963, (1972/3), 1984, 1995 (Shannon et al. 1986; Gammelsrød et al. 1998). Field measurements of the 1995 Benguela Niño were reported by Gammelsrød et al. (1998). More recently, Florenchie et al. (2003) and Florenchie et al. (2004) have examined the nature of the 1984 and 1995 Benguela Niños using an ocean general circulation model together with satellite derived sea surface temperature (SST) and sea surface height (SSH) data, to show how they can be related to local and remote wind forcing. Their results suggest that a possible forecast lead time of two months exists for anticipating strong positive SST anomalies propagating from the equatorial region, polewards beyond the Angola Benguela Frontal Zone (ABFZ). Benguela Niños represent the lowest frequency, largest-scale instance of variability in the BCLME. The main large scale physical features of the BCLME are summarised in the cartoon in Fig. 1-1 in Chapter 1. The main purpose of this review is to set the scene for a discussion of potentially forecastable aspects of major importance in the BCLME. Questions that are central to this discussion are formulated below. • What proportion of the BCLME large scale variability is associated with the seasonal forcing, and what part is related to inter-decadal events such as the Benguela Niño? • What is the role of other large scale modes such as ENSO and the Southern Annular Mode in driving variability in the BCLME region ?

Large scale physical variability of the Benguela Current LME

51

• Can monitoring the remote and/or local wind forcing in the western equatorial Atlantic region give an acceptable lead time for nowcasting/forecasting Benguela Niños at or south of the Angola Benguela Frontal Zone? • How is the Benguela Niño signal transmitted/influenced by the poleward flowing Angola Current and what is the nature of the interaction with the northern part of the Benguela Upwelling at the ABFZ at ~15-17ºS? • What is the nature and importance of the Angola Dome area for the large scale formation of low oxygen water, and is this water responsible for low oxygen water in the northern Benguela? • What is the effect and importance of the variability of the outflow of the main large rivers (e.g. Congo/Zaire and Orange/Gariep Rivers) on the BCLME? • Is it possible to nowcast/forecast changes in the extremely vigorous wind driven upwelling in the Lüderitz region, which tends to persist throughout the year? • What is the nature of the predictability of the remote and local forcing of the southern Benguela Upwelling System from the poleward end via the influence of the Agulhas Current and its retroflection? • Is there seasonality in the Agulhas Current and the shedding of Agulhas retroflection rings? Is the shedding of rings predictable or capable of being monitored sufficiently far “upstream in the Agulhas Current” to provide advance warning of interaction with the Benguela upwelling front? Major intrusions of sub-Antarctic water have been observed to interact with the outer boundary of the southern Benguela Ecosystem in 1987 (see Shannon et al. 1990 as cited in Hardman-Mountford et al. 2003). Are such intrusions of sub-Antarctic water into the southern Benguela System important sources of variability and are they predictable? MAJOR PHYSICAL PROCESSES IN THE BCLME The main dynamic processes in the BCLME are similar to other major eastern boundary upwelling systems (Hill et al. 1998). They include: • Dominant equatorward wind stress inducing Ekman offshore transport of surface water, which is replaced by cool, nutrient rich subsurface central water (see detailed section below on recently measured water masses in the central Benguela). The upwelling process leads to the surfacing of cool, coastal nutrient rich water; the subsequent growth and decay, and instability of oceanic fronts, filaments and frontal jets e.g. Shillington (1998). • A poleward undercurrent along the continental shelf break which later intrudes onto the shallower continental shelf in various places. The detailed mechanism responsible for this is not clear at present. • Poleward propagating coastal trapped waves on the shelf, which are easily detected in coastal tide gauge recordings of sea level, e.g. Brundrit et al. (1987). • Kelvin wave like disturbances travelling eastwards along the Atlantic Ocean equatorial waveguide, travel from South America to Africa, and later turn polewards along the Angolan coast. Temperature anomalies can give rise to either

52


local warm events, or in some cases, Benguela Niños that reach as far south as Walvis Bay (~22°S), e.g. Florenchie et al. (2004). • Agulhas ring formation after the Agulhas Current Retroflection south and west of Cape Town, and the subsequent interaction of these rings with the southern Benguela upwelling frontal system, e.g. Duncombe Rae et al. (1992), Shillington (1998). This process is unique to the BCLME, and not found in any of the other major eastern boundary current systems. • The seasonal and inter-annual meridional movement of the ABFZ, and the quasidecadal variability of Benguela Niños, e.g. Veitch et al. (2006). This process appears to be unique to the BCLME, and not found in any of the other major eastern boundary current upwelling systems. The main transboundary areas of the BCLME are: the northern Angola Current border with the equatorial currents; the ABFZ; the Lüderitz-Orange River cone area (Duncombe Rae 2005); the Agulhas Current-Benguela upwelling interaction at the southern boundary of the Benguela upwelling area; the coastal transition zone between the cold upwelling coastal and continental shelf region and the deeper ocean. ATMOSPHERIC FORCING OF THE BCLME The atmospheric circulation of the BCLME region is dominated by the South Atlantic subtropical anticyclone which gives rise to southerly wind stress near the west coast of Africa. To the south of Africa and the region, there is generally a westerly flow, with changes in wind direction associated with west to east travelling mid-latitude cyclones. During austral summer, surface heat induced low pressure systems develop over western South Africa, enhancing the zonal pressure gradient and leading to an intensification of the southerly wind stress off the west coast. A separate heat induced low pressure system develops over southern Angola/northern Namibia with an associated westerly windstress off the tropical SE Atlantic that feeds into the confluence between the ITCZ and the Congo air boundary. In winter, the major atmospheric circulation features shift north so that most of the BCLME region is dominated by low level southerlies. The exception to this is south of about 30°S, which is subjected to frequent atmospheric frontal activity (e.g. Hardman-Mountford et al. 2003). Superimposed on these seasonal changes in the low level winds is considerable mesoscale, synoptic, intra-seasonal, inter-annual and longer time scale variability. On synoptic time scales, the predominant anticyclonic equatorward wind flow along southern Namibian and South African west coast is perturbed by cold fronts, coastally trapped low pressure systems, “cut off lows”, and mesoscale features such as “berg winds” and sea breezes. Sometimes berg winds are followed by “coastal lows” (Reason and Jury 1990) which tend to significantly perturb the coastal wind fields. Cold fronts are most common in the winter half of the year whereas “cut off lows” may occur in any season but tend to be more likely in the austral spring and autumn.


53

West coast troughs may affect the entire coast south of about 10°S but are more common in the north (south) during winter (summer). All these weather systems interrupt the upwelling-favourable winds at a variety of space and time scales. Risien et al. (2004) examined sixteen months of QuikScat satellite derived windstress data in the Benguela System, using an artificial neural network (the Kohonen self organising map) to divide the region into six discrete regions, and wavelet analysis to extract the spatial and temporal variability scales between four and sixty four days. Chelton et al. (2004) have located significant time independent narrow bands of cyclonic curl (see Fig. 4-1; negative in the Southern Hemisphere) with large alongshore scales, adjacent to the western coastline of southern Africa from an analysis of four years of Quikscat windstress. The detailed structures and evolutions of these nearshore curl and divergence features were previously poorly resolved by historical ship observations. The implication of the long term averaged cyclonic curl of windstress is that the shallow eastern boundary Atlantic Ocean thermocline will be elevated upwards towards the surface, while the divergence will modulate the upwelling along the coast.

Figure 4-1. Four year average windstress curl (left) and divergence (right) calculated from Quikscat. Units are N m-3 x 107. (After Chelton et al. 2004)

54


LARGE SCALE MODES OF VARIABILITY A number of large-scale modes of variability influence the atmospheric circulation over the South Atlantic, and hence the BCLME region. These include ENSO, which primarily influences the region via the Pacific-South America pattern (Mo and Paegle 2001; Colberg et al. 2004), the semi-annual oscillation (van Loon 1967), the Antarctic Oscillation or Southern Annular Mode (Kidson 1988), and large-scale modulations of the subtropical anticyclone which may be locally forced (Venegas et al. 1996; 1997; 1998) or related to near-hemispheric modulations of the subtropical high pressure belt (Jones and Allan 1998; Reason 2000). Modulations in the trades over the western tropical Atlantic may generate Benguela Ninos (Florenchie et al. 2004) which may then influence the atmospheric circulation over the northern Benguela region (Rouault et al. 2003). In addition, shifts in the atmospheric wave number three pattern can often produce dipole-like SST variability in the South Atlantic and South Indian Oceans (Fauchereau et al. 2003; Hermes and Reason 2005) that tends to occur during the summer. There are several other large-scale modes that are important for the tropical Atlantic (meridional gradient mode, zonal mode, North Atlantic Oscillation) and whose potential influence on the BCLME region needs to be assessed (see Chapter 10: Reason et al. this volume). WATER MASSES AND VERTICAL STRUCTURE OF THE BCLME The major oceanic influences on the Benguela upwelling system are derived from the equatorial Atlantic in the north and the South Atlantic/South Indian to the south. Direct water mass analysis in the BCLME (Figs. 4-2 and 4-3) can be used to discriminate the influence of tropical water entering from the Angola Basin and the northern Benguela, from that being upwelled in the southern Benguela. A recent comparative study of the historical record of nutrients and hydrographic properties of the Benguela has been made by Kearns and Carr (2003). Antarctic Intermediate Water (AAIW) that is formed at the surface in the sub-polar and polar regions has a salinity minimum deep in the water column, with distinct characteristics in the northern and southern Benguela (Shannon and Hunter 1988; Talley, 1996). From the Angola Basin a high (relative to the southern Benguela water type) salinity AAIW (HSAIW) enters the northern Benguela in a poleward undercurrent along the shelf edge. The southern Benguela has a low salinity AAIW (LSAIW) close to the Subtropical Front. Similarly the South Atlantic central water in the Benguela has a relatively High Salinity component (HSCW) originating in the tropical Angola Basin and relatively Low Salinity Central Water (LSCW) in the Cape Basin. Above the central waters there is higher salinity, warm Oceanic Surface Water (OSW). The surface water is subject to the influence of precipitation and continental runoff from rivers into the Angola Basin resulting in low salinities at the surface (Mohrholz et al. 2001). In the southern Benguela the run-off from the Orange River is intermittent


55

and controlled by dams and therefore less evident in extent and persistence than in the north. The central water on the shelf is upwelled near the coast by the persistent equatorward component of the wind. Because of the atmospheric modification of temperature and salinity it is designated Modified Upwelled Water (MUW). In general terms, the intermediate, central, and upper waters can be summarised as having either (a) a high salinity, high temperature character indicating a tropical influence; or (b) a low salinity, low temperature character indicating an Antarctic or sub-Antarctic influence. Appropriate modifications of the surface and upwelled water occur during contact with the atmosphere due to solar heating and turbulent mixing processes.

Figure 4-2. Sampling stations on the BENEFIT cruises of RV Africana in 1999 (lines GG, WB) and 2002 (lines E, H, L), and from the ASTTEX deployment cruise of RV Melville in 2003 (line R). The dashed contour represents the 200 m isobath. Other isobaths are labelled.

56


Figure 4-3. Θ-S diagram of the water column profiles from the northern and southern Benguela. Note the clear salinity difference between the central waters of the two extremes. The water mass definitions used in the text are superimposed. The water mass definitions used in the text are superimposed. The water masses labelled are: ABW – Antarctic Bottom Water; NADW– North Atlantic Deep Water; LSAIW – Low Salinity Antarctic Intermediate Water; HSAIW – High Salinity Antarctic Intermediate Water; LSCW – Low Salinity Central Water; HSCW – High Salinity Central Water; MUW – Modified Upwelled Water; OSW–Oceanic Surface Water. The very low salinity seen in the surface water of some stations is due to continental run-off. Water masses below the isopycnal shown (σt= 27.75 kg.m.-3) are not discussed in detail in the text.

Figure 4-4. Monthly temperature fields in April at level 20 (surfaceleft panel) and at level 16 (approximating the mixed layer, right panel) from ROMS model.


57

The waters of the South Atlantic Ocean thermocline (central water) layer originate in two source water mass regimes (Poole and Tomczak 1999): Eastern South Atlantic Central Water (ESACW) and Western South Atlantic Central Water (WSACW). The ESACW is derived from the Indian Central Water through the Agulhas Current, and the WSACW is derived from the Brazil Current through the Brazil/Malvinas Confluence in the western South Atlantic subtropical gyre. In the eastern basins of the South Atlantic, the WSACW is present in the Angola Basin while the ESACW is found in the Cape Basin. The characteristics of WSACW are modified in the region of the ABFZ from their source water characteristics by upper layers processes in the equatorial Atlantic (Mohrholz et al. 2001). Below the main thermocline, the AAIW on the west coast of southern Africa has a salinity of 34.35, rising to 34.50 near the ABFZ (Talley 1996; Duncombe Rae 1998; Shannon and Hunter 1988; Mohrholz et al. 2001). Higher salinities are found in the intermediate water on the east coast, and in the Agulhas Current, due to the influence of occasional intrusions of Red Sea Water (Gründlingh 1985). These latter high salinity AAIW sources, however, appear not to influence the intermediate water of the central Benguela. The broad circulation of the water masses (described by Shannon and Nelson, 1996, after Chapman and Shannon, 1985) is indicated in detail by the steric height anomaly at 500 dbar (Reid 1989) and shows two opposing gyres within the South Atlantic which have a confluence in the region of the Lüderitz upwelling cell (Mercier et al. 2003). In vertical sections across the shelf the high salinity water appears constrained to the shelf edge, consistent with a poleward undercurrent of Angola Basin origin. In the region of the Lüderitz upwelling centre, consistent with Monteiro (1996), the southward moving water in the poleward undercurrent appears directed off-shore at about the same level as a local oxygen minimum in the central water of the Cape Basin gyre. Discontinuity in water masses between this latitude and the Orange River Mouth suggests that the Lüderitz upwelling cell at 26°40’S diverts the southward movement of high salinity central water in the poleward undercurrent. As an indication of the extent of the exchange between the two kinds of central water, the proportion of the HSCW within the water column was determined as a fraction of the central water as defined above. The distribution of this proportion shows the exchange between the two extremes of the system occurring between Lüderitz and Cape Frio. As the vertical sections of water mass show, the water masses remain separable showing little mixing. It is only the extent of the denser high salinity portion that becomes less as the Lüderitz cell is approached. NUMERICAL OCEAN MODELLING IN THE BCLME The numerical modelling of the oceanic properties is a central aspect for oceanic forecasting in the BCLME. During the last 25 years, several models have been applied

58


to the Benguela Current System. While a number of these models concentrate uniquely on a limited portion of the ocean around southern Africa, others include ocean basins such as the South Atlantic or the Indian Ocean. With the tremendous increase in power of the supercomputers, Ocean General Circulation Models (OGCMs) might now be able to possess sufficient resolution to resolve the major processes in the Benguela (see the data and modelling animations in folder B of the CD ROM). Coastal models Van Foreest and Brundrit (1982) designed the first model for the South African west coast. The originality of their approach lay in the decomposition of the equations of motion into two vertical modes. The model domain extended from 70 km south of Cape Peninsula to North of St Helena Bay and to more than 150 km offshore. Open boundaries were applied at the connection with the open ocean. Although the model was forced with a constant wind and the duration of the simulation was brief (3 days), the solution showed some interesting spatial variability. The ocean modelling group at the CSIR is currently developing a high resolution model for the circulation in St Helena Bay (Monteiro and Kemp, personal communication). They use the Delft3D-FLOW ocean model with a variable grid resolution that ranges from a few kilometres offshore, to a few hundreds meters close to the coastline. The vertical grid is decomposed into 8 sigma layers and the model is forced by real time winds calibrated by a coastal weather station. This model is expected to resolve the coastal poleward flow during upwelling relaxations. To understand the retention of fish larvae in St Helena Bay, an idealized barotropic model (Penven et al. 2000) was designed and implemented during the VIBES-IDYLE project. The shallow water equations were solved on a 5 km resolution grid in a periodic channel forced by a constant alongshore wind. In the lee of Cape Columbine, the model produced a cyclonic recirculation that is able to retain biological elements. To extend the analysis to the different oceanic processes which might affect pelagic fish recruitment along the South Africa West Coast, a 3D regional configuration based on the Regional Ocean Modelling System (ROMS) has been implemented by Penven et al. (2001a). The model grid followed the coastline from Cape St Francis, 100 km west of Port Elizabeth to Lüderitz (see Fig. 1-1 in Chapter 1). The horizontal resolution ranges from 9 km at the coast to 18 km offshore. On the vertical, 20 sigma levels are stretched to keep a sufficient resolution close to the surface. The information at the open boundaries is provided by a basin scale model, and the atmospheric forcing was derived from the comprehensive ocean and atmosphere dataset (COADS) climatology. Using this model configuration, Blanke et al. (2002) quantified the wind contribution to inter-annual SST variability. They found that while the west coast is affected by mesoscale activity, the wind appears to be the dominant driving for the variability over the Agulhas Bank. By coupling the physical model to an individual based model, Parada et al. (2003) quantified the influence of eggs floatability on the transport from the Agulhas Bank to St Helena Bay, while Mullon et al. (2002) tested the "obstinate nature" hypothesis for the selection of the spawning zone, and Hugget et al. (2003) examined the influence of different


59

environmental factors on the transport of fish eggs and larvae in the Southern Benguela. Larger scale models At a regional scale, Skogen (1999) adapted NORWECOM, a model based on the Princeton Ocean Model, to simulate the ocean around the whole south-western southern Africa (i.e. including the coasts of Angola, Namibia and South Africa). The resolution was 20 km, and the model was forced by the National Centre for Environmental Prediction (NCEP) winds, with a surface nudging of SST and no surface salinity flux. Below 500 m, a relaxation towards Levitus climatology prevented the solution from drifting numerically. The physical model has been coupled to a biogeochemical model and to an Individual Based Model to simulate the fate of sardine larvae in the Northern Benguela (Stenevik et al. 2003). Speich et al. (2004) have especially designed a model to explore the Agulhas Retroflection, and its influence on the BCLME. They used ROMS at 1/6° and 1/10° degree resolution, with 32 vertical levels, forced by a monthly wind climatology derived from QuickSCAT scatterometer data and OGCM data for the open boundaries. Their simulations show the sensitivity of the Agulhas Current to the bottom slope steepness and its variations. Basin scale and global models Barnier et al. (1998) performed one of the first basin scale experiments using a sigma coordinate model. They applied SPEM for the Southern Atlantic at 1.375° resolution. The model had 20 vertical sigma levels and was forced by the Hellerman and Rosenstein wind stress climatology. Although very coarse, this model was able to capture some of the large scale features in the Benguela region. Biastoch and Krauß (1999) took advantage of the curvilinear coordinate in MOM2 to design a model at coarse resolution over the South Atlantic and the South Indian Oceans, but with an increase of resolution in the South African waters (1/3°). The model was forced by ECMWF winds and used data from an OGCM for its lateral boundary conditions. From this simulation, Reason et al. (2003) derived a heat export into the Southern Atlantic at 20°E of 1 PW in winter and 0.7 PW in summer. The behaviour of the Agulhas retroflection has been extensively studied in the Fine Resolution Antarctic Model (FRAM) simulation (Lutjeharms and Webb, 1995). FRAM is based on the Bryan-Cox-Semtner ocean model, it encompasses the totality of the Southern Ocean from 24° S to the Antarctic at a resolution of 1/2° in longitude and 1/4° in latitude (i. e. approximately 27 km around 60°S). FRAM appeared to be able to reproduce several observed patterns of the Agulhas retroflection, but displayed too much regularity in the subsequent path of the Agulhas Rings into the Atlantic Ocean, compared with observations.

60


For the same period, simulation experiments of the basin scale circulation were also conducted by Florenchie and Veron (1998) with an eddy-resolving 1/6° quasigeostrophic model. By means of a nudging data assimilation procedure along satellite tracks, Topex/Poseidon and ERS1 altimeter measurements were introduced in the model to control the simulation. The assimilation procedure enabled to produce schematic diagrams of the circulation in which patterns ranging from basin-scale currents to mesoscale eddies were portrayed in a realistic way. Treguier et al. (2003) have analyzed the generation and the fate of cyclonic and anticyclonic eddies from the Agulhas retroflection in an eddy resolving simulation of the whole Atlantic Ocean (CLIPPER). The model employed is OPA, at 1/6° resolution, with 42 z-levels, and forced by ECMWF ERA15 data from 1979-1993. OGCMs now start to have sufficient resolution to be relevant for the Benguela region. The United States National Research Laboratory (NRL) Layered Ocean Model (NLOM) and the NRL Coastal Ocean Model (NCOM) are presently running globally in real-time at respectively 1/16° and 1/8° resolution (Rhodes et al. 2002). In Japan, the enormous computational power of the “earth simulator” made it possible to run a global simulation at a resolution of 1/10° (Masumoto 2004). The respective role for the transport of heat and salt of cyclonic and anticyclonic eddies that are generated in the Agulhas region has been quantified in a global simulation based on POCM (Parallel Ocean Circulation Model) at 1/4° resolution (Matano and Beier 2003). SCHEMATIC CIRCULATION DEDUCED FROM A NUMERICAL MODEL There is a dearth of observations in the northern BCLME. Therefore one of the BCLME projects has examined the output from a numerical ocean model such as ROMS (e.g. Fig 4-4) and CLIPPER. These model outputs could then be used as a cost effective method to test various hypotheses, and to guide the observational programme of monitoring the environment in this region of the BCLME. The CLIPPER numerical simulation model The most recent CLIPPER experiment is a simulation of the global Atlantic oceanic circulation (http://www.ifremer.fr/lpo/clipper/present.html) based on the OPA model (http://www.lodyc.jussieu.fr/opa/). From 1990-1992 (the period of the spin up), the model is forced by a windstress climatology based on the European Earth Resources Satellite (ERS) derived wind fields. For the period 1993-2000, the model is forced directly by the more realistic varying direct ERS wind field products. The European Centre for Medium Range Weather Forecasting (ECMWF) heat and freshwater fluxes are used in combination with the Reynolds sea surface temperature (SST) for the heat feedback term for the period 1990-2000. The model domain covers most of the Atlantic Ocean and extends from 60°S-60°N. The model output has been examined to determine aspects of the seasonal circulation from 0-30°S.


61

Surface layers (0 to 30m) The modelled shelf circulation appears to be dominated by a narrow coastal current flowing northward all year long from about 30°S to the ABFZ near 18°S. Its intensity is higher during austral summer (January-March) and lower in winter (June to August). The coastal area between 26-22°S is somewhat different: the northward flow is less intense and it exhibits a weaker seasonal cycle. In fact there is an abrupt change in the current field immediately north of Lüderitz (28°S). From 30-26°S, the maximum velocity in the core of the current remains constant, with a value of about 25 cm s-1 in summer. At 26°S, the current speed maximum decreases abruptly to values of about 15 cm s-1 and the current intensity fluctuates as far southwards as 22°S. North of 22°S, the current speed increases once again. Figure 4-5 is a schematic representation of the model circulation at the surface (Fig. 4-5a – Lev 01) and at a 40m depth (Fig. 4-5b – Lev 04): Figure 4-5a is representative of the perennial modelled coastal circulation. However the northward coastal current (1 and 3 on the figure) is weaker during austral winter.

Figure 4-5. Schematic circulation reproduced by the CLIPPER model at the surface (a) and at 40 m (b).

The current is always stronger south of Lüderitz (branch 1). Then it bifurcates partly westward (branches 2) leading to a decrease of its transport and intensity north of 26°S. The orientation of the coast also changes near Lüderitz from northwest to north. Windstress in the area shows a regular northwestward direction all year long and it is stronger in summer south of 26°S. As a result, the surface discontinuity observed in the coastal current at 26°S might originate from the orientation of the coast, the wind field strength and its direction. The westward circulation in (4) does not seem to interact with the coastal circulation pattern. Circulation at 40m depth (lev04) The model circulation at 40m is globally identical to the surface, although weaker (Figure 4-5b). The maximum current speed in branch 1 is about 20 cm s-1 in summer. The northward coastal current experiences a similar seasonal cycle with minimum

62


intensity during austral winter. The discontinuity at 26°S is still present. The dotted line of branch 3 means that the current is not always clearly defined. The main change in Figure 4-5b concerns a new seasonal pattern in the circulation that occurs twice a year in February-March and October: a poleward current centred at 14°E develops off shore in the north and reaches a latitude of about 25-26°S at its maximum in October (branch 4). The dotted arrow indicates that the current is not permanent throughout the year. Its temperature is about 4°C higher than the coastal current with slightly higher salinity levels. It does not seem to interact much with the northward current. Circulation at 80m and 130m depths (lev07 and lev10) Figure 4-6 represents diagrams based on the model circulation at 80m and 130m depths for the same area. Dotted arrows indicate that the current is intermittent and shows some seasonal variability. It is at this model level in which the main major differences occur when a comparison is made with the surface layers. The northward coastal current is much weaker with maximum speeds of about 10 cm s-1 and it is not as clearly identifiable. It still shows a seasonal cycle but its intensity is higher in July and August, instead of summer. Branches feeding the current south of Lüderitz (1) are unstable and not well defined. Despite this, the northward current still reaches its maximum intensity in the Lüderitz area. The southward current (branch 4) intensifies in comparison with upper levels and is now noticeable from September-April. It exhibits two maxima: one in October and another in March, respectively. On these occasions the current meets the westward branch (2) of the coastal current near 26°S. It is more saline than the surrounding water and its temperature is about 2°C higher than the coastal water temperature. The circulation at model levels 08 through 10 (Figure 4-6b) reproduces the circulation patterns encountered at levels 04 and 07 with a marked seasonal shift. The cycle divides the whole area in two separate domains; September-March, the circulation is dominated by a southward flow north of Lüderitz (current 3). This flow develops along the coast as well and the northward current (branch 2) disappears. The southernmost extent of this flow occurs in February (26°S) and in October (27.5°S) with maximum speeds of about 10 cm s-1. From April to August, the situation is somewhat reversed. There is no more poleward flow. The cold northward current intensifies in the south (branch 1) with speeds of about 5 cm s-1. It reaches the Lüderitz area and its intensity north of 26°S remains very weak. At this depth, the Lüderitz area displays a natural border between two opposing seasonal regimes, a northern one associated with warmer and more saline waters flowing southward from October-March, a southern one concerning cold and fresher waters flowing northward from April/May-August. The cold pool shown on Figure 46b underlines the fact that along the coast the Lüderitz area constitutes a transition between the warm and the cold regimes. Temperature in the cold pool is about 1011°C whereas it is about 13°C north of 26°S. The cold pool appears to be a permanent feature with quite regular shape and size all year long.


63

Figure 4-6. Schematic circulation reproduced by the CLIPPER model at 80m (a) and at 130 m (b) depths.

Figure 4-7. Schematic circulation reproduced by the CLIPPER model at 230m (a) and at 350 m (b) depths.

Figure 4-8. Schematic circulation reproduced by the CLIPPER model at 575m depth.

64


Circulation at 230m and 350m depths (lev13 and lev15) At deeper model levels, the circulation along the shelf is poleward almost all year long with maxima occurring in February and October. Speeds are of the order of a few centimetres per second. The two diagrams (Figure 4-7) illustrate the model circulation at 230m and 350m depths. The poleward current brings warmer waters south of the Lüderitz area until 28°S where it reaches the cold pool. At this depth this permanent feature is less developed and its temperature is about 2°C lower than the poleward flow temperature. The northward current along the shelf (branch 1) develops in winter with very low speeds. Once again the current represented by the branch 3 does not seem to interact with it. Circulation at 570m depth (lev17) The circulation is dominated by the permanent northwestward flow associated with relatively warmer water masses (Figure 4-8, red arrow). Along the coast, a poleward coastal current develops from July to September during the winter period. Its maximum southward extent occurs in August. It meets the northward dominant flow near 30 and retroflects northward. In terms of temperature the whole area can be divided in two persistent parts; north of the warm current (red arrow) the water is about 1.5°C colder compared to the south (respectively 5.5°C and 7°C). The temperature variability throughout the year is small.

DISCUSSION AND CONCLUSIONS Processes with forecasting potential (see also Monteiro and Van Der Plas , this Volume – Chapter 5) Good progress has been made recently in the study of the Benguela Current Large Marine Ecosystem, as evidenced by the reviews in part two of this volume. In particular, considerable effort has been invested in trying to understand the mechanisms underlying the formation and evolution Benguela Niños (Florenchie et al. 2003; Florenchie et al. 2004). From our present understanding, by using appropriate observing systems in the equatorial region, it may be possible to get a forecast lead time of about two months for major warm events arriving at, and progressing polewards beyond the Angola Benguela Frontal zone. It is expected that aspects of the large scale variability of the BCLME are likely to be amenable to near real time observation and/or short term forecast. The most likely processes that have been identified to have potential in an early warning/forecasting system in the BCLME are set out in Table 4-1. The table is divided into an area (Domain), the most important forcing component, the main type of forcing process responsible for the variability, the approximate time scale of variability, the potential for being able to observe the phenomenon in near real time, and the subjective forecast/early warning potential with present limited capacity and resources. A three point scale: poor, fair and good is used. The scheme notes that anomalous signals propagate both from the equatorial Atlantic Ocean and into the northern BCLME, and


65

from the Agulhas Current in the Indian Ocean, into the southern BCLME. The discussion starts with the remote wind forcing in the western equatorial Atlantic Ocean, and its likely effects on the northern part of the BCLME (Angola and Namibia), via warm and cool anomalous signals propagating southwards along the Angolan coast. Severe warm SST anomalies at or south of the ABFZ are classed as Benguela Niños, the last well documented one occurring in 1995 (e.g. Florenchie et al. 2003). The main variability influence on the southern BCLME is from Agulhas Current ring shedding, early retroflection and intrusions of subantarctic cold water. The most difficult processes to forecast are the local influences on the upwelling centres at relatively short time scales of days-months. It is vital for the sustainable management of the BCLME, that extreme events (e.g. Roy et al. 2001) are recognised and understood, and if possible, forecast with a reasonable lead time. A regular state of the environment (SOE) reporting system, together with better communication for the BCLME would improve BCLME management advice. NUMERICAL MODELLING OF THE PHYSICAL PROCESSES IN THE BCLME In the past five years, there has been a sharp increase in the hydrodynamic modelling of the southern Benguela Upwelling ecosystem by implementing the 3-D ROMS numerical code and using seasonal wind forcing, (Penven et al. 2001a; Penven et al., 2001b) and then by refining the wind forcing with realistic winds from ERS (Blanke et al. 2002). The influence of the Agulhas Current shear edge instabilities on the southern border of the BCLME has been partially addressed (Lutjeharms et al. 2003). A ten-year model run with a time resolution of two days, and variable horizontal grid spacing from 9-18 km has provided the community with output for use of a number of individual based model (IBM) configurations (Field and Shillington 2005). With the advent of the BCLME, a dedicated group is presently modelling both the large scale influences on the BCLME, and using a nested approach to gain a better understanding of the local variability. Good synergy is maintained between the BCLME project and the IRD Upwelling Ecosystems project which is undertaking a comparative study of the Benguela, Canary and Humboldt Upwelling Systems. Project: SAfE (Southern Africa Experiment) Around the Southern African coasts, several different questions can be posed to the numerical ocean modeller. For example, how do the Benguela Niños propagate into the BCLME? Or, what is the role of Mozambique channel eddies in the shedding of Agulhas rings into the Atlantic Ocean, and their subsequent interaction with the BCLME? Or, why is there a cool ridge on the Agulhas Bank? To address each of these questions, the modeller needs a high degree of spatial resolution in the model region of interest, as well as a correct representation of the large scale ocean dynamics. To do this, a modelling platform under the auspices of the BCLME has been set up for

66


Table 4-1. Processes that are likely to have a cost effective observational capacity (satellite remote sensing or large scale in situ measurements). The forecasting potential is judged on a subjective scale of poor, fair and good. The forecasting potential of the large scale BCLME variability depends mainly on how well the linkages and processes that transfer the equatorial signals, and those from the Agulhas Retroflection, to the Benguela are understood.

Domain

Remote

Forcing System

Eastern Tropical South Atlantic

Processes

Scales of variability

Observing Potential

Forecast potential

Equatorial upwelling

Seasonal interannual

-

Good: Altimetry, Ocean Colour

Good

Intensity and timing of trade winds

Seasonal Interannual

-

Quikscat, PIRATA, GCM

Good

Equatorial stratification

Interannual – decadal (Benguela Niño)

Good: Ocean Buoys

Fair

Angola Current


Fair: Altimetry and AVHRR

Poor

AngolaBenguela Frontal Zone

Twice annual

Good: SST, colour

Good

Good: Altimetry, SST

Good

Fair: Ocean Buoy

Fair

-

Remote

Agulhas Retroflection

Ring shedding

Episodic: few times per annum

Local

Upwelling centres

Benguela Poleward transport


Upwelling wind variability

Days - weeks

Good: wind forecasts

Fair

Relaxation events in the southern Benguela

Days - weeks

Good: SST, colour

Fair

-

the simulation of the ocean around Southern Africa (SAfE: Southern Africa Experiment). The model is based on ROMS and takes advantage of its nesting capabilities. The parent grid includes the ocean around Southern Africa at a reasonable resolution (i.e. ~20-25 km). Several levels of child grids can be embedded into the parent grid, to reach locally a resolution of a few kilometers to a few hundred meters (see for example the grid set up for the ABFZ in Figure 4-4. The Parent model is inexpensive to run: 30 hours of computing for 1 year of simulation on a PC


67

workstation. Hence, it is possible to rapidly test new configurations and developments. Once the parent solution is satisfactory in its representation of the large scale solution, the presently one way nested high resolution “child” model configurations are added to provide the fine scale information. These simulations will be coupled to biogeochemical models (Monteiro 2005, pers. Com) in one of the BCLME projects. ACKNOWLEDGEMENTS The first author (FAS) acknowledges ongoing support funding from the NRF and UCT, while PF acknowledges funding from the BCLME project. PP has been seconded by the IRD (France) to Cape Town for the period 2004-2006. REFERENCES Ajao, E.A. and R.W. Houghton. 1998. Coastal ocean of Equatorial West Africa from 10°N to 10°S. 605631 in Robinson, A.R. and K.H. Brink, eds. The Sea, Vol. 11, The global coastal ocean, regional studies and syntheses Wiley, New-York. Barnier, B., P. Marchesiello, A.P. De Miranda, J.M. Molines, and M. Coulibaly. 1998. A sigmacoordinate primitive equation model for studying the circulation in the south Atlantic. Part I: Model configuration with error estimates.Deep-Sea Res. Part I, 45:543-572. Biastoch, A. and W. Krauß. 1999. The role of mesoscale eddies in the source regions of the Agulhas Current, J. Phys. Oceanogr. 29:2303-2317. Blanke, B., C. Roy, P. Penven, S. Speich, J. McWilliams, and G. Nelson. 2002. Linking wind and upwelling interannual variability in a regional model of the southern Benguela. Geophys. Res. Lett., 29, 2188, doi :10.1029/2002GL015718. Boyd, A.J. and F.A. Shillington. 1994. The Agulhas Bank: A review of the physical processes. S. Afr. J. Sci. 90:114-122. Boyer, D., J. Cole , and C. Bartholomae. 2000. Southwestern Africa: Northern Benguela Current Region. In Sheppard, C.R.C., ed. Seas at the Millenium: An environmental evaluation, Vol 1, Elsevier Science Ltd. 821-840. Brundrit, G. B., B.A. De Cuevas, and A.M. Shipley. 1987. Long-term sea-level variability in the eastern south Atlantic and comparison with that in the eastern Pacific, S. Afr. J. Mar. Sci. 5:73–78. Chapman, P. and L.V. Shannon. 1985. The Benguela Ecosystem Part II. Chemistry and related processes. Oceanogr. Mar. Biol. Ann. Rev. 23:183-251. Chelton, D.B., M.G. Schlax, M.H. Freilich, and R.F. Milliff. 2004. Satellite Measurements Reveal Persistent Short-Scale Features in Ocean Winds. Science, 303, Issue 5660: 978-983. Colberg, F., C.J.C. Reason, and K. Rodgers. 2004. South Atlantic Response to El Nino-Southern Oscillation induced Climate Variability in an Ocean General Circu;lation Model. J. Geophys. Res.,. 109, C12015, doi:10.1029/2004JC002301. Demarcq, H., R. Barlow, and F. A. Shillington. 2003. Climatology and variability of sea surface temperature and surface chlorophyll in the Benguela and Agulhas ecosystems as observed by satellite imagery. Afr. J. mar. Sci. 25: 363-372. Duncombe Rae, C.M., F.A. Shillington, J.J. Agenbag, J. Taunton-Clark and M. L. Grundlingh. 1992. An Agulhas ring in the South Atlantic Ocean and its interaction with the Benguela upwelling frontal system. Deep-Sea Research 39, 2009-2027. Duncombe Rae, C.M. 1998. Antarctic Intermediate and Central Waters in the Angola-Benguela Front region: results from the first BENEFIT cruise, April 1997. In: International Symposium on Environmental Variability in the South-East Atlantic, 30 March to 1 April 1998, Swakopmund, Namibia. p.14. Duncombe Rae, C.M. 2005. A demonstration of the hydrographic partition of the Benguela upwelling ecosystem at 26 40'.S .African Journal of Marine Science 27(3): 617-628.

68


Fauchereau, N., S. Trzaska, M. Rouault, and Y. Richard. 2003. Rainfall variability and changes in southern Africa during the 20th century in the global warming context. Natural Hazards 29: 139-154. Field, J.G. and F.A. Shillington. 2005. Variability of the Benguela Current System. 833-860 In Robinson, A.R. and K.H. Brink, eds. The Sea, Vol. 14, The global coastal ocean, Interdisciplinary regional studies and syntheses. Harvard University Press. Florenchie, P. and J. Verron. 1998. South Atlantic Ocean circulation: Simulation experiments with a quasi-geostrophic model and assimilation of Topex/Poseidon and ERS1 altimeter data, J. Geo. Res., Vol 103, NO. C11, 24737-24758. Florenchie, P., C.J.C. Reason, J.R.E. Lutjeharms, M. Rouault, C. Roy, and S. Masson. 2004. Evolution of Interannual Warm and Cold Events in the Southeast Atlantic Ocean. J. Climate 17: 2318-2334. Florenchie, P., J.R.E. Lutjeharms, C.J.C. Reason, S. Masson, and M. Rouault. 2003. The source of Benguela Niños in the South Atlantic Ocean, Geophys. Res. Lett. 30, (10), 1505, doi:10.1029/2003GL017172. Gammelsrød, T., C.H. Batholomae, D.C. Boyer, V.L.L. Filipe, and M.J.O’Toole. 1998. Intrusion of warm surface water along the Angolan-Namibian coast in February – March 1995: The 1995 Benguela Niño, S. Afr. J. Mar. Sci. 19: 51– 56. Gründlingh, M.L. 1985. Occurrence of Red Sea water in the south western Indian Ocean, 1981. J. phys. Oceanog. 15(2): 207-212. Hardman-Mountford, N.J., A.J. Richardson, J.J Agenbag, E. Hagen, L. Nykjaer, F.A. Shillington and C. Villacastin. 2003. Ocean Climate of the South East Atlantic observed from satellite data and wind models. Progress in Oceanography 59: 181-221. Hermes, J.C., and C.J.C. Reason. 2005. Ocean model diagnosis of interannual co-evolving SST variability in the South Indian and Atlantic Oceans. J. Climate 18: 2864-2882. Hill, A.E., B.M. Hickey, F.A. Shillington, P.T. Strub, K.H. Brink, E.D. Barton and A.C. Thomas. 1998. Eastern Ocean Boundaries. 29-68 in Robinson A.R. and K.H. Brink, eds. The Sea, Vol 11,The Global Coastal Ocean, Regional Studies and Syntheses. John Wiley and Sons, New York. Huggett, J., P. Freon, C. Mullon, and Penven. 2003. Modelling the transport success of anchovy (Engraulis encrasicolus) eggs and larvae in the southern Benguela: The effect of spatio-temporal spawning patterns. Marine Ecology Progress Series 250: 247-262. Hutchings, L., L.E. Beckley, M.H. Griffiths, M.J. Roberts, S. Sundby and C. van der Lingen. 2002. Spawning on the edge: spawning grounds and nursery areas around the southern African coastline. Mar. Freshwater Res. 53: 307–318. Jones, P.D. and R.J. Allan. 1998: Climate change and long-term climatic variability. In Karoly, D. and D. Vintcent, eds. Meteorology of the Southern Hemisphere. Amer. Meteorol. Soc., Boston, Massachussetts. 337-363. Kearns, E.J. and M-E. Carr. 2003. Seasonal climatologies of nutrients and hydrographicproperties on quasi-neutral surfaces for four coastal upwelling systems. Deep-Sea Research, II, 50, 3171-3197. Kidson, J.W. 1988. Interannual variations in the Southern Hemisphere circulation. J. Climate 1:11771198. Lutjeharms J.R.E., P. Penven and C. Roy. 2003. Modelling the shear edge eddies of the southern Agulhas Current. Continental Shelf Research 23:1099-1115. Lutjeharms, J.R.E., and D.J. Webb. 1995. Modelling the Agulhas Current system with FRAM (Fine Resolution Antarctic Model), Deep Sea Res. Part I, 42: 23-551. Masumoto, Y. 2004. Generation of Small Meanders of the Kuroshio South of Kyushu in a HighResolution Ocean General Circulation Model. Journal of Oceanography 60:13-320. Matano, R.P. and E.J. Beier. 2003. A kinematic analysis of the Indian/Atlantic interocean exchange, Deep-Sea Res. Part II, 50:229-249. Mercier, H., M. Arhan and J.R.E. Lutjeharms. 2003. Upper-layer circulation in the eastern Equatorial and South Atlantic Ocean in January-March 1995. Deep-Sea Res. I 50: 863-887. Mo, K.C. and J.N. Paegle. 2001. The Pacific-South American modes and their downstream effects. Int. J. Climatology 21(10): 1211-1229. Mohrholz, V., M. Schmidt, and J.R.E. Lutjeharms. 2001. The hydrography and dynamics of the AngolaBenguela Frontal Zone and environment in April 1999. S. Afr. J. Sci. 97:199-208. Monteiro, P.M.S. 1996. The oceanography, the biogeochemistry and the fluxes of carbon dioxide in the Benguela upwelling system. Ph.D. thesis, Univ.Cape Town, S. Afr, 354pp. Mullon, C., P. Cury, and P. Penven. 2002. Evolutionary individual-based model for the recruitment of the anchovy in the southern Benguela. Can. J. Fish. Aquat. Sci. 59: 910-922.


69

Parada, C., C.D. Van der Lingen, C. Mullon and P. Penven. 2003. Modelling the effect of buoyancy on the transport of anchovy (Engraulis capensis) eggs from spawning to nursery grounds in the southern Benguela: An IBM approach. Fish. Oceanogr. 12:170-184. Penven, P., C. Roy, A. Colin de Verdiere and J. Largier. 2000. Simulation and quantification of a coastal jet retention process using a barotropic model. Oceanol. Acta 23:615-634. Penven, P., C. Roy, G.B. Brundrit, A. Colin de Verdière, P. Fréon, A.S. Johnson, J.R.E. Lutjeharms, and F.A. Shillington. 2001a. A regional hydrodynamic model of the Southern Benguela upwelling. S. Afr. J. Sci. 97: 472-475. Penven, P, J.R.E. Lutjeharms, P. Marchesiello, C. Roy, and S.J. Weeks. 2001b. Generation of cyclonic eddies by the Agulhas Current in the lee of the Agulhas Bank. Geophys. Res.Lett. 27:1055-1058. Peterson, R.G. and L. Stramma. 1991: Upper-level circulation in the South Atlantic Ocean. Progress in Oceanography 26, 1-73. Poole, R. and M. Tomczak. 1999. Optimum multiparameter analysis ofthe water mass structure in the Atlantic Ocean thermocline. Deep-Sea Res. Part I, 46:1895-1921. Reason, C.J.C., J.R.E. Lutjeharms, J. Hermes, A. Biastoch. 2003. Inter-ocean fluxes south of Africa in an eddy-permitting model, Deep-Sea Research Part II, 50:281-298. Reason, C.J.C. and M.R. Jury. 1990. On the generation and propagation of the southern African coastal low. Quarterly Journal of the Royal Meteorological Society, vol. 116 (495): 1133-1151. Reason, C.J.C. 2000. Multidecadal climate variability in the subtropics/midlatitudes of the Southern Hemisphere oceans. Tellus, 52A: 203-223. Reid, J.L. 1989. On the total geostrophic circulation of the South Atlantic Ocean: Flow patterns, tracers, and transports. Prog. Oceanog. 23: 149-244. Rhodes, R.C., H.E. Hurlburt, A.J. Wallcraft, C.N. Barron, P.J. Martin, E.J. Metzger, J.F. Shriver, D.S. Ko, O.M. Smedstad, S.L. Cross and A.B. Kara. 2002. Navy real-time global modeling systems. Oceanogr. 15: 29-43. Risien, C.M., C.J.C Reason, F.A Shillington, and D.B. Chelton. 2004. Variability in satellite winds over the Benguela upwelling system during 1999–2000. J. Geophys. Res. 109: C3, C0301010.1029/2003JC001880. Rouault, M., P. Florenchie, N. Fauchereau and C.J.C. Reason. 2003. South East Atlantic warm events and southern African rainfall. Geophys. Res. Lett. 30 (5): 8009, doi:10.1029/2002GL014840. Roy, C., S.J. Weeks, M. Rouault, G. Nelson, R. Barlow, and C.D. van der Lingen. 2001. Extreme oceanographic events recorded in the southern Benguela during the 1999-2000 summer season. S. Afr. J. Sci. 97:465-471. Schumann, E.H. and K.H. Brink. 1990. Coastal-Trapped Waves off the Coast of South Africa: Generation, Propagation and Current Structures. Journal of Physical Oceanography 20(8): 1206–1218. Schumann, E.H. 1998. The coastal ocean off southeast Africa, including Madagascar. 557-581 in Robinson, A.R. and K.H. Brink, eds. The Sea, Vol. 11, The global coastal ocean, regional studies and syntheses. Wiley, New-York. Shannon, L.V, A.J. Boyd, G.B. Brundrit, and J. Taunton-Clark. 1986. On the existence of an El-Niño type phenomenon in the Benguela system. J. Mar. Res. 44(3): 495-520. Shannon, L.V. and G. Nelson. 1996. The Benguela: Large scale features and processes and system variability. 163-210 in Wefer, G. W.H. Berger, G. Siedler, and D.J. Webb, editors. The South Atlantic Past and Present Circulation. Springer Verlag, Berlin, Heidelberg. Shannon, L.V., and M.J. O’Toole. 2003. Sustainability of the Benguela: ex Africa semper aliquid novi. In: K. Sherman and G. Hempel, Large Marine Ecosystems of the World – Trends in Exploitation, Protection and Research. Elsevier B.V. 227-253. Shannon, L.V. and D. Hunter. 1988. Notes on Antarctic Intermediate Water around southern Africa. S. Afr. J. mar. Sci. 6:107-117. Shannon, L.V., J.J. Agenbag, M.E.L. Buys., 1987. Large and mesoscale features of the Angola-Benguela front. In: Payne, A.I.L., J.A. Gulland and K.H. Brink, eds. The Benguela and Comparable Ecosystems, S. Afr. J. mar. Sci. 5:11-34. Shillington, F.A. 1998. The Benguela upwelling system off southwestern Africa. 583-604 in Robinson , A.R.and K.H. Brink. The Sea, Vol. 11, The Global Coastal Ocean, Regional Studies and Syntheses. Wiley, New-York. Skogen, M.D. 1999. A biophysical model applied to the Benguela upwelling system. S. Afr. J.. Mar. Sci. 21: 235-249.

70


Speich, S., P. Penven and B. Blanke. 2004. On the Cape Cauldron dynamics: some physical insights on the turbulent Indo-Atlantic exchange and impact of Agulhas waters in the Southern Africa upwelling region from a hierarchy of regional numerical simulations. European Geosciences Union 2004, Geophysical Research Abstracts. 6:4815. Stenevik, E.K., M. Skogen, S. Sundby, and D. Boyer. 2003. The effect of vertical and horizontal distribution on retention of sardine (Sardinops sagax) larvae in the Northern Benguela - observations and modelling. Fish. Oceanogr. 12(3): 185-200. Talley, L.D. 1996. Antarctic Intermediate Water in the South Atlantic. In G. Wefer, W.H.Berger, G. Siedler, and D.J. Webb, eds. The South Atlantic: Present and Past Circulation. Berlin Heidelberg: Springer-Verlag, pp. 219-238. Treguier, A.M., O. Boebel, B. Barnier, and G. Madec. 2003. Agulhas eddy fluxes in a 1/6 degree Atlantic model. Deep-Sea Res., Part II, 50: 251-280. Van Foreest, D. and G.B. Brundrit. 1982. A two mode numerical model with application to coastal upwelling, Prog. Oceanogr. 11:329-392. van Loon, H. 1967. The half-yearly oscillation in middle and high southern latitudes and the coreless winter. J. Atmos. Sci. 24: 472-486. Veitch, J.A., P. Florenchie and F.A. Shillington. 2006. Seasonal and interannual fluctuations of the Angola Benguela Frontal Zone (ABFZ) using 4.5 km resolution satellite imagery from 1982 to 1999. International Journal of Remote Sensing 27: 989-1000. Venegas, S.A., L.A. Mysak, and D.N. Straub. 1996. Evidence for interannual and interdecadal climate variability in the South Atlantic. Geophysical Research Letters 23(19): 2673-2676. Venegas, S.A, L.A. Mysak and D.N. Straub. 1997. Atmosphere-ocean coupled variability in the South Atlantic. J. Climate 10: 2904-2920. Venegas, S.A., L.A. Mysak and D.N. Straub. 1998. An interdecadal climate cycle in the South Atlantic and its links to other ocean basins. J. Geophys. Res. 103, No. C11, 24,723-24,736.


71

5 Low Oxygen Water (LOW) Variability in the Benguela System: Key Processes and Forcing Scales Relevant to Forecasting Pedro M.S. Monteiro and Anja K. van der Plas INTRODUCTION Low oxygen water (LOW) is an endemic characteristic of the Benguela system (Chapman and Shannon 1985; Bailey 1991; Monteiro et al. 2004). The ecological impacts of LOW were identified in the early research work undertaken in the system (Copenhagen 1953; Pieterse and van der Post 1967) and its close association to the incidence of elevated sulphide concentrations was also noted in a qualitative sense (Marchand 1928; Copenhagen 1953; Hart and Currie 1960; Pieterse and van der Post 1967). Events that resulted in significant losses of both demersal and bottom species have occurred in both the central (Namibia) and southern (South Africa) Benguela system (see case study). At present, stock assessment models treat environmental factors as random sources of mortality that can be parameterised by a mortality factor - this assumes that there are no systematic shifts in the forcing and response to LOW. Similarly, ecosystem models are typically less than sensitive to environmental forcing which can impact fisheries and ecosystem behaviour, distribution and mortality (Shannon and Jarre-Teichmann 1999). The most recent time series data analysis supports the view that LOW variability is characterised by regime shifts in both remote forcing and local forcing factors which interact non-linearly to create LOW conditions and events of hitherto unpredicted magnitude (Monteiro et al. 2004). The BCLME Transboundary Diagnostic Analysis (TDA) identified LOW as one of the key environmental factors governing the variability and commercial viability of fisheries and ultimately the ecosystem (www.bclme.org). Its implementation plan requires that not only should the causes of LOW variability be understood but the BCLME should also invest in developing a forecasting capability which could assist the optimal ecosystem management, anticipate its impacts, provide better understanding of the underlying complexity and support fisheries management. The forecasting goal for LOW in the Benguela requires that the processes and the forcing scales that drive events and their variability be better characterized and understood.

72

P.M.S. Monteiro and A.K. van der Plas

The impact of LOW variability on hake fisheries: Namibia 1992 - 1994

Hake recruit mortalities off Namibia in 1992/3 (Woodhead et al. 1997a) and the hake recruitment failure off Walvis Bay in 1994 (Hamukuaya et al. 1998; Woodhead et al. 1997b) are examples of the decimating effect that LOW can have on the marine resources of the Benguela ecosystem. Although Cape hake Merluccius capensis have adapted both behaviourally and physiologically to tolerate hypoxic conditions to a degree (Woodhead et al. 1998) the severity and the prolonged duration of the hypoxic conditions over the central Benguela continental shelf between 1992 and 1994 is thought to have led to mass mortalities of the hake recruits during 1993 and 1994. In austral summer of 1992 to 1993 the juvenile hake were thought to have been trapped by the expansion of hypoxic conditions leading to loss of half the recruits (Woodhead et al. 1997a). During 1994 the juvenile Cape hake that did not succumb to the oxygendepleted waters sought to avoid the LOW offshore but cannibalism by the adults that frequent the deeper shelf waters as well as discarding by trawlers targeting these adults are thought to have lead to a recruit mortality of 70-84% (Hamukuaya et al. 1998; Woodhead et al. 1997b). Thus, at certain scales, LOW variability affects the abundance, distribution, availability and catchability of commercially fished stocks through modification of both behavioural and mortality responses. The non-random character of LOW impacts on fisheries also challenges the assumptions in fish stock assessment models of the relationship between mortality and environmental variability.

SYNTHESIS OF SYSTEM PROCESSES AND VARIABILITY The importance of an advection link between the tropical eastern Atlantic low oxygen reservoir and the Benguela was proposed by Moroshkin et al. (1970), and Bubnov (1972) who also suggested that the oceanic LOW reservoir was generated by productivity associated with the Angola Dome. This latter idea was challenged by the work of Voituriez and Herbland (1982) that pointed to the equatorial upwelling zone as the main source of production, which is consistent with remote sensing data (Monteiro and van der Plas 2004). The parallel models of remote forcing and local biological production as the main drivers for low oxygen variability were reviewed in detail by Chapman and Shannon 1985. Monteiro et al. (2004) have recently suggested that low oxygen variability in the Benguela system is forced by the interaction on varying scales of both large- (basin) and local- (shelf) scale forcing. The processes at these two scales form the core of the discussion. Furthermore, LOW variability in the Benguela system can be further divided into three physically characterised regimes: Northern (Angola): LOW variability is completely advection controlled and tightly coupled to upwelling that peaks in June – August.

Benguela low oxygen variability

73

Central (Namibia) LOW variability is governed by a complex interaction between the remotely forced shelf boundary conditions, seasonal thermocline variability and biogeochemical carbon fluxes. Southern (South Africa) LOW variability is largely driven by local seasonal wind characteristics and minimal remote forcing. This synthesis is based on the following two foundation papers: a review of BCLME LOW formation assessing the importance of both remote and local forcing for the Benguela region (Monteiro et al. 2004) and a paper on sediment vs. water column hypoxia coupling (van der Plas et al. 2005). As neither of these papers is yet published, the essence of the thinking in both is reflected in this synthesis. The physical oceanography focuses on processes that are directly relevant to the formation or advection of LOW. More general syntheses of large scale and shelf physical processes are provided elsewhere (Hardman-Mountford et al. 2003; Shillington et al. 2005). One of the key requirements of forecasting schemes is their ability to translate predicted LOW temporal and spatial characteristics into robust ecological risk categories. A revised set of categories using the most recent observational data is given in Table 5-1.

Table 5-1. Oxygen concentration thresholds that are of relevance to the linkages between predicted oxygen concentrations and their ecological consequences. These should be seen as guidelines to be interpreted more closely on a case by case basis because exposure times and frequencies are also relevant.

Oxygen State

Oxygen Concentrations

Impacts

Super Saturated

> 100% saturation

Out-gassing to the Atmosphere f (t,S): typical in high surface primary production

Saturated

100% Saturation

Equilibrium with the atmosphere f (t,S)

Under saturated

3 – 100% Saturation

Range over which biological responses should be insignificant

Depleted

2 - 3 ml l-1

Biological impacts felt at behavioural level

Critical Hypoxia

-1

1 – 2 ml l

Hypoxic

0.5 – 1 ml l-1

Anoxic

-1

< 0.5 ml l

Threshold that enables the system to go anoxic under a flux of bloom detritus. Organisms require physiological adaptation to survive Extreme stress and mortality in organisms. (denitrification) Respiration dominated by anaerobes and sulphide / methane fluxes

74


REMOTE FORCING: EASTERN TROPICAL SOUTHEAST ATLANTIC (ETSA – BENGUELA LINKAGE) (Further supporting information in CD-ROM: LOWCH5.htm)

The Eastern Tropical Southeast Atlantic (ETSA) region is recognised as the main reservoir of LOW in the region but its internal processes and its linkages to the Benguela are weakly understood (Chapman and Shannon, 1985; Voituriez and Herbland 1982). While the importance of temporal variability was understood early on (Chapman and Shannon 1987; Voituriez and Herbland 1982), limited progress has been made in understanding the processes that govern the scales of variability. The basin thermocline shallows eastwards and gets to within 50m of the surface in the ETSA zone, which is commonly referred to as the Angola gyre (Stramma and Schott 1999). It is here that occur the dominant processes of primary production, stratification and retention, which govern LOW formation, transport and ultimately the boundary conditions of the Benguela shelf. The following processes are essential to the formation and maintenance of the ETSA LOW reservoir: • The scale and variability of phytoplankton new production which provides the required electron donating capacity to the oxygen sink processes. • A thermocline that limits the downward flux of oxygen across the thermocline to below the biogeochemical uptake rate. • A retention zone that limits the rate of sub-thermocline ventilation by advected aerated water In order to understand the generation and change of LOW in the ETSA circulation zone it is essential to characterise the scales of spatial and temporal variability. The main features and flows of the Eastern Tropical South Atlantic (ETSA) – Benguela region that are relevant to LOW variability (Figure 5-1) are briefly noted below. The spatial and temporal characteristics of LOW oceanography in the tropical South Atlantic are governed by the cyclonic part of its circulation (Reid 1989). The core of LOW within the ETSA zone extends from the equatorial zone to two southern boundaries: one at 16 - 17°S and a second at 25 - 26°S (Figure 5-1). These two boundaries correspond to the southern edge of the Angola gyre and the southern edge of the sub-equatorial cyclonic circulation respectively. The sharp oxygen gradient across the latter boundary defines the transition between the two South Atlantic Central Water masses derived respectively from the hypoxic ETSA and the aerated Cape Basin (Figure 5-1). In the north, the equatorial divergence zone and its associated upwelled nutrient flux are driven by the seasonal easterly trade winds. This system supplies the main phytoplankton export production flux that creates the subthermocline oxygen demand within the Angola gyre. Although there are several intermittent divergent flow features here, such as the Angola dome, the only significant export flux of carbon is due to upwelling activity in the austral winter (Voituriez and Herbland 1982).


75

The current system in the region is fairly complex (Figure 5-1). The eastward flowing Equatorial Under Current (EUC) and South Equatorial Under Current (SEUC) converge, particularly during the seasonal (Dec - April) weakening in the trade winds, when the EUC intensifies to form the Guinea-Congo Under Current (GCUC; Stramma and Schott 1999). The combination of the GCUC and SEUC forms the southward flowing Angola Current (16 Sv), which is also the eastern boundary of the Angola gyre (Mohrholz et al. 2001; Mercier et al. 2003). The South Equatorial Counter Current (SECC) provides an additional inflow to the Angola Current during the austral winter.

Eastern Tropical South Atlantic System: ETSA

EUC

EDZ

GCUC

SEUC SECC

AC

sSEC

BPUC

Cape Basin SACW

Figure 5-1. A diagrammatic view of the main components of the Eastern Tropical South Atlantic System (ETSA) cyclonic circulation zone that are relevant to LOW variability in the Benguela. It shows the core cyclonic circulation, also known as the Angola gyre, supplied with three eastward flows, the Equatorial Under Current (EUC), South Equatorial UC (SEUC) and South Equatorial Counter Current (SECC). The eastern boundary comprises the seasonal Guinea-Congo UC (GCUC) (July – Sept), the southward coastal Angola Current (AC)(16 Sv), the Benguela southward extension as the Poleward Under Current (BPUC) (2 – 5 Sv) which defines the boundary conditions for the shelf upwelling system.

The Angola current splits into two flows, the main one (14 Sv) closing the Angola gyre while its southward extension becomes the Benguela Poleward Under Current (BPUC) along the Namibian shelf as far south as 27°S (Mercier et al. 2003). This southerly extension of ETSA-generated LOW establishes the boundary conditions for the northern and central Benguela system. The poleward undercurrent also feeds further south into the southern branch of the South Equatorial Current (SSEC), which

76


together with the Benguela Current, closes the basin scale cyclonic gyre of the South Atlantic thermocline waters (Stramma and England 1999). The BPUC also becomes the poleward undercurrent on the slope of the Benguela that acts as the main advective link for LOW between the ETSA and the Benguela systems. The closure of the Angola gyre by the main flow of the Angola current creates the recirculation retention zone that, together with its thermocline dynamics, establishes the conditions necessary for LOW formation. The main sources of ventilation for the sub-thermocline waters of the ETSA retention zone are the EUC, SEUC and SECC. These currents also may play an important role in the transport of the new production flux from the EDZ into the retention zone. Within the Angola gyre, the Angola dome is a seasonally transient feature with apparently only limited impact on low oxygen variability. Although previously thought to be the main source of divergent transport that supported phytoplankton production (Chapman and Shannon, 1985), its contribution to the overall oxygen demand is likely to be small compared to the upwelling at the equatorial divergence zone (EDZ). (Note that Chapman and Shannon did not consider the role of the EDZ in their paper.) At the southern end of the Angola gyre is a surface feature known as the AngolaBenguela Front (ABF). The spatial and intensity characteristics of the ABF are governed by the seasonal relaxation of the equatorial easterly winds in the late austral summer (Feb - April), which drives the eastward and southward propagation of warm surface water probably as a baroclinic Kelvin wave (Stramma and Schott 1999; Lass et al. 2000; Mohrholz et al. 2001). The relevance of this process to LOW variability in the Benguela is that the resulting intensification of the thermocline intensifies the poleward transport of LOW in the slope and on the shelf. Large perturbations of the Atlantic equatorial thermocline occur at approximately decadal intervals. These perturbations propagate eastward as a Kelvin wave and surface at the ABF. The effect is an anomalous warming of the surface layer, known as a Benguela Niño, that can then propagate onto the Namibian shelf (Florenchie et al. 2003). This process impacts LOW by intensifying the thermocline and increasing the poleward flow below it. Other associated but less well understood effects of the Benguela Niño include weakening of the equatorial thermocline and the EUC which impact on the ventilation of the ETSA and the ETSA Benguela linkage respectively. The Benguela Niño warming at the ABF should not be confused with the annual (late summer) warming that results from the seasonal relaxation of the equatorial easterly winds. These ETSA-derived features combine to drive the spatial and temporal characteristics of the LOW boundary conditions for the northern and central Benguela system, although not including the Lüderitz upwelling cell. The Lüderitz upwelling cell and the southern Benguela (28oS - 35oS) are defined by the SACW in the Cape Basin characterised by the well aerated boundary conditions driven from the sub-Polar domain.


77

BENGUELA SHELF VARIABILITY LOW variability in the Benguela (5oS – 35oS) has been separated into three recognizable domains according to the extent to which the variability is externally forced or locally generated. These are: • Northern Benguela: Congo – Angola sub-system • Central Benguela: Namibian sub-system • Southern Benguela: South African sub-system Whereas in the northern sub-system a narrow shelf results in a spatially extensive upwelling, in the central and southern sub-systems the slope – shelf link is at discrete sites also termed “gates” (Monteiro, 1996; Duncombe-Rae, 2004). Three main upwelling “gates” have been suggested to govern the slope – shelf exchange of SACW, Cape Frio (17 – 18°S), Lüderitz (25 – 26°S) and Oliphants Valley (33°S) (Monteiro 1996). Meridional and vertical shifts in the ETSA derived LOW core control the oxygen characteristics of upwelled water at the Lüderitz and the Cape Frio upwelling centres, the main slope – shelf exchange “gates” in the central Benguela system (Monteiro 1996; Duncombe-Rae 2004). Influxes of more oxygenated SACW at the Oliphants Valley zone will shift the southern Benguela shelf system away from hypoxic conditions even under intense upwelling derived new production fluxes. Northern Benguela: Congo – Angola sub-system The temporal variability of LOW in the narrow Congo – Angola shelf system (Figure 5-2) shows that it is strongly driven by the boundary conditions characteristic of the ETSA region along the shelf. The narrow shelf means that ETSA- LOW is upwelled from the slope onto the shelf along the entire coastal system and LOW seasonal variability is strongly correlated to temperature. This correlation indicates that the variability is governed by the advection of upwelled water rather than by any shelf domain processes. Seasonal variability in the northern sector of the system (Figure 52) shows that LOW intensifies in the 3rd and 4th quarter of the year linked to the intensification of the equatorial easterlies. Moreover, higher oxygen conditions of the system (Figure 5-2) are driven by downwelling and aeration linked to the southward advection of tropical warm water during the relaxation of the equatorial easterlies in the 1st quarter. For the remainder of the year the combined effects of the narrow shelf and proximity to the domed core of the ETSA LOW system mean that the oxygen concentrations are low and closely correlated to temperature (Figure 5-2). Central Benguela: Namibian sub-system LOW variability in the Central Benguela shelf is governed primarily by the boundary characteristics at the two main upwelling centres of Cape Frio and Lüderitz. The linkage between these boundaries and the ETSA is of key importance. We suggest that the relative contribution from these two sources of shelf water is strongly


78

Dep th (m)

Lobito (12°S, 110m to 150m bottom depth) 0 -50 -100 1995

a)

1996

1997

1998

1999

2000

2001

2002

2003

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 °C

De pth (m)

0 -50 -100 1996

1995

1996

b) Depth (m)

1997

1998

1999

2000

2001

2002

2003

2002

2003

36 PSU 35. 9 35 . 8 35. 7 35. 6 35. 5 35 . 4 35. 3 35. 2 35. 1 35 34 . 9 34 . 8 34. 7 34. 6 34 . 5 34 . 4 34. 3 34 . 2 34 . 1 34 33 . 8 33. 6 33 . 4 33. 2 33

1995

0 -50 -100

c)

1997

0

0.5

1998

1

2

1999

3

4

2000

5

6

2001

7 ml/l

Figure 5-2. Time series of temperature (a), salinity (b) and oxygen (c) variability on the Angolan shelf for the period 1994 – 2003. It highlights the strong relationship between the incidence of low oxygen waters (< 2mll-1) and cold upwelled water (< 16oC). Because of the narrow shelf the incidence of LOW is driven almost completely by the upwelling driven advection of ETSA LOW rather than any shelf based modification. In this part of the system oxygen behaves conservatively with temperature. Sampling periods and depths are indicated on the diagrams.

dependent on the characteristics and the poleward extent of the warm tropical surface water and the impact it has on the thermocline characteristics on the shelf. As stated above, the poleward extent of the warm tropical surface water governs the strength of the sub-thermocline poleward flow which regulates the spatial scale of the impact of the hypoxic waters upwelled at Cape Frio. Under conditions of weak stratification and south easterly wind stress, typical of the early upwelling season in the 3rd and 4th quarters of the year, the dominant flow on the Namibian shelf is equatorward, driven by the barotropic pressure gradient and a weak or non existent poleward flow on the shelf. When stratification intensifies, either as a result of


79

seasonal or interannual warm events, the sub-thermocline poleward flow strengthens due to the increasing forcing of the baroclinic pressure gradient. While the former condition favours a larger contribution of mostly aerated water (Cape Basin SACW) derived from the Lüderitz gate, the latter favours a greater magnitude from the hypoxic Cape Frio flux. Thus, we believe that the LOW environment on the Namibian shelf is modulated by the changing contributions of water from the two input fluxes driven indirectly by the strength of the warm water events. This dynamic is suggested to govern the magnitude of both the seasonal and the interannual LOW signal in this part of the system. Combining these ideas of stratification and shelf transport allows LOW variability over a 10 year period (1995 – 2004) within the central part of the Central Benguela to be better understood (Figure 5-3a-c). The time series of oxygen concentration at the outer shelf in the mid-Central Benguela (Figure 5-3) shows that the variability of the hypoxic water is driven by both the stratification as well as the LOW boundary conditions, with the strength of the stratification, which according to the model drives the poleward transport, modulating the boundary condition LOW signal on the shelf. In periods when the stratification weakens the hypoxic signal is also weakened because there is a greater contribution from water upwelled at Lüderitz and moving equatorward. This happens every year in the winter – spring upwelling period and occasionally, such as in 1997 – 1998, it covers an interannual scale when stratification remains weak and water column oxygen concentrations are relatively higher (< 2ml l-1 ; Figure 5-3). In this period salinities remained low, supporting the prediction that the system would under these conditions have a stronger forcing from Lüderitz. Salinities then increase as predicted from the result of the increasing contribution from the Cape Frio upwelling centre. The data shows that there are consistent differences in oxygen content between the inner and outer shelf areas of the central Benguela. The inner shelf concentrations are consistently lower. Differences in LOW variability between the inner and outer shelf zones are due to the lag effect caused by the biogeochemical oxygen demand driven by the respiration rates in the inner shelf mud belt where much of the surface derived new production accumulates (Monteiro et al. 2005; Monteiro and Roychoudhury 2005). The sediment ecosystem in the mud belt can exist in two redox states, aerobic and anaerobic. Both states create oxygen demand fluxes but whereas in the aerobic condition this is directly related to the metabolism of the flux of organic carbon, in the anaerobic condition it includes also the additional oxygen demand fluxes driven by reduced metabolic products such as HS-, CH4 and NH4 +. Once the system switches to the anaerobic condition, the lagged flux of reduced products driven by accumulated organic carbon maintains an oxygen sink that increases the persistence of hypoxic / anoxic conditions. The lag in the consumption of the electron donors as well as the flux of reduced products damp the variability in the inner shelf region of Namibia. However, it is not the upwelling-derived flux of organic carbon that governs the shift from aerobic to anaerobic conditions, but the boundary derived LOW signal of O2 < 1.5ml l-1. If this condition is not achieved, either because of boundary conditions at Cape Frio or an increased contribution from Lüderitz, the anaerobic fluxes weaken and

80


the system will after one or two seasons switch to an aerobic state ( e.g. 1997-1998). Despite the lag effect of the locally forced anaerobic conditions, LOW variability is still characterised by a seasonality where water column hypoxia is deepened in the later summer – autumn period and weakened in the winter – early spring period (Figure 5-3). Walvis Bay (23°S, 320m bottom depth)

Depth (m)

0 -100 -200 -300

1995

a)

1996 5

1997 6

7

8

1998

1999

2000

2001

2002

2003

2004

9 10 11 12 13 14 15 16 17 18 19 20 21 °C

Depth (m)

0 -100 -200 -300

1995

1997

1998

1999

2000

2001

2002

2003

2004

36 PSU 35.9 35. 8 35.7 35.6 35.5 35. 4 35 .3 35 .2 35 .1 35 34. 9 34.8 34 .7 34.6 34. 5 34.4 34.3 34.2 34. 1 34 33.8 33.6 33. 4 33.2 33

b)

1996

Depth (m)

0 -100 -200 -300 1995

c)

1996

1997 0

0.5

1998 1

2

1999 3

4

2000 5

6

2001

2002

2003

2004

7 ml/l

Figure 5-3. Variability of temperature (a), salinity (b) and oxygen (c) at an outer shelf location at 23oS in the Central Benguela between 1994 and 2004. The oxygen variability is modulated by both seasonal (summer / late summer) and interannual (1996 – 1999 vs 2000 – 2002) scales. The significant point is to link the period of enhanced LOW (2000 – 2002) to increased surface warming and higher salinities. In contrast, the 1996 – 1999 periods reflect weaker hypoxia. Sampling periods and depths are indicated on the diagrams.

The importance of this finding is that it supports the view that the toxic events driven by methane and sulphide are a response to boundary forcing rather than a forcing factor. The unexpected aspect is just how weak the local generation signal really is. It


81

controls persistence, intensity as measured by water column depth, and toxicity to local fauna but not the incidence. In summary, while LOW variability in the outer shelf is governed by both the boundary conditions and dynamic interaction of fluxes between the Lüderitz and Cape Frio upwelling centres, the variability in the inner shelf is the result of the same factors as well as the local biogeochemical processes. Measured oxygen concentrations reflect the spatially separated inputs from the Cape Frio and Lüderitz upwelling cells as well as the poleward transport of warm tropical surface water which exerts its impact through the baroclinic pressure gradient. Southern Benguela: South African sub-system In contrast to both the central and northern sub-systems, LOW variability in the southern sector (e.g. see Figure 5-4) is largely governed by a combination of local physical (stratification, recirculation-retention and advection) and biogeochemical processes (upwelling driven new production). Moreover, both northern and central sub-systems have shelf boundary conditions characterised by ETSA-derived LOW whereas the boundary conditions in the southern sub-system are those of aerated subAntarctic SACW (O2 > 4ml l-1) - see Chapman and Shannon 1987. Therefore, rather than being “primed” with remote sourced LOW, local formation has to rely on the physics of retention and stratification to bring down the oxygen concentrations of newly upwelled water. This is, in principle, the same set of processes that govern the ETSA zone on a larger spatial scale. The main LOW generation zone is the St Helena Bay retention zone (31 - 33oS) downwind from the Cape Columbine upwelling centre (Bailey and Chapman 1985; Penven et al. 2000). The hydrodynamics of this system drive a seasonal cyclonic circulation that gives rise to a strongly stratified two layer system sustained with cold upwelled water and a sun-warmed surface layer (Waldron and Probyn 1991). These conditions persist over the upwelling season (September – April) and support a highly productive nitrate-driven biological pump (Touratier et al. 2003; Monteiro et al. 2005; Monteiro and Roychoudhury 2005), which coupled to the physically driven nutrient fluxes, lead to high rates of sedimentation of POC (Bailey 1983). The remineralization of the POC coupled to fluxes of HS- creates an environment where, with strong stratification that reduces the aeration rates of sub-thermocline waters, the seasonal LOW is generated. The detailed interactive dynamics that govern LOW variability and which form the basis to a possible forecasting system are described in greater detail in the section dealing with LOW forecasting scales (see Chapter 13 this volume). LOW variability in the remainder of the southern Benguela sub-system shelf is the result of equatorward advection of LOW formed in St Helena Bay. The relatively low salinity values (S < 34.9) over a decade-long time series support the view that ETSA waters do not make a significant contribution to the water and LOW in the southern Benguela. The northward transport is depicted in the distribution of the integrated surface chlorophyll from St Helena Bay (see composite images in Figures 7-2 and 7-4

82


De pth (m)

Cape Columbine (100m to 150m bottom depth) 0 -50 -100 1959

1960

1961

1962

1963

1964

1965

1966

1967

Dep th (m)

0

-50 -100 -150 1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

De pth (m)

0 -50 -100 -150 1994

1995

1996

5

6

7

1998

8

1999

2000

2001

2002

2003

9 10 11 12 13 14 15 16 17 18 19 20 21 °C

0 -50

36 PSU 35.9 35 .8 35.7 35 .6 35.5 35 .4 35.3 35 .2 35.1 35 34.9 34 .8 34.7 34 .6 34.5 34 .4 34.3 34 .2 34.1 34 33.8 33 .6 33.4 33 .2 33

Dep th (m)

a)

1997

-100

1959

1960

1961

Dep th (m)

0 -50 -100 -150

1984

1985

1986

1987

1988

1989

1990

1991

1992

1993

1994

1993

1994

Depth (m)

0 -50 -100 -150 1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

De pth (m)

b) 0 -50 -100

1959

1960

1961

1962

1963

1964

1965

1966

1967

De pth (m)

0 -50 -100 -150

1984

1985

1986

1987

1988

1989

1990

1991

1992

De pth (m)

0 -50 -100 -150 1994

c)

1995

1996

1997

0

1998

0.5

1

1999

2

3

4

2000

5

6

2001

2002

2003

7 ml/l

Figure 5-4. The variability of temperature (a), salinity (b) and oxygen (c) at a mid-shelf position in the period 1984 – 2004. It shows a remarkable contrast in oxygen regimes between the 1980’s (aerated) and the 1990’s which were oxygen deficient / hypoxic. The explanation lies in the quasi-decadal scale changes in the upwelling wind regimes. The 1980s were characterised by relatively weak winds whereas the 1990’s by strong upwelling conditions. Hypoxia is related to changes in the retention characteristics of the St Helena Bay retention area. temperature. Its variability is driven exclusively by the variability of the ETSA characteristics


83

in Pitcher and Weeks – Chapter 7 this volume). LOW variability off Hondeklip Bay (30oS) is closely correlated to temperature (Monteiro et al. 2004) which is an expected outcome from advection-controlled variability. We agree with Johnson and Nelson (1999) that interannual LOW variability in the southern Benguela is governed mainly by the interannual variability in the equatorward component of the seasonal upwelling winds. This is in contrast to the controls on the boundary conditions of the northern and central Benguela system that are exerted by the seasonal, interannual and decadal shifts in the easterly equatorward winds. Summary of characteristics of LOW variability The characteristics of LOW variability in the Benguela can be summarised into three modes: Northern Benguela: The Angolan shelf system is directly coupled to the boundary conditions and the variability in LOW is largely predicted by its strong correlation to temperature. Its variability is driven exclusively by the variability of the ETSA characteristics. Central Benguela: LOW variability on the Namibian shelf is non-linear in respect of upwelling because it is dependent on a conjunction of processes and conditions that are not directly linked. The factors that govern LOW variability on the Namibian shelf are thought to be ETSA characteristics that set the boundary conditions, the incidence and strength of warm surface events, and upwelling rates at both Cape Frio and Lüderitz. These are amplified by the local production fluxes in the inner shelf. Southern Benguela: LOW variability in the Southern Benguela is largely governed by the interannual variability in the equatorward component of the seasonal upwelling winds. The importance of this characterisation is that it helps to define key scales of forcing and response that are sub-system specific and perhaps result in a more sensitive forecasting or at least predictive system.

PROCESSES REQUIRING DIAGNOSTIC ASSESSMENT The recently completed review identified a number of new possible processes that may govern low oxygen variability over a wide range of space and time scales (Monteiro et al., 2004). While these new proposed explanations were consistent with the data sets used in the review it is not certain whether the dynamics proposed to account for their impact on LOW variability are consistent.

84


The most important process uncertainties driving remote forcing are: • The combination of processes that govern the formation and variability of LOW in the ETSA zone • The coupling of warm surface flow and sub-thermocline poleward flow which may be the mechanism that transports LOW from the ETSA to the Benguela boundaries • The Slope – Shelf coupling which transports LOW onto the shelf at preferential sites such as the Cape Frio or Lüderitz upwelling centres • The coupling between warm surface flow and sub thermocline transport on the shelf through the strengthening of the thermocline • Coupling between remote and local forcing The dynamic consistency of these proposed mechanisms needs to be evaluated using appropriately set up hydrodynamic models through a set of modelling experiments. These do not need to be undertaken in simulation mode but in synthetic domains set up at scales that are comparable to the actual mechanism in question. This is the proposed approach in the follow up Chapter 13 that focuses on processes and scales that are amenable to forecasting. Coupled remote and local forcing The dependence of LOW variability on the coupling between remote and local forcing is a key finding which makes the forecasting potential of LOW variability and its impacts a possibility (van der Plas et al. 2006). This is because it is remote forcing that defines the regime modes that govern variability in the northern and central Benguela through the boundary conditions. Regime mode shifts on a basin scale that eventually impact on the Benguela boundaries may not only be forecast on a time scale of months but their impact on a time scale of years – decades may perhaps be evaluated through scenario modelling. However this forecasting potential depends sensitively on the proposed biogeochemical coupling between remote and local forcing (van der Plas et al. 2006). It is important that this hypothesised link be tested using a combination of modelling and observational data. The coupling was proposed using steady state assumptions and its incidence in a time varying sense needs to be tested (van der Plas et al. 2006). PROCESSES WITH FORECASTING POTENTIAL The table of processes below (Table 5-2) shows that the cost effective observational capacity of the individual processes is mostly good but the forecasting of LOW variability depends largely on how well the linkages that transfer the equatorial signal to the Benguela are understood and modelled.

Three different temporal or forecasting scales of LOW variability are evident from the table: • the short term events of a few days with localised impact


85

Table 5-2. Physical processes that have a bearing on the variability of Low Oxygen Waters in the Benguela and may be worth forecasting.

Domain

Forcing System

Processes

Scales of variability

Observational Potential

Worthwhile to Forecast

Remote

ETSA

Equatorial upwelling and new production


Good: Ocean Colour

Maybe

Intensity and timing of trade winds


GCM

Yes

Equatorial stratification

Interannual – decadal (Benguela Niño)

Good: Ocean Buoys

Yes

Angola Current


Good: Altimetry and AVHRR

Yes

LOW in the ETSA

Interannual decadal

Good: Ocean Buoy

Yes

Depth of the upper boundary of O2 < 2ml l-1

Interannual decadal

Good: Ocean Buoy

Yes

Depth range of O2 < 2ml l-1

Interannual decadal

Good: Ocean Buoy

Yes

Poleward transport of LOW into Benguela


Good: Ocean Buoy

Yes

Upwelling driven new production

Days - weeks

Good (ocean colour remote sensing)

Yes

Upwelling wind variability

Days - weeks

Good

Maybe

Relaxation events in the southern Benguela

Days - weeks

Good

Maybe

Spatial scales of depositional areas

10 – 1000km

Good

Yes

Transport and dispersion of LOW

10 – 1000km

Poor

Yes

Transport and dispersion of sulphide in water column

10 – 100km

Poor

Yes

Local

Upwelling centres

Ecosystem responses

86

• •


medium term events of a few months duration and with shelf wide impact long lasting LOW variability of year to decade scale and with system wide impact (scenarios)

Any future operational LOW forecasting system will need to combine modelling with data assimilation and verification platforms. It is envisaged that such a system would make use of “real time large scale models” running predominantly with data assimilation into which will be nested the LOW region specific model domains. These would derive their boundary conditions from the large scale models, advect the signal and drive the internal processes that govern LOW variability in critical habitat areas. Forecasting in region specific domains would be based on “free running” models rather than on data assimilation and would be verified against real time data sets. The LOW scales and processes that are most amenable to forecasting whether for scientific reasons or because they are relevant to ecosystem management perspectives are addressed in detail in the companion Chapter 13 (Monteiro et al. this volume). WHAT ARE THE GAPS? Time series observations The Lüderitz upwelling centre plays a pivotal role in forcing the system by supplying upwelled water to both the central and southern Benguela shelf. However, the attempts to understand this role are severely limited by the paucity of data from this area. The most important forcing point has the weakest data set. The temporal resolution of the data from the second most important upwelling “gate” in the Benguela, Cape Frio, is also quarterly at best. (Refer also to Shillington et al. and Reason et al. Chapters 4 and 10 respectively, this volume) Slope - Shelf exchange

An observational programme should be put in place that will elucidate the mechanisms of slope-shelf exchange of LOW. The proposed observationally based early warning system should make use of the understanding derived from both the literature review and the data based advances. The observational programme to support a first early warning system should aim to make use of existing freely available data products. These should include modelling, remote sensing and observational programmes that are already in place. The processes that need to be monitored include: • The thermocline characteristics in the ETSA area which governs the LOW characteristics for the Benguela; • Poleward advection of warm tropical water:


87

• This was shown to be one of the most sensitive indicators of the southward propagation of LOW • from the ETSA to the Benguela off the shelf • the shelf based southward displacement of tropical surface waters e.g.: ABF; and • The thermocline depth and strength on the slope (1000 – 2000m) and on the shelf (100m) The recommended additional observational programmes are the monitoring of oxygen and temperature in the ETSA region and at the two main upwelling centres that cover the Central and Southern Benguela. This is most likely best done using large ocean buoys with temperature records at 50 m intervals and oxygen observations at 50m intervals in the upper 200m and 100m spacing below 200m. The buoys should be located on the slope in the zone of the 1000 – 2000m depth range. These should also provide telemetry based data streams that allow the data quality to be assessed and test linkages with response scales at the monthly monitoring sites off Walvis Bay and St Helena Bay. It is recommended that, when this proposed programme is accepted, the BCLME commission be the regional facility to provide the products to the community. This first phase early warning system is expected to be operational for a period of two years by which time the modelling platforms for the BCLME should be operational and providing a second and later third phase forecasting. Remote - local coupling Advection vs. local formation: Local formation of hypoxic or anoxic LOW depends on the boundary conditions being at or below a critical threshold (approximately 1.5ml l-1 O2) at which the physical supply rate of dissolved oxygen falls below the biogeochemical demand and the system rapidly switches to anaerobic respiration (van der Plas et al. 2005). Thus the magnitude and persistence of LOW variability on the Benguela shelf is primarily the result of the degree of oxygen depletion in incoming water across the boundary and only secondarily the local oxygen demand driven by the sedimenting flux of upwelling-linked new production. The latter is, however, responsible for modulating the response of the system to boundary forcing (Monteiro et al. 2004). It has been proposed that state of environment (SOE) indicators be devised to monitor the LOW status over the Benguela shelf. The SOE indicator effort should be focussed on the areas where regular spatial monitoring can take place at least once a month. The indicators are a measure of the response of the system and should therefore also ideally be located in areas where measuring that response is of relevance. The present monthly monitoring lines in the southern, central and northern Benguela partially fulfil these requirements as the location of the lines is for historical reasons pragmatically close to the sponsoring institution. The following thresholds are suggested: depth of oxygen under- saturation in two zones of the shelf, namely at depth < 100m and depth > 100m.

88


SUMMARY LOW variability in the Benguela is governed by varying scales of remote and local forcing linked to both Equatorial and Cape Basin systems. The nature of these nonlinear interactions is not clearly understood because scales are large and their elucidation through observational programmes alone is not cost effective. Models are required to characterise the complexity of the most important forcing and response scales in both time and space. It will be necessary to approach this as a multi-phase process, beginning with a diagnostic emphasis which evolves to a forecasting system through hindcasting focussed specifically on large scale events of the past. It is clear that not all the variability scales are amenable to forecasting either because the driving process scales are too uncertain or because they are of little management of policy interest. Two scales were defined as being of interest to both these criteria: • Short term (7 day) scale related to forecasting conditions leading to the walkout or mortality of rock lobster in the southern Benguela • Medium term (2 month) forecasting of the intensification of the remote forcing of ETSA derived LOW which has a bearing on the Namibian hake fishery These two scales are discussed in detail in the companion Chapter 13, this volume. ACKNOWLEDGEMENTS We acknowledge the inputs from our collaborators Geoff Bailey and Quilanda Fidel as well as the constructive comments from Profs. Geoff Brundrit and Vere Shannon and Dr Piers Chapman of Louisiana State University. Our participation is made possible through the support of our respective institutions CSIR, South Africa and MFMR, Namibia and the work in general was only possible through the data contributions from both MCM, South Africa and IIM, Angola. REFERENCES Bailey, G.W. 1983. Pilot study of the vertical flux of POC and PON in St Helena Bay., South African Journal of Science. 79: 145-146. Bailey, G.W. 1991. Organic carbon flux and development of oxygen deficiency on the modern Benguela continental shelf south of 22°S spatial and temporal variability. 171-183 in Tyson, R.V. and T.H. Pearson, editors. Modern and Ancient Continental Shelf Anoxia. Bailey, G.W. and P. Chapman. 1985. Nutrient Status in the St Helena Bay region in February 1979. 125145 in Shannon, L.V., ed. The South African Ocean Colour and Upwelling Experiment. Cape Town, Sea Fisheries Research Institute. Bubnov, V.A. 1972. Structure and characteristics of the oxygen minimum layer in the southeastern Atlantic. Oceanology 12:193-200. Chapman, P. and L.V. Shannon. 1985. The Benguela Ecosystem Part II. Chemistry and Related Processes. Oceanography and Marine Biology: An Annual Review. 23:183-251. Chapman, P. and L.V. Shannon. 1987. Seasonality in the oxygen minimum layers at the extremities of the Benguela system. South African Journal of Marine Science. 5:85-94. Copenhagen, W.J. 1953: The periodic mortality of fish in the Walvis region; a phenomenon within the Benguela Current. Investigational Report of the Division of Fisheries South Africa 14, 35pp. Duncombe Rae, C. M. 2004. A demonstration of the hydrographic partition of the Benguela upwelling ecosystem at 26˚40'S. African Journal of Marine Science, (in press).


89

Florenchie, P., R.E. Lutjeharms, and C.J.C. Reason. 2003. The source of Benguela Niños in the South Atlantic Ocean. Geophysical Research Letters 30(10): 12. Hamukuaya, H., M. O’Toole and P.M.J. Woodhead . 1998. Observations of severe hypoxia and offshore displacement of Cape Hake over the Namibian shelf in 1994. In: Benguela dynamics: Impacts of variability on shelf-sea environments and their living resources. South African Journal of Marine Science 19:57-61. Hardman-Mountford, H.J. A.J. Richardson, J.J. Agenbag, E. Hagen,L. Nykjaer e, F.A. Shillington, C. Villacastin. 2003. Ocean climate of the South East Atlantic observed from satellite data and wind models. Progress in Oceanography. 59:181 – 221 Hart, T. J. and R. I. Currie. 1960. The Benguela Current. Discovery Reports 31: 123-298. Johnson, A. and G. Nelson. 1999. Ekman estimates of upwelling at Cape Columbine based on measurments of longshore wind from a 35 year time-series. South African Journal of Marine Science 99: 433 – 436. Lass, H.U., M. Schmidt, V. Mohrholz and G. Nausch. 2000. Hydrographic and current measurements in the area of the Angola-Benguela Front. Journal of Physical Oceanography 30: 2589-2609. Marchand, J. M. (1928). The Nature of the Sea-Floor Deposits in certain Regions of the West Coast. Fish and Marine Biological Survey, Department of Mines and Industries Annual Report 6(5): 1-11. Mercier, H., M. Arhan and J.R.E. Lutjeharms. 2003. Upper-layer circulation in the eastern Equatorial and South Atlantic Ocean in January-March 1995. Deep-Sea Research I 50: 863-887. Mohrholz, V., M. Schmidt and J.R.E. Lutjeharms. 2001. The hydrography and dynamics of the AngolaBenguela Frontal Zone and environment in April 1999. South African Journal of Science 97: 199-208. Monteiro, P.M.S. (1996) The oceanography and biogeochemistry of CO2 in the Benguela upwelling system. PhD Thesis, University of Cape Town, South Africa Monteiro, P.M.S., A.K. van der Plas, G.W. Bailey and Q. Fidel. 2004. Low oxygen variability in the Benguela ecosystem: a review and new understanding. CSIR Report (Internationally Peer Reviewed), ENV-S-C 2004-075, 67pp. Monteiro P.M.S. and A. Roychoudhury. 2005. Spatial Distribution of Trace Metals in an Eastern Boundary Upwelling Retention Area (St. Helena Bay, South Africa): A Hydrodynamic-Biological Pump Hypothesis. Estuarine, Coastal & Shelf Science 65:123-134. Monteiro, P.M.S., G. Nelson, A. van der Plas, E. Mabille, G.W. Bailey, E. Klingelhoeffer. 2005. Internal tide-shelf topography interactions as a potential forcing factor governing the large scale sedimentation and burial fluxes of particulate organic matter (POM) in the Benguela upwelling system. Continental Shelf Research 25:1864-1876. Monteiro, P.M.S., A. van der Plas, G.W. Bailey, P. Rizzoli, C. Duncombe Rae, D. Byrnes, G. Pitcher, P. Florenchie, P. Penven, J. Fitzpatrick, U. Lass. 2006. Low Oxygen Water (LOW) forcing scales amenable to forecasting in the Benguela ecosystem. Chapter 13, p67-89 in Shannon, V., G. Hempel, P. Malanotte-Rizzoli, C. Moloney and J. Woods, eds. The Benguela: predicting a large marine ecosystem. Elsevier (this volume). Moroshkin, K.V., V.A. Bubnov, and R.P. Bulatov. 1970. Water circulation in the eastern South Atlantic Ocean. Oceanology 10:27-37. Penven P., C. Roy, A. Colin de Verdiere, and J.L. Largier. 2000. Simulation of a coastal jet retention process using a barotropic model. Oceanologica Acta 23(5): 616-634 Pieterse, F. and D.C. van der Post. 1967. Oceanographical Conditions Associated with Red- Tides and Fish Mortalities in the Walvis Bay Region. Investigational Report of the Administration of South West Africa Marine Research Laboratory 14, 125pp. Pitcher, G.C. and S.J. Weeks. 2005. Variability and potential for prediction of harmful algal blooms in the southern Benguela ecosystem. (this volume) Reason et al., this volume Reid, J. L. 1989. On the total geostrophic circulation of the South Atlantic Ocean: Flow patterns, tracers, and transports. Progress in Oceanography 23:149-244. Shannon L.J. and A. Jarre-Teichemann. 1999. A model of trophic flows in the northern Benguela upwelling system during the 1980’s. South African Journal of Marine Science 21: 349 – 366. Shillington, F.A. et al. 2005. Large Scale Physical Processes. This volume 67-89. Stramma, L. and M. England. 1999. On the water masses and mean circulation of the south Atlantic ocean. Journal of Geophysical Research (Oceans), 104: 20863 – 20883. Stramma, L. and F. Schott. 1999. The mean flow field of the tropical Atlantic Ocean. Deep Sea Research II, 46: 279–303.

90


Touratier F., J.G. Field and C.L. Moloney. 2003. Simulated carbon and nitrogen flow of the planktonic food web during an upwelling relaxation period in St Helena Bay (southern Benguela ecosystem). Progress in Oceanography 58:1-41. van der Plas, A.K., P.M.S. Monteiro, and A. Pascall. 2005. The cross shelf biogeochemical characteristics of sediments in the central Benguela and its relationship to overlying water column hypoxia. Continental Shelf Research, in review. Voituriez, B. and A. Herbland. 1982. A Comparative Study of the productive systems of the tropical east Atlantic: Thermal Domes, Coastal Upwellings and Equatorial Upwelling. Rapp .P-v. Reun. Cons. perm. int. Explo. Mer. 180:114-130. Waldron, H.N. and T.A. Probyn. 1991. Short term variability during an anchor station study in the southern Benguela upwelling system: Nitrogen supply to the euphotic zone during a quiescent phase of the upwelling cycle. Progress in Oceanography 28:153 – 166 Woodhead, P.M., H. Hamukuaya, M.J. O’Toole, and M. McEnroe. 1998. Effects of oxygen depletion in shelf waters on hake populations off central and northern Namibia. In Shannon, L.V. and M.J. O’Toole, editors. International Symposium, Environmental variability in the South East Atlantic., 10 pp. NATMIRC, Namibia. Woodhead, P.M., H. Hamukuaya, M.J. O’Toole, T. Stroemme, G. Saetersdal and M. Reiss. 1997a. Catastrophic loss of two billion Cape hake recruits during widespread anoxia in the Benguela Current off Namibia. In: ICES International Symposium, Recruitment dynamics of exploited marine populations. Physical-biological interactions. pp. 105-106. Woodhead, P.M., H. Hamukuaya, M.J. O’Toole, T. Stroemme and S. Kristmannsson. 1997b. Recruit mortalities in Cape hake, following exclusion from shelf habitat by persistent hypoxia in the Benguela Current, Namibia. In: ICES International Symposium, Recruitment dynamics of exploited marine populations. Physical-biological interactions. pp. 26-27.

91


6 Variability of Plankton with Reference to Fish Variability in the Benguela Current Large Marine Ecosystem – An Overview Larry Hutchings, Hans M. Verheye, Jenny A. Huggett, Hervé Demarcq, Rudi Cloete, Ray G. Barlow, Deon Louw, Antonio da Silva

ABSTRACT This article reviews the variability of plankton over time scales ranging from mesoscale upwelling events of a few days’ duration to decadal scale changes in the northern and southern subsystems in the Benguela Current. It focuses on the plankton that are considered important for fish, particularly the crustacean zooplankton. The southern Benguela is strongly pulsed over periods of 4-12 days with a series of upwelling events modulated by passing cyclonic weather systems. The northern Benguela is less pulsed with short-term variability linked to continental shelf waves. Upwelling is particularly active at seven major sites in the Benguela system. Dense phytoplankton blooms develop in the cool nutrient-rich plumes, which merge and blend with surrounding waters, creating a broad band of phytoplankton-rich water over the shelf. Species succession from small to large diatoms, dinoflagellates and small flagellates occurs as the waters mature after upwelling and generally move offshore, although numerous exceptions occur, with small-celled communities occasionally dominant in nearshore waters. Much regeneration and recycling of nutrients occurs, resulting in lower than expected f-ratios. Frontal zone aggregations provide important feeding opportunities in the transport phase of ichthyoplankton between the Agulhas Bank spawning grounds and the nursery grounds on the South African West Coast. The Angola-Benguela front in the northern Benguela is also an important region for pelagic fish spawning. Seasonal changes in wind forcing indicate maximum upwelling in spring and autumn throughout the Benguela, with a tendency for a summer maximum in the south. Lüderitz (25oS) and Cape Frio (17oS) are particularly active upwelling regions. Phytoplankton biomass, estimated as chlorophyll a, shows a winter maximum in the northern Benguela and a summer maximum in the southern Benguela. The Lüderitz area shows perennial phytoplankon minima, possibly due to strong turbulence. The central Namibian shelf and the South African west coast shelf have persistently high phytoplankton biomass. A seasonal intrusion of warm oligotrophic water from Angola in late summer (December to March) results in strong contrasts between winter and

92

L. Hutchings et al.

summer in the extreme northern Benguela. Zooplankton biomass shows different cycles along the coast, with spring, summer and autumn maxima in the south, and a slight maximum during the second half of the year (July to December) off central Namibia. The dominant fish spawning period is spring–summer throughout the region. Long-term changes in the southern Benguela include a significant increase in zooplankton over the past five decades, with a decline since 1995. Fish abundance has declined in the northern Benguela but remained reasonably stable in the southern Benguela until 2000, when pelagic fish biomass increased dramatically with concomitant declines in zooplankton biomass. A range of modelling exercises, including expert systems, statistical models and linked IBM-hydrodynamic models, has been compared to or derived from field data, and has stimulated new observational programmes at improved space and time scales. Observational data at pertinent time and space scales are lacking in the northern Benguela system, which will hamper validation of prognostic and diagnostic models. INTRODUCTION From space, viewed at a coarse scale (Demarcq et al. 2003), the Benguela Current looks like a cool, broad sluggish drift, the eastern portion of the South Atlantic gyre, which is characterized by a narrow belt of cold, phytoplankton-rich water along the coastline (Figure 6-1a). Sharp discontinuities occur at the southern boundary with the Agulhas Current (32˚-37˚S) and at the northern boundary in the Angola-Benguela Front region (12˚-18˚S). Intensive mixing and high variability characterize these boundary zones, matched only by the Brazil/Malvinas current interactions in the SW Atlantic (Bakun 1996). Within the major upwelling region of the Benguela, there are seven particularly active sites (Shannon and Nelson 1996), of which the Lüderitz cell (25˚-26˚S) is by far the most powerful in terms of Ekman transport and turbulent mixing (Parrish et al. 1983). Event-scale variability is often dominant, with significant seasonal modulation of upwelling winds in the northern and southern extremities. In a wind-driven upwelling system such as the Benguela, the plankton variability is driven by complex non-linear interactions between the driving forces of winds and solar radiation, and stabilization, sinking and the response times of the individual organisms. The time scales vary from hours and days through to inter-decadal shifts, with corresponding spatial scales. Fish populations that inhabit this ecosystem are subject to high variability in the triad of factors affecting recruitment, i.e. enrichment, retention and concentration (Bakun 1996), particularly during the early life-history stages. This paper is intended to give a brief description of the variability of the planktonic components within the Benguela Current and some of the problems associated with predicting future events, with particular reference to the dynamics of fish populations, such as growth, recruitment and fish condition. As such, this review focuses on the crustacean components, principally the copepods and the euphausiids. Pitcher and

Variability of plankton with reference to fish variability, BCLME

(a)

93

(b)

(e) (c)

13 15 17

Latitude (°S)

19

(d)

21 23 25 27 29 31 33

Figure 6-1. (a) Annual average chlorophyll concentration computed from monthly composites derived from Global Area Coverage (GAC) data at 4.5 km resolution from SeaWiFS ocean colour sensors between September 1997 and April 2002 and the annual mean position of the 1 mg m-3 offshore limit between 12 and 34˚S; (b) time-series of monthly averages of chlorophyll biomass integrated between the coast and this limit, showing seasonal and interannual variability and spatial patterns within the Benguela system, September 1997 – July 2003; (c) monthly variations in an enrichment index (chlorophyll content summed between the coast and the 1 mg.m-3 isoline) in the northern and southern Benguela, showing opposing trends, January 1998 – March 2002; mean seasonality of indices of (d) upwelling (upwellingfavourable winds, low-moderate-high) and (e) enrichment (mean integrated chlorophyll a) in the Benguela Current system (from Demarcq, unpubl. data).

94

L. Hutchings et al.

Weeks (2006) cover the major variability in phytoplankton, with particular reference to the formation of harmful algal blooms close inshore. However, fish populations such as anchovy (Engraulis encrasicolus), sardine (Sardinops sagax) and hake (Merluccius spp.) spawn over extensive areas of the continental shelf, and serially spawn for prolonged periods, possibly as an adaptation to the high mesoscale and microscale variability which characterizes the Benguela. As always, the major problem has been integrating small-scale events over a sufficiently long time and large area to provide indices of the suitability of a particular zone that is pertinent to fish population dynamics. EVENT-SCALE VARIABILITY Southern Benguela Characteristic time scales of phytoplankton and zooplankton in relation to the wind forcing and pelagic fish early life history were first described by Hutchings and Nelson (1985). Upwelling in the southern Benguela is typically pulsed over a period of approximately six days, varying between two and 12 days of upwelling-favourable winds interspersed with calms or wind reversals (Nelson 1992). The patterning of wind variability in the southern Benguela is best illustrated by Roy et al. (2001) (Figure 6-2a). Twelve upwelling events are seen over a six-month period, and each results in a “spring bloom!” For each wind event, cold source water originating from South Atlantic Central Water (Shannon 1985) rises at the upwelling site and moves offshore, warming, dispersing and mixing with existing surface waters. Essentially, this water injects new nitrogen into the euphotic zone, where it is retained by heating and stabilization of the upwelled water as a lens of warm water overlying cooler waters. Phytoplankton seed cells, either resting cells or viable cells present in source water or cells mixed laterally into the newly-upwelled water mass, grow and divide rapidly, forming dense blooms (Pitcher and Weeks, 2006). By following drogues placed in the coldest, newly-upwelled waters, the development and decline of phytoplankton in individual “boluses” of water could be followed over periods of 6-10 days (Figure 6-2b; Brown and Hutchings 1985). This period is short in relation to the 2-4 weeks development time of a dominant West Coast upwelling copepod, Calanoides carinatus, at temperatures typical of sun-warming upwelled waters (Figure 6-2c; Peterson and Painting 1990; Hutchings 1992), resulting in a potential for a major mismatch between phytoplankton and grazers. Only when chlorophyll a rises above 36 mg.m-3 do large herbivorous copepods respond in terms of increased egg production (Figure 6-2d; Armstrong et al. 1991) and somatic growth rates (Richardson and Verheye 1998; 1999). This further shortens the overlapping optimal time period between plants and grazers, since by the time the juvenile stages are developed the bloom has declined. Behavioural adaptations such as vertical migration may help to prolong the residence time of zooplankters in developing phytoplankton patches.


95

Figure 6-2. Typical time scales for the Benguela region. (a) Daily time-series of north-south wind speed at Cape Columbine, 1 November 1999 – 30 April 2000 (top) and cumulative divergence per upwelling event for the same period (bottom); the episode number is indicated for each major upwelling event (after Roy et al 2001). (b) Development and decay of phytoplankton blooms over 6-13 days, tracked by drogues placed in newly upwelled water (after Brown and Hutchings 1985). (c) Development of Calanoides carinatus populations at 15ºC over 18-20 days, indicating a basic temporal mismatch with phytoplankton blooms in a strongly-pulsed upwelling system (after Peterson and Painting 1990 and Hutchings 1992). (d) Daily egg production by female Calanoides carinatus in response to variable chlorophyll a concentrations (after Armstrong et al. 1991 and Hutchings and Field 1997)

96

L. Hutchings et al.

Figure 6-3. Vertical sections of daily variation in (a) temperature; (b) nitrate concentration; (c) chlorophyll a concentration; (d) major phytoplankton groups and successional changes of dominant taxa; and (e) primary production during an anchor station time-series study in St Helena Bay, 20 March – 15 April 1987. Arrows at the top indicate the advection of newly-upwelled water (after Mitchell-Innes and Pitcher 1992)


97

A 27-day anchor station study carried out in St Helena Bay (33˚S) in March-April 1987 demonstrated a sequence of two upwelling cycles as the upwelled water stabilized (Figure 6-3), and some succession of phytoplankton types, from small diatoms to large diatoms and then to small flagellates (Mitchell-Innes and Pitcher 1992). The same changes appear to occur with distance offshore on transects from the coast (Barlow et al. 2005). Using diagnostic pigment indicators, Barlow et al. (2005) showed diatoms and dinoflagellates were dominant in cool, maturing upwelled waters over the shelf, while small flagellates were dominant offshore but were prominent on occasions in the nearshore zone and likely to contribute significantly to the primary productivity of the Benguela. As copepods develop, there are ontogenetic changes in their vertical migration behaviour (Verheye and Field 1992; Huggett and Richardson 2000), which minimize predation and facilitate retention within the phytoplankton patch and within the nearshore zone (Verheye et al. 1992; Huggett 2003). The amplitude of migration is related to body size, developmental stage and food concentration (Verheye and Field 1992; Huggett 2003), and interspecific depthpartitioning has been demonstrated for some of the smaller copepod species in the southern Benguela (Stuart and Verheye 1991). Ontogenetic depth layering and migration have also been shown for local euphausiid communities (Figure 6-4a; Pillar et al. 1989), which, combined with differential flow patterns with depth, facilitate retention over the shelf. These life-history features provide important inputs to models of copepod or euphausiid abundance, as well as to individual-based models, such as that of Parada (2003), who modelled the vertical behaviour of anchovy larvae in the southern Benguela nursery area. Since peak upwelling occurs in summer, the frequency of upwelling events increases through the season, and successive plumes of stabilizing upwelled waters merge and mingle alongshore and across-shore, resulting in a broadening of the plankton-rich belt along the coast (Figure 6-1a). Much regeneration, recycling and secondary blooming (Figure 6-5) appear to occur in the mature upwelled waters, resulting in lower than anticipated f-ratios in the upwelling region (Probyn 1992). The seaward boundary of the inshore productive belt is a strong convergent front, where positively buoyant or phototropic organisms will aggregate in the gradually descending mature upwelled water as it sinks beneath the warm offshore water. Larval fish transported in the frontal jet currents over summer (Figure 6-6; Huggett et al. 1998) will therefore be afforded concentrated food organisms. The stronger the upwelling winds, the more convergence and concentration; essentially, this is a vertically orientated thermocline, which extends along much of the West Coast shelf as far as Lüderitz at about 25˚S.

Northern Benguela Upwelling appears to be less pulsed off Namibia, since the influence of the eastwardmoving cyclones is diminished with decreasing latitude (Shannon 1985). Upwelling centres are more dispersed and belt-like, so an inshore-offshore gradient in phytoplankton and zooplankton is apparent along most of the Namibian coast. Few

98

L. Hutchings et al.

Figure 6-4. (a) Weighted mean depth plots of eggs and early and late larval stages of Euphausia lucens in the southern Benguela (after Pillar et al. 1992). (b) Vertical distribution of Stylocheiron longicorne, Nematoscelis megalops, Thysanoessa gregaria, Euphausia gibboides, E. hanseni and E. americana. Data are averaged from nine night-time samples taken at a fixed station in the northern Benguela (after Barange 1990). (c) Conceptual three-dimensional model of cross-shelf circulation during upwelling in the northern Benguela. Encircled crosses denote equatorward flow, encircled dots indicate poleward flow (after Barange and Pillar 1992)


99

Figure 6-5. Daily nitrogen transfers (mmol N m-3 day-1) for (a) a net-phytoplankton-dominated and (b) a pico- and nanophytoplankton-dominated system (after Probyn 1992). Nitrogen uptake rates and ammonia regeneration rates were measured during February 1991. Phytoplankton compartments are based on particulate N concentrations. Dissolved nitrogen fluxes are represented by shaded arrows and predatorprey transfers by clear arrows. Broken arrows indicate the sinking by phytoplankton and egestion by zooplankton. Fluxes are balanced assuming steady state to produce a potential zooplankton production P. Note the greatly increased mesozooplankton production and flux during the net-phytoplankton dominant state.

100

L. Hutchings et al.

Figure 6-6. Three-dimensional representation of mean monthly abundances and distribution of (a) anchovy eggs, (b) anchovy larvae, (c) sardine eggs, and (d) sardine larvae along the SARP transect off the Cape Peninsula from August 1995 to July 1996 (after Huggett et al. 1998). The transect extended 65km offshore with stations 01-04 situated over the shelf.

event-scale studies have been made here, a notable exception being repeated transects of 15 stations at 36-hour intervals off 20˚S by the R.V. Alexander von Humboldt in the spring of 1979, at 30-170 km offshore (Hagen 1985; Postel 1990). The physical variability appeared to peak at about 5.6 days up to 100 km offshore, driven by continental shelf waves; beyond that a 13-day interval was indicated. Phytoplankton peaked close inshore whereas zooplankton wet mass peaked at 100-130 km from the coast. Barange (1990) demonstrated differential vertical distribution among six species of euphausiid (Figure 6-4b), suggesting niche partitioning of potentially competitive species. In contrast to the one-celled, cross-shelf circulation model proposed for the dominant euphausiid (Euphausia lucens) in the southern Benguela, a two-celled circulation model comprising both nearshore and offshore cells was proposed to facilitate euphausiid retention and niche partitioning in the northern Benguela (Figure 6-4c; Barange and Pillar 1992). Only Verheye et al. (2005) have reported on shortterm changes in zooplankton in Angolan waters.


101

SEASONAL CHANGES The clearest exposition of Benguela-wide seasonal variability in indices of upwelling (wind) and enrichment (chlorophyll a) is that of Demarcq (unpubl. data) (Figure 6-1b, c, d, e). Indices of enrichment were based on a time-series of Global Area Coverage (GAC) data at 4.5 km resolution from SeaWiFS ocean colour sensors between September 1997 and July 2003. Upwelling-favourable winds appear to have two maxima (in spring and autumn) throughout the Benguela upwelling region, tending towards a summer maximum in the extreme south. Stronger, more perennial winds blow at Lüderitz (26˚S) and Cape Frio (17˚S), with a slight winter maximum in northcentral Namibia. However, chlorophyll a maxima occur in the winter/spring months in the northern Benguela whereas the reverse pattern is seen in the southern Benguela, separated at Lüderitz where, paradoxically, a relative chlorophyll-minimum is evident throughout the year, possibly due to extreme turbulence. Persistent, strong upwelling also occurs at 17˚S (off the Cunene River), but a seasonal intrusion of warm Angola Current water in December-March overrides the phytoplankton signal. Off central Namibia and off the west coast of South Africa high phytoplankton concentrations persist throughout the year. Southern Benguela Another seasonal feature superimposed on the upwelling signal in the southern Benguela is a stabilization of the water column in early spring over the Agulhas Bank and offshore on the West Coast, after deep mixing during winter months (Shannon et al 1984). The nutrients isolated in the euphotic zone after stabilization generally do not result in extensive local phytoplankton blooms, since the water masses and phytoplankton are advected rapidly westwards and offshore, and surface waters are replaced by oligotrophic subtropical waters in filaments and eddies from the Agulhas Current. There is a general but very weak increase in phytoplankton in spring in the waters surrounding the upwelling region. This results in extremely food-poor environments immediately offshore of the upwelling zone over the summer months, unless eddies, filaments or shear-edge features increase productivity in the offshore zone adjacent to the frontal jet. This implies that any larval fish displaced offshore by whichever mechanism may have diminished survivorship, a central tenet of the recruitment hypotheses in the southern Benguela (Cochrane and Hutchings 1995). Seasonal cycles of mesozooplankton (primarily copepods) differ along the coast (Figure 6-7a). On the western Agulhas Bank (35˚S) zooplankton populations peak in late autumn and spring with a minimum in mid-summer. Off the Cape Peninsula (34˚S) the peak is in summer, with considerable monthly variability (Figure 6-7b), while in St Helena Bay (32˚S) the peak is in late summer with a marked decline in autumn. Verheye et al. (1992) suggested that the zooplankton annual cycle, although tightly coupled to the seasonality in upwelling and productivity (see e.g. Richardson et al. 2003a), is not solely driven from the bottom up, but may be altered, confounded or obscured by the top-down effect of predator-prey interactions. Thus, the distribution of different zooplanktivorous life-history stages (larvae, recruits, adults) of pelagic fish

102

L. Hutchings et al.

along different parts of the coast at different times of the year (Hampton 1992, Barange et al 1999)is likely to cause the observed spatial deviations from the expected zooplankton annual cycle. This makes both interpretation and prediction difficult.

Figure 6-7a. Seasonal variability in the biomass of copepods in the St Helena Bay area (top), the Cape Columbine-Cape Point area (centre), and the western Agulhas Bank (bottom); data points are 3-month running means between August 1977 and August 1978 (after Pillar 1986); peak recruitment and spawning seasons of anchovy are also shown (after Verheye et al. 1992).


103

Figure 6-7b. Seasonal variation of mean zooplankton standing stock (top) in response to primary production (bottom) along an Upwelling Monitoring Line off the Cape Peninsula, October 1970 – March 1973 (after Andrews and Hutchings 1980).

Compared with 1977/78, the seasonal signal of zooplankton in St Helena Bay has amplified dramatically in recent years, with an increase of the summer to winter ratio from approximately 2:1 in the late 1970s to well over 5:1 in the early 2000s (Figure 68a), while phytoplankton seasonal variability is not that pronounced (Demarcq et al. 2003; Barlow et al. 2005). During those recent years, anchovy recruitment was at record levels and, as a consequence, winter levels of zooplankton along the west coast have been depleted due to intense predation by these record concentrations of anchovy (see below). Fluctuations in the abundance of Calanoides carinatus, one of the most characteristic copepods of the upwelled waters along the African west and northeast coasts, have been studied extensively (see Verheye et al. 1991). At lower latitudes, this species exhibits a striking seasonality; its life history is characterized by a deep-living state of temporary developmental arrest (dormancy or diapause) at the pre-adult copepodite

104

L. Hutchings et al.

stage C5, and its extremely reduced metabolism (Auel et al 2005) allows it to bridge unfavourable conditions during the prolonged, warm non-upwelling season. In contrast, in the southern Benguela it maintains large populations on the shelf perennially, with only slight seasonal variability in its abundance. It is uncertain whether these populations enter true diapause, as suggested by Verheye et al. (1991), although observations of pre-adult C5s in deep (>600 m) water (De Decker and Mombeck 1964; Borchers and Hutchings 1986) lend support to this hypothesis. This remains an area for future investigation. Likewise, in contrast to the northern Benguela, there is no distinct seasonality in euphausiid biomass in the southern Benguela, where Euphausia lucens is one of the dominant macrozooplankters (Pillar et al. 1992). Many dominant fish species spawn seasonally in the Benguela, resulting in strong seasonal changes in ichthyoplankton abundance. Anchovies have a distinct summer maximum, whereas sardines spawn through most of the year, with a slight minimum in winter months and slight maxima in early spring and late summer, bracketing the anchovy spawning. Round herring (Etrumeus whiteheadii) spawn in late winter/early spring. All three species, and several others, spawn on the Agulhas Bank and their eggs and larvae are advected to the West Coast offshore. By mechanisms that are not well understood, recruits appear on the west coast close inshore. The adult spawning, and juvenile nursery areas are clearly separated (Hutchings et al 1998). There have been marked shifts in the spatial location of spawning, from 32ºS on the West Coast to east of Cape Agulhas (Barange et al 1999, van der Lingen and Huggett 2003, ), with variable interannual recruitment success (Boyd et al ,1998, Hutchings et al, 1998). The utility of the Agulhas Bank as a nursery ground for sardine juveniles is currently being evaluated (Miller et al., in press). Little spawning occurs at the Lüderitz upwelling cell, a site of high offshore losses and turbulence, and low phytoplankton concentrations. Horse mackerel (Trachurus trachurus capensis) spawn in late winter/early spring on the central Agulhas Bank (Barange et al. 1998), with juveniles appearing in the West Coast nursery area over midsummer. Hake appear to spawn all year round, but the deep-spawned eggs have not been well sampled. Most of these species also appear to utilize the inshore West Coast area as a nursery ground (Hutchings et al. 2001). Northern Benguela Seasonal trends off central Namibia (Walvis Bay, 23ºS) indicate a broad division into seasons of maximum (April to December) and minimum upwelling. Phytoplankton are widespread across the shelf in winter months but maximum concentrations are usually found between 10 and 20 n. miles offshore (Louw and Barlow, unpubl. data). In summer, phytoplankton appear to be concentrated inshore as upwelling intensity declines and the surface waters stratify. Zooplankton off Walvis Bay (Figure 6-8b) increase slightly in the second half of the year but, unlike the southern Benguela (e.g. Richardson et al. 2003a), there is no clear signal coupling upwelling activity, phytoplankton concentration and zooplankton abundance in the northern Benguela.


105

a 12 SHBML 2000-2003 (< 200 m)

-2

BIOMASS (g C m )

10 8 6 4 2 0

CELP 1977/78

A M

J

J

A

S O

N

D

J

F M A M

J

J

A

S O

N

D

J

F M A M

J

J

A

S O

N

D

J

F M

o

b

2000

2001

Dec

Oct

Aug

Jun

Apr

Feb

Dec

Oct

Aug

Jun

Apr

Feb

Dec

Oct

Aug

Jun

Apr

2500 2000 1500 1000 500 0 -500 -1000 -1500 -2000 Feb

Nos/m³

Walvis Bay Monitoring Line - 23 S Febr. 2000 - Dec. 2002

2002

Figure 6-8. A comparison of the seasonal variability of zooplankton in the northern and southern Benguela. (a) Mesozooplankton (primarily copepods) biomass (gC m-2) on the continental shelf during monthly SHBML (St Helena Bay Monitoring Line) surveys in 2000-2003 in the St Helena Bay region, South Africa (from Koch and Hutchings, unpubl. data); data from monthly CELP (Cape Egg and Larvae Programme) surveys in 1977-1978 (calculated from Pillar 1986) are superimposed. (b) Copepod abundance (expressed as anomalies from the 2000-2002 time-series mean; No. m-3) on the monthly Walvis Bay Monitoring Line surveys at 23˚S, Namibia (from Cloete, unpubl. data).

In addition to the presence of an active population of C. carinatus throughout the year in the upper water column on the Namibian shelf (Unterüberbacher 1964; Timonin et

106

L. Hutchings et al.

al. 1992) and off southern Angola, there is a resting component of diapausal C5s in deep (300-1000 m) water offshore (Timonin et al. 1992; Arashkevich et al. 1996; Timonin 1997; Loick et al. 2005; Verheye et al. 2005). Whereas abundance of the neritic population varies considerably depending on the upwelling phase, there is very little seasonal variability in abundance of the deep-living animals (Timonin 1990, cited in Arashkevich et al. 1996), suggesting a permanent pool of diapausing animals throughout the year. The triggers that terminate diapause remain poorly understood (Verheye et al. 2005). Interestingly, despite the reversal between the southern and northern Benguela of seasonality in phytoplankton abundance (Figure 6-1c), pelagic fish in the northern Benguela have the same spawning seasons as in the southern Benguela; sardines are widespread spawners over the period September to March, while anchovy spawn in mid- to late-summer in the Angola-Benguela frontal region. Hake spawn in the September-October period from just north of the Lüderitz upwelling cell to northern Namibia. The central Namibian shelf appears to be an important nursery ground for juveniles, where low oxygen concentrations, warm water intrusions and strong upwelling are thought to exert major influences on recruitment success. Strong intrusions of very warm Angola Current water occur about every ten years with marked effects on the distribution and recruitment of Namibian fish stocks. A large intrusion of very low-oxygen water in 1994 appeared to have resulted in very poor hake recruitment and marked changes in the entire northern Benguela ecosystem (e.g. Boyer et al. 2001). INTERANNUAL AND DECADAL CHANGES In contrast to plankton variability, which has for long been regarded as trivial and not directly relevant to fisheries, decade-scale variability is a primary feature of fish stocks. For instance, several of the world’s productive upwelling regions have experienced extensive fluctuations in pelagic fish yields and regime shifts of fish populations, which are echoed in the sediment record of fish scale deposits over periods of 50-60 years (Lluch-Belda et al. 1989; 1992; Schwartzlose et al. 1999). Recent research (see Colijn et al. 1998) has, however, shown that long-term variability in plankton is closely linked to climate change and that foodweb changes are also manifested in long-term variations in the abundance, distribution and species composition of the plankton (see Perry et al. 2004 and references therein). There are three mechanisms that control trophic levels in marine ecosystems (Cury and Shannon 2004). The first mechanism is bottom-up control, the conventional trophic flow control that seems to dominate most ecosystems, with the environment as the controlling agent. A marked change in the environment will alter the primary productivity of the ecosystem and its availability to higher trophic levels. The second mechanism is top-down control, by which lower foodweb components are regulated by higher-level predators. Thus, the abundance of predators will determine that of lower trophic levels, leading to alternating up- and down patterns of abundance, or trophic cascades. The third mechanism is wasp-waist control, where environmental changes


-3

CONCENTRATION (mg m )

b

3 2

Pseudo wind

1

-3

Mean surface

20 15 10 5 0 1950

1960

1970

1980

1990

2000

(c ) To ta l c o p e p o d s

7.0

LOG (ABUNDANCE)

1990

-2

25

(b)

c

1970

0 -1

1950

ANOMALY (100 m2 s-2)

a

107

6.0

5.0

4.0

3.0

2.0 1950

1960

1970

1980

1 990

2000

5.0

(d ) L a rg e c a la n o id s

d LOG (ABUNDANCE)

4.5

4.0

3.5

3.0

2.5

2.0 1950

1955

1960

1965

1970

1975

1980

1985

1990

1995

2000

Figure 6-9a to d. Bottom-up control of zooplankton in the southern Benguela: long-term time-series of: (a) an upwelling index, as annual averaged anomalies of equatorward pseudo wind stress on the west coast of South Africa, representing the southerly, upwelling-favourable component, 1951 – 1989 (redrawn from Shannon et al. 1992); (b) phytoplankton biomass, as annual (March – June) mean surface chlorophyll a concentration measured in St Helena Bay, 1974 – 1996 (from Verheye 2000); Mesozooplankton abundance, as annual mean abundance of total copepods (c) and large calanoid copepods (d) within ca 50 n. miles from the coast in the St Helena Bay region during autumn (March – June) of 1951-2004 (b, c and d updated from Verheye and Richardson 1998, Verheye et al. 1998, Verheye 2000).

108

L. Hutchings et al.

may initiate ecosystem changes via direct effects on dominant middle trophic level pelagic fish species, which in turn affect other pelagic fish species as well as higher and lower trophic levels. Southern Benguela Over the past 4-5 decades, declining trends in zooplankton have been observed in upwelling regions of most eastern boundary current systems (Cury et al. 1999). In contrast, in St Helena Bay on the west coast of South Africa, where pelagic fish recruit most intensely during the austral autumn/winter, zooplankton abundance has increased 100-fold since the 1950s (Verheye and Richardson 1998; Verheye et al. 1998). At the same time, there was a long-term shift in zooplankton community size structure, from large to small, coincident with a decade-scale change in dominance from sardine to anchovy, two planktivorous species with different prey size selectivities. Parallel long-term trends across consecutive lower trophic levels in the southern Benguela have been documented (Verheye 2000), which suggest upward-propagating effects of oceanographic and biological processes in response to increased wind stress (Figure 6-9a). This leads to intensified upwelling, nutrient enrichment and enhanced phytoplankton (Figure 6-9b) and zooplankton production (Figure 6-9c). Conversely, alternating up- and downward long-term trends observed across consecutive higher trophic levels (Verheye 2000) reveal trophic cascading effects of predator-prey interactions, from piscivorous apex predators (including fishing activities) to planktivorous small pelagic forage species to zooplankton. Beginning in the mid-1990s, there has been a reversal in the long-term trend in zooplankton of St Helena Bay (Figure 6-9c). Declining abundances are particularly evident in the large calanoid copepods (2-5 mm TL) (Figure 6-9d), which are selectively preyed upon by anchovy recruits (James, 1988; James and Findlay 1989). The biomass of these fish has increased substantially since the mid 1990s (see Figure 6-10a), with their predatory effect reducing prey abundances to below the time-series minimum of the 1950s (Figure 6-9d). That top-down mechanisms may indeed play an appreciable role in controlling coastal zooplankton populations is supported by negative relationships found between copepod abundance and pelagic fish stocks, both on the west coast recruit grounds and the south coast spawning grounds. Such predator-prey relationships are particularly evident since 1988, when hydro-acoustics were introduced for more accurate fish stock assessments and when zooplankton were concomitantly monitored along the coast (Figure 6-10). The best relationships were found between the abundance of large calanoid copepods in St. Helena Bay and the biomasses of total pelagic fish (r2 = 0.46) and anchovy recruits (r2 = 0.49; Figure 6-10b) on the west coast during autumn (March-June) of 1988 - 2004.


a

109

Recruit surveys: Winter - West Coast

BIOMASS ('000 tons)

4000

RED EYE 3000

SARDINE

ANCHOVY

2000 1000

2500

4.5

2000

4.0

1500

3.5

1000

3.0

500

2.5

0

2004

ANCHOVY BIOMASS ('000 tons)

2002

2000

1998

1996

1994

1992

5.0

1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004

LOG ABUNDANCE LARGE CALANOIDS

b

1990

1988

0

c ANCHOVY BIOM AS S ('000 tons)

2500

y = -2609.6Ln(x) + 4167.7

2000

2

R = 0.7047 (excluding 2000 data point)

1500 1000 500 0 2.5

3.0

3.5

4.0

4.5

5.0

LOG ABUNDANCE LARGE CALANOIDS

Figure 6-10 a, b and c. Top-down control of zooplankton in the southern Benguela: (a) time-series of recruit biomass estimates of pelagic fish during annual hydro-acoustic stock assessment surveys on the west coast of South Africa, May – June of 1988 – 2004 (from Coetzee, unpubl. data); (b, c) relationship between large calanoid copepod abundance in St Helena Bay and anchovy recruit biomass on the West Coast, suggesting predator-prey interactions (from Verheye, unpubl. data);

110

L. Hutchings et al.

Spawner surveys: Summer - South Coast

d Biomass (000 tons)

7500

Round herring Sardine Anchovy

5000

2500

e

A N C H O VY SPA W N ER B IO M A S S ( '0 0 0 t o n s )

3500 3000 2500 2000 1500 1000 500 0

2002

2001

2000

1999

1998

1997

1996

1995

1994

1993

1992

1991

1990

1989

1988

1987

0

y = -986.69Ln(x) + 974.63 R2 = 0.6054

0.0

0.5

1.0

1.5

2.0

2.5

CALANUS BIOMASS (g dry wt m -2)

Figure 6-10 d and e. (d) time-series of spawner biomass estimates of pelagic fish during annual hydroacoustic stock assessment surveys on the south coast, November of 1988 – 2002 (from Coetzee, unpubl. data); (e) predator-prey relationship between anchovy spawner and Calanus agulhensis biomasses on the south coast for the period 1988 to 2000 (after Huggett 2003).

Exclusion of the data for 2000, a year when zooplankton levels remained comparatively high despite exceptionally high fish recruitment (the highest on record), improved the relationships substantially (r2 = 0.61 and 0.70 respectively; Figure 610c). Likewise, based on data from the south coast during summer (1988-2000), Huggett (2003) described significant predator-prey relationships between the biomasses of the dominant copepod Calanus agulhensis and adult anchovy on the western (r2 = 0.61) and central Agulhas Bank (r2 = 0.46) (Figure 6-10d, e), where these fish spawn each year and rely on this large calanoid copepod as a primary energy source (Richardson et al 1997, 1998). This copepod’s biomass was also negatively related to total pelagic fish biomass (r2 = 0.62 and 0.34 respectively). Interestingly, similar to its recruit biomass on the west coast, the spawner biomass of anchovy on the western Bank was extremely high during 2000. However, unlike recruit biomass, this peak spawner biomass did coincide with the lowest copepod prey biomass on record over the period 1988 – 2000.


111

Northern Benguela The past decade has seen a series of warm events, but uncertainties in the early life history of the majority of Namibian fish stocks have not allowed more than speculation as to mechanisms causing changes in productivity of these stocks. Sardine stocks have been depleted since 1972 and an apparent recovery in the early 1990s was reversed abruptly when a combination of low-oxygen and warm waters occurred over a large area in central and northern Namibia, possibly constraining the spawning habitat of pelagic fish and impacting on the development and survival of their early life-history stages (e.g. Ekau and Verheye 2005). Sardine stocks are still depleted, despite stringent curtailment of fishing activities, in contrast to the southern Benguela stocks, which are at record highs. Shallow-water hake (Merluccius capensis) appeared to have experienced low recruitment through the late 1990s and early 2000s, while deepwater hake (M. paradoxus) have expanded their distribution range northwards in the Lüderitz region (van der Lingen et al. 2006). An extensive zooplankton sample repository (referred to as the SWAPELS collection – South West African Pelagic Egg and Larval Surveys) exists for the region between the Cunene River (17ºS) and Lüderitz (26ºS), collected monthly during the 1970s and 1980s. However, to date no comprehensive analysis of this enormous archive has been undertaken. It has, therefore, not been possible to quantify (suspected) long-term changes in the northern Benguela Current region. Recently however, Hansen et al. (2005) suggested long-term changes in zooplankton community structure since the early 1960s, based on sporadic published accounts of copepod species composition or species dominance in the region (Table 1). A pilot programme of retrospective analyses of some SWAPELS samples from late 1970s and early 1980s was recently initiated through the BENEFIT (Benguela Environment and Fisheries Interactions and Training) Programme (e.g. Tsotsobe et al. 2003; 2004; Mainoane, unpubl. data), and is currently being fast-tracked under the aegis of the BCLME Programme. It allows a preliminary reconstruction of decadescale variability of zooplankton since 1959 for the region off Walvis Bay, historically one of the main fish spawning areas. Although the data are only crude estimates of zooplankton biomass (expressed as settled volume), they do however suggest an overall increasing, long-term trend in coastal zooplankton abundance, as observed in the southern Benguela. After an initial decline over the first two decades, from roughly 200 ml m-2 in the late 1950s (calculated from Kollmer 1963) to 32°S summer. Weak in N, Moderate-strong to S. Very low

Large variation in pelagic stock abundance and long period of collapse of the sardine stock Importance of salmonid stocks (northern part, coastal zone)

Dominance of sardine (north) or sardinella (south) compared to anchovy. Frequent burst of secondary species. Partial replacement of longer lived bottom fish species, by short lived small pelagics and cephalopods, in the commercial fisheries. Low abundance of top predators (birds, large sharks and mammals)

Most extended system. Most extended, most superficial and most depleted MOL. Dominance of anchovy versus sardine. Frequent burst of secondary species. Strong effect of ENSO events on all living organisms

Ecosystem features

f = -0.4 to -0.9 x 10

Physical features

Freshwater input

15-37°S -4

Mainly narrow off west coast (30°S summer. Strong in S, moderate to N Generally low, with occasional incursions of Congo R. water, and seasonal incursions of Orange, Cunene and Angolan river waters. Alternation between anchovy dominated and sardine dominated regimes Spawning area of many species located upstream of their nursery area. Observed mortality of fish and shellfish due to H2S and/or low oxygen concentration and/or sulfide emissions.

The abundance of some top predators (pinnipeds, birds) is presently low in the CanCE compared to the other three and this ecosystem is also characterised by the dominance of sardine and sardinella, whereas anchovies are also important in the other

Modelling forecasting and scenarios in upwelling LMEs

211

ecosystems. The CanCE and the HCE display more frequent outbursts of secondary or rare species than the two other systems. Because many rivers flow into the northern CalCE, several species of salmon occur in coastal waters of this system. Finally, the biological variability in the two Pacific systems is largely dominated by El Niño/La Niña events, whereas in both Atlantic regions El Niño has weaker influence complemented by NAO in the north. In all regions, coastal trapped waves play a significant if not fully understood role. The way in which climate changes affect ecosystems, on both land and ocean, is complex and difficult to forecast, although efforts in meta-analysis, process studies and modeling with data assimilation are promising (Drinkwater and Myers 1987; DeAngelis and Cushman 1990; Bakun and Broad 2002). Uncertainty persists in the physical effects of climate change. Model based scenarios of the impact of global warming on the physical components of the earth system still remain coarse and uncertain. The poor resolution of these simulations introduces major uncertainties when trying to resolve regional scales. Further work is needed before global climate change scenarios will be directly applicable to the regional responses of the upwelling and other ecosystems. Furthermore, under any given scenario, additional uncertainty surrounds the biological responses to physical forcing. The effects of anthropogenic forcing, particularly increasing fishing pressure but also habitat invasions, eutrophication and diseases, complicate the issue (Verity et al. 2002). Increased fishing effort, and consequent increased fish mortality, will likely increase the relative abundance of low trophic levels in these ecosystems (Pauly et al. 1998) and might favour or exacerbate regime shifts and possibly population outbursts as observed in the CanCE. The most promising example of environmental forecasting from a model including data assimilation is provided by Chen et al. (2004) on El Niño/La Niña predictions. There is a clear need to develop similar environmental models coupling atmospheric and oceanographic processes in the other upwelling systems, with data assimilation to make them more realistic. One can expect that this kind of physical prediction will help in forecasting biological responses of the ecosystem based on our understanding of processes. In the mean time, biological or fishery related predictions can be made simply by using autoregressive properties of time series (e.g. Stergiou et al. 1997) but the resulting forecast is likely to fail whenever a major change (regime shift, modification of the exploitation pattern) occurs (Ulltang 1996). Simple or multiple regression models suffer from the same limitation, aggravated by the poor predictive power of these models due to limited degrees of freedom, the difficulty of selecting the right explanatory variables and the uncertainty in the functional form of variable relationships (Fréon et al. 2005). Furthermore, for purely mathematical reasons, when r2

Benguela: Predicting a Large Marine Ecosystem, Volume 14 (Large Marine Ecosystems)

The Gulf of Guinea Large Marine Ecosystem (Large Marine Ecosystems)

Large Marine Ecosystems of the North Atlantic (Large Marine Ecosystems)

Biological Invasions in Marine Ecosystems

Marine Ecosystems and Global Change

Advances in Marine Biology, Volume 14

Marine Optics (Elsevier Oceanography Series, Volume 14)

Marine Optics (Elsevier Oceanography Series, Volume 14)

Advances in Marine Biology, Volume 14

Marine

Marine

Marine

Large Herbivore Ecology, Ecosystem Dynamics and Conservation

Responsible fisheries in the marine ecosystem

Protecting the Gulf's Marine Ecosystems from Pollution

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Large deviations

A Marine for Christmas

Advances In Marine Biology, Volume 46 (Advances in Marine Biology)

Advances in Marine Biology, Volume 44 (Advances in Marine Biology)

Advances In Marine Biology, Volume 51 (Advances in Marine Biology)

Benguela: Predicting a Large Marine Ecosystem, Volume 14 (Large Marine Ecosystems)

The Gulf of Guinea Large Marine Ecosystem (Large Marine Ecosystems)

Large Marine Ecosystems of the North Atlantic (Large Marine Ecosystems)

Biological Invasions in Marine Ecosystems

Marine Ecosystems and Global Change

Advances in Marine Biology, Volume 14

Marine Optics (Elsevier Oceanography Series, Volume 14)

Marine Optics (Elsevier Oceanography Series, Volume 14)

Advances in Marine Biology, Volume 14

Marine

Marine

Marine

Large Herbivore Ecology, Ecosystem Dynamics and Conservation

Responsible fisheries in the marine ecosystem

Protecting the Gulf's Marine Ecosystems from Pollution

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Seeing a Large Cat

Large deviations

A Marine for Christmas

Advances In Marine Biology, Volume 46 (Advances in Marine Biology)

Advances in Marine Biology, Volume 44 (Advances in Marine Biology)

Advances In Marine Biology, Volume 51 (Advances in Marine Biology)

Recommend Documents